·ÇÇÖÈëʽ¸ººÉʶ±ð:ÀíÂÛ¡¢¼¼ÊõÓëÓ¦ÓÃ:theory, technologies and applications

1 Introduction 1.1 Overview of the Non-intrusive Load Monitoring 1.1.1 The Non-intrusive Load Monitoring 1.1.2 Overview of Recent Research in Non-intrusive Load Monitoring 1.2 Fundamental Key Problems of Non-intrusive Load Monitoring 1.2.1 Event Detection in Non-intrusive Load Monitoring 1.2.2 Feature Extraction in Non-intrusive Load Monitoring 1.2.3 Load Identification in Non-intrusive Load Monitoring 1.2.4 Energy Forecasting in Smart Buildings 1.3 Scope of This Book References 2 Detection of Transient Events in Time Series 2.1 Introduction 2.2 Cumulative Sum Based Transient Event Detection Algorithm 2.2.1 Mathematical Description of Change Point Detection 2.2.2 Parametric CUSUM Algorithm 2.2.3 Non-parametric CUSUM Algorithm 2.2.4 Sliding Windows Based on Two-Sided CUSUM Algorithm 2.2.5 Original Dataset 2.2.6 Evaluation Criteria and Results Analysis 2.3 Generalized Likelihood Ratio 2.3.1 The Theoretical Basis of GLR 2.3.2 Comparison of Event Detection Results 2.4 Sequential Probability Ratio Test 2.4.1 The Theoretical Basis of SPRT 2.4.2 Comparison of Event Detection Results 2.5 Experiment Analysis 2.5.1 The Results of Three Kinds of Algorithms. 2.5.2 Conclusion References 3 Appliance Signature Extraction 3.1 Introduction 3.1.1 Background 3.1.2 Feature Evaluation Indices 3.1.3 Classification Evaluation Indices 3.1.4 Data Selection 3.2 Features Based on Conventional Physical Definition 3.2.1 The Theoretical Basis of Physical Definition Features 3.2.2 Feature Extraction 3.2.3 Feature Evaluation 3.2.4 Classification Results 3.3 Features Based on Time-Frequency Analysis 3.3.1 The Theoretical Basis of Harmonic Features 3.3.2 Feature Extraction 3.3.3 Feature Evaluation 3.3.4 Classification Results 3.4 Features Based on VI Image 3.4.1 The Theoretical Basis of VI Image Features 3.4.2 Feature Extraction 3.4.3 Feature Evaluation 3.4.4 Classification Results 3.5 Features Based on Adaptive Methods 3.5.1 The Theoretical Basis of Adaptive Features 3.5.2 Feature Extraction 3.5.3 Classification Results 3.6 Experimental Analysis 3.6.1 Comparative Analysis of Classification Performance 3.6.2 Conclusion References 4 Appliance Identification Based on Template Matching 4.1 Introduction 4.1.1 Background 4.1.2 Data Preprocessing of the PLAID Dataset 4.2 Appliance Identification Based on Decision Tree 4.2.1 The Theoretical Basis of Decision Tree 4.2.2 Steps of Modeling 4.2.3 Classification Results 4.3 Appliance Identification Based on KNN Algorithm 4.3.1 The Theoretical Basis of KNN 4.3.2 Steps of Modeling 4.3.3 Classification Results 4.4 Appliance Identification Based on DTW Algorithm 4.4.1 The Theoretical Basis of DTW 4.4.2 Steps of Modeling 4.4.3 Classification Results 4.5 Experiment Analysis 4.5.1 Model Framework 4.5.2 Comparative Analysis of Classification Performance 4.5.3 Conclusion References 5 Steady-State Current Decomposition Based Appliance Identification 5.1 Introduction 5.2 Classical Steady-State Current Decomposition Models 5.2.1 Model Framework 5.2.2 Classical Features of Steady-State Decomposition and the Feature Extraction Method 5.2.3 Classical Methods in Steady-State Current Decomposition 5.2.4 Performance of the Various Features and Models 5.3 Current Decomposition Models Based on Harmonic Phasor 5.3.1 Model Framework 5.3.2 Novel Features of Steady-State Current Decomposition 5.3.3 Multi-objective Optimization Methods in Steady-State Current Decomposition 5.3.4 Performance of the Novel Features and Multi-objective Optimization Models 5.4 Current Decomposition Models Based on Non-negative Matrix Factor 5.4.1 Model Framework 5.4.2 Reconstruction of the Data 5.4.3 Non-negative Matrix Factorization Method of the Current Decomposition 5.4.4 Evaluation of the NMF Method in Current Decomposition 5.5 Experiment Analysis 5.5.1 Data