LEADER 10863nam 2200541 450 001 9910767537603321 005 20220616124902.0 010 $a981-16-2778-9 035 $a(CKB)4100000012025944 035 $a(MiAaPQ)EBC6725064 035 $a(Au-PeEL)EBL6725064 035 $a(OCoLC)1268205603 035 $a(PPN)258055227 035 $a(EXLCZ)994100000012025944 100 $a20220616d2021 uy 0 101 0 $aeng 135 $aurcnu|||||||| 181 $ctxt$2rdacontent 182 $cc$2rdamedia 183 $acr$2rdacarrier 200 00$aData-driven analytics for sustainable buildings and cities $efrom theory to application /$fXingxing Zhang, editor 210 1$aSingapore :$cSpringer,$d[2021] 210 4$dİ2021 215 $a1 online resource (446 pages) 225 1 $aSustainable Development Goals Series 311 $a981-16-2777-0 327 $aIntro -- Preface -- Contents -- 1 The Evolving of Data-Driven Analytics for Buildings and Cities Towards Sustainability -- Abstract -- 1.1 Introduction -- 1.1.1 Background -- 1.1.2 Data-Driven Analytics for Sustainability Goals -- 1.2 Aim and Objectives -- 1.3 Motivations and Novelties -- 1.4 Structure and Contents -- References -- Energy in Buildings -- 2 Data-Driven Approaches for Prediction and Classification of Building Energy Consumption -- Abstract -- 2.1 Introduction -- 2.1.1 The Need for Energy Consumption Analysis -- 2.1.2 Advantage and Motivation -- 2.1.3 Usage of Building Energy and Performance Data -- 2.1.4 Proposed Methodologies for Building Energy Consumption -- 2.1.5 Data-Driven Approaches -- 2.1.6 Data-Driven Prediction Models -- 2.1.6.1 Artificial Neural Networks -- 2.1.6.2 Support Vector Machine -- 2.1.6.3 Statistical Regression -- 2.1.6.4 Decision Tree -- 2.1.6.5 Genetic Algorithms -- 2.1.7 Data-Driven Classification Approaches -- 2.1.7.1 K-Means Cluster -- 2.1.7.2 Self-organizing Map -- 2.1.7.3 Hierarchical Clustering -- 2.2 Practical Application of Data-Driven Approaches -- 2.2.1 R & -- D Works and Practical Applications -- 2.2.1.1 Prediction -- Prediction Application of ANNs -- Prediction Application of SVM -- Prediction Application of Statistical Regression -- Prediction Application of Decision Tree -- Prediction Application of Genetic Algorithms -- 2.2.1.2 Profile -- Profile Application of Cluster Method -- Profile Application of Regression -- 2.2.1.3 Energy Mapping -- 2.2.1.4 Benchmarking of Buildings -- 2.2.1.5 Retrofit of Buildings -- 2.2.2 Analyses of the Review Works -- 2.3 Opportunities for Further Works -- 2.4 Conclusion -- References -- 3 Prediction of Occupancy Level and Energy Consumption in Office Building Using Blind System Identification and Neural Networks -- Abstract -- 3.1 Introduction. 327 $a3.1.1 State of the Art -- 3.1.2 Statements of Contribution -- 3.2 Occupancy Estimation Methodologies -- 3.2.1 Frequentist Maximum Likelihood (ML) Approach -- 3.2.2 Bayesian Estimation Approach -- 3.2.3 Evaluation Criterion -- 3.3 Energy-Consumption Prediction Methodologies -- 3.3.1 Architecture of FFNN Model -- 3.3.2 Architecture of ELM Model -- 3.3.3 Architecture of Ensemble Model -- 3.3.4 Evaluation Criteria -- 3.4 Reference Room and Dataset -- 3.5 Parameter Selection Analysis -- 3.5.1 Principal Component Analysis -- 3.5.2 Effect of Structure Parameters of FFNN Model -- 3.6 Prediction Results -- 3.6.1 Occupancy-Estimation Results -- 3.6.2 Energy-Prediction Result with True Occupant Counts -- 3.6.3 Energy Prediction Result with Estimated Occupant Counts -- 3.6.4 Limitations -- 3.7 Conclusion -- References -- 4 Cluster Analysis for Occupant-Behaviour Based Electricity Load Patterns in Buildings: A Case Study in Shanghai Residences -- Abstract -- 4.1 