Final Report Summary - MEMSCON (Radio Frequency Identification Tags Linked to on Board Micro-Electro-Mechanical Systems in a Wireless, Remote and Intelligent Monitoring and Assessment System for Maintenance of CONstructed Facilities)
Executive Summary:
During the project activities and in the first year of the project, the activities started with the hardware development including the definition and reporting of the end-user requirements and system specifications that lead to the actual hardware development of the base-station, strain and acceleration sensor and wireless communication units. At the same time developments of the Decision Support System (DSS) started including all its different modules. After the development of the sensorial, wireless and testing activities, the system integration took place, combining thus the hardware with the software developments. Towards the end of the final running year of the project, the laboratory evaluations of the sensors and systems started. This concluded and gave its way to the execution of the actual field validations that took place during the last year of the project.
During the project validation phase, full-scale tests were performed over the whole system including the validation of the sensorial systems (strain and acceleration sensors) as embedded in the actual building. Monitoring of the structure and actual measurements using the integrated system as a whole proved the validity and applicability of the developed technologies.
At the same time, training, dissemination and exploitation activities run in parallel with the RTD activities to ensure project results diffusion while at the same time ensure proper outcomes copyright arrangements.
Project Context and Objectives:
The MEMSCON objectives can be summarized in the bullets that follow:
a) To integrate MEMS-based sensors and an RFID tag in a single package of small size that will be attached to reinforced concrete (r.c.) buildings for life cycle measurements of acceleration in 3 dimensions or strain in 1 dimension that will be transmitted to a remote base station using a wireless interface;
Each sensor node will be individually supplied by a battery. In order to keep the required maintenance level low enough, the battery must have a lifetime of at least a couple of years. This requires not only efficient batteries, but demands also power optimised sensors and RFID tags.
b) To develop a Decision-Support-System (DSS) for proactive rehabilitation and rehabilitation after earthquake damage in r.c. buildings. This DSS will accept input from the sensors in order to assess the structural condition of the monitored building and to select optimal remedial measures and
c) To evaluate an integrated package of the system both in experimental and field conditions.
The general architecture of the DSS under development can be seen in the figure below. It includes an Expert System that is connected to the Knowledge Base and the Sensor, History and External Data databases and Modules 1, 2 and 3.
On completion of the project there will be the following main deliverables which deserve special attention for exploitation:
1. MEMS-type micro sensors and a RFID tag in a single package of small size that can be embedded or attached to constructed facilities. The sensors will measure acceleration in three axes or strain in one axis. The acceleration sensors will be activated only in case of seismic events and the strain sensors on a periodic basis, thus achieving low power consumption. The bandwidth at which the acceleration sensors will be interrogated will be better than 25 Hz which is more than adequate for seismic monitoring. The system will have wireless communication capability.
2. Software that can assess the current condition of a reinforced concrete building locally and globally based on strain measurements.
3. Software that can assess the local and global structural condition of a reinforced concrete building based on the history of accelerations of the building during an earthquake.
4. Software that can determine the best remedial measures in the case of proactive rehabilitation.
5. Software that can assess the best remedial measures in the case of seismic upgrading.
6. An integrated package that contains the sensing and data acquisition system all of the aforementioned software.
Project Results:
The MEMSCON main S/T results have been included in the paragraphs that follow:
4.1.3.1 MEMSCON Accelerometers
IMEC-NL has developed a versatile low-power solution to be used as readout for accelerometers and strain sensors. Developed on standard CMOS 0.25µm technology the ASIC readout can process signals from MEMS-based comb-finger accelerometers and strain sensors. Its uniqueness lies in the fact that the same ASIC can be used for both types of sensors, therefore making the system simple, cost-effective and versatile. The 3-channel system can be used for a 3-axis accelerometer system or 2-axis accelerometer in combination with 1-channel strain sensor or any combination of them. DC signals, as those from a strain sensor, are easily sensed and amplified. The dynamic input range of the ASIC readout has been tuned for optimum usage of both the accelerometers and strain sensors used in the project. The user can connect either of the two sensors to the readout and, a maximum output range of the two sensed signals, namely ±2 g or ±30,000 micro-strains appear at the output. No extra trimming or tuning is required. The whole system including biasing stages, excitation of the MEMS sensors and a strong output buffer to send data over meter long cables, draws only 120 micro amps (on a 3.0V supply), thus making it comparable to the most advanced commercial devices available in the market. Its multiplexed analogue output allows the end user to readout 3 signals at 200Hz each, in a sequence with a minimum of overhead. The linearity of the ASIC readout plus sensors gives a deviation of less than 1%.
A capacitive sensing principle with mechanical elements realized in a surface micromachining technology has been selected for the MEMSCON accelerometer. The sensor is using independent mechanical elements for each axis. For the in plane sensing, inter-digitated comb structures and for the out of plane sensing a pendulum have been realized. The fabrication technology for the three elements is identical; therefore the three elements can be placed on one die. We have also chosen a hybrid integration of the MEMS and its ASIC within a hermetically sealed ceramic package. Under normal environmental pressure conditions, the MEMS would run into an over-damped regime. In that case the functionality of the accelerometer could not be guaranteed especially regarding the linearity in the bandwidth. Therefore, some level of vacuum is needed within the micro-cavity. The wafer level packaging (WLP) technique consists in bonding a cap wafer onto the MEMS wafer once the moveable structures are released. This technique offers several advantages compared to traditional packaging techniques. First, it allows the protection of the fragile MEMS structures from the early stages of the manufacturing process and eases the problematic step of the singulation of the MEMS dies. Furthermore, since the bonding step can easily be performed under controlled pressure, the vacuum inside the MEMS accelerometer is set at the desired value. And finally, WLP takes full advantage of the parallel manufacturing to drastically reduce the cost associated to the packaging. In addition, WLP simplifies the implementation of the out-of-plane acceleration sensor since the counter electrode can be integrated into this cap layer. For the in plane accelerations the sensor is using inter-digitated comb structures, which form a differential capacitor. The fixed plates of the differential capacitor are formed by fingers attached to the substrate, whereas the movable part is a mass with fingers attached. Two of these structures rotated by 90 degrees are placed on the die. For the measurement of out of plane accelerations (z axis) a pendulum with asymmetric mass distribution is used, which is forming the movable part of a differential capacitor. The fixed part is realized by counter electrode on top of the mechanical element.
Although the basic structure of the x- and y-sensors and the z-sensor is different, a suitable design allows the achievement of similar sensor parameters. An SOI wafer is used for the MEMS sensors. The mechanical structures are defined by a combination of DRIE etching and sacrificial layer etching. The cap is formed by two silicon wafers bonded together. The top wafer of the cap stack is etched with KOH to allow the electrical contact to the cap bottom. For the bottom part of the cap, deep trenches are etched all around the different areas to ensure the electrical isolation. For the top part, the isolation is guaranteed by a thick surface oxidation (1-2?m) of the silicon wafer. The cap itself is mounted to the MEMS wafer by using a metal-silicon eutectic bonding process, used also to realize the electrical connections between the cap and the MEMS wafer.
The MEMS sensors are integrated into wireless sensor modules to form a monitoring system which communicates the measurement data to a remote base station and is controlled by this station. The accelerometers are integrated using standard printed circuit board technology and assembled into a rigid protective housing that allows fixation to the building. The strain sensor modules consist of two parts. The wireless communication and processing part is implemented in a similar rigid housing and remains accessible for potential battery replacement. The front-end sensing module combining the MEMS strain sensor, read-out ASIC and supporting oscillator circuitry and passives, is embedded inside the concrete structure, applied to the metal reinforcing bars. It uses a special package using PDMS silicone molding and a polyimide carrier in order to allow the sensor to be glued onto the metal bar similarly to a traditional strain gage while efficiently transferring the strain from the metal bar to the MEMS sensor.
