Detection of Oil Condition on Motorcycle Based on Sound Signal Analysis

Oil is an important part of motor vehicles, including motorcycles. Detect the condition of oil on a motorcycle is a thing to do. In this research, an oil detector system has been made for a motorcycle by using sound sensor and MyRio microcontroller. The sound sensor uses to capture the sound generated from motorcycles with two conditions, motorcycle with oil condition that does not need to be replaced and motorcycle with oil condition that has to be replaced. The captured signal then goes through signal processing processes, there are feature extraction and pattern recognition. Feature extraction process is done by using Mel Frequency Cepstral Coefficient (MFCC) method. The extracted MFCC coefficients been the input to pattern recognition process by using Support Vector Machine (SVM) method to determine the oil condition of the motorcycle are the oil has to be replaced or not. In this research, the system has been successfully detected oil condition of motorcycle with 100% accuracy.