Abstract:The key to fault diagnosis of rotating machinery is to extract fault features effectively and select the appropriate classification algorithm. As a common signal decomposition method, the effect of wavelet packet decomposition (WPD) largely depends on the applicability of the wavelet basis function (WBF). In this paper, a novel fault diagnosis approach for rotating machinery based on feature importance ranking and selection is proposed. Firstly, a two-step principle is proposed to select the most suitable WBF for the vibration signal, based on which an optimized WPD (OWPD) method is proposed to decompose the vibration signal and extract the fault information in the frequency domain. Secondly, FE is utilized to extract fault features of the decomposed subsignals of OWPD. Thirdly, the categorical boosting (CatBoost) algorithm is introduced to rank the fault features by a certain strategy, and the optimal feature set is further utilized to identify and diagnose the fault types. A hybrid dataset of bearing and rotor faults and an actual dataset of the one-stage reduction gearbox are utilized for experimental verification. Experimental results indicate that the proposed approach can achieve higher fault diagnosis accuracy using fewer features under complex working conditions.

What problem does this paper attempt to address?

This paper attempts to solve several key problems in the fault diagnosis of rotating machinery: 1. **Effective extraction of fault features**: Traditional feature extraction methods may not fully explore deep - level fault features. This paper proposes a new method combining optimized wavelet packet decomposition (OWPD) and fuzzy entropy (FE) to extract fault features from vibration signals. 2. **Selection of appropriate classification algorithms**: In order to improve the performance of fault pattern recognition, this paper introduces the CatBoost algorithm, which is used not only for fault pattern recognition but also for feature selection to filter redundant features, reduce model training time, and improve classification accuracy. 3. **Optimization of hyper - parameters**: The Bayesian optimization (BO) algorithm is used to solve the problem of hyper - parameter optimization in the CatBoost algorithm to find the optimal combination of hyper - parameters, thereby further enhancing the performance of the model. ### Specific steps 1. **Acquisition of original vibration signals**: The original vibration signals are obtained through accelerometers and data acquisition systems. 2. **Selection of the optimal wavelet basis function (WBF)**: The most suitable WBF is selected according to the proposed two - step principle, and then the original vibration signal is decomposed using the optimized wavelet packet decomposition (OWPD) method. 3. **Calculation of fuzzy entropy values**: Calculate the fuzzy entropy values of the decomposed sub - signals to form a fault feature set \( F \). 4. **Ranking of feature importance**: Use the CatBoost algorithm to calculate the importance of each feature in the feature set \( F \), and select candidate features in turn according to the ranking results of feature importance, and combine them with the corresponding labels to form a data set \( S \). 5. **Division of data sets**: Divide the data set \( S \) into training sets and test sets according to a certain proportion. 6. **Model training and hyper - parameter optimization**: Use the training set to train the CatBoost classifier and use the Bayesian optimization algorithm to optimize the main hyper - parameters. 7. **Output of fault diagnosis results**: Input the test set into the trained CatBoost classifier and output the final diagnosis results. ### Experimental verification In order to verify the effectiveness of the proposed method, the author uses a mixed data set of bearing and rotor faults collected by the mechanical fault simulator (MFS) platform. The experimental results show that the proposed method can achieve a higher fault diagnosis accuracy with fewer features in complex working conditions.

Fault Diagnosis Approach for Rotating Machinery Based on Feature Importance Ranking and Selection

Fault diagnosis of rotating machinery based on improved wavelet package transform and SVMs ensemble

Fault Diagnosis Method for Rotating Machinery Based on Hierarchical Amplitude-Aware Permutation Entropy and Pairwise Feature Proximity

Application of an intelligent classification method to mechanical fault diagnosis

A Two-Stage Feature Selection and Intelligent Fault Diagnosis Method for Rotating Machinery Using Hybrid Filter and Wrapper Method.

A Novel Fault Feature Selection and Diagnosis Method for Rotating Machinery with SI-IR2CMSE and SSGMM-SR

Multiscale Slope Feature Extraction for Rotating Machinery Fault Diagnosis Using Wavelet Analysis

Fault diagnosis of key components in the rotating machinery based on Fourier transform multi-filter decomposition and optimized LightGBM

A fault diagnosis method for rotating machinery based on improved variational mode decomposition and a hybrid artificial sheep algorithm

A Fault Diagnosis Method for Rotating Bearings Based on EWT Multi-Scale Entropy and PSO Algorithm to Optimize SVM

Fault Diagnosis for Rotating Machinery: A Method based on Image Processing

An Approach to Performance Assessment and Fault Diagnosis for Rotating Machinery Equipment.

Rotating Machinery Fault Diagnosis Using Signal-Adapted Lifting Scheme

Rotating Machinery Fault Diagnosis Based on Improved Multiscale Amplitude-Aware Permutation Entropy and Multiclass Relevance Vector Machine

Bearing Fault Feature Extraction and Fault Diagnosis Method Based on Feature Fusion

A Fault Diagnosis Method for Rotating Machinery Based on PCA and Morlet Kernel SVM

Dominant feature selection for the fault diagnosis of rotary machines using modified genetic algorithm and empirical mode decomposition

Fault diagnosis of rotating machines based on modified hierarchical fluctuation dispersion entropy and multi-cluster feature selection

A hybrid fault diagnosis approach based on mixed-domain state features for rotating machinery.

Bearing Fault Feature Selection Method Based on Weighted Multidimensional Feature Fusion.

Feature Extraction Based on DWT and CNN for Rotating Machinery Fault Diagnosis