For the wireless communication an IEEE 802.15.4 radio module in the 900MHz band is used in combination with a custom-designed patch antenna, in order to obtain robust links in real-world conditions inside a building. The wireless sensor modules are powered by an 8.5Ah C-cell long operating life primary lithium battery. The modules also include a low power processor and a 64Kx16 bit RAM memory to record earthquake events or strain data. A low power network architecture was implemented on top of an 802.15.4 MAC using indirect data transfers which allows the sensor modules radios to remain powered down most of the time except during brief polling moments and during periodic or even-triggered transmission of recorded data. The strain sensor modules measure periodically and use a radio polling interval of 60 seconds. The 3D accelerometer is constantly running at 3 x 200Hz sample rate with the measurements recorded in a 54-second loop buffer. This is required to be able to record the early onset of an earthquake event, even before it reaches a trigger threshold, and requires an ultra-low power sensor and readout as we have developed in the MEMSCON project. It also enables the triggering to be done remotely by the base station based on the combined information from several monitoring nodes across the network. In that way data can be obtained from all accelerometers in the network during an earthquake or other global event, even if locally at some sensor nodes the chosen earthquake threshold was not exceeded. It also allows the wake-up monitoring function to be enabled only on a select amount of sensor modules which do not give many false alarms, e.g. due to street traffic or other local interfering signals. To respond timely without data loss to an event triggered from the base station, the radio polling interval of the accelerometer modules is limited to 15 seconds. The strain sensor modules show an average power consumption of 0.274mW which results in a operational time of 12 years before the batteries need to be replaced. The accelerometer modules consume 1.73mW resulting in 2 years of battery life.
4.1.1.1 MEMSCON Strain Sensors
The MEMS Strain Sensor chip was designed using a capacitive sensing principle. Hence a strain in the rebars of the building under study has to be transferred to a change of capacitance in the sensing part of the sensor. In our case, this is done by a relative displacement of the fingers of a capacitive comb drive structure. The design of the sensor is based on the inputs of the end users, and is designed to measure strain over a +/- 30000µ? range with a resolution of 10µ?. It has a size of 3600x 8400x1050µm. The MEMS sensor is fabricated using a SOI (Silicon on Insulator) wafer as substrate. SOI wafer consist of a Device layer, a BOX layer and a Handle. Both the Handle and the Device layer are made out of doped Silicon (for the MEMSCON Strain Sensor) and the BOX layer out of Silicon Oxide (glass). The purpose of using such a type of wafer is that it simplifies the production of the moving capacitor (in the Device layer). With such wafers, the desired structure can be patterned in the device layer and released by selectively etching the oxide layer under it.
The layout of the structure is such that 2 differential capacitors are created and each of those capacitors consists of 4 comb drives in parallel. Each comb drives consist of 50 fingers. Two anchors are located on the back side of the MEMS Sensor. They define the position where the strain will be transfer from the rebar to the sensor. The distance between the 2 anchor points define the relation between strain in the structure and displacement in the comb drives. In our case, the distance between the anchors (center to center) is 3mm, which means that the strain of 30000µ? in the rebar produces a displacement of 90µm in the comb drives. A challenge was in designing the large stiff spring that holds the moving parts together. An optimum needed to be found between having enough stiffness so that the offset due to packaging and assembly stays within the allowed limit (20µm), not to stiff to limit creep in the glue during operation and finally the stress at full elongation/compression shall not exceed the yields strength of the silicon. The theoretical yield strength of monocrystalline silicon is around 7GPa, however, due to defect in the crystal the actual yield strength is closer to 3.5GPa. For the design of the spring, we want to stay in a region of 700MPa-1GPa (5-3.5 time lower that the actual yield strength) in order to have a sufficient security margin. The design was modeled with a 3D CAD drawing software (IronCAD) and simulations were performed with a Finite Element Model Analysis software (ALGOR) to check that the design doesnt exceed the maximum stress and have enough stiffness.
4.1.1.2 MEMSCON Integrated System
The MEMS sensors are integrated into wireless sensor modules to form a monitoring system which communicates the measurement data to a remote base station and is controlled by this station. The accelerometers are integrated using standard printed circuit board technology and assembled into a rigid protective housing that allows fixation to the building. The strain sensor modules consist of two parts. The wireless communication and processing part is implemented in a similar rigid housing and remains accessible for potential battery replacement. The front-end sensing module combining the MEMS strain sensor, read-out ASIC and supporting oscillator circuitry and passives, is embedded inside the concrete structure, applied to the metal reinforcing bars. It uses a special package using PDMS silicone molding and a polyimide carrier in order to allow the sensor to be glued onto the metal bar similarly to a traditional strain gage while efficiently transferring the strain from the metal bar to the MEMS sensor.
For the wireless communication an IEEE 802.15.4 radio module in the 900MHz band is used in combination with a custom-designed patch antenna, in order to obtain robust links in real-world conditions inside a building. The wireless sensor modules are powered by an 8.5Ah C-cell long operating life primary lithium battery. The modules also include a low power processor and a 64Kx16 bit RAM memory to record earthquake events or strain data. A low power network architecture was implemented on top of an 802.15.4 MAC using indirect data transfers which allows the sensor modules radios to remain powered down most of the time except during brief polling moments and during periodic or even-triggered transmission of recorded data. The strain sensor modules measure periodically and use a radio polling interval of 60 seconds. The 3D accelerometer is constantly running at 3 x 200Hz sample rate with the measurements recorded in a 54-second loop buffer. This is required to be able to record the early onset of an earthquake event, even before it reaches a trigger threshold, and requires an ultra-low power sensor and readout as we have developed in the MEMSCON project. It also enables the triggering to be done remotely by the base station based on the combined information from several monitoring nodes across the network. In that way data can be obtained from all accelerometers in the network during an earthquake or other global event, even if locally at some sensor nodes the chosen earthquake threshold was not exceeded. It also allows the wake-up monitoring function to be enabled only on a select amount of sensor modules which do not give many false alarms, e.g. due to street traffic or other local interfering signals. To respond timely without data loss to an event triggered from the base station, the radio polling interval of the accelerometer modules is limited to 15 seconds. The strain sensor modules show an average power consumption of 0.274mW which results in a operational time of 12 years before the batteries need to be replaced. The accelerometer modules consume 1.73mW resulting in 2 years of battery life.
4.1.1.3 Software for the Assessment of the Damage Due to Differential Settlement between Foundations
To calculate the resulting internal forces at critical cross-sections of the structural members, measurements of strains are needed at four points in each cross-section. Then, from the constitutive laws of the structural materials and on the assumption of linear distribution of strains over the area of the cross-section, one can estimate the stresses and their resultants, the internal forces. For the above we need on the average 6 strain sensors per structural element. The implication is that for an average 5 storey building with 25 columns and 40 beams per storey one needs approximately 2,000 strain sensors. In this work the required number of strain sensors has been dramatically reduced. Thus, strain sensors are only needed at the bottom cross-section of the columns at the ground floor level. The implication of this is that for an average 5 storey building only about 60 (as opposed to 2000) sensors are needed. Then, under operating conditions, the internal forces in each structural member as well as their structural adequacy and the differential settlement between foundations are being assessed through a commercially available finite element program that accepts as input the measured values of the above critical parameters. In more detail the above are executed in the following steps: From the strain measurements the axial forces and the bending moments in two principal directions at the bottom cross sections of the columns at the ground floor are calculated. The variations of these forces with respect to those developed at the initial state after the finishing of the structure are calculated. The differential displacements between the foundations of the columns are calculated from the compatibility condition with the above variations in internal forces. The total structure, represented by a finite element space model, is analysed for the actual applied loads and the estimated differential settlements, by using a non-linear structural analysis computer program. The output is the values of internal forces and displacements for the totality of the structural members. The values of the safety factors of the critical cross sections for the estimated values of internal forces with respect to their available ultimate strength are calculated and compared to their permissible values, in order to estimate their structural adequacy.
4.1.1.4 Software for the Assessment of Seismic Damage
The recording system selected in this work involved the installation of a small number of three-dimensional accelerometers, two per storey of the building, in combination with a commercially available finite element program for structural analyses. The assessment of local stability conditions of the structure involves the detection of the cross-sections where plastic hinges are formed and the estimation of the corresponding damage degree. From the successive imposed displacements and the corresponding structural analyses the bending moment-curvature diagrams for the cross-sections are developed. These diagrams representing the development of the pair of values for bending moment-curvature during the successive cycles of the seismic oscillation, constitute the hysteresis loop pattern. The various areas inside the loops represent the different modes of energy (dissipated, restoring, released, elastic and complimentary). These, in the form of a modified Parc and Ang Damage Criterion, are used to determine the damage degree for the cross-sections in terms of energy quantities. Global instability (total failure) of the structure ensues when both end cross-sections of the totality of the columns in a storey develop plastic hinges and a storey kinematic mechanism is formed. From the relative translations of the two accelerometers on the floor one can estimate the coordinates of the instantaneous pole of rotation. The safety factor against global instability defines the degree of storey mechanism formation for each storey in the building.
