Skip to main content

A novel decentralized edge-enabled PREsCriptivE and ProacTive framework for increased energy efficiency and well-being in residential buildings

Periodic Reporting for period 1 - PRECEPT (A novel decentralized edge-enabled PREsCriptivE and ProacTive framework for increased energy efficiency and well-being in residential buildings)

Reporting period: 2020-10-01 to 2021-09-30

As the severity of environmental issues continuously increases, the need to reduce overall energy consumption becomes imperative. The focus is more fixated on the building sector, where the largest potential for cost-efficient energy savings has been identified. PRECEPT ambitiously aims to enable the “smooth” and at near-zero operational and maintenance costs transformation of conventional residential buildings into highly efficient proactive residential buildings by introducing a Pred(scr)ictive and Proactive building energy management system (PP-BMS) installed locally at building level. The transformation of conventional reactive residential buildings into proactive ones leads to increased energy efficiency and lower energy consumption as well as optimal levels of comfort while improving the residents’ well-being, as optimal levels of temperature, humidity, moisture, air quality, and lighting are constantly maintained in a building. Moreover, increased operational efficiency is achieved through asset management, as well as predictive maintenance which leads to lower maintenance costs, equipment downtime, fewer breakdowns, and conversely increased time between maintenance cycles and the overall life of the equipment.
To achieve its ambitious goal, the PRECEPT consortium had to set well-defined objectives. PRECEPT utilizing its innovative PP-BMS will enable the building’s proactive behaviour and will fully exploit Renewable Energy Sources (RES), with minimum need for human intervention, to autonomously configure itself and proactively optimize its actions. The PRECEPT PP-BMS can rapidly and efficiently self-manage, self-monitor, self-heal and self-optimize itself, whenever there are system malfunctions, significant deviations in the forecasted weather conditions affecting both thermal and visual comfort as well as RES generation, occupancy schedules, flexibility forecasts, energy tariffs, loads-scheduling, and level of utilization. The use of IoT sensors and machine learning sources will undoubtedly lead to an exponential growth in data volume, that will be needed to be collected and analyzed. PRECEPT aims to provide novel techniques for supporting the collection and analysis of the expected large volume of data. Furthermore, PRECEPT will deliver and integrate digital twin technology tools that will be continuously updated with real-life data for early prediction, automatic repair, and optimal building operation. PRECEPT also realizes the importance of user communication and user communities. Towards this objective PRECEPT plans to develop a social collaboration platform on which, users may exchange information, experiences, and practices. PRECEPT will also introduce standardized procedures for data management and building proactiveness that will contribute to the development of the required Smart Proactiveness Indicators (SPIs) of the building. PRECEPT also aims to deliver AI algorithms dedicated to prescriptive maintenance. These algorithms will exploit historical building information, as well as local climatic data, creating appliance/ device behavioural models. Lastly, PRECEPT will introduce, implement, and demonstrate novel sustainable business models for transforming traditional reactive buildings into proactive buildings. These models will offer diverse solution sets that will include all relevant metrics and indicators.
The PRECEPT work initiated with the composition of the core management procedures and tools used for the effective monitoring and management of the project. The quality assessment plan was developed and provided the framework that allows high-quality results, describing the risk-management procedures and mitigation actions. An occupant-centred approach was followed by performing 77 interviews, which led to the identification of 3 stakeholders’ categories, 11 stakeholders’ groups and the limiting factors for the wide adoption of smart building technologies. Internal documents were produced regarding all impact and excellent indicators as well as the first draft of the key performance indicators (KPIs). The procedure for the establishment of the new Smart Proactiveness Indicators (SPIs) was initiated with an overview of the status and the study of the methodology of the existing Smart Readiness Indicators (SRIs). Field research on the Existing European and national legislation was conducted and emphasis was also given to the conceptual architecture design of the PRECEPT solution by crafting an overview of the solution along with its components and sub-components, their interfaces, and the connections with external systems. A thorough analysis of features affecting human behaviour and comfort was performed and the development of the PRECEPT inference engine was initiated. Innovative advancements in anomaly detection methods and algorithms were investigated while researching approaches for supervised and unsupervised anomaly detection. BIM workflow has been also established during the reporting period. A 3D BIM model of the Netherlands, Germany and CERTH’s smart home pilots were developed, based on which the remaining pilots will be modelled. The OneClickLCA 6D BIM modelling software was tested using a developed pilot model along with the development of a modelling strategy for BIM and BEPS. State-of-the-art algorithms on RES and storage control as well as on management and grid optimization were evaluated while a household DR (Demand-Response) Reinforcement Learning algorithm was developed. Additionally, initial mockups were created and used to design an initial design of the PRECEPT social collaboration platform after extensive discussions about the possible use cases and features of the platform. Parallel to the technical activities, an exploitation and business plan is being developed that will facilitate the commercialization phase of the project. The project’s dissemination and promotion were also main priorities, establishing an effective DC strategy and the delivery and utilization of the first marketing materials.
PRECEPTgoes beyond state of the art on multiple fronts, first and foremost by introducing a completely new self-adapted, -learned, -managed, -monitored, -healing and -optimized Pred(scr)ictive and Proactive building energy management system (PP-BMS). Within the work to achieve this ambitious task, there are already significant achievements that offer cutting edge advantages in the relevant markets. State of the art analysis was conducted to describe prescriptive maintenance algorithm development by applying different machine learning techniques as well as to test the algorithms for detection and prediction of the potential malfunctions in the selected building appliances/devices by using open-access datasets. State of the art analysis was also conducted to describe both a non-intrusive load monitoring algorithm for disaggregating the overall building’s energy consumption to the level of individual devices/ appliances energy consumption and Building Behaviour Inference Engine development by applying supervised and unsupervised machine learning techniques. Lastly, state of the art analysis and bibliography research was conducted on NILM algorithms, existing Federated Learning (FL) frameworks, Building Elasticity & Integrated Proactive Algorithms, and error diagnosis.