Periodic Reporting for period 2 - SENSIBLE (SENSors and Intelligence in BuLt Environment)
Reporting period: 2019-01-01 to 2022-06-30
Buildings contribute over 40% of the total energy demand, half of which is due to non-residential buildings. The increase in working population, hours spent in offices, and growth in demand for building services and improved comfort, escalates building energy demand. According to the European Commission (EC) 2010 Energy Performance of Buildings Directive, improving the energy efficiency of buildings will decrease by 5-6% the total EU energy consumption. All EU countries must set minimum energy performance requirements for new buildings, and for major renovation or retrofit work in old buildings. Based on the EC 2012 Energy Efficiency Directive, EU countries must draw-up long-term national building renovation strategies. Information and communications technology will play an essential role in reducing energy consumption of buildings as well as improving occupants’ comfort levels, through advanced sensing, simulation, modelling, analytical, and visualization tools.
SENSIBLE attempts to answer the question of whether smart buildings can deliver on their energy efficiency and sustainability targets without compromising, but rather improve, their occupants’ perceived and sensed comfort. SENSIBLE will seamlessly incorporate multiple heterogeneous subsystems, tackling sensor design, communications and data processing to provide a comfortable working and living environment for all occupants as well as deliver on energy efficiency targets. This will impact how business services evolve under globalisation and sustainability challenges.
SENSIBLE attempts to answer the question of whether smart buildings can deliver on their energy efficiency and sustainability targets without compromising, but rather improve, their occupants’ perceived and sensed comfort. SENSIBLE will seamlessly incorporate multiple heterogeneous subsystems, tackling sensor design, communications and data processing to provide a comfortable working and living environment for all occupants as well as deliver on energy efficiency targets. This will impact how business services evolve under globalisation and sustainability challenges.
The goal of SENSIBLE is to develop a novel integrated decision support mechanism embedding intelligent sensing, communications and data processing methodology for improving sustainability of smart buildings through new insights, approaches and technologies for acquisition, communications, and extraction of useful information from the sheer volume of sensed data.
The work carried out towards achieving this goal comprises a blend of structured research, training, and knowledge sharing activities. In particular, the research objectives have been tackled during the secondments, by hands-on research work by the secondees supported by mentors and hosting research teams, including sharing access to relevant equipment, joint PhD student supervision, information sharing, analysis, testing, and cross-utilization of anonymised/processed data.
At the beginning of the project, five smart building-related applications were identified, namely, smart chair, smart lighting, smart transport, energy, health and wellbeing, all underpinned by the scientific work conducted across technical Work Packages, beneficiaries and partners, to support theoretical, algorithmic and practical implementation aspects of these applications.
The specific research activities include: (a) developing advanced sensing for smart buildings via a combination of integrated micro-spectrometers for local environmental monitoring and wearable, paper-like flexible sensors based on organic electronics; (b) energy-efficient IoT-enabled fusion of data from physical and virtual sensors; (c) data mining and learning algorithms; and (d) integration of the developed sensing, communications and data analysis into an overall decision support system.
The main research achievements so far include:
1) New sensors for smart buildings encompassing environment, occupants, transport and building infrastructure, including SHFT microspectrometers incorporating hardware athermalization and paper-like sensors based on organic electronics and ferroelectric materials
2) Energy-efficient IoT-based architecture for acquiring signals from smart objects
3) Novel near real-time data science approaches for extracting meaningful information from the acquired data, including data representation methods for dimensionality reduction and computationally-efficient, data mining algorithms for large, distributed and heterogeneous datasets.
4) Integration of sensor design, communication and information processing results via predictive data analytics, scalable data management and decision support tools, into a smart chair, smart lighting, smart parking, energy and wellbeing systems.
Two SENSIBLE schools were organised as planned, with a range of courses delivered that contributed to expanding research and transferable skills of researchers. In addition, joint student mentoring and research output co-production supports exchanging best practice and improving teaching and research delivery at individual institutions.
The work carried out towards achieving this goal comprises a blend of structured research, training, and knowledge sharing activities. In particular, the research objectives have been tackled during the secondments, by hands-on research work by the secondees supported by mentors and hosting research teams, including sharing access to relevant equipment, joint PhD student supervision, information sharing, analysis, testing, and cross-utilization of anonymised/processed data.
At the beginning of the project, five smart building-related applications were identified, namely, smart chair, smart lighting, smart transport, energy, health and wellbeing, all underpinned by the scientific work conducted across technical Work Packages, beneficiaries and partners, to support theoretical, algorithmic and practical implementation aspects of these applications.
The specific research activities include: (a) developing advanced sensing for smart buildings via a combination of integrated micro-spectrometers for local environmental monitoring and wearable, paper-like flexible sensors based on organic electronics; (b) energy-efficient IoT-enabled fusion of data from physical and virtual sensors; (c) data mining and learning algorithms; and (d) integration of the developed sensing, communications and data analysis into an overall decision support system.
