Brake fault diagnosis using histogram features and artificial immune recognition system (AIRS)

Brakes are one of the most important components in automobiles because they allow the vehicle to stop or slow down. It requires extra caution in terms of safety and dependability. As a result, it is critical to monitor the brake system’s condition in order to assure safety. Vibration signals play an important function in detecting brake system faults. A machine learning approach was employed in this work to identify brake defects under various scenarios. A piezoelectric type transducer and data collecting system were used to collect vibration signals. The vibration signals were used to obtain the relevant histogram features. The feature selection and feature classification were done using the vibration signals obtained from the transducer. An artificial immune recognition system was used to classify the extracted features (AIRS). The classification accuracy as well as the classifier’s performance level have been reported.