Intelligent Data-Analytics for Condition Monitoring

Chapter 1: Advances in Machine Learning and Data Analytics 1.1 Introduction 1.2 Data and it relation 1.3 Data pre-processing 1.4 Data visualization and correlation representation1.5 Application area 1.5.1 Clustering 1.5.2 Regression 1.5.3 Classification 1.5.4 forecasting 1.6 Softwares and techniques used for data analytics 1.7 Sources of datasets for data analytics 1.8 Conclusion References PART-A: Intelligent Data Analytics for Classification in Smart Grid Chapter 2: Intelligent Data Analytics for PV Fault diagnosis Using Deep Convolutional Neural Network (ConvNet/CNN)2.1 Introduction2.2 PV Image Data set collection2.2 Values of data2.3 Experimental design, materials, and methods2.3 Proposed Approach2.4 Deep Convolutional Neural Network (ConvNet/CNN)2.5 Results and Discussion2.6 ConclusionReferences Chapter 3: Intelligent Data Analytics for Power Transformer Health Monitoring Using Modified Fuzzy Q Learning (MFQL)3.1 Introduction3.2 Conventional Techniques Used For DGA Interpretation3.3 Dataset Collection3.3.1 Dataset : Credible Literature3.3.2 Practical DGA Dataset3.3.3 Accuracy Analysis of DGA Performance3.4 MFQL Framework3.4.1 Q-Learning (QL)3.4.2 Fuzzy Q-Learning (FQL)3.4.3 Modified FQL3.5 Input Variable Selection using J48 Algorithm3.6 Fault Classification Using MFQL3.6.1 DGA Training & Testing DATA3.6.2 MFQL Based Fault Classification Model Formation3.7 Results and Discussion3.8 ConclusionReferences Chapter 4: Intelligent Data Analytics for Induction Motor Using Gene Expression Programming (GEP)4.1 Introduction4.2 GEP Methodology and Data Sources4.2.1 Database Used for Study4.2.2 Gene Expression Programming (GEP)4.3 External Fault Classifier based on GEP4.3.1 Data Set: Training and Testing4.3.2 The GEP Approach4.3.3 GEP fault classification model4.4 Results and Discussion4.5 ConclusionReferences Chapter 5: Intelligent Data Analytics for Power Quality Disturbance Analysis Using Multi-Class ELM5.1 Introduction5.2 Model Description5.3 Proposed Framework5.4 Feature Extraction5.5 Most Relevant Input Variable Selection5.6 Multi-Class ELM Framework5.7 Results and Discussion5.8 ConclusionReferences Chapter 6: Intelligent Data Analytics for Transmission Line Fault Diagnosis Using EEMD Based Multiclass SVM and PSVM6.1 Introduction6.2 Methodology6.2.1 Proposed Approach6.2.2 Model Formulation6.2.3 Feature Extraction Using EEMD6.2.4 Support Vector Machine (SVM)6.2.5 Proximal Support Vector Machine (PSVM)6.2.6 SVM and PSVM Based Transmission Line Fault Classification Model Formation6.3 Results and Discussions6.3.1 SVM Based Transmission Line Fault Classification6.3.2 PSVM Based Transmission Line Fault Classification6.3.3 Comparative Results Analysis of SVM and PSVM Based Fault Classification Models6.4 ConclusionReferences PART-B: Intelligent Data Analytics for Forecasting in Smart Grid Chapter 7: Intelligent Data Analytics for Global Solar Radiation Forecasting for Solar Power Production Using Deep Learning Neural Network (DLNN)7.1 Introduction7.2 Related Work7.3 Solar Irradiance Forecasting Methods7.3.1 Conventional Methods7.3.2 AI and Machine Learning Based Methods7.4 Dataset Used for Study7.4.1 Dataset7.4.2 Data Pre-processing7.4.3 Data Analysis7.5 The Structure of Proposed Model7.5.1 Deep Learning Neural Network7.5.2 Performance Evaluation Measures7.6 Results and Discussion7.7 Model Validation7.8 Conclusion References Chapter 8: Intelligent Data Analytics for Wind Speed Forecasting for Wind Power Production Using Long Short-Term memory (LSTM) Network8.1 Introduction8.1.1 Review of Related Works and Motivation8.1.2 Objective and Key Contributions8.2 Proposed Framework