Cavitation recognition of axial piston pumps in noisy environment based on Grad‐CAM visualization technique

Abstract

The cavitation in axial piston pumps threatens the reliability and safety of the overall hydraulic system. Vibration signal can reflect the cavitation conditions in axial piston pumps and it has been combined with machine learning to detect the pump cavitation. However, the vibration signal usually contains noise in real working conditions, which raises concerns about accurate recognition of cavitation in noisy environment. This paper presents an intelligent method to recognise the cavitation in axial piston pumps in noisy environment. First, we train a convolutional neural network (CNN) using the spectrogram images transformed from raw vibration data under different cavitation conditions. Second, we employ the technique of gradient-weighted class activation mapping (Grad-CAM) to visualise class-discriminative regions in the spectrogram image. Finally, we propose a novel image processing method based on Grad-CAM heatmap to automatically remove entrained noise and enhance class features in the spectrogram image. The experimental results show that the proposed method greatly improves the diagnostic performance of the CNN model in noisy environments. The classification accuracy of cavitation conditions increases from 0.50 to 0.89 and from 0.80 to 0.92 at signal-to-noise ratios of 4 and 6 dB, respectively.

Estimating sliding drop width via side-view features using recurrent neural networks

“…Gradient-weighted Class Activation Mapping (Grad-CAM) is an interpretation model that was originally designed for image processing 47 . Grad-CAM is versatile and is extended for text and sound analysis 48 – 50 . Similar to Grad-CAM, which calculates the gradients of a target output with respect to convolutional layer activations to identify important regions in an image, we used gradient-based feature importance to assess the influence of input features on the output of the LSTM model.…”

A Comprehensive Approach for Detecting Brake Pad Defects Using Histogram and Wavelet Features with Nested Dichotomy Family Classifiers

“…The selected features should be classified for predicting performance. The commonly used classifiers are Decision Tree [23,24], Fuzzy [25], Artificial Neural Network [20], Bayes Net [26], Naive Bayes [20], Support Vector Machine (SVM) [27], Convolutional Neural Network (CNN) [28] and the Hidden Markov Model [29]. Different classifiers monitored the tool wear using vibration signals with the discrete wavelet feature [20].…”