Periodic Reporting for period 2 - MATRYCS (Modular Big Data Applications for Holistic Energy Services in Buildings)
Reporting period: 2022-04-01 to 2023-09-30
The decentralization of the energy system coupled with advancements on IoT, big data, AI and distributed computing are creating a new momentum for exploiting data-driven services to improve buildings energy efficiency.
Despite a large number of buildings data hubs and vocabularies have become available, some barriers hamper the exploitation of that potential, such as the lack of interoperability among heterogeneous static (e.g.BIM) building automation and IoT dynamic data sources and ontologies, and the lack of interoperable big data architectures fully tailored to smart buildings.
In that respect MATRYCS will
i) deliver an open Reference Architecture for Smart Energy Efficient Buildings, which aligns BDVA SRIA, FIWARE architecture, SAREF, HAYSTACK, and BRICK schema vocabularies (among the many others), and enable B2B sovereignty preserving multi-party data exchange, while providing full interoperability of big data enablers with smart buildings standards and addressing privacy and cyber-security constraints
ii) upscale a number of TRL 5-6 technology enablers, such as sovereignty-preserving DLT/off-chain data governance, big data pipeline orchestration, IoT/edge AI-based federated learning and visual analytics and deploy them within the TRL 7-8 MATRYCS workbench
iii) deliver a TRL8 open modular big data cloud analytic toolbox as front-end for one-stop-shop analytics services development
iv) validate such framework through the deployment of analytics services focusing on digital building twins, improved buildings operation, building infrastructure design, EU/ national policy assessment for energy efficiency investments on 11 large scale pilots by different stakeholders (facility managers, ESCOs, financial institutions, construction companies, municipalities, electricity grid and DH operators, policy makers)
Despite a large number of buildings data hubs and vocabularies have become available, some barriers hamper the exploitation of that potential, such as the lack of interoperability among heterogeneous static (e.g.BIM) building automation and IoT dynamic data sources and ontologies, and the lack of interoperable big data architectures fully tailored to smart buildings.
In that respect MATRYCS will
i) deliver an open Reference Architecture for Smart Energy Efficient Buildings, which aligns BDVA SRIA, FIWARE architecture, SAREF, HAYSTACK, and BRICK schema vocabularies (among the many others), and enable B2B sovereignty preserving multi-party data exchange, while providing full interoperability of big data enablers with smart buildings standards and addressing privacy and cyber-security constraints
ii) upscale a number of TRL 5-6 technology enablers, such as sovereignty-preserving DLT/off-chain data governance, big data pipeline orchestration, IoT/edge AI-based federated learning and visual analytics and deploy them within the TRL 7-8 MATRYCS workbench
iii) deliver a TRL8 open modular big data cloud analytic toolbox as front-end for one-stop-shop analytics services development
iv) validate such framework through the deployment of analytics services focusing on digital building twins, improved buildings operation, building infrastructure design, EU/ national policy assessment for energy efficiency investments on 11 large scale pilots by different stakeholders (facility managers, ESCOs, financial institutions, construction companies, municipalities, electricity grid and DH operators, policy makers)
The report covers the period M1-M18 and shows the regular course of the project. Many KPIs, both concerning the technical and scientific objectives as well as the business and communication ones, have already been achieved. Moreover, all the deliverables expected in this period have been submitted, and some milestones have been reached.
One of the main value-added challenges achieved in this period is the capability to elicit synergies, both internally among the partners and externally with similar projects. MATRYCS is collaborating with other initiatives and events presenting its results in different contexts, also outside the EC funded projects ecosystem. Another important point is the user-oriented approach in designing the MATRYCS reference architecture: as a matter of fact, within the 11 project Large Scale Pilots, 23 targets groups have been identified, 67 user stories have been produced for 73 usage scenarios and 28 use cases, defining 92 user requirements.
Noteworthy is that the Consortium has started up a new community in these months, named Big Data Alliance (BDA), to leverage project results visibility and create a new ecosystem where to discuss about "Building and Big Data". The BDA is a running structured collaborative space for selected and active stakeholders that together will solve the practical challenges relating to building lifecycle management.
Below a brief synthesis of the work performed in the WPs during the first 18 months is reported:
Activity in terms of definition of the scenarios, use cases, specifications, requirements as well as the overall view of the reference architecture. The activities led to the release of the first version of the MATRYCS reference architecture (D2.3) was submitted at the end of M12, whereas the release of the final version of the MATRYCS reference architecture has been presented in D2.4 at the end of M18, as well as the final version of use cases, scenarios, and requirements.
The objective of WP3 is to implement the Data Services & Semantic Enrichment Layer. Between M3 and M18 the activities have mainly focused on the definition and implementation of the MATRYCS-GOVERNANCE layer in order to reach the 1st technologies release (D3.1) and the deployment of the GOVERNANCE components on EGI Cloud Infrastructure. WP3 is currently dedicated to ongoing activities for the implementation of the 2nd technology release.
With respect to design and implement the Big Data Management & AI Services Layer, the activities, conducted until now, have been reported in the deliverables D4.1 concerning the 1st technologies release of the MATRYCS PROCESSING (submitted in M11) and D4.4 about the 1st technology release of MATRYCS – ANALYTICS and Open Cloud-based Analytics Toolbox.
The objective of WP5 is to create different types of services, according to the following main topics: MATRYCS – PERFORMANCE, MATRYCS – DESIGN, MATRYCS – POLICY and MATRYCS – FUND. Together with these groups, other two tools will be developed to cover fundamental aspects of building design and management: Digital Twins and the geoclustering tool. This WP started in M8, and the task 5.7 related to analytics platform definition started in M10.