Formation8.2.1 Proposed Approach Formation8.2.2 Dataset Collection8.2.3 Dataset Pre-processing8.2.4 Feature Extraction8.2.5 Feature Selection8.2.6 Design of LSTM Network 8.2.7 Performance Measure Indices 8.3 Case Study: Experiments and Discussion8.3.1 The Description of Experimental Dataset8.3.2 Results and Comparisons8.3.3 Comparative Experiments8.4 Conclusion and Future ScopeReferences Chapter 9: Intelligent Data Analytics for Time-Series Load Forecasting Using Fuzzy Reinforcement Learning (FRL)9.1 Introduction9.2 Methodology9.2.1 Proposed Approach9.2.2 Brief Detail of FRL Approach9.2.3 Data Collection 9.3 Time-Series Load Forecasting Model9.3.1 Data Pre-processing Using Different ML Approaches9.3.2 Conventional Model9.3.3 AI and ML Based Model9.3.4 Hybrid Model9.4 Case Studies: Performance Evaluation9.4.1 Minute-Ahead Forecasting9.4.2 Hour-Ahead Forecasting9.4.3 Day-Ahead Forecasting9.4.4 Month-Ahead Forecasting9.5 Conclusion and Future WorkReferences Chapter 10: Intelligent Data Analytics for Battery Charging/Discharging Forecasting Using Semi-supervised and Unsupervised Extreme Learning Machines10.1 Introduction10.2 Methodology10.2.1 Formation of Proposed Approach10.2.2 Health Indication (HI) Extraction10.2.3 Box-Cox Transformation (BCT)10.3.1 BC Transformation10.3.2 BCT Parameter Evaluation Using ML Method10.2.4 Correlation Analysis Using PCA and SRCA Methods Pearson Correlation Analysis (PCA) Spearman Rank Correlation Analysis (SRCA)10.2.5 RUL Estimation Approach10.3 Verification of LIB HI Evaluation10.3.1 LIB Dataset Used for Study10.3.1.1 Charging Condition10.3.1.2 Discharging Condition10.3.1.3 Impedance Measurement Condition10.3.2 Correlation Analysis and Evaluation10.3.2.1 Qualitative Analysis10.3.2.2 Quantitative Analysis10.3.3 HI Performance Evaluation10.4 RUL Estimation Validation10.5 ConclusionReferences

Chapter 1:　Advances in Machine Learning and Data Analytics　1.1　Introduction　1.2　Data and it relation　1.3　Data pre-processing　1.4　Data visualization and correlation representation1.5　Application area　1.5.1　Clustering　1.5.2　Regression　1.5.3　Classification　1.5.4　forecasting　1.6　Softwares and techniques used for data analytics　1.7　Sources of datasets for data analytics　1.8　Conclusion　References　　　　PART-A: Intelligent Data Analytics for Classification in Smart Grid　　　Chapter 2:　Intelligent Data Analytics for PV Fault diagnosis Using Deep Convolutional Neural Network (ConvNet/CNN)2.1　Introduction2.2　PV Image Data set collection2.2　Values of data2.3　Experimental design, materials, and methods2.3　Proposed Approach2.4　Deep Convolutional Neural Network (ConvNet/CNN)2.5　Results and Discussion2.6　ConclusionReferences　　　Chapter 3:　Intelligent Data Analytics for Power Transformer Health Monitoring Using Modified Fuzzy Q Learning (MFQL)3.1　Introduction3.2　Conventional Techniques Used For DGA Interpretation3.3　Dataset　 Collection3.3.1　Dataset : Credible Literature3.3.2　Practical DGA Dataset3.3.3　Accuracy Analysis of DGA Performance3.4　MFQL Framework3.4.1　Q-Learning (QL)3.4.2　Fuzzy Q-Learning (FQL)3.4.3　Modified FQL3.5　Input Variable Selection using J48 Algorithm3.6　Fault Classification Using MFQL3.6.1　DGA Training & Testing DATA3.6.2　MFQL Based Fault Classification Model Formation3.7　Results and Discussion3.8　ConclusionReferences　　　Chapter 4:　Intelligent Data Analytics for Induction Motor Using Gene Expression Programming (GEP)4.1　Introduction4.2　GEP Methodology and Data Sources4.2.1　Database Used for Study4.2.2　Gene Expression Programming (GEP)4.3　External Fault Classifier based on GEP4.3.1　Data Set: Training and Testing4.3.2　The GEP Approach4.3.3　GEP fault classification model4.4　Results and Discussion4.5　