4.1.1.5 Software for the Selection of remedial Measures and the Estimation of their Costs
This software accepts input from the software on damage due to differential settlements on the magnitude of differential settlements as a function of time. Thus, it can be established, timely, whether foundation movement is stabilising or is progressive and threatening to the building reducing dramatically the expense of remedial measures. Input from the above software additionally includes the safety factors for all structural members. Based on these methods for strengthening or repairs are selected and their cost estimated. In the case of an earthquake, this software accepts input from the software on seismic damage on the structural elements that have been damaged and their degree of damage. Based on the latter degree it presents common repair methods to bring the element to its pre-earthquake state. The user selects one of the suggested repair methods for each damaged element and then the model estimates the cost to repair each element and sum it up to get the total direct structural repair cost.
4.1.1.6 The Integrated DSS
All of the above software was integrated in a DSS for proactive rehabilitation and rehabilitation following seismic damage. This DSS also includes an expert system with a friendly user interface and knowledge and data DBs. A graphical user interface provides the graphical environment with which the end-user can retrieve current and historical data from the DBs while also provides real-time alerts and warnings in case of unsafe situations and allows the end-user to examine different scenarios for hypothetical situations.
4.1.1.7 Laboratory Validation
MEMSCON sensors were tested in laboratory conditions on both reduced scale specimens, to assess their performance, and full-scale specimens, to assess the reliability of the whole system. Performance of the MEMSCON accelerometers were assessed mounting three wireless nodes on a shaking table, driven by a function generator connected to an amplifier, back to back with high sensitivity wired piezoelectric accelerometers (model 393B12 and 393C produced by PCB Piezotronics), producing several vibration tests with excitation of various shape, frequency and amplitude. In particular the tests focused on frequencies (0-20 Hz) and amplitudes (100-500 mg) of the shaking on the range of relevance for seismic monitoring. Several tests conducted producing harmonic excitation to the shaking table were conducted, in order to estimate calibration coefficients of each device, directly comparing MEMSCON and reference measurements, along each axis. Moreover, the wireless sensors were mounted also on a two story metal frame, again back to back with reference accelerometers, reproducing to the shaking table a spectrum compatible ground acceleration time history, to simulate operative conditions during an earthquake. Tests highlighted that the sensitivity of the MEMSCON accelerometers differs among the 3 axes, particularly between axes X and Y (finger comb technology) with a mean sensitivity of 200 mV/g and Z axis (asymmetric pendulum technology) with a mean sensitivity of 100 mV/g. Considering the maximum voltage range (0-2.5V) and the 16 bit resolution, the nominal resolution of the devices in terms of acceleration is estimated as 0.1-0.4 mg, therefore better than system requirements. The accuracy of the devices is directly correlated to the background noise contained in the signals, estimated as about 20 mg. The clock seems to be stable, with a sampling rate of 202 Hz.
MEMSCON accelerometers were also validated performing a experimental campaign on a full scale three dimensional concrete frame, with dimensions 3.15x3.15x3.90 meters realized inside laboratory. The frame, realized selecting materials widely used in construction as B450C steel reinforcing bars and concrete C25/30 class, reproduced in total a portion of the ground floor of a prototype building intended for public use and designed in order to develop plastic hinges at the interface between foundation and column during the earthquake and to keep into the elastic range all the elements except obviously the columns at the bottom end. The frame consisted of four 350x700 mm orthogonal beam post-tensioned foundations, four 300x300 mm columns 2.80m high, two 300x500 mm longitudinal beams and two 300x400 mm transversal beams at the top and a 120 mm concrete slab. The response of the first floor of the prototype building to a spectrum compatible earthquake was numerically estimated and applied dynamically to the concrete frame slab using an horizontal actuator controlled in displacements. Accelerations at the top of the concrete frame were recorded using MEMSCON accelerometers and piezoelectric wired accelerometers as reference, mounted also in this case in back-to-back configuration, fastening all the instruments directly to the concrete slab. The dynamic tests conducted at different amplitude highlighted that MEMSCON accelerometers are suitable to monitor a real concrete structure during a seismic event, with discrepancies with the wired accelerometers less than 20 mg in terms of root mean square. Acceleration measures were used also to estimate the displacement time history at the top of the frame, by double integrating the accelerations in respect of the time. This estimation was compared with the time history of displacements as produced by the horizontal actuator. The comparison shown that the error committed estimating displacement using double integration of the acceleration as recorded by the MEMSCON accelerometers is in the order of 0.5mm.
The performance of MEMSCON strain sensors were firstly assessed carrying out tensile and compressive tests performed on steel bars and reinforced concrete small scale specimens, adopting various load protocols, all consisting of load-unload cycles with increasing intensity, up to the failure of either the specimen or the sensor. The sensors performances were assessed directly comparing the devices with commercially available reference strain gauges (HBM LY41-3/700). Since one of the project target was to guarantee an high survival rate of the sensors also after a severe seismic event, a particular gluing procedure consisting in develop a chemically clean surface having appropriate roughness and a surface alkalinity suitable for gluing, was developed. Tests verified that most part of the sensors survive the installation procedure and the concrete hardening. Sensitivity of the sensors after gluing them on the reinforcing bars was estimated at 20 ??/mV while the resolution was about 0.75 ??.??Comparing MEMSCON and reference observed an accuracy of 5-15 ??/mV during test on bare bars was observed, while the precision is of the same order of magnitude of accuracy (10-20 ??/mV). Concerning the maximum strain recordable, during test on bare bars, strains up to 2% easily reached and recorded without any debonding, confirming that the developed gluing procedure is suitable to realize a system which survives after a severe earthquake.
Strain sensors were embedded also into a concrete frame, in order to monitor the strain evolution inside the columns due to gravity loads, thermal effects and shrinkage/creep of the concrete. Examining the response of the sensors is possible to detect the moment when the slab was casted, the daily thermal variation of 5-10 micro-strains due to a daily thermal excursion inside laboratory of about 0.5 °C, the creep/shrinkage effects in the form as a slow contraction of the reinforcing bars. The reliability of the system by using the reference sensors was therefore demonstrated.
Potential Impact:
4.1.2.1 MEMSCON Potential Impact
The impact of MEMSCON can be summarised to the following:
Transform the building rehabilitation sector (currently dominated by SMEs) into an advanced knowledge sector through the promotion of integrated monitoring systems.
Enhance the competitiveness of European SMEs in building design, construction, inspection and rehabilitation by increasing their productivity through new processes and high added value products: With the developed integrated package, the engineers will have the ability to precisely determine the precise structural state of the building.
Decrease the time needed for engineers to evaluate the structural condition of a building in service and decide on remedial measures.
Promote the Proactive Condition-Based Maintenance of buildings which is based on measurements aimed at early detection of degradation, thereby allowing degradation to be eliminated or controlled prior to significant physical deterioration. The result is a significant decrease in maintenance cost (since problems are less expensive to fix when they are first developing) and increase in building safety.
Promote a new concept of civil engineered smart structures conceived and designed as high performance systems, embedding novel, low cost and highly reliable sensors, capable of self-diagnosis and assessment of optimal rehabilitation.
Enable the MEMS European SMEs to enter the large building market: Provisions of the building codes in most cities in seismic zones no 3 and no 4 require the installation of accelerographs in new high buildings while legislation on building code regulation is bound to become more pervasive in the future, necessitating the installation of monitoring stations on all types of new public and private buildings as well as bridges and the other infrastructure. The ideal implementation of the code requirements would require 3-axes monitoring platforms in multiple locations within the building and the MEMSCON package is ideally suited for this.
Boost the already high value of MEMS technologies in the field (size, power consumption, performance, and overall costs) enabling a wide spread of the system, with a direct impact on safety in our everyday life. This will enable the MEMS European SMEs to enter this market by leveraging their existing know-how in high-end products, increase their worldwide leadership in such fields, and contribute actively to their employment and turnover growths.
