在物联网迅速发展的当下，利用数据科学实现非侵入式的电气设备辨识对能源节约、机电控制技术发展等具有重要意义。本书详细介绍了设备辨识的智能分类方法，包括机器学习、深度学习、智能聚类、优化模型、集成学习、单标签和多标签识别模型等，并进行了大量的实验仿真对不同的设备辨识方法进行合理的评价，为数据科学技术在非侵入式设备识别中的发展提供了重要的参考。此外，本书还对传统的基于物理和模板匹配的解决方案进行了比较，并分析了智能设备辨识在工业中的巨大应用潜力，对智能设备辨识方法在工业中的应用有较高的参考价值。

1 Introduction
1.1 Overview of Ubiquitous Electric Intemet of Things (UEIOT)
1.1.1 Features of Ubiquitous Electric Intemet of Things
1.1.2 Composition of Ubiquitous Electric Intemet of Things
1.1.3 Application Prospect and Value of Ubiquitous Electric Internet of Things
1.2 Key Techniques of UEIOT
1.2.1 Smart Electric Device Recognition
1.2.2 Intemet of Things
1.2.3 Big Data Analysis
1.2.4 Cloud Platforms
1.2.5 Computational Intelligence
1.2.6 Smart Model Embedding
1.2.7 Others
1.3 Smart Device Recognition in UEIOT
1.3.1 Data Acquisition Module
1.3.2 Event Detection Module
1.3.3 Feature Extraction Module
1.3.4 Load Identification Module
1.4 Different Strategies for Smart Device Recognition
1.4.1 Clustering Strategies for Device Recognition
1.4.2 Optimizing Strategies for Device Recognition
1.4.3 Ensemble Strategies for Device Recognition
1.4.4 Deep Learning Strategies for Device Recognition
1.5 Scope of the Book
References
2 Smart Non-intrusive Device Recognition Based on Physical Methods
2.1 Introduction
2.2 Device Recognition Method Based on Decision Tree
2.2.1 Evaluation Criteria
2.2.2 Basic Definitions of Physical Features
2.2.3 Original Dataset
2.2.4 The Theoretical Basis of Decision Tree
2.3 Device Recognition Method Based on Template Matching Method
2.3.1 The Basic Content of the Template Matching Method
2.3.2 Device Recognition Based on KNN Algorithm
2.3.3 Device Recognition Based on DTW Algorithm
2.4 Device Recognition Method Based On Current Decomposition
2.4.1 Introduction of the Current Decomposition Method
2.4.2 Physical Features of Current Decomposition
2.5 Experiment Analysis
2.5.1 Common Optimization Algorithms
2.5.2 Classification Results
2.5.3 Summary
References
3 Smart Non-intrusive Device Recognition Based on Intelfigent Single-Label Classification Methods
3.1 Introduction
3.2 Device Recognition Method Based on Support Vector Machine
3.2.1 Feature Extraction
3.2.2 Steps of the Model Based on SVM
3.2.3 Performance Evaluation
3.3 Device Recognition Method Based on Extreme Learning Machine
3.3.1 Data Process and Feature Extraction
3.3.2 Steps of the Model Based on Extreme Learning Machine
3.3.3 Performance Evaluation
3.4 Device Recognition Method Based on Artificial Neural Network
3.4.1 Data Process and Feature Extraction
3.4.2 Steps of the Multi-layer Perceptron Based Model
3.4.3 Performance Evaluation
3.5 Experiment Analysis
References
4 Smart Non-intrusive Device Recognition Based on Intelligent Multi-label Classification Methods
4.1 Introduction
4.1.1 Background
4.1.2 Dataset Used in the Chapter
4.2 Device Recognition Method Based on Ranking Support Vector Machine
4.2.1 Model Framework
4.2.2 Data Labeling
4.2.3 Feature Extraction and Reconstruction
4.2.4 The Basic Theory of the Ranking Support Vector Machine
4.2.5 Multi-label Classification Evaluation Indices
4.2.6 Evaluation of Ranking SVM in Terms of Multi-label Device Recognition
4.3 Device Recognition Method Based on Multi-label K-Nearest Neighbors Algorithm
4.3.1 Model Framework
4.3.2 Data Preprocessing
4.3.3 The Basic Theory of Multi-label K-Nearest Neighbors
4.3.4 Evaluation of MLKNN in Terms of Multi-label Device Recognition
4.4 Device Recognition Method Based on Multi-label Neural Networks
4.4.1 Model Framework
4.4.2 Preprocessing of the Raw Data
4.4.3 The Basic Theory of Backpropagation Multi-labe Learning
4.4.4 Evaluation of BPMLL in Terms of Multi-label Device Recognition
4.5 Experiment Analysis
References
5 Smart Non-intrusive Device Recognition Based on Intelligent
Clustering Methods
5.1 Introducti