Faced with the challenges posed by global warming and the depletion of fossil fuels, all the acto... more Faced with the challenges posed by global warming and the depletion of fossil fuels, all the actors in society<br> (individuals, industries, communities, and governments) are led to find ways to reduce their energy consumption.<br> As residential buildings contribute significantly to the overall energy consumption, a particular effort is requested<br> from buildings occupants to reduce their energy consumption without having to compensate on their quality of<br> life. The energy behaviour of buildings is directly influenced among others by the behaviour of their occupants.<br> This paper describes a new approach based on historical data to sensitize occupants for their actions' impact on<br> building energy comportment without the need to develop a physical model thanks to case-based reasoning<br> (CBR). We report on an experimental evaluation of the approach based on real-world data collected from an<br> office building including ...
2021 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS), 2021
Collective action and citizen involvement in the energy sector represent the new platform upon wh... more Collective action and citizen involvement in the energy sector represent the new platform upon which the novel energy communities are developed. From a scientific point of view, we consider modeling and simulation represent important instruments to shape the future energy communities and assess their impact on the society. In this context, this paper presents an initial conceptual analysis upon the most important requirements and challenges that can be expected in this new research direction, where human actors must be directly involved in the technical solution. We also propose a general model of an energy community, emphasizing the key entities and limitations, along with a concept for modeling the human behaviour. Furthermore, we consider several metrics for assessing the impact of the community with respect to the internally consumed energy.
2018 15th International Multi-Conference on Systems, Signals & Devices (SSD), 2018
Buildings are complex, even more complex than plants because humans are not only controlling them... more Buildings are complex, even more complex than plants because humans are not only controlling them but they are part of them. Recently, complexity still increased with the numerous sensors deployed in new building environment. Diagnosis process were yet dedicated to industrial plants but these approaches are now also meaningful for building systems. The objective of this work is to analyse new challenges and to out point the problems that really affect the energy efficiency of buildings as well as make a comparative study of methods useful for diagnosis.
2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019
Smart buildings entail several challenges including how to integrate occupants and improve their ... more Smart buildings entail several challenges including how to integrate occupants and improve their consciousness towards the energy management systems. For this purpose, occupancy estimation problem has gained increasing attention in smart buildings during the past few years and many approaches have been proposed. Greater number of these approaches has been based on supervised learning techniques. Interactive learning is a new sort of supervised learning approach used to estimate the number of occupants in buildings. This paper proposes a new solution to estimate the number of occupants based on interactive learning concept using the Multi-Layer Perceptron (MLP) regressor and statistical process control charts to avoid useless interactions with users. Statistical process control charts are deployed to ensure data analysis robustness for the purpose of deciding if an interaction with the occupant is necessary or not. Experiments in an office equipped with 30 sensors have been conducted and proved the merits of the proposed framework.
This paper proposes a new general approach based on Bayesian networks to model the human behaviou... more This paper proposes a new general approach based on Bayesian networks to model the human behaviour. This approach represents human behaviour with probabilistic cause-effect relations based on knowledge, but also with conditional probabilities coming either from knowledge or deduced from observations. This approach has been applied to the co-simulation of the CO2 concentration in an office coupled with human behaviour.
A general approach is proposed to estimate the number of occupants in a zone using different kind... more A general approach is proposed to estimate the number of occupants in a zone using different kinds of measurements such as motion detection, power consumption or CO2 concentration. The proposed approach is inspired from machine learning. It starts by determining among different measurements those that are the most useful by calculating the information gains. Then, an estimation algorithm is proposed. It relies on a C4.5 learning algorithm that yields human readable decision trees using measurements to estimate the number of occupants. It has been applied to an office setting.
2019 International Conference on Control, Automation and Diagnosis (ICCAD)
Similar to other engineering systems, buildings are also complex systems. The whole building syst... more Similar to other engineering systems, buildings are also complex systems. The whole building system is highly vulnerable to different kinds of faults, failures and human misbehavior that could affect the building performance. The available fault diagnosis tools and methodologies particularly rely on rule or pure model-based approaches. It is assumed that the model could be applied to any situation without taking into account actual contexts. Setting a complete model with all contextual representations is a challenging job. On the other hand, unreliable instrumentation and non-formalization of the expert’s knowledge are also challenging ones. The objective of this paper is to highlight the challenges that lie in this problem domain, as well as to provide a strategy about how to solve them.A real case study has been investigated to exemplify the new trends in buildings: a classroom in the University of Southern Denmark.