Provide solution that integrates building design, construction, monitoring, operation and maintenance thus promoting sustainability (extend the life of buildings and reduce resource consumption at the same time that will offer increased safety) and offer full services with a high value added for clients.
Promote a breakthrough monitoring technology that will permit the timely input of objective, reliable, quantitative data from all points of interest in the building structure. This will make it possible for the SME dominated building management sector to provide better quality services (e.g. higher safety) and dramatically reduce life span inspection and maintenance costs.
Provide an essential way for an accurate and quick assessment of the building structural condition for several reasons: (a) dissemination of information to emergency response officials on building collapses within minutes after the occurrence of the earthquake resulting in many lives saved and prudent allocation of resources and (b) quick and accurate estimates of the level of damage that can be used to indicate loss of function and help officials decide whether the school, hospital, etc., should be evacuated or remain in service.
Act as a catalyst for new high-tech sensing systems (e.g. systems including chemical sensors), structural control systems and software for the structural assessment, performance and management of additional structures (e.g. retaining walls, bridges, dams), additional materials (e. g. steel) and groups of structures (e.g. in an intelligent city), with the subsequent creation of growth and employment of construction SMEs in the Union.
? MEMSCON Competitive Advantage
Using proprietary state-of-the-art modeling know-how for reinforced concrete structures, innovative decision support algorithms, and breakthrough low power wireless sensor networks including accelerometers and strain sensors, the MEMSCON consortium has developed a full monitoring system which enables building structural monitoring and assessment.
The consortium has identified an opportunity to leverage this activity in order to further develop market and sell the monitoring system, specifically in earthquake prone areas.
In the above application, at the present, rather expensive monitoring systems could be used in seismic prone areas. They are rather expensive and complex to implement using wired communication between a limited numbers of sensors and are power supplied by cables. They are rather expensive and are therefore mainly used in expensive state of the art structures in civil engineering (expensive bridges and buildings).
Such systems are sold today in two different ways:
the first sale occurs when civil engineering companies design the building and integrate in their plans the network of sensors and monitoring system: this is a rather expensive activity and increases the cost of ownership of a building/bridge by a few percents
the second sales strategy is coming from a service attached to such monitoring system. The cost of the system, the people used to take decisions and monitor the building quality is then spread out over time as an additional charge (typ. Monthly) like security, video surveillance, etc.
MEMSCON monitoring systems core value proposition to address the above market segments is many folds:
- the limited number and low costs sensors used to monitor the infrastructure
- the low power consumption of these self powered sensors enabling low maintenance of the sensor network
- the easiness with which the sensor network is installed and communicates (wireless system)
- the innovative modeling of structures which enables a limited number of measurement nodes in a building to assess its integrity or the need for repair
- the decision support system which enables a low experienced person to trigger appropriate actions with a limited amount of technical expertise.
? MEMSCON European Added Value
The most important trend in construction in ALL EU countries is the shift from construction costs to life cycle costs. As part of it non-destructive evaluation technologies, like the one offered by MEMSCON, need to be incorporated into buildings throughout Europe. Further to this, the MEMSCON related research is of interest to ALL earthquake prone countries in the southern part of the Union.
The developed integrated system is of interest to ALL EU countries: Knowledge of the structural condition and safety of in-service structures on a continuous time basis is an ultimate objective for owners and maintenance authorities throughout Europe. In all EU countries available funds are insufficient to maintain the existing building infrastructure. Moreover, in all EU countries there is an increasing demand for structural engineers to deal with the issues of life extension of existing structures. To meet this challenge the profession must be able to provide defendable estimates of the safety and performance of existing structural systems.
Building retrofitting and maintenance is (an SME dominated) low margin industry and this legitimates that R&D be supported at a European level, as companies have little financial capacity to invest in technology. Moreover, this is an industry with heterogeneous practices that do require EC working methods to be confronted, adapted and harmonised. Additionally, it is an industry that is lacking in innovation. The delivered package of results provides much needed data on the demand placed on buildings under operating conditions and seismic disturbances and on structures behaving in the nonlinear realm. These data are immediately usable throughout the Union.
4.1.2.2 MEMSCON Dissemination & Exploitation Activities
The results of MEMSCON can find large scale applications in the safety assessment of relevant buildings. To this aim, although the partners also made efforts to disseminate the project results to the scientific community, the main effort has been focused on a market oriented approach, to end-users, building structural rehabilitation contractors, rehabilitation engineers, companies offering structural monitoring services, companies in RC building construction, companies in building facilities management, owners of RC buildings, the Building Maintenance Departments of local and federal government, representatives of relevant associations, consortia, organizations and societies, insurers, standardization bodies, public safety officials, relevant forums in construction.
In particular, the dissemination strategy focused on:
Inducing among technical specialists a change regarding processes, procedures, and methods used for the monitoring of structures. The feasibility of the system in detecting damage will be put in evidence, as well as the benefit for the safety level of the whole community.
Particular stress is given to the economic convenience of the proposed system, when fully integrated in the construction phases of new buildings, thus minimizing direct and indirect costs.
In the course of the project, dissemination strategy was directed to the accomplishment of three major tasks:
Task 1: Activity of promotion: communication the MEMSCON Project
Task 2: Technology diffusion: definition of service for potential users
Task 3: Communication of scientific results
In order to accomplish these tasks, the partners have been involved in the following activities:
Presentation of MEMSCON to international conferences and symposia.
Further advertising the project results through appropriate channels.
Implementation and Maintenance of the MEMSCON Web Site.
Training of the staff that will be using the proposed system in the field tests.
Organization of MEMSCON Workshops.
Following the dissemination activities a large set of dissemination material was produced within the framework of the project in order to describe the project and its results. The material is both in digital format, such as Newsletters, website, etc. and in hard copy, such as leaflets. The primary function of dissemination material is to provide information on project concept, participants, aims, results, activities, etc. Since not all results have been available from the beginning of the project, they have been continuously updated with new information about the project, realized activities, events, findings, etc. All dissemination material is designed according to a common principle following the same concept and logo specifically designed for MEMSCON. In detail the dissemination means were the following:
Project Logo
The project logo is used for the cover page of all documents of MEMSCON project, internal reports, technical reports, deliverables, project plans. It is also inserted in all dissemination material such as leaflets, website, newsletters, videos, etc. in order their relation to MEMSCON to be directly identifiable.
Project Website
Available at the address www.memscon.com/ the website contains all the necessary information on the structure of the Project, its aims and goals, important news and milestones reached throughout the whole duration of the Project as well as project dissemination and public material.
Project Newsletter and press releases
Six Newsletters were published during the project execution. The Newsletters have been distributed to a wide audience through the MEMSCON mailing list, and distributed in printed copies at conferences, workshops and events.
Project Leaflets and Posters
Leaflets and posters have been created providing a brief presentation of the projects objectives and activities, distributed during various events to all interested people
Project Workshops (1st: Bucharest and 2nd: Athens)
Two workshops were organized by the members of the consortium, to present the results of the project.
Project Presentations and Submissions
During the duration of MEMSCON project research activities, a large number of publications were issued, covering the projects concept, activities and results. These include: 4 papers in scientific journals, 15 papers in conference proceedings, plus the papers published by the partners at the two MEMSCON Workshops.
Detailed descriptions of the MEMSCON dissemination plans and activities can be found in the deliverable D5.1 (Plan for the dissemination and use of the results Final).
Regarding exploitation activities, the MEMSCON consortium has produced a detailed IPR protection and IPR plan documented in the MEMSCON Technology Implementation Plan (TIP). In this document, the IPR issues are fully described including the agreements on ownership rights. The industrialization efforts required for each of the MEMSCON outcome, are also described in the same document.