In the first 18 months of the MATRYCS project has devised a Pilot Operation and Measurement and Verification plan, which will be implemented in the subsequent stages of the project and will contribute to the adequate handling of the Large-Scale Pilots and their assessment. Also, a baseline assessment of all pilots has been implemented, and guidelines given to all pilots to follow in the current Cycle 2 (Functional implementation – up to M27).
The work carried out within the first 18 months of the project involves also the first steps of the business and exploitation plan, the identification of relevant projects and initiatives to be reached by MATRYCS, and the management of the BDA. The goal of the first 18 months of the ecommunication part was to build a strong foundation on which the project could clearly and effectively communicate.
One of the main value-added challenges achieved in this period is the capability to elicit synergies, both internally among the partners and externally with similar projects. MATRYCS is collaborating with other initiatives and events presenting its results in different contexts, also outside the EC funded projects ecosystem. Another important point is the user-oriented approach in designing the MATRYCS reference architecture: as a matter of fact, within the 11 project Large Scale Pilots, 23 targets groups have been identified, 67 user stories have been produced for 73 usage scenarios and 28 use cases, defining 92 user requirements.
Noteworthy is that the Consortium has started up a new community in these months, named Big Data Alliance (BDA), to leverage project results visibility and create a new ecosystem where to discuss about "Building and Big Data". The BDA is a running structured collaborative space for selected and active stakeholders that together will solve the practical challenges relating to building lifecycle management.
Below a brief synthesis of the work performed in the WPs during the first 18 months is reported:
Activity in terms of definition of the scenarios, use cases, specifications, requirements as well as the overall view of the reference architecture. The activities led to the release of the first version of the MATRYCS reference architecture (D2.3) was submitted at the end of M12, whereas the release of the final version of the MATRYCS reference architecture has been presented in D2.4 at the end of M18, as well as the final version of use cases, scenarios, and requirements.
The objective of WP3 is to implement the Data Services & Semantic Enrichment Layer. Between M3 and M18 the activities have mainly focused on the definition and implementation of the MATRYCS-GOVERNANCE layer in order to reach the 1st technologies release (D3.1) and the deployment of the GOVERNANCE components on EGI Cloud Infrastructure. WP3 is currently dedicated to ongoing activities for the implementation of the 2nd technology release.
With respect to design and implement the Big Data Management & AI Services Layer, the activities, conducted until now, have been reported in the deliverables D4.1 concerning the 1st technologies release of the MATRYCS PROCESSING (submitted in M11) and D4.4 about the 1st technology release of MATRYCS – ANALYTICS and Open Cloud-based Analytics Toolbox.
The objective of WP5 is to create different types of services, according to the following main topics: MATRYCS – PERFORMANCE, MATRYCS – DESIGN, MATRYCS – POLICY and MATRYCS – FUND. Together with these groups, other two tools will be developed to cover fundamental aspects of building design and management: Digital Twins and the geoclustering tool. This WP started in M8, and the task 5.7 related to analytics platform definition started in M10.
In the first 18 months of the MATRYCS project has devised a Pilot Operation and Measurement and Verification plan, which will be implemented in the subsequent stages of the project and will contribute to the adequate handling of the Large-Scale Pilots and their assessment. Also, a baseline assessment of all pilots has been implemented, and guidelines given to all pilots to follow in the current Cycle 2 (Functional implementation – up to M27).
The work carried out within the first 18 months of the project involves also the first steps of the business and exploitation plan, the identification of relevant projects and initiatives to be reached by MATRYCS, and the management of the BDA. The goal of the first 18 months of the ecommunication part was to build a strong foundation on which the project could clearly and effectively communicate.
Main progress and value added of MATRYCS project will be the design and implementation of the MATRYCS Reference Architecture and it consists of three main pillars:
i) the MATRYCS-GOVERNANCE encompassing modules related to data collection, semantic annotation and distributed storage,
ii) the MATRYCS-PROCESSING including ML and DL models and
iii) the MATRYCS-ANALYTICS providing a set of analytics tools as a service (SaaS/PaaS/IaaS models). On top of the MATRYCS platform, the project defines four main categories of buildings’ services that will be developed and tested in the context of Large-Scale Pilots (LSPs).
MATRYCS will enable reliable and effective policymaking, as well as support the creation and exploitation of innovative services through the utilisation of a wide variety of data, for the safe and effective operation of buildings.
The MATRYCS Modular Toolbox will be able to handle heterogeneous types of data (open data, sensor/IoT, historical data, etc.) from multiple domains and sources. The collected data will be processed to create different business cases considering the interaction of different stakeholders and scales that will be emerged through the LSPs (grouped under the MATRYCS-PERFORMANCE / DESIGN / POLICY / FUND categories).
i) the MATRYCS-GOVERNANCE encompassing modules related to data collection, semantic annotation and distributed storage,
ii) the MATRYCS-PROCESSING including ML and DL models and
iii) the MATRYCS-ANALYTICS providing a set of analytics tools as a service (SaaS/PaaS/IaaS models). On top of the MATRYCS platform, the project defines four main categories of buildings’ services that will be developed and tested in the context of Large-Scale Pilots (LSPs).
MATRYCS will enable reliable and effective policymaking, as well as support the creation and exploitation of innovative services through the utilisation of a wide variety of data, for the safe and effective operation of buildings.
The MATRYCS Modular Toolbox will be able to handle heterogeneous types of data (open data, sensor/IoT, historical data, etc.) from multiple domains and sources. The collected data will be processed to create different business cases considering the interaction of different stakeholders and scales that will be emerged through the LSPs (grouped under the MATRYCS-PERFORMANCE / DESIGN / POLICY / FUND categories).