Foreword to the Book: Towards Energy Smart Homes
The Global Issues for Smart Buildings (SBs)
SBs: The Biggest Source of Consumption
SBs: One of the Possible Biggest Producer of Renewable Energy
SBs: The Key Issue Regarding Flexibility in Energy Demand and Consumption
But There Are No Smart Buildings Without Smart Users: The Need of a “Human-in-the-Loop” Approach
About the Desire of the Final Users for Being Involved from the Individual Level to the Collective Level
About the Individual Level
About the Emergence at Collective Level
About the Necessity to Involve the Final User from the Individual Level to the Collective Level
About the Individual Level
To the Collective Level
That Is Why It Is a Timely Book for Exploring Innovative and High-Level Solutions Putting Humans in the Loop Approach
About a Formal and Linear Description of the Content of the Book
To a Transversal Analysis of the Main Messages of the Book
With Some of the Main Scientific and Fundamental Debate Linked on SBs put on the Table by the Book
Compromise Between Data-Driven Approach and Theory-Driven Approach?
Compromise Between Complete Automatic Delegation to Numerical System, or Buildings and Dwellings Entirely (or only) Controlled by Inhabitants?
The Need to Introduce an Interdisciplinary Approach: The Book Is a Main Contribution of the Cross-Disciplinary Program Eco-SESA
Contents
Urban Modeling and Analytics in a Smart Context
1 Introduction
2 Definitions
2.1 Smart Cities and Smart Urban Contexts
2.2 Urban Energy Modeling Tools
3 Urban Energy Modeling Tools
3.1 Framework
3.2 UBEM Tools
3.2.1 CEA
3.2.2 UMI
3.2.3 CityBES
4 Case Studies
4.1 CEA
4.2 UMI
4.3 CItyBES
5 The Evolution of UBEM Systems for Smart Urban Contexts: The Cognitive UBEMs
5.1 The Use of 3D and Time
5.2 Use of Real Time Data
5.3 Integration of Machine Learning Models with UBEM
5.4 UBEM for Operational Real Time Energy Management
6 Conclusions
References
Energy Sobriety: A Behaviour Measurement Indicator for Fuel Poverty Using Aggregated Load Readings from Smart Meters