List of Websites:
? The project website URL is the following: www.memscon.com
? Contact details:
Project Coordinator:
Dr Angelos Amditis
Research Director
Institute of Communication and Computer Systems
a.amditis@iccs.gr
Technical Manager:
Dr Matthaios Bimpas
Institute of Communication and Computer Systems
mbibas@esd.ece.ntua.gr
Dissemination Manager:
Professor Daniele Zonta
University of Trento
daniele.zonta@ing.unitn.it
European Commission Project Officer
Dr. Ir. Dominique Planchon
Project Officer, European Commission
dominique.planchon@ec.europa.eu
During the project activities and in the first year of the project, the activities started with the hardware development including the definition and reporting of the end-user requirements and system specifications that lead to the actual hardware development of the base-station, strain and acceleration sensor and wireless communication units. At the same time developments of the Decision Support System (DSS) started including all its different modules. After the development of the sensorial, wireless and testing activities, the system integration took place, combining thus the hardware with the software developments. Towards the end of the final running year of the project, the laboratory evaluations of the sensors and systems started. This concluded and gave its way to the execution of the actual field validations that took place during the last year of the project.
During the project validation phase, full-scale tests were performed over the whole system including the validation of the sensorial systems (strain and acceleration sensors) as embedded in the actual building. Monitoring of the structure and actual measurements using the integrated system as a whole proved the validity and applicability of the developed technologies.
At the same time, training, dissemination and exploitation activities run in parallel with the RTD activities to ensure project results diffusion while at the same time ensure proper outcomes copyright arrangements.
Project Context and Objectives:
The MEMSCON objectives can be summarized in the bullets that follow:
a) To integrate MEMS-based sensors and an RFID tag in a single package of small size that will be attached to reinforced concrete (r.c.) buildings for life cycle measurements of acceleration in 3 dimensions or strain in 1 dimension that will be transmitted to a remote base station using a wireless interface;
Each sensor node will be individually supplied by a battery. In order to keep the required maintenance level low enough, the battery must have a lifetime of at least a couple of years. This requires not only efficient batteries, but demands also power optimised sensors and RFID tags.
b) To develop a Decision-Support-System (DSS) for proactive rehabilitation and rehabilitation after earthquake damage in r.c. buildings. This DSS will accept input from the sensors in order to assess the structural condition of the monitored building and to select optimal remedial measures and
c) To evaluate an integrated package of the system both in experimental and field conditions.
The general architecture of the DSS under development can be seen in the figure below. It includes an Expert System that is connected to the Knowledge Base and the Sensor, History and External Data databases and Modules 1, 2 and 3.
On completion of the project there will be the following main deliverables which deserve special attention for exploitation:
1. MEMS-type micro sensors and a RFID tag in a single package of small size that can be embedded or attached to constructed facilities. The sensors will measure acceleration in three axes or strain in one axis. The acceleration sensors will be activated only in case of seismic events and the strain sensors on a periodic basis, thus achieving low power consumption. The bandwidth at which the acceleration sensors will be interrogated will be better than 25 Hz which is more than adequate for seismic monitoring. The system will have wireless communication capability.
2. Software that can assess the current condition of a reinforced concrete building locally and globally based on strain measurements.
3. Software that can assess the local and global structural condition of a reinforced concrete building based on the history of accelerations of the building during an earthquake.
4. Software that can determine the best remedial measures in the case of proactive rehabilitation.
5. Software that can assess the best remedial measures in the case of seismic upgrading.
6. An integrated package that contains the sensing and data acquisition system all of the aforementioned software.
Project Results:
The MEMSCON main S/T results have been included in the paragraphs that follow:
4.1.3.1 MEMSCON Accelerometers
IMEC-NL has developed a versatile low-power solution to be used as readout for accelerometers and strain sensors. Developed on standard CMOS 0.25µm technology the ASIC readout can process signals from MEMS-based comb-finger accelerometers and strain sensors. Its uniqueness lies in the fact that the same ASIC can be used for both types of sensors, therefore making the system simple, cost-effective and versatile. The 3-channel system can be used for a 3-axis accelerometer system or 2-axis accelerometer in combination with 1-channel strain sensor or any combination of them. DC signals, as those from a strain sensor, are easily sensed and amplified. The dynamic input range of the ASIC readout has been tuned for optimum usage of both the accelerometers and strain sensors used in the project. The user can connect either of the two sensors to the readout and, a maximum output range of the two sensed signals, namely ±2 g or ±30,000 micro-strains appear at the output. No extra trimming or tuning is required. The whole system including biasing stages, excitation of the MEMS sensors and a strong output buffer to send data over meter long cables, draws only 120 micro amps (on a 3.0V supply), thus making it comparable to the most advanced commercial devices available in the market. Its multiplexed analogue output allows the end user to readout 3 signals at 200Hz each, in a sequence with a minimum of overhead. The linearity of the ASIC readout plus sensors gives a deviation of less than 1%.
A capacitive sensing principle with mechanical elements realized in a surface micromachining technology has been selected for the MEMSCON accelerometer. The sensor is using independent mechanical elements for each axis. For the in plane sensing, inter-digitated comb structures and for the out of plane sensing a pendulum have been realized. The fabrication technology for the three elements is identical; therefore the three elements can be placed on one die. We have also chosen a hybrid integration of the MEMS and its ASIC within a hermetically sealed ceramic package. Under normal environmental pressure conditions, the MEMS would run into an over-damped regime. In that case the functionality of the accelerometer could not be guaranteed especially regarding the linearity in the bandwidth. Therefore, some level of vacuum is needed within the micro-cavity. The wafer level packaging (WLP) technique consists in bonding a cap wafer onto the MEMS wafer once the moveable structures are released. This technique offers several advantages compared to traditional packaging techniques. First, it allows the protection of the fragile MEMS structures from the early stages of the manufacturing process and eases the problematic step of the singulation of the MEMS dies. Furthermore, since the bonding step can easily be performed under controlled pressure, the vacuum inside the MEMS accelerometer is set at the desired value. And finally, WLP takes full advantage of the parallel manufacturing to drastically reduce the cost associated to the packaging. In addition, WLP simplifies the implementation of the out-of-plane acceleration sensor since the counter electrode can be integrated into this cap layer. For the in plane accelerations the sensor is using inter-digitated comb structures, which form a differential capacitor. The fixed plates of the differential capacitor are formed by fingers attached to the substrate, whereas the movable part is a mass with fingers attached. Two of these structures rotated by 90 degrees are placed on the die. For the measurement of out of plane accelerations (z axis) a pendulum with asymmetric mass distribution is used, which is forming the movable part of a differential capacitor. The fixed part is realized by counter electrode on top of the mechanical element.
Although the basic structure of the x- and y-sensors and the z-sensor is different, a suitable design allows the achievement of similar sensor parameters. An SOI wafer is used for the MEMS sensors. The mechanical structures are defined by a combination of DRIE etching and sacrificial layer etching. The cap is formed by two silicon wafers bonded together. The top wafer of the cap stack is etched with KOH to allow the electrical contact to the cap bottom. For the bottom part of the cap, deep trenches are etched all around the different areas to ensure the electrical isolation. For the top part, the isolation is guaranteed by a thick surface oxidation (1-2?m) of the silicon wafer. The cap itself is mounted to the MEMS wafer by using a metal-silicon eutectic bonding process, used also to realize the electrical connections between the cap and the MEMS wafer.
The MEMS sensors are integrated into wireless sensor modules to form a monitoring system which communicates the measurement data to a remote base station and is controlled by this station. The accelerometers are integrated using standard printed circuit board technology and assembled into a rigid protective housing that allows fixation to the building. The strain sensor modules consist of two parts. The wireless communication and processing part is implemented in a similar rigid housing and remains accessible for potential battery replacement. The front-end sensing module combining the MEMS strain sensor, read-out ASIC and supporting oscillator circuitry and passives, is embedded inside the concrete structure, applied to the metal reinforcing bars. It uses a special package using PDMS silicone molding and a polyimide carrier in order to allow the sensor to be glued onto the metal bar similarly to a traditional strain gage while efficiently transferring the strain from the metal bar to the MEMS sensor.