Generation 5.5.2 Comparison Analysis of the Features Used in the Steady-State Decomposition 5.5.3 Comparison Analysis of the Models Used in the Steady-State Decomposition 5.5.4 Conclusion References 6 Machine Learning Based Appliance Identification 6.1 Introduction 6.2 Appliance Identification Based on Extreme Learning Machine 6.2.1 The Theoretical Basis of ELM 6.2.2 Steps of Modeling 6.2.3 Classification Results 6.3 Appliance Identification Based on Support Vector Machine 6.3.1 The Theoretical Basis of SVM 6.3.2 Steps of Modeling 6.3.3 Classification Results 6.4 Appliance Identification Based on Random Forest 6.4.1 The Theoretical Basis of Random Forest 6.4.2 Steps of Modeling 6.4.3 Classification Results 6.5 Experiment Analysis 6.5.1 Model Framework 6.5.2 Feature Preprocessing for Non-intrusive Load Monitoring 6.5.3 Classifier Model Optimization Algorithm for Non-intrusive Load Monitoring 6.6 Conclusion References 7 Hidden Markov Models Based Appliance 7.1 Introduction 7.2 Appliance Identification Based on Hidden Markov Models 7.2.1 Basic Problems Solved by HMM 7.2.2 Data Preprocessing 7.2.3 Determination of Load Status Information 7.3 Appliance Identification Based on Factorial Hidden Markov Models 7.3.1 The Theoretical Basis of the FHMM 7.3.2 Load Decomposition Steps Based on FHMM 7.3.3 Load Power Estimation 7.3.4 Decomposition Experiment Based on FHMM 7.3.5 Evaluation Criteria and Result Analysis 7.4 Appliance Identification Based on Hidden Semi-Markov Models 7.4.1 Hidden Semi-Markov Model 7.4.2 Improved Viterbi Algorithm 7.4.3 Evaluation Criteria and Result Analysis 7.5 Experiment Analysis References 8 Deep Learning Based Appliance Identification 8.1 Introduction 8.1.1 Deep Learning 8.1.2 NILM Based on Deep Learning 8.2 Appliance Identification Based on End-to-End Decomposition 8.2.1 Single Feature Based LSTM Network Load Decomposition 8.2.2 Multiple Features Based LSTM Network Load Decomposition 8.3 Appliance Identification Based on Appliance Classification 8.3.1 Appliance Identification Based on CNN 8.3.2 Appliance Identification Based on AlexNet 8.3.3 Appliance Identification Based on LeNet-SVM Model 8.4 Experiment Analysis 8.4.1 Experimental Analysis of End-to-End Decomposition 8.4.2 Experimental Analysis of Appliance Classification References 9 Deterministic Prediction of Electric Load Time Series 9.1 Introduction 9.1.1 Background 9.1.2 Advance Prediction Strategies 9.1.3 Original Electric Load Time Series 9.2 Load Forecasting Based on ARIMA Model 9.2.1 Model Framework 9.2.2 Theoretical Basis of ARIMA 9.2.3 Modeling Steps of ARIMA Predictive Model 9.2.4 Predictive Results 9.2.5 The Theoretical Basis of EMD 9.2.6 Optimization of EMD Decomposition Layers 9.2.7 Predictive Results 9.3 Load Forecasting Based on Elman Neural Network 9.3.1 Model Framework 9.3.2 Steps of Modeling 9.3.3 Predictive Results 9.3.4 Optimization of EMD Decomposition Layers 9.3.5 Predictive Results 9.4 Experiment Analysis 9.4.1 Comparative Analysis of Predictive Performance 9.5 Conclusion References 10 Interval Prediction of Electric Load Time Series 10.1 Introduction 10.2 Interval Prediction Based on Quantile Regression 10.2.1 The Performance Evaluation Metrics 10.2.2 Original Sequence for Modeling 10.2.3 The Theoretical Basis of Quantile Regression 10.2.4 Quantile Regression Based on the Total Electric Load Time Series 10.2.5 Quantile Regression Based on Additional Time and Date Information 10.2.6 Quantile Regression Based on Information from Sub-meters 10.3 Interval Prediction Based on Gaussian Process Filtering 10.3.1 The Theoretical Basis of Gaussian Process Regression 10.3.2 Gaussian Process Regression Based on the Total Electric Load Time Series 10.3.3 Gaussian Process Regression Based on Different Input Features 10.3.4 Gaussian Process Regression Based on Feature Selection 10.4 Experiment Analysis References