International audienceThis paper proposes a new general approach based on Bayesian networks to mo... more International audienceThis paper proposes a new general approach based on Bayesian networks to model the human behaviour. This approach represents human behaviour withprobabilistic cause-effect relations based not only on previous works, but also with conditional probabilities coming either from expert knowledge or deduced from observations. The approach has been used in the co-simulation of building physics and human behaviour in order to assess the CO 2 concentration in an office
2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI), 2021
People spend the majority of their time in buildings which are a major contributor to overall ene... more People spend the majority of their time in buildings which are a major contributor to overall energy consumption. Both buildings and their occupants can therefore be viewed as an opportunity but also a genuine challenge to energy saving. A classical approach in this context is to build knowledge models which are based on expert knowledge to model the energy behavior of buildings. However, designing knowledge models requires a great deal of expert time, significant knowledge of building specifications and a thorough physical background. Once designed, the calibration phase based on sensor data from the building is also a long and complex task. Finally, the slightest change in the environment, such as adding a ventilation system, would necessitate a parameters adjustment or even a re-design of the model. At the same time, while the occupants are not able to analyze and apprehend the energetic behavior of their housing, due to these complexities, they can decide on their comfort and ex...
Building system is highly vulnerable to different kinds<br> of faults and human misbehavior... more Building system is highly vulnerable to different kinds<br> of faults and human misbehaviors. Energy efficiency and user comfort<br> are directly targeted due to abnormalities in building operation. The<br> available fault diagnosis tools and methodologies particularly rely on<br> rules or pure model-based approaches. It is assumed that model or<br> rule-based test could be applied to any situation without taking into<br> account actual testing contexts. Contextual tests with validity domain<br> could reduce a lot of the design of detection tests. The main objective of this paper is to consider fault validity when<br> validate the test model considering the non-modeled events such<br> as occupancy, weather conditions, door and window openings and<br> the integration of the knowledge of the expert on the state of the<br> system. The concept of heterogeneous tests is combined with test<br> validity to generate fau...
Building systems are highly vulnerable to different<br> kinds of faults and failures. In fa... more Building systems are highly vulnerable to different<br> kinds of faults and failures. In fact, various faults, failures and human<br> behaviors could affect the building performance. This paper tackles<br> the detection of unreliable sensors in buildings. Different literature<br> surveys on diagnosis techniques for sensor grids in buildings have<br> been published but all of them treat only bias and outliers. Occurences<br> of data gaps have also not been given an adequate span of attention<br> in the academia. The proposed methodology comprises the automatic thresholding<br> for data gap detection for a set of heterogeneous sensors in<br> instrumented buildings. Sensor measurements are considered to be<br> regular time series. However, in reality, sensor values are not<br> uniformly sampled. So, the issue to solve is from which delay each<br> sensor become faulty? The use of time series is required for detectio...
Faced with the challenges posed by global warming and the depletion of fossil fuels, all the acto... more Faced with the challenges posed by global warming and the depletion of fossil fuels, all the actors in society<br> (individuals, industries, communities, and governments) are led to find ways to reduce their energy consumption.<br> As residential buildings contribute significantly to the overall energy consumption, a particular effort is requested<br> from buildings occupants to reduce their energy consumption without having to compensate on their quality of<br> life. The energy behaviour of buildings is directly influenced among others by the behaviour of their occupants.<br> This paper describes a new approach based on historical data to sensitize occupants for their actions' impact on<br> building energy comportment without the need to develop a physical model thanks to case-based reasoning<br> (CBR). We report on an experimental evaluation of the approach based on real-world data collected from an<br> office building including ...
2021 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS), 2021
Collective action and citizen involvement in the energy sector represent the new platform upon wh... more Collective action and citizen involvement in the energy sector represent the new platform upon which the novel energy communities are developed. From a scientific point of view, we consider modeling and simulation represent important instruments to shape the future energy communities and assess their impact on the society. In this context, this paper presents an initial conceptual analysis upon the most important requirements and challenges that can be expected in this new research direction, where human actors must be directly involved in the technical solution. We also propose a general model of an energy community, emphasizing the key entities and limitations, along with a concept for modeling the human behaviour. Furthermore, we consider several metrics for assessing the impact of the community with respect to the internally consumed energy.
2018 15th International Multi-Conference on Systems, Signals & Devices (SSD), 2018
Buildings are complex, even more complex than plants because humans are not only controlling them... more Buildings are complex, even more complex than plants because humans are not only controlling them but they are part of them. Recently, complexity still increased with the numerous sensors deployed in new building environment. Diagnosis process were yet dedicated to industrial plants but these approaches are now also meaningful for building systems. The objective of this work is to analyse new challenges and to out point the problems that really affect the energy efficiency of buildings as well as make a comparative study of methods useful for diagnosis.
2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019
Smart buildings entail several challenges including how to integrate occupants and improve their ... more Smart buildings entail several challenges including how to integrate occupants and improve their consciousness towards the energy management systems. For this purpose, occupancy estimation problem has gained increasing attention in smart buildings during the past few years and many approaches have been proposed. Greater number of these approaches has been based on supervised learning techniques. Interactive learning is a new sort of supervised learning approach used to estimate the number of occupants in buildings. This paper proposes a new solution to estimate the number of occupants based on interactive learning concept using the Multi-Layer Perceptron (MLP) regressor and statistical process control charts to avoid useless interactions with users. Statistical process control charts are deployed to ensure data analysis robustness for the purpose of deciding if an interaction with the occupant is necessary or not. Experiments in an office equipped with 30 sensors have been conducted and proved the merits of the proposed framework.