For the wireless communication an IEEE 802.15.4 radio module in the 900MHz band is used in combination with a custom-designed patch antenna, in order to obtain robust links in real-world conditions inside a building. The wireless sensor modules are powered by an 8.5Ah C-cell long operating life primary lithium battery. The modules also include a low power processor and a 64Kx16 bit RAM memory to record earthquake events or strain data. A low power network architecture was implemented on top of an 802.15.4 MAC using indirect data transfers which allows the sensor modules radios to remain powered down most of the time except during brief polling moments and during periodic or even-triggered transmission of recorded data. The strain sensor modules measure periodically and use a radio polling interval of 60 seconds. The 3D accelerometer is constantly running at 3 x 200Hz sample rate with the measurements recorded in a 54-second loop buffer. This is required to be able to record the early onset of an earthquake event, even before it reaches a trigger threshold, and requires an ultra-low power sensor and readout as we have developed in the MEMSCON project. It also enables the triggering to be done remotely by the base station based on the combined information from several monitoring nodes across the network. In that way data can be obtained from all accelerometers in the network during an earthquake or other global event, even if locally at some sensor nodes the chosen earthquake threshold was not exceeded. It also allows the wake-up monitoring function to be enabled only on a select amount of sensor modules which do not give many false alarms, e.g. due to street traffic or other local interfering signals. To respond timely without data loss to an event triggered from the base station, the radio polling interval of the accelerometer modules is limited to 15 seconds. The strain sensor modules show an average power consumption of 0.274mW which results in a operational time of 12 years before the batteries need to be replaced. The accelerometer modules consume 1.73mW resulting in 2 years of battery life.
4.1.1.1 MEMSCON Strain Sensors
The MEMS Strain Sensor chip was designed using a capacitive sensing principle. Hence a strain in the rebars of the building under study has to be transferred to a change of capacitance in the sensing part of the sensor. In our case, this is done by a relative displacement of the fingers of a capacitive comb drive structure. The design of the sensor is based on the inputs of the end users, and is designed to measure strain over a +/- 30000µ? range with a resolution of 10µ?. It has a size of 3600x 8400x1050µm. The MEMS sensor is fabricated using a SOI (Silicon on Insulator) wafer as substrate. SOI wafer consist of a Device layer, a BOX layer and a Handle. Both the Handle and the Device layer are made out of doped Silicon (for the MEMSCON Strain Sensor) and the BOX layer out of Silicon Oxide (glass). The purpose of using such a type of wafer is that it simplifies the production of the moving capacitor (in the Device layer). With such wafers, the desired structure can be patterned in the device layer and released by selectively etching the oxide layer under it.
The layout of the structure is such that 2 differential capacitors are created and each of those capacitors consists of 4 comb drives in parallel. Each comb drives consist of 50 fingers. Two anchors are located on the back side of the MEMS Sensor. They define the position where the strain will be transfer from the rebar to the sensor. The distance between the 2 anchor points define the relation between strain in the structure and displacement in the comb drives. In our case, the distance between the anchors (center to center) is 3mm, which means that the strain of 30000µ? in the rebar produces a displacement of 90µm in the comb drives. A challenge was in designing the large stiff spring that holds the moving parts together. An optimum needed to be found between having enough stiffness so that the offset due to packaging and assembly stays within the allowed limit (20µm), not to stiff to limit creep in the glue during operation and finally the stress at full elongation/compression shall not exceed the yields strength of the silicon. The theoretical yield strength of monocrystalline silicon is around 7GPa, however, due to defect in the crystal the actual yield strength is closer to 3.5GPa. For the design of the spring, we want to stay in a region of 700MPa-1GPa (5-3.5 time lower that the actual yield strength) in order to have a sufficient security margin. The design was modeled with a 3D CAD drawing software (IronCAD) and simulations were performed with a Finite Element Model Analysis software (ALGOR) to check that the design doesnt exceed the maximum stress and have enough stiffness.
4.1.1.2 MEMSCON Integrated System
The MEMS sensors are integrated into wireless sensor modules to form a monitoring system which communicates the measurement data to a remote base station and is controlled by this station. The accelerometers are integrated using standard printed circuit board technology and assembled into a rigid protective housing that allows fixation to the building. The strain sensor modules consist of two parts. The wireless communication and processing part is implemented in a similar rigid housing and remains accessible for potential battery replacement. The front-end sensing module combining the MEMS strain sensor, read-out ASIC and supporting oscillator circuitry and passives, is embedded inside the concrete structure, applied to the metal reinforcing bars. It uses a special package using PDMS silicone molding and a polyimide carrier in order to allow the sensor to be glued onto the metal bar similarly to a traditional strain gage while efficiently transferring the strain from the metal bar to the MEMS sensor.
For the wireless communication an IEEE 802.15.4 radio module in the 900MHz band is used in combination with a custom-designed patch antenna, in order to obtain robust links in real-world conditions inside a building. The wireless sensor modules are powered by an 8.5Ah C-cell long operating life primary lithium battery. The modules also include a low power processor and a 64Kx16 bit RAM memory to record earthquake events or strain data. A low power network architecture was implemented on top of an 802.15.4 MAC using indirect data transfers which allows the sensor modules radios to remain powered down most of the time except during brief polling moments and during periodic or even-triggered transmission of recorded data. The strain sensor modules measure periodically and use a radio polling interval of 60 seconds. The 3D accelerometer is constantly running at 3 x 200Hz sample rate with the measurements recorded in a 54-second loop buffer. This is required to be able to record the early onset of an earthquake event, even before it reaches a trigger threshold, and requires an ultra-low power sensor and readout as we have developed in the MEMSCON project. It also enables the triggering to be done remotely by the base station based on the combined information from several monitoring nodes across the network. In that way data can be obtained from all accelerometers in the network during an earthquake or other global event, even if locally at some sensor nodes the chosen earthquake threshold was not exceeded. It also allows the wake-up monitoring function to be enabled only on a select amount of sensor modules which do not give many false alarms, e.g. due to street traffic or other local interfering signals. To respond timely without data loss to an event triggered from the base station, the radio polling interval of the accelerometer modules is limited to 15 seconds. The strain sensor modules show an average power consumption of 0.274mW which results in a operational time of 12 years before the batteries need to be replaced. The accelerometer modules consume 1.73mW resulting in 2 years of battery life.
4.1.1.3 Software for the Assessment of the Damage Due to Differential Settlement between Foundations
To calculate the resulting internal forces at critical cross-sections of the structural members, measurements of strains are needed at four points in each cross-section. Then, from the constitutive laws of the structural materials and on the assumption of linear distribution of strains over the area of the cross-section, one can estimate the stresses and their resultants, the internal forces. For the above we need on the average 6 strain sensors per structural element. The implication is that for an average 5 storey building with 25 columns and 40 beams per storey one needs approximately 2,000 strain sensors. In this work the required number of strain sensors has been dramatically reduced. Thus, strain sensors are only needed at the bottom cross-section of the columns at the ground floor level. The implication of this is that for an average 5 storey building only about 60 (as opposed to 2000) sensors are needed. Then, under operating conditions, the internal forces in each structural member as well as their structural adequacy and the differential settlement between foundations are being assessed through a commercially available finite element program that accepts as input the measured values of the above critical parameters. In more detail the above are executed in the following steps: From the strain measurements the axial forces and the bending moments in two principal directions at the bottom cross sections of the columns at the ground floor are calculated. The variations of these forces with respect to those developed at the initial state after the finishing of the structure are calculated. The differential displacements between the foundations of the columns are calculated from the compatibility condition with the above variations in internal forces. The total structure, represented by a finite element space model, is analysed for the actual applied loads and the estimated differential settlements, by using a non-linear structural analysis computer program. The output is the values of internal forces and displacements for the totality of the structural members. The values of the safety factors of the critical cross sections for the estimated values of internal forces with respect to their available ultimate strength are calculated and compared to their permissible values, in order to estimate their structural adequacy.
4.1.1.4 Software for the Assessment of Seismic Damage
The recording system selected in this work involved the installation of a small number of three-dimensional accelerometers, two per storey of the building, in combination with a commercially available finite element program for structural analyses. The assessment of local stability conditions of the structure involves the detection of the cross-sections where plastic hinges are formed and the estimation of the corresponding damage degree. From the successive imposed displacements and the corresponding structural analyses the bending moment-curvature diagrams for the cross-sections are developed. These diagrams representing the development of the pair of values for bending moment-curvature during the successive cycles of the seismic oscillation, constitute the hysteresis loop pattern. The various areas inside the loops represent the different modes of energy (dissipated, restoring, released, elastic and complimentary). These, in the form of a modified Parc and Ang Damage Criterion, are used to determine the damage degree for the cross-sections in terms of energy quantities. Global instability (total failure) of the structure ensues when both end cross-sections of the totality of the columns in a storey develop plastic hinges and a storey kinematic mechanism is formed. From the relative translations of the two accelerometers on the floor one can estimate the coordinates of the instantaneous pole of rotation. The safety factor against global instability defines the degree of storey mechanism formation for each storey in the building.