1 Introduction
2 Measuring Fuel Poverty
2.1 Expenditure-Based Indicators
2.2 Consensual-Based Indicators
2.3 Limitations
3 Smart Meters
3.1 Smart Meter Infrastructure
3.2 Smart Meter Sampling Frequencies
3.3 Load Disaggregation
3.4 Electrical Device Types
4 BMI: A Behaviour Measurement Indicator for Fuel Poverty Assessments
4.1 BMI Framework
4.2 Data Collection
4.3 Data Pre-processing
4.4 CAD NILM Machine Learning Model for Appliance Disaggregation
4.5 Measuring Behaviour
4.5.1 Vectors for Behavioural Analysis
5 Discussion
6 Conclusions
References
Standards and Technologies from Building Sector, IoT, and Open-Source Trends
1 From Home Automation to Smart Home
2 Standards and Performance Indicators from Building Sector
2.1 Standards to Deploy Energy-Efficient Technologies
2.2 The Benefits of Energy Monitoring
2.2.1 Measure and Verification (M&V), from Design to Real Performances
2.2.2 Energy Performance Contracting
2.2.3 Impact of End-User Energy Consumption Feedbacks
2.2.4 Beyond the Building, Interaction with the Power Grid
2.3 Standardized Performance Indicators
2.3.1 Disaggregation of Overall Consumption and Categorization
2.3.2 Standards on Performance Indicators
2.3.3 Smart Readiness Indicator (SRI)
3 Building Automation and Control System
3.1 Introduction to BACS
3.2 Building Management System Functions
3.2.1 Monitoring
3.2.2 Supervision
3.2.3 Energy Efficiency Continuous Improvement
3.3 Energy Management Algorithms
3.3.1 Overall Energy Performance Assessment
3.3.2 Impact of Building Automation
3.3.3 Control of Energy Systems
3.3.4 Adaptive Behavior, Predictive Automation, Control, and Maintenance
3.4 Technical Building Management
3.4.1 Architecture
3.4.2 Communication Technologies
4 Home Automation Technologies
4.1 Home Automation Market
4.1.1 Home-Specific Constraints
4.1.2 Smart Home Key Technologies and Market
4.2 Internet of Things (IoT) Technology
4.2.1 IoT Definition and Characteristics
4.2.2 Context-Aware IoT
4.2.3 IoT Interoperability
4.2.4 IoT Security
4.2.5 IoT Privacy
4.3 IoT Architecture
4.3.1 Four Layers Architecture
4.3.2 Cloud-Based Architecture
4.3.3 Typical IoT for Home Energy Monitoring
4.4 Wireless Communication Energy Consumption
4.4.1 Wireless Characteristics
4.4.2 Will There Be Only One Standard?
4.4.3 Energy Harvesting
5 Open-Source Home Automation
5.1 Open-Source Projects
5.1.1 Home Automation Software
5.1.2 Low-Cost Hardware
5.1.3 Definition and History of Open-Source Projects
5.1.4 Efficient Open-Source Projects
5.1.5 How to Protect This Model?
5.2 Review of Some Smart Home Projects
5.2.1 Smart Citizen Kit
5.2.2 Open Energy Monitor
5.2.3 A4H Smart Home
5.2.4 G2Elab Smart Home and Open-Source Tutorials
6 Conclusions
References
Formalization of the Energy Management Problem and Related Issues
1 Introduction
2 Sobriety and Flexibility Issues at Dwelling Scale
3 Illustrative Examples
4 Problem Statement of Energy Management in Smart Buildings
5 The Model Issue
5.1 Modeling from Knowledge
5.2 Learning Parameters of Knowledge Models
5.3 Learning Parameters of Regressive Models
6 Mirroring Inhabitant Service
7 Input-Output Model Based Inhabitant Services
8 Case-Based Inhabitant Services
8.1 Proposed Approach
8.2 Results for H358 Office
8.3 Validation for H358 Office
9 Conclusion
References
Dynamic Models for Energy Control of Smart Homes
1 Introduction
2 Thermal Networks
2.1 Heat Sources
2.1.1 Temperature Sources
2.1.2 Heat Flow Rate Sources
2.2 Heat Resistances (or Conductances)
2.2.1 Conduction
2.2.2 Convection
2.2.3 Long-Wave Radiation
2.2.4 Advection
2.3 Heat Capacities
3 Assembling of Thermal Networks
3.1 Defining the Problem of Circuit Assembling
3.2 Algebraic Description of the Thermal Circuits
3.3 Numbering the Thermal Circuits
3.3.1 Numbering Elementary Circuits
3.3.2 Numbering the Assembled Circuit
3.4 Assembling the Circuits
3.5 Algorithm
3.5.1 Obtaining the Global Indexes of the Assembling Matrix
3.5.2 Obtaining the Disassembling Matrix
3.5.3 Algorithm for the Disassembling Matrix
3.5.4 Assembled Circuits
3.5.5 Global Assembled Indexes
4 Transforming Thermal Circuits into State-Space Representation
4.1 Obtaining the System of Differential-Algebraic Equations