ConclusionReferences　　　Chapter 5:　Intelligent Data Analytics for Power Quality Disturbance Analysis Using Multi-Class ELM5.1　Introduction5.2　Model Description5.3　Proposed Framework5.4　Feature Extraction5.5　Most Relevant Input Variable Selection5.6　Multi-Class ELM Framework5.7　Results and Discussion5.8　ConclusionReferences　　　Chapter 6:　Intelligent Data Analytics for Transmission Line Fault Diagnosis Using EEMD Based Multiclass SVM and PSVM6.1　Introduction6.2　Methodology6.2.1　Proposed Approach6.2.2　Model Formulation6.2.3　Feature Extraction Using EEMD6.2.4　Support Vector Machine (SVM)6.2.5　Proximal Support Vector Machine (PSVM)6.2.6　SVM and PSVM Based Transmission Line Fault Classification Model Formation6.3　Results and Discussions6.3.1　SVM Based Transmission Line Fault Classification6.3.2　PSVM Based Transmission Line Fault Classification6.3.3　Comparative Results Analysis of SVM and PSVM Based Fault Classification Models6.4　ConclusionReferences　　　PART-B: Intelligent Data Analytics for Forecasting in Smart Grid　　　Chapter 7:　Intelligent Data Analytics for Global Solar Radiation Forecasting for Solar Power Production Using Deep Learning Neural Network (DLNN)7.1　Introduction7.2　Related Work7.3　Solar Irradiance Forecasting Methods7.3.1　Conventional Methods7.3.2　AI and Machine Learning Based Methods7.4　Dataset Used for Study7.4.1　Dataset7.4.2　Data Pre-processing7.4.3　Data Analysis7.5　The Structure of Proposed Model7.5.1　Deep Learning Neural Network7.5.2　Performance Evaluation Measures7.6　Results and Discussion7.7　Model Validation7.8　Conclusion　References　　　Chapter 8:　Intelligent Data Analytics for Wind Speed Forecasting for Wind Power Production Using Long Short-Term memory (LSTM) Network8.1　Introduction8.1.1　Review of Related Works and Motivation8.1.2　Objective and Key Contributions8.2　Proposed Framework Formation8.2.1　Proposed Approach Formation8.2.2　Dataset Collection8.2.3　Dataset Pre-processing8.2.4　Feature Extraction8.2.5　Feature Selection8.2.6　Design of LSTM Network 8.2.7　Performance Measure Indices 8.3　Case Study: Experiments and Discussion8.3.1　The Description of Experimental Dataset8.3.2　Results and Comparisons8.3.3　Comparative Experiments8.4　Conclusion and Future ScopeReferences　　　Chapter 9:　Intelligent Data Analytics for Time-Series Load Forecasting Using Fuzzy Reinforcement Learning (FRL)9.1　Introduction9.2　Methodology9.2.1　Proposed Approach9.2.2　Brief Detail of FRL Approach9.2.3　Data Collection 9.3　Time-Series Load Forecasting Model9.3.1　Data Pre-processing Using Different ML Approaches9.3.2　Conventional Model9.3.3　AI and ML Based Model9.3.4　Hybrid Model9.4　Case Studies: Performance Evaluation9.4.1　Minute-Ahead Forecasting9.4.2　Hour-Ahead Forecasting9.4.3　Day-Ahead Forecasting9.4.4　Month-Ahead Forecasting9.5　Conclusion and Future WorkReferences　　　Chapter 10:　Intelligent Data Analytics for Battery Charging/Discharging Forecasting Using Semi-supervised and Unsupervised Extreme Learning Machines10.1　Introduction10.2　Methodology10.2.1　Formation of Proposed Approach10.2.2　Health Indication (HI) Extraction10.2.3　Box-Cox Transformation (BCT)10.3.1　BC Transformation10.3.2　BCT Parameter Evaluation Using ML Method10.2.4　Correlation Analysis Using PCA and SRCA Methods　Pearson Correlation Analysis (PCA)　Spearman Rank Correlation Analysis (SRCA)10.2.5　RUL Estimation Approach10.3　Verification of LIB HI Evaluation10.3.1　LIB Dataset Used for Study10.3.1.1　Charging Condition10.3.1.2　Discharging Condition10.3.1.3　Impedance Measurement Condition10.3.2　Correlation Analysis and Evaluation10.3.2.1　Qualitative Analysis10.3.2.2　Quantitative Analysis10.3.3　HI Performance Evaluation10.4　RUL Estimation Validation10.5　ConclusionReferences