Advanced Fault Detection in Hydraulic Actuator Cylinders of Heavy Machinery Using Neural Networks

Abstract This work presents for detecting internal leakage faults in hydraulic actuator cylinders using signal analysis and a supervised artificial neural network classifier. An artificial leakage is introduced and a signal-based fault detection method is employed to process and transform the signals for internal leakage detection. The analysis focuses on extracting features from the pressure signal, particularly the peaks, which include information as location, height, and width. Once the neural network is trained, it is utilized to classify the fault level into three categories: healthy system, system with low fault, and system with high fault. The proposed technique utilizes pressure signals and extracts features from the peak signals to reduce dimensionality. This method offers advantages such as reduced computational cost through feature extraction and dimensionality reduction, and it is capable of detecting multiple leakage classes. This effective technique based on artificial neural networks for detecting internal leakage faults in hydraulic cylinders.