4.1.1.5 Software for the Selection of remedial Measures and the Estimation of their Costs
This software accepts input from the software on damage due to differential settlements on the magnitude of differential settlements as a function of time. Thus, it can be established, timely, whether foundation movement is stabilising or is progressive and threatening to the building reducing dramatically the expense of remedial measures. Input from the above software additionally includes the safety factors for all structural members. Based on these methods for strengthening or repairs are selected and their cost estimated. In the case of an earthquake, this software accepts input from the software on seismic damage on the structural elements that have been damaged and their degree of damage. Based on the latter degree it presents common repair methods to bring the element to its pre-earthquake state. The user selects one of the suggested repair methods for each damaged element and then the model estimates the cost to repair each element and sum it up to get the total direct structural repair cost.
4.1.1.6 The Integrated DSS
All of the above software was integrated in a DSS for proactive rehabilitation and rehabilitation following seismic damage. This DSS also includes an expert system with a friendly user interface and knowledge and data DBs. A graphical user interface provides the graphical environment with which the end-user can retrieve current and historical data from the DBs while also provides real-time alerts and warnings in case of unsafe situations and allows the end-user to examine different scenarios for hypothetical situations.
4.1.1.7 Laboratory Validation
MEMSCON sensors were tested in laboratory conditions on both reduced scale specimens, to assess their performance, and full-scale specimens, to assess the reliability of the whole system. Performance of the MEMSCON accelerometers were assessed mounting three wireless nodes on a shaking table, driven by a function generator connected to an amplifier, back to back with high sensitivity wired piezoelectric accelerometers (model 393B12 and 393C produced by PCB Piezotronics), producing several vibration tests with excitation of various shape, frequency and amplitude. In particular the tests focused on frequencies (0-20 Hz) and amplitudes (100-500 mg) of the shaking on the range of relevance for seismic monitoring. Several tests conducted producing harmonic excitation to the shaking table were conducted, in order to estimate calibration coefficients of each device, directly comparing MEMSCON and reference measurements, along each axis. Moreover, the wireless sensors were mounted also on a two story metal frame, again back to back with reference accelerometers, reproducing to the shaking table a spectrum compatible ground acceleration time history, to simulate operative conditions during an earthquake. Tests highlighted that the sensitivity of the MEMSCON accelerometers differs among the 3 axes, particularly between axes X and Y (finger comb technology) with a mean sensitivity of 200 mV/g and Z axis (asymmetric pendulum technology) with a mean sensitivity of 100 mV/g. Considering the maximum voltage range (0-2.5V) and the 16 bit resolution, the nominal resolution of the devices in terms of acceleration is estimated as 0.1-0.4 mg, therefore better than system requirements. The accuracy of the devices is directly correlated to the background noise contained in the signals, estimated as about 20 mg. The clock seems to be stable, with a sampling rate of 202 Hz.
MEMSCON accelerometers were also validated performing a experimental campaign on a full scale three dimensional concrete frame, with dimensions 3.15x3.15x3.90 meters realized inside laboratory. The frame, realized selecting materials widely used in construction as B450C steel reinforcing bars and concrete C25/30 class, reproduced in total a portion of the ground floor of a prototype building intended for public use and designed in order to develop plastic hinges at the interface between foundation and column during the earthquake and to keep into the elastic range all the elements except obviously the columns at the bottom end. The frame consisted of four 350x700 mm orthogonal beam post-tensioned foundations, four 300x300 mm columns 2.80m high, two 300x500 mm longitudinal beams and two 300x400 mm transversal beams at the top and a 120 mm concrete slab. The response of the first floor of the prototype building to a spectrum compatible earthquake was numerically estimated and applied dynamically to the concrete frame slab using an horizontal actuator controlled in displacements. Accelerations at the top of the concrete frame were recorded using MEMSCON accelerometers and piezoelectric wired accelerometers as reference, mounted also in this case in back-to-back configuration, fastening all the instruments directly to the concrete slab. The dynamic tests conducted at different amplitude highlighted that MEMSCON accelerometers are suitable to monitor a real concrete structure during a seismic event, with discrepancies with the wired accelerometers less than 20 mg in terms of root mean square. Acceleration measures were used also to estimate the displacement time history at the top of the frame, by double integrating the accelerations in respect of the time. This estimation was compared with the time history of displacements as produced by the horizontal actuator. The comparison shown that the error committed estimating displacement using double integration of the acceleration as recorded by the MEMSCON accelerometers is in the order of 0.5mm.
The performance of MEMSCON strain sensors were firstly assessed carrying out tensile and compressive tests performed on steel bars and reinforced concrete small scale specimens, adopting various load protocols, all consisting of load-unload cycles with increasing intensity, up to the failure of either the specimen or the sensor. The sensors performances were assessed directly comparing the devices with commercially available reference strain gauges (HBM LY41-3/700). Since one of the project target was to guarantee an high survival rate of the sensors also after a severe seismic event, a particular gluing procedure consisting in develop a chemically clean surface having appropriate roughness and a surface alkalinity suitable for gluing, was developed. Tests verified that most part of the sensors survive the installation procedure and the concrete hardening. Sensitivity of the sensors after gluing them on the reinforcing bars was estimated at 20 ??/mV while the resolution was about 0.75 ??.??Comparing MEMSCON and reference observed an accuracy of 5-15 ??/mV during test on bare bars was observed, while the precision is of the same order of magnitude of accuracy (10-20 ??/mV). Concerning the maximum strain recordable, during test on bare bars, strains up to 2% easily reached and recorded without any debonding, confirming that the developed gluing procedure is suitable to realize a system which survives after a severe earthquake.
Strain sensors were embedded also into a concrete frame, in order to monitor the strain evolution inside the columns due to gravity loads, thermal effects and shrinkage/creep of the concrete. Examining the response of the sensors is possible to detect the moment when the slab was casted, the daily thermal variation of 5-10 micro-strains due to a daily thermal excursion inside laboratory of about 0.5 °C, the creep/shrinkage effects in the form as a slow contraction of the reinforcing bars. The reliability of the system by using the reference sensors was therefore demonstrated.
Potential Impact:
4.1.2.1 MEMSCON Potential Impact
The impact of MEMSCON can be summarised to the following:
Transform the building rehabilitation sector (currently dominated by SMEs) into an advanced knowledge sector through the promotion of integrated monitoring systems.
Enhance the competitiveness of European SMEs in building design, construction, inspection and rehabilitation by increasing their productivity through new processes and high added value products: With the developed integrated package, the engineers will have the ability to precisely determine the precise structural state of the building.
Decrease the time needed for engineers to evaluate the structural condition of a building in service and decide on remedial measures.
Promote the Proactive Condition-Based Maintenance of buildings which is based on measurements aimed at early detection of degradation, thereby allowing degradation to be eliminated or controlled prior to significant physical deterioration. The result is a significant decrease in maintenance cost (since problems are less expensive to fix when they are first developing) and increase in building safety.
Promote a new concept of civil engineered smart structures conceived and designed as high performance systems, embedding novel, low cost and highly reliable sensors, capable of self-diagnosis and assessment of optimal rehabilitation.
Enable the MEMS European SMEs to enter the large building market: Provisions of the building codes in most cities in seismic zones no 3 and no 4 require the installation of accelerographs in new high buildings while legislation on building code regulation is bound to become more pervasive in the future, necessitating the installation of monitoring stations on all types of new public and private buildings as well as bridges and the other infrastructure. The ideal implementation of the code requirements would require 3-axes monitoring platforms in multiple locations within the building and the MEMSCON package is ideally suited for this.
Boost the already high value of MEMS technologies in the field (size, power consumption, performance, and overall costs) enabling a wide spread of the system, with a direct impact on safety in our everyday life. This will enable the MEMS European SMEs to enter this market by leveraging their existing know-how in high-end products, increase their worldwide leadership in such fields, and contribute actively to their employment and turnover growths.