4.2 Obtaining the State-Space Representation from the Thermal Circuit
5 Conclusions
References
Machine Learning for Activity Recognition inSmart Buildings: A Survey
1 Introduction
2 Activity Recognition in Smart Buildings
2.1 Classification
2.1.1 General Classification Approaches
2.1.2 Hidden Markov Models
2.2 Regression
2.3 Clustering
2.4 Miscellaneous
3 Case Study
3.1 Estimating Occupancy with a Set of Sensors, and Possible Manual Labeling by an Expert
3.2 Resulting Occupancy Estimators
3.3 Designing Estimators from Knowledge and Adjusting from Data
3.4 Designing Estimators from Interactive Learning
3.4.1 The Principle of Interactive Learning
4 Conclusion
References
Characterization of Energy Demand and Energy Services Using Model-Based and Data-Driven Approaches
1 Introduction
2 Engineering Models
2.1 Building Energy Simulation Models
2.2 Technological Models
2.3 Time-of-Use-Surveys Models
2.4 Using Smart Meter Data to Improve Engineering Models
3 Data-Driven Models
3.1 Intrusive Load Monitoring
3.2 Non-intrusive Load Monitoring (NILM)
3.2.1 NILM for High-Sapling Rate
3.2.2 NILM for Low Sampling Rate
3.2.3 New Approaches Based on Machine Learning Algorithms
4 Case Study: Application to Residential Energy Consumption in France
4.1 Engineering Models
4.1.1 Building Energy Simulation Model
4.1.2 Technological Models
4.1.3 Time-of-Use Survey Model
4.2 Data-Driven Models
4.2.1 Intrusive Load Monitoring
4.2.2 Non-intrusive Load Monitoring
4.3 A New Approach to Develop Data-Driven Models: Interactive Learning
5 Discussion and Conclusions
References
Occupant Actions Selection Strategies Based on Pareto-Optimal Schedules and Daily Schedule for Energy Management in Buildings
1 Introduction
2 Brief Review on Decision-Making Strategies in Presence of Multiple Compromises of Interest
3 Determining the Best Schedule of Occupant Actions
3.1 Description of Experimental Testbed
3.2 Physical Knowledge Models for Simulation
3.3 Formulation of the Optimization Problem
3.4 Proposed Framework to Generate Relevant Compromises
3.5 Proposed Schedule Selection Approach
4 Discussion on Various Schedule Selection Strategies
5 Conclusion and Scope of Further Research
References
Generation of Optimal Strategies for Energy Management of Living Area Depicted by Thousands of Constraints
1 Introduction
2 Principle of Regular/Centralized Solving Approach
2.1 The Concept of Service
2.1.1 Definition of Service
2.1.2 Type of Services
2.1.3 Service Qualification
2.2 Principle of Control Mechanism
2.3 Modeling and Solving Approach
2.3.1 Modeling Services
2.3.2 Modeling Behavior of Services
Finite-State Machines (FSM)
2.3.3 Modeling the Performance of Services
End-User Services
2.3.4 Formulation of the Anticipative Problem as a Linear Problem
2.3.5 Formalizing Time Shifting
3 Model Predictive Control for Energy Management
4 Automatic Model Generation for Energy Management
4.1 Introduction
4.2 Problem Formulation
4.2.1 Concept of Model Transformation
4.3 Concept of MDE
4.4 Concept of Pivot Model
4.5 Transformation Process Principles
4.5.1 Composition Process
4.5.2 Projection Process
4.6 Application on PREDIS/MHI
4.6.1 Transformation of the PREDIS/MHI Pivot Model to the Fast Simulated Annealing Optimization Model
5 Conclusion
References
Distributed and Self-learning Approaches for Energy Management
1 Principle of Distributed Solving Approach
2 Principle of Mixed Solving Approach
2.1 The Solving Strategy in Mixed Solving Approach
2.1.1 One Step Solving
2.1.2 Solver's Role
2.1.3 Role of the Agents
2.1.4 Agent Solving Algorithm
2.2 Results of the Implementation
3 Principle of Solving Approach Using Reinforcement Learning
3.1 Reinforcement Learning
3.1.1 Markov Decision Process
3.1.2 Model-Based vs Model-Free RL
3.1.3 Q-learning
3.1.4 Deep Reinforcement Learning (DRL)
3.2 Optimizing HVAC Systems Using RL
3.2.1 Learning a Simple Thermostat Controller
3.2.2 Zone Air Flow Controller
3.2.3 Cooling Optimization of a Simulated Data Center
3.3 Discussions
References
Model Predictive Control Based on Stochastic Grey-Box Models
Acronyms
1 Introduction
2 Grey-Box Models
2.1 A Simple Linear Grey-Box Model
3 Identification of Grey-Box Models
3.1 Initial Model Structure Identification
