Research on Fault Diagnosis of PST Electro-Hydraulic Control System of Heavy Tractor Based on Support Vector Machine

Due to the harsh working environment of tractor, transmission can often be faulty. In order ensure reliability its operation, it must monitored and fault discovered. this paper, support vector machine (SVM) method is used. The eigenvector conversion original data uses following eigenvectors: Three modes (leakage shift clutch hydraulic cylinder, blockage oil passage, proportional valve spool) are identified in matrix laboratory (MATLAB) with help library for machines (LibSVM) toolkit, classification accuracy test samples 90%. normal mode PST electro-hydraulic system three kinds mentioned above discriminated against, correct rate diagnosis reaches 95%, which meets needs practical engineering. Analysis recording power shifting solenoid shows that difference between pressure small, value traffic greater. This realize online recognition based on controller area network (CAN) communication, research results provide a theoretical basis control system.