Introduction -- 4.2 Cluster Concept -- 4.3 Case Study of Occupant Behaviour Patterns in Residential Buildings -- 4.3.1 Data Source -- 4.3.2 Methodology -- 4.3.3 Results and Analysis -- 4.3.3.1 Analysis of Daily Consumption -- 4.3.3.2 Analysis of Seasonal Consumption -- 4.3.3.3 Analysis of Weekly Consumption -- 4.4 Further Work -- 4.5 Conclusion -- References -- 5 A Data-Driven Model Predictive Control for Lighting System Based on Historical Occupancy in an Office Building: Methodology Development -- Abstract -- 5.1 Introduction -- 5.2 System Description and Research Methodology -- 5.2.1 Description of the Proposed Control System -- 5.2.2 Overall Methodology -- 5.2.2.1 Data Composition -- 5.2.2.2 Overall Methodology Elaboration -- 5.2.3 Data Acquisition and Processing -- 5.2.3.1 PIR Sensor Calibration -- 5.2.3.2 Noise Removal -- 5.2.3.3 Occupant Feedback -- 5.2.4 Model Predictive Control Method. 327 $a5.2.4.1 Multi-layer Perceptron Algorithm -- L-BFGS Method -- SGD Method -- ADAM Method -- 5.2.4.2 Temporal Sequential-Based ANN Method -- 5.2.5 Evaluation of Predictive Control -- 5.3 Case Study of an Office Building -- 5.3.1 Historical Occupancy Data -- 5.3.2 Control Method -- 5.3.3 Control Performance Comparison -- 5.4 Discussion and Further Work -- 5.5 Summary -- References -- 6 Tailoring Future Climate Data for Building Energy Simulation -- Abstract -- 6.1 Introduction -- 6.2 Methodology -- 6.2.1 Overview -- 6.2.2 Generation of Future Weather Using the Morphing Method -- 6.2.2.1 Typical Meteorological year (TMY) -- 6.2.2.2 Global Climate Model (GCM) Selection -- 6.2.2.3 Morphing Method -- 6.2.3 Investigation of Climate Change Impacts on NZEB Lifecycle Performance -- 6.2.3.1 NZEB System Sizing Using TMY Data -- 6.2.3.2 Performance Indicators -- Energy-Balance Indicator -- Thermal-Comfort Indicator -- Grid-Interaction Indicator -- 6.2.4 Evaluation of Different Mitigation Measures' Effectiveness on NZEB Performance -- 6.3 Study Platform -- 6.3.1 Climate Regions and Selected Cities -- 6.3.2 Building Model -- 6.3.3 Air-Conditioning System Model -- 6.3.4 Renewable Energy System Model -- 6.4 Results and Discussions -- 6.4.1 Future Weather Analysis -- 6.4.2 Climate Change Impacts on NZEB Lifecycle Performance -- 6.4.2.1 Climate Change Impacts on Energy Balance -- 6.4.2.2 Climate Change Impacts on Thermal Comfort -- 6.4.2.3 Climate Change Impacts on Grid Interaction -- 6.4.3 Different Mitigation Measures' Effectiveness on NZEB Performance -- 6.4.3.1 Effectiveness of Mitigation Measures on Energy Balance -- 6.4.3.2 Effectiveness of Mitigation Measures on Thermal Comfort -- 6.4.3.3 Effectiveness of Mitigation Measures on Grid Interaction -- 6.5 Conclusions -- References. 327 $a7 A Solar Photovoltaic/Thermal (PV/T) Concentrator for Building Application in Sweden Using Monte Carlo Method -- Abstract -- 7.1 Market Analysis of Swedish PV/T Industry -- 7.2 Development of Techno-Economic Model -- 7.2.1 Reference PV/T Concentrator -- 7.2.2 Energy Generation Model -- 7.2.3 Economical Evaluation Metrics -- 7.2.4 Key Input Parameters -- 7.2.5 Simulation Process -- 7.3 Sensitivity and Reliability Analysis -- 7.4 Optimization Analysis -- 7.4.1 Impact of Average Daily Solar Irradiance -- 7.4.2 Impact of Debt to Equity Ratio -- 7.4.3 Impact of Heating Price for Household -- 7.4.4 Impact of Concentrator Capital Price -- 7.4.5 Impact of Discount Rate -- 7.5 Future Work to Improve the Model -- 7.6 Summary -- References -- Thermal Comfort and Air