3.2 Estimation of Model Parameters
3.3 Uncertainty of Parameter Estimates
3.4 Selection of Model Structure
3.5 Model Validation
3.6 Comparison of Models
4 Smart Building-Related Models
4.1 The Heat Pump Model
4.2 The Electrical Heater Model
4.3 Buildings with Stationary Batteries and Electrical Vehicles
5 Disturbance Modelling
5.1 Cloud Cover
5.1.1 Discrete State-Space Cloud Cover Model
5.1.2 Continuous State-Space Model Based on Stochastic Differential Equations
5.1.3 Transformation into a State-Independent Diffusion Process
5.1.4 Estimation of Parameters Embedded in the SDE
5.2 Solar Radiation
5.2.1 Modification of the Cloud Cover Data
5.2.2 Solar Radiation Components and Modelling Approach
5.2.3 Describing the Deviation and Autocorrelation
5.3 Net Radiation
5.4 Ambient Air Temperature
6 Model-Based Predictive Control
6.1 Constrained Model Predictive Control
6.1.1 Rewriting the State Equations of the Optimisation Problem
6.1.2 Rewriting the Constraints in the Optimisation Problem
6.2 Offset-Free Control Without Separate Disturbance Model
7 Predictive Control with Embedded Disturbance Models
7.1 Comparison of Advanced Disturbance Forecasts and Persistent Forecasts
8 Simulation Results
9 Hierarchical Control
9.1 Two-level Control for Utilising Energy Flexibility
9.2 Multi-Level Control and Markets
10 Summary
References
Explanations Generation with Knowledge Models
1 Problem Statement and General Solving Principles
2 Generating Explanations
2.1 Differential Explanations
2.2 Differential Explanations with Contextual Causality
2.3 Model Fragment
3 Direct Explanations
4 Validation Scenario for the Generated Explanations
4.1 Context and Goals
4.2 Method
4.3 Participants
4.4 Independent Variables
4.5 Tasks
4.6 Scenario of the Interview
4.7 Results
5 Second Validation
6 Conclusion
References
The Mondrian User Interface Pattern: Inspiring Eco-responsibility in Homes
1 Introduction
2 Domestic Environment and Design Implications
3 Ambient Displays and Aesthetics
3.1 Design for the Periphery
3.2 Aesthetic Representations
3.2.1 Abstract Representations
3.2.2 Metaphorical Representations
3.2.3 Informative Art
3.3 Combining Aesthetics with Pragmatic Representations
4 Zoomable User Interface and Focus+Context Techniques
4.1 Focus+Context Techniques
4.2 Semantic Zoom
5 The Mondrian User Interface Pattern
5.1 Overall Description
5.2 Mondrian's Abstract Compositions and Their Benefits
5.3 Mondrian's Compositions Augmented with Multilevel Interaction
5.3.1 Three Levels of Interaction: At-A-Glance, At-One-Click, At-Additional Zoom
5.3.2 Transitioning Between the Mondrian's Tiles
6 The E-coach Mondrian User Interface
6.1 Overall Structure of the e-Coach UI
6.2 Spatial Eco-information
6.2.1 At-A-Glance Spatial Eco-information
6.2.2 At-One-Click Spatial Eco-representation
6.2.3 Additional Zoom-In of Spatial Eco-information
6.3 Temporal Eco-information Combined with Utilitarian Information
6.3.1 At-A-Glance Temporal Eco-information
6.4 Social Eco-information
6.5 Human Control and Collaboration
6.5.1 At-A-Glance Human Control and Recommendations
6.5.2 At-One-Click Human Control and Recommendations
6.5.3 Additional Zoom-In of the Human Control and Recommendations
7 Conclusion
7.1 Personalization
7.2 Long-Term Study
References
Faults and Failures in Smart Buildings: A New Tool for Diagnosis
1 Introduction
2 Fault Diagnosis in Buildings: State of the Art
2.1 Faults in Buildings
2.2 Overview of General Diagnosis Methods
2.2.1 Building Fault Diagnosis Using Model-Based Techniques
2.2.2 Building Fault Diagnosis Using Rule-Based Techniques
2.2.3 Building Fault Diagnosis Using Signal-Based Techniques
3 Diagnosis in Buildings: New Challenges
3.1 Complexity
3.2 No Universal Model
3.3 Unreliable Sensors in Buildings
4 Need for New Services for Diagnosis in Buildings
4.1 Need for Testing in Specific Context Under the Hypothesis of Fault Modeling
4.2 Need for Indicators to Assess a Level of Validity of a Test and a Confidence Level for Global Diagnosis
4.3 Need to Know the Periods of Good Operation of Sensors
5 Application Example
5.1 Presentation of the Platform
5.2 Diagnosis Challenges in Danish Platform
6 Diagnostic Analysis in Danish Application
