Machine Learning-based Fault Diagnosis for Industrial Engineering Systems (Advances in Intelligent Decision-making, Systems Engineering, and Project Management)

Description

This book provides advanced techniques for precision compensation and fault diagnosis of precision motion systems and rotating machinery. Techniques and applications through experiments and case studies for intelligent precision compensation and fault diagnosis are offered along with the introduction of machine learning and deep learning methods. Cover Half Title Series Page Title Page Copyright Page Table of Contents Preface Authors Chapter 1: Background and Related Methods 1.1 Background 1.2 Related Methods 1.2.1 Back Propagation Neural Network 1.2.2 Convolutional Neural Network 1.2.3 Recurrent Neural Network 1.2.4 Generative Adversarial Networks 1.2.5 Bagging Algorithm 1.2.6 Classification and Regression Tree 1.2.7 Random Forest 1.2.8 Density-Based Spatial Clustering of Applications with Noise 1.2.9 Safe-Level Synthetic Minority Over-Sampling Technique Bibliography Chapter 2: Fault Diagnosis Method Based on Recurrent Convolutional Neural Network 2.1 Introduction 2.2 Model Establishment and Theoretical Derivation 2.2.1 One-Dimensional Convolutional Neural Network 2.2.2 Convolutional Recurrent Neural Network Model 2.2.3 Dropout in Neural Network Model 2.3 Diagnostic Flow of the Proposed Method 2.4 Experimental Research Based on The Proposed Method 2.4.1 Experiment Platform 2.4.2 Experimental Setup 2.4.3 Summary of Experimental Results Bibliography Chapter 3: Fault Diagnosis of Rotating Machinery Gear Based on Random Forest Algorithm 3.1 Introduction 3.2 Fault Diagnosis of Rotating Machinery Gear Based on Random Forest Algorithm 3.3 Experimental Verification 3.3.1 Experiment Platform 3.3.2 Experimental Results 3.3.3 Comparison Study Bibliography Chapter 4: Bearing Fault Diagnosis under Different Working Conditions Based on Generative Adversarial Networks 4.1 Introduction 4.2 Model Establishment and Theoretical Derivation 4.2.1 Wasserstein Generative Adversarial Network 4.2.2 Maximum Mean Discrepancy 4.2.3 Establishment of Fault Diagnosis Model 4.2.4 Fault Diagnosis Procedures of the Proposed Method 4.3 Experimental Results Bibliography Chapter 5: Rotating Machinery Gearbox Fault Diagnosis Based on One-Dimensional Convolutional Neural Network and Random Forest 5.1 Introduction 5.2 Model Establishment and Theoretical Derivation 5.2.1 One-Dimensional Convolutional Neural Network 5.2.2 Random Forest Algorithm 5.2.3 The Proposed Fault Diagnosis Model 5.2.4 Error Back Propagation of the Proposed Model 5.2.5 Weights Optimization Using Adaptive Moments 5.3 Experimental Results 5.3.1 Experimental Platform 5.3.2 Experimental Setup 5.3.3 Analysis of Experimental Results Bibliography Chapter 6: Fault Diagnosis for Rotating Machinery Gearbox Based on Improved Random Forest Algorithm 6.1 Introduction 6.2 Improved Random Forest Algorithm 6.2.1 Semi-Supervised Learning 6.2.2 Improved Random Forest Classification Algorithm 6.3 Experimental Verification Bibliography Chapter 7: Imbalanced Data Fault Diagnosis Based on Hybrid Feature Dimensionality Reduction and Varied Density-Based Safe-Level Synthetic Minority Oversampling Technique 7.1 Introduction 7.2 Design of Hybrid Feature Dimensionality Reduction Algorithm 7.2.1 Sensitive Feature Selection 7.2.2 Dimension Reduction of Features 7.3 Design of Varied Density-Based Safe-Level Synthetic Minority Oversampling Technique 7.4 Experiment and Results 7.4.1 Data Classification Method 7.4.2 Experiment Platform 7.4.3 Feature Extraction 7.4.4 Data Acquisition 7.4.5 Results Analysis Bibliography Index