The main research achievements so far include:
1) New sensors for smart buildings encompassing environment, occupants, transport and building infrastructure, including SHFT microspectrometers incorporating hardware athermalization and paper-like sensors based on organic electronics and ferroelectric materials
2) Energy-efficient IoT-based architecture for acquiring signals from smart objects
3) Novel near real-time data science approaches for extracting meaningful information from the acquired data, including data representation methods for dimensionality reduction and computationally-efficient, data mining algorithms for large, distributed and heterogeneous datasets.
4) Integration of sensor design, communication and information processing results via predictive data analytics, scalable data management and decision support tools, into a smart chair, smart lighting, smart parking, energy and wellbeing systems.
Two SENSIBLE schools were organised as planned, with a range of courses delivered that contributed to expanding research and transferable skills of researchers. In addition, joint student mentoring and research output co-production supports exchanging best practice and improving teaching and research delivery at individual institutions.
Progress beyond state of the art includes:
- Two spectral retrieval algorithms for thermal-drift mitigation developed and experimentally demonstrated.
- A novel polarization-independent monomode waveguide with temperature resilience developed
- An experimental setup for testing broadband integrated devices for wavelength conversion of 10 Gb/s RZ-OOK signals based on cross phase modulation in an integrated nonlinear optical loop mirror fabricated on silicon-on-insulator
- Optical sensor design based on asymmetric Mazh-Zehnder interferometer
- Smart positioning framework via LED-based Visible Light Communications
- Modelling of multi-state appliances (washing machine, dish-washer) by explicit duration models together with computationally efficient optimal classification and performance characterization in the large sample regime via error (mis-classification) exponents.
- Deep learning NILM algorithms suitable for transfer learning for understanding energy demand in buildings
- Gait phase classification methods using Random Forest for improving posture and well-being
- Robust deep graph learning based on graph regularisation tailored to noisy or insufficient training set
- Novel person-specific smart lighting platform that can maximise user experience by simultaneously satisfying multiple user requirements in terms of level of light in an open-space office environment
- Novel smart chair system based on sensing the sitting posture and performing data analytics to classify the posture and provide siting advice
SENSIBLE is working towards a holistic decision support system that includes: smart lighting system that can automatically adjust level of light based on user requirements; smart chair system that provides posture advice in real time; smart energy system that provide energy advice based on disaggregating total consumption down to individual appliances used; all supported by advanced sensing including ambient, structural sensing, and wearable electronics, for reliable and secure signal acquisition and pre-processing; energy efficient IoT and M2M architecture for communicating the readings; and advanced semi-supervised classification methods that provide high accuracy even when training data are scarce or noisy.
Achieved Impact so far includes:
1. A major contract established with mobile operator to test IoT devices in real-world environment, expected impact novel services via NB-IoT designs, education provision and future research
2. Smart lighting and smart chair systems are developed through close interaction with industrial partners aiming to impact the future of open-space office designs
3. Smart parking and smart building testbeds will impact education offerings at the local universities and beyond and will contribute to further research in the area via novel datasets, methodology and research findings.
- Two spectral retrieval algorithms for thermal-drift mitigation developed and experimentally demonstrated.
- A novel polarization-independent monomode waveguide with temperature resilience developed
- An experimental setup for testing broadband integrated devices for wavelength conversion of 10 Gb/s RZ-OOK signals based on cross phase modulation in an integrated nonlinear optical loop mirror fabricated on silicon-on-insulator
- Optical sensor design based on asymmetric Mazh-Zehnder interferometer
- Smart positioning framework via LED-based Visible Light Communications
- Modelling of multi-state appliances (washing machine, dish-washer) by explicit duration models together with computationally efficient optimal classification and performance characterization in the large sample regime via error (mis-classification) exponents.
- Deep learning NILM algorithms suitable for transfer learning for understanding energy demand in buildings
- Gait phase classification methods using Random Forest for improving posture and well-being
- Robust deep graph learning based on graph regularisation tailored to noisy or insufficient training set
- Novel person-specific smart lighting platform that can maximise user experience by simultaneously satisfying multiple user requirements in terms of level of light in an open-space office environment
- Novel smart chair system based on sensing the sitting posture and performing data analytics to classify the posture and provide siting advice
SENSIBLE is working towards a holistic decision support system that includes: smart lighting system that can automatically adjust level of light based on user requirements; smart chair system that provides posture advice in real time; smart energy system that provide energy advice based on disaggregating total consumption down to individual appliances used; all supported by advanced sensing including ambient, structural sensing, and wearable electronics, for reliable and secure signal acquisition and pre-processing; energy efficient IoT and M2M architecture for communicating the readings; and advanced semi-supervised classification methods that provide high accuracy even when training data are scarce or noisy.
Achieved Impact so far includes:
1. A major contract established with mobile operator to test IoT devices in real-world environment, expected impact novel services via NB-IoT designs, education provision and future research
2. Smart lighting and smart chair systems are developed through close interaction with industrial partners aiming to impact the future of open-space office designs
3. Smart parking and smart building testbeds will impact education offerings at the local universities and beyond and will contribute to further research in the area via novel datasets, methodology and research findings.