This paper proposes a new general approach based on Bayesian networks to model the human behaviou... more This paper proposes a new general approach based on Bayesian networks to model the human behaviour. This approach represents human behaviour with probabilistic cause-effect relations based on knowledge, but also with conditional probabilities coming either from knowledge or deduced from observations. This approach has been applied to the co-simulation of the CO2 concentration in an office coupled with human behaviour.
A general approach is proposed to estimate the number of occupants in a zone using different kind... more A general approach is proposed to estimate the number of occupants in a zone using different kinds of measurements such as motion detection, power consumption or CO2 concentration. The proposed approach is inspired from machine learning. It starts by determining among different measurements those that are the most useful by calculating the information gains. Then, an estimation algorithm is proposed. It relies on a C4.5 learning algorithm that yields human readable decision trees using measurements to estimate the number of occupants. It has been applied to an office setting.
2019 International Conference on Control, Automation and Diagnosis (ICCAD)
Similar to other engineering systems, buildings are also complex systems. The whole building syst... more Similar to other engineering systems, buildings are also complex systems. The whole building system is highly vulnerable to different kinds of faults, failures and human misbehavior that could affect the building performance. The available fault diagnosis tools and methodologies particularly rely on rule or pure model-based approaches. It is assumed that the model could be applied to any situation without taking into account actual contexts. Setting a complete model with all contextual representations is a challenging job. On the other hand, unreliable instrumentation and non-formalization of the expert’s knowledge are also challenging ones. The objective of this paper is to highlight the challenges that lie in this problem domain, as well as to provide a strategy about how to solve them.A real case study has been investigated to exemplify the new trends in buildings: a classroom in the University of Southern Denmark.
International audienceThis paper proposes a new general approach based on Bayesian networks to mo... more International audienceThis paper proposes a new general approach based on Bayesian networks to model the human behaviour. This approach represents human behaviour withprobabilistic cause-effect relations based not only on previous works, but also with conditional probabilities coming either from expert knowledge or deduced from observations. The approach has been used in the co-simulation of building physics and human behaviour in order to assess the CO 2 concentration in an office
2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI), 2021
People spend the majority of their time in buildings which are a major contributor to overall ene... more People spend the majority of their time in buildings which are a major contributor to overall energy consumption. Both buildings and their occupants can therefore be viewed as an opportunity but also a genuine challenge to energy saving. A classical approach in this context is to build knowledge models which are based on expert knowledge to model the energy behavior of buildings. However, designing knowledge models requires a great deal of expert time, significant knowledge of building specifications and a thorough physical background. Once designed, the calibration phase based on sensor data from the building is also a long and complex task. Finally, the slightest change in the environment, such as adding a ventilation system, would necessitate a parameters adjustment or even a re-design of the model. At the same time, while the occupants are not able to analyze and apprehend the energetic behavior of their housing, due to these complexities, they can decide on their comfort and ex...
Building system is highly vulnerable to different kinds<br> of faults and human misbehavior... more Building system is highly vulnerable to different kinds<br> of faults and human misbehaviors. Energy efficiency and user comfort<br> are directly targeted due to abnormalities in building operation. The<br> available fault diagnosis tools and methodologies particularly rely on<br> rules or pure model-based approaches. It is assumed that model or<br> rule-based test could be applied to any situation without taking into<br> account actual testing contexts. Contextual tests with validity domain<br> could reduce a lot of the design of detection tests. The main objective of this paper is to consider fault validity when<br> validate the test model considering the non-modeled events such<br> as occupancy, weather conditions, door and window openings and<br> the integration of the knowledge of the expert on the state of the<br> system. The concept of heterogeneous tests is combined with test<br> validity to generate fau...
Building systems are highly vulnerable to different<br> kinds of faults and failures. In fa... more Building systems are highly vulnerable to different<br> kinds of faults and failures. In fact, various faults, failures and human<br> behaviors could affect the building performance. This paper tackles<br> the detection of unreliable sensors in buildings. Different literature<br> surveys on diagnosis techniques for sensor grids in buildings have<br> been published but all of them treat only bias and outliers. Occurences<br> of data gaps have also not been given an adequate span of attention<br> in the academia. The proposed methodology comprises the automatic thresholding<br> for data gap detection for a set of heterogeneous sensors in<br> instrumented buildings. Sensor measurements are considered to be<br> regular time series. However, in reality, sensor values are not<br> uniformly sampled. So, the issue to solve is from which delay each<br> sensor become faulty? The use of time series is required for detectio...
Uploads