Provide solution that integrates building design, construction, monitoring, operation and maintenance thus promoting sustainability (extend the life of buildings and reduce resource consumption at the same time that will offer increased safety) and offer full services with a high value added for clients.
Promote a breakthrough monitoring technology that will permit the timely input of objective, reliable, quantitative data from all points of interest in the building structure. This will make it possible for the SME dominated building management sector to provide better quality services (e.g. higher safety) and dramatically reduce life span inspection and maintenance costs.
Provide an essential way for an accurate and quick assessment of the building structural condition for several reasons: (a) dissemination of information to emergency response officials on building collapses within minutes after the occurrence of the earthquake resulting in many lives saved and prudent allocation of resources and (b) quick and accurate estimates of the level of damage that can be used to indicate loss of function and help officials decide whether the school, hospital, etc., should be evacuated or remain in service.
Act as a catalyst for new high-tech sensing systems (e.g. systems including chemical sensors), structural control systems and software for the structural assessment, performance and management of additional structures (e.g. retaining walls, bridges, dams), additional materials (e. g. steel) and groups of structures (e.g. in an intelligent city), with the subsequent creation of growth and employment of construction SMEs in the Union.
? MEMSCON Competitive Advantage
Using proprietary state-of-the-art modeling know-how for reinforced concrete structures, innovative decision support algorithms, and breakthrough low power wireless sensor networks including accelerometers and strain sensors, the MEMSCON consortium has developed a full monitoring system which enables building structural monitoring and assessment.
The consortium has identified an opportunity to leverage this activity in order to further develop market and sell the monitoring system, specifically in earthquake prone areas.
In the above application, at the present, rather expensive monitoring systems could be used in seismic prone areas. They are rather expensive and complex to implement using wired communication between a limited numbers of sensors and are power supplied by cables. They are rather expensive and are therefore mainly used in expensive state of the art structures in civil engineering (expensive bridges and buildings).
Such systems are sold today in two different ways:
the first sale occurs when civil engineering companies design the building and integrate in their plans the network of sensors and monitoring system: this is a rather expensive activity and increases the cost of ownership of a building/bridge by a few percents
the second sales strategy is coming from a service attached to such monitoring system. The cost of the system, the people used to take decisions and monitor the building quality is then spread out over time as an additional charge (typ. Monthly) like security, video surveillance, etc.
MEMSCON monitoring systems core value proposition to address the above market segments is many folds:
- the limited number and low costs sensors used to monitor the infrastructure
- the low power consumption of these self powered sensors enabling low maintenance of the sensor network
- the easiness with which the sensor network is installed and communicates (wireless system)
- the innovative modeling of structures which enables a limited number of measurement nodes in a building to assess its integrity or the need for repair
- the decision support system which enables a low experienced person to trigger appropriate actions with a limited amount of technical expertise.
? MEMSCON European Added Value
The most important trend in construction in ALL EU countries is the shift from construction costs to life cycle costs. As part of it non-destructive evaluation technologies, like the one offered by MEMSCON, need to be incorporated into buildings throughout Europe. Further to this, the MEMSCON related research is of interest to ALL earthquake prone countries in the southern part of the Union.
The developed integrated system is of interest to ALL EU countries: Knowledge of the structural condition and safety of in-service structures on a continuous time basis is an ultimate objective for owners and maintenance authorities throughout Europe. In all EU countries available funds are insufficient to maintain the existing building infrastructure. Moreover, in all EU countries there is an increasing demand for structural engineers to deal with the issues of life extension of existing structures. To meet this challenge the profession must be able to provide defendable estimates of the safety and performance of existing structural systems.
Building retrofitting and maintenance is (an SME dominated) low margin industry and this legitimates that R&D be supported at a European level, as companies have little financial capacity to invest in technology. Moreover, this is an industry with heterogeneous practices that do require EC working methods to be confronted, adapted and harmonised. Additionally, it is an industry that is lacking in innovation. The delivered package of results provides much needed data on the demand placed on buildings under operating conditions and seismic disturbances and on structures behaving in the nonlinear realm. These data are immediately usable throughout the Union.
4.1.2.2 MEMSCON Dissemination & Exploitation Activities
The results of MEMSCON can find large scale applications in the safety assessment of relevant buildings. To this aim, although the partners also made efforts to disseminate the project results to the scientific community, the main effort has been focused on a market oriented approach, to end-users, building structural rehabilitation contractors, rehabilitation engineers, companies offering structural monitoring services, companies in RC building construction, companies in building facilities management, owners of RC buildings, the Building Maintenance Departments of local and federal government, representatives of relevant associations, consortia, organizations and societies, insurers, standardization bodies, public safety officials, relevant forums in construction.
In particular, the dissemination strategy focused on:
Inducing among technical specialists a change regarding processes, procedures, and methods used for the monitoring of structures. The feasibility of the system in detecting damage will be put in evidence, as well as the benefit for the safety level of the whole community.
Particular stress is given to the economic convenience of the proposed system, when fully integrated in the construction phases of new buildings, thus minimizing direct and indirect costs.
In the course of the project, dissemination strategy was directed to the accomplishment of three major tasks:
Task 1: Activity of promotion: communication the MEMSCON Project
Task 2: Technology diffusion: definition of service for potential users
Task 3: Communication of scientific results
In order to accomplish these tasks, the partners have been involved in the following activities:
Presentation of MEMSCON to international conferences and symposia.
Further advertising the project results through appropriate channels.
Implementation and Maintenance of the MEMSCON Web Site.
Training of the staff that will be using the proposed system in the field tests.
Organization of MEMSCON Workshops.
Following the dissemination activities a large set of dissemination material was produced within the framework of the project in order to describe the project and its results. The material is both in digital format, such as Newsletters, website, etc. and in hard copy, such as leaflets. The primary function of dissemination material is to provide information on project concept, participants, aims, results, activities, etc. Since not all results have been available from the beginning of the project, they have been continuously updated with new information about the project, realized activities, events, findings, etc. All dissemination material is designed according to a common principle following the same concept and logo specifically designed for MEMSCON. In detail the dissemination means were the following:
Project Logo
The project logo is used for the cover page of all documents of MEMSCON project, internal reports, technical reports, deliverables, project plans. It is also inserted in all dissemination material such as leaflets, website, newsletters, videos, etc. in order their relation to MEMSCON to be directly identifiable.
Project Website
Available at the address www.memscon.com/ the website contains all the necessary information on the structure of the Project, its aims and goals, important news and milestones reached throughout the whole duration of the Project as well as project dissemination and public material.
Project Newsletter and press releases
Six Newsletters were published during the project execution. The Newsletters have been distributed to a wide audience through the MEMSCON mailing list, and distributed in printed copies at conferences, workshops and events.
Project Leaflets and Posters
Leaflets and posters have been created providing a brief presentation of the projects objectives and activities, distributed during various events to all interested people
Project Workshops (1st: Bucharest and 2nd: Athens)
Two workshops were organized by the members of the consortium, to present the results of the project.
Project Presentations and Submissions
During the duration of MEMSCON project research activities, a large number of publications were issued, covering the projects concept, activities and results. These include: 4 papers in scientific journals, 15 papers in conference proceedings, plus the papers published by the partners at the two MEMSCON Workshops.
Detailed descriptions of the MEMSCON dissemination plans and activities can be found in the deliverable D5.1 (Plan for the dissemination and use of the results Final).
Regarding exploitation activities, the MEMSCON consortium has produced a detailed IPR protection and IPR plan documented in the MEMSCON Technology Implementation Plan (TIP). In this document, the IPR issues are fully described including the agreements on ownership rights. The industrialization efforts required for each of the MEMSCON outcome, are also described in the same document.
List of Websites:
? The project website URL is the following: www.memscon.com
? Contact details:
Project Coordinator:
Dr Angelos Amditis
Research Director
Institute of Communication and Computer Systems
a.amditis@iccs.gr
Technical Manager:
Dr Matthaios Bimpas
Institute of Communication and Computer Systems
mbibas@esd.ece.ntua.gr
Dissemination Manager:
Professor Daniele Zonta
University of Trento
daniele.zonta@ing.unitn.it
European Commission Project Officer
Dr. Ir. Dominique Planchon
Project Officer, European Commission
dominique.planchon@ec.europa.eu