6.1 Design of Partial Valid Tests
6.2 Diagnosis Reasoning for Danish Application
6.2.1 Visual Diagnostic Analysis
6.2.2 Diagnostic Analysis by Singh et al. singh2019advancing
6.2.3 Proposed Diagnostic Analysis Najehh2019
7 Conclusion
References
Analyzing Load Profiles in Commercial Buildings Using Smart Meter Data
1 Introduction
2 Literature Survey
3 Proposed Method
3.1 Segmentation
3.2 Symbolic Representation
3.3 Time Complexity Analysis
4 Experimental Setup
4.1 Data Description
4.2 Quality Measure
5 Results and Analysis
5.1 Planted Patterns
5.2 Comparative Analysis
5.3 Most Frequently Occurring Pattern (MFOP)
5.3.1 Analysis of MFOP for Sequential Encoding Technique
5.3.2 Analysis of MFOP for Simultaneous Encoding Technique
5.4 Analysis of MFOP of Individual Buildings
5.5 Applications Toward Demand Side Management (DSM)
6 Summary and Conclusions
References
A Modern Approach to Include Representative Behaviour Models in Energy Simulations
1 Introduction
2 Inclusion of Occupants' Behaviour in Buildings Energy Management
3 Multi-Agent Based Approach for Dynamic Behaviour Model Generation and Validation
3.1 Data Collection, Pre-Processing and Analysis (Step 1)
3.1.1 Case Study of a Fridge
3.1.2 Fridge Freezer On-Cycle Durations Computation
3.1.3 Impact of Seasons, Day Type and Cooking Activity
3.1.4 How the Impact of Cooking Activity on Fridge On-Cycles Is Computed
3.1.5 Heuristic Approach to Compute Fridge Consumption During Cooking Activity
3.2 Physical Behaviour Modelling (Step 2)
3.2.1 Building Envelop Modelling
3.2.2 Appliance's Behaviour Modelling
3.3 Tune Parameters of Inhabitant's Behaviour Models (Step 3)
3.4 Clustering Houses with Similar Behaviours (Step 4)
3.4.1 Identifying Representative Behaviours
3.5 Inhabitant's Reactive, Deliberative Behaviour Modelling (Step 5)
3.6 Implementation and Co-simulation of Behaviour Models (Step 6)
3.6.1 Scenario Implementation
3.7 Co-simulate the Complex Behaviour with Physical Models (Step 6)
3.8 Tune Parameters of Inhabitant's Behaviour Models (Step 7)
3.8.1 Tune Parameters of Inhabitant's Behaviour Models (Step 7)
4 Validate the Models with Building System and BEMS
4.1 Inhabitants' Behaviour Simulation
4.2 Fanger's Thermal Comfort Model and Inhabitants' Behaviour
4.3 Co-Simulation Environment
4.4 Eco vs Non-Eco Behaviours
4.5 Eco Agent Controls the Environment Without BEMS
4.6 Non-Eco Agent Controls the Environment with BEMS
4.7 Eco vs Non-Eco Behaviours with and Without BEMS
5 Conclusions and Discussions
References
Stochastic Prediction of Residents' Activities and Related Energy Management
1 Introduction
1.1 Context
1.2 DBES Reliability
1.3 Various Occupants' Behaviour Modelling
1.3.1 Agent-Based Approach
1.3.2 Stochastic Approach
1.4 Chapter Outline/Objectives
2 Modelling of Presence, Activities and Related Electrical Equipment
2.1 Introduction
2.2 Presence Modelling
2.2.1 Transition Probabilities
2.2.2 Presence Duration
2.2.3 First-Order Model Selection
2.2.4 Presence Model Results
2.3 Residential Activities Modelling
2.3.1 Activities' Probabilities
2.3.2 Activities' Duration
2.3.3 Evaluation of the Models and Selection
2.4 Presence and Activities Simulation Results
2.4.1 Algorithm
2.4.2 Simulation Results
2.5 Creation of a Household
2.5.1 Introduction
2.5.2 Household Model Description
2.5.3 Number of Household Members
2.5.4 Ownership Status
2.5.5 Type of Household and Reference Household Member's Characteristics
2.5.6 Remaining Household Member's Characteristics
2.6 Activities' Location
2.6.1 Association of Occupants and Zones
2.6.2 Association of Occupants' Activities and Zones
2.7 Discussion on Presence and Activities Modelling
3 Electrical Equipment Modelling
3.1 General Principles
3.2 Simulation Results
3.2.1 Single Dwelling Electricity Load
3.2.2 Aggregated Electricity Load
3.2.3 Internal Heat Input
4 Adaptive Behaviour
4.1 Windows Opening
4.2 Temperature Setpoint
4.2.1 State of the Art
4.2.2 Temperature Setpoint Model Principles
4.2.3 Temperature Setpoint Data
4.2.4 Thermal Zones Temperature Setpoint
4.2.5 Temperature Setpoint Management
4.2.6 Temperature Setpoint Results
5 Application
5.1 Implementation
5.2 Case Study
6 Conclusion
References
Index