Quality in Buildings -- 8 Influencing Factors for Occupants' Window-Opening Behaviour in an Office Building Through Logistic Regression and Pearson Correlation Approaches -- Abstract -- 8.1 Method of Study -- 8.1.1 Building Description -- 8.1.2 Measured Factors -- 8.1.3 Measuring Devices -- 8.2 Data Processing and Results Analysis -- 8.2.1 Environmental Factors -- 8.2.2 Outdoor Air Temperature -- 8.2.3 Indoor Air Temperature -- 8.2.4 Outdoor PM2.5 Concentrations -- 8.2.5 Correlation Analysis for All Factors -- 8.3 Non-environmental Factors -- 8.3.1 Seasonal Change -- 8.3.2 Time of Day -- 8.3.3 Personal Preference -- 8.4 Conclusions and Discussion -- References -- 9 Reinforcement Learning Methodologies for Controlling Occupant Comfort in Buildings -- Abstract -- 9.1 Overview of Comfort Control in Buildings -- 9.1.1 Background -- 9.1.2 Necessity of New Methods for the Building Control System (BCS) -- 9.1.3 Review Methods -- 9.2 Summary of Relevant Review Works and Their Indications -- 9.3 The Reinforcement Learning Method -- 9.3.1 Elements of Reinforcement Learning and MDPs. 327 $a9.3.2 Policies and Functions -- 9.3.3 Bellman Optimality Equation -- 9.3.4 Categorisation of RL Algorithms -- 9.3.5 Multi-agent Systems -- 9.4 Applications of Reinforcement Learning Methods for Comfort Control in Buildings -- 9.4.1 Comfort Factors -- 9.4.2 Thermal Comfort -- 9.4.2.1 IAQ -- 9.4.2.2 Lighting -- 9.4.2.3 Combinations of Factors -- 9.4.2.4 Algorithm Class -- 9.4.2.5 Value-Based -- 9.4.2.6 Actor-Critic -- 9.4.2.7 Exploration Versus Exploitation Strategies -- 9.4.3 Agent Perspectives -- 9.4.4 Physical Implementations -- 9.5 Discussions -- 9.6 Chapter Summary -- References -- 10 A Novel Reinforcement Learning Method for Improving Occupant Comfort via Window Opening and Closing -- Abstract -- 10.1 Introduction -- 10.2 Behaviours of Window Opening and Closing -- 10.2.1 Drivers of Window Opening and Closing -- 10.2.2 Occupant Comfort and Intelligent Controllers -- 10.2.3 Building Environment -- 10.3 RL and Algorithms -- 10.3.1 Markov Decision Processes -- 10.3.2 Policies and Value Functions -- 10.3.3 Q-learning and SARSA -- 10.4 Data and Methods -- 10.4.1 Data -- 10.4.2 Methods -- 10.5 Results -- 10.5.1 RNN Predictions -- 10.5.2 Performance of RL Agents -- 10.6 Conclusions -- References -- 11 Development of an Adaptation Table to Enhance the Accuracy of the Predicted Mean Vote Model -- Abstract -- 11.1 Introduction -- 11.2 Factors Contributing to the Discrepancies Between PMV and Thermal Sensation -- 11.2.1 Season -- 11.2.2 Climate -- 11.2.3 Building Type -- 11.2.4 Age Group -- 11.2.5 Gender -- 11.3 Methodology -- 11.3.1 Data Source and Sampling Technique -- 11.3.2 Analytical Methods for Categorical Variables -- 11.3.3 Analytical Methods to Quantify the Relationships -- 11.3.4 Workflow of the Methodology -- 11.4 Results -- 11.4.1 Accuracy of the PMV Model for TS Prediction -- 11.4.2 The Effect of Variables on the Discrepancy -- 11.4.2.1 Season. 327 $a11.4.2.2 Climate. 410 0$aSustainable development goals series. 606 $aSustainable buildings$xData processing 606 $aSustainable buildings$xStatistical methods 606 $aSmart cities 615 0$aSustainable buildings$xData processing. 615 0$aSustainable buildings$xStatistical methods. 615 0$aSmart cities. 676 $a720.47 702 $aZhang$b Xingxing 801 0$bMiAaPQ 801 1$bMiAaPQ 801 2$bMiAaPQ 906 $aBOOK 912 $a9910767537603321 996 $aData-driven Analytics for Sustainable Buildings and Cities$92587432 997 $aUNINA