Research on Engine Weakness Fault Diagnosis Method Based on Commercial Vehicle Driving Data

A unique diagnostic approach based on data mining technology is presented to replace the conventional technique of diagnosing engine powerlessness in commercial vehicles, which involves on-site test drives and a loss of human and material resources. On the basis of analyzing the signals of commercial vehicle-carrying terminals, we select the features closely related to engine powerlessness faults including vehicle speed, acceleration, and the rate of change of throttle opening, etc., and take the lead in proposing the acceleration-intent dual-model fault diagnosis technology scheme. The data are divided into two categories based on the rate of change of the throttle opening: those with acceleration intention and those without obviously apparent acceleration intention. The former uses a neural network model to diagnose the vehicle's power performance; the latter determines whether the engine is powerless by setting up a benchmark model that describes the engine's power performance and calculating the difference between the benchmark value and the actual value. The experimental findings indicate that the method can identify the engine powerlessness fault with 82 % accuracy, 92 % specificity, high precision, and potential for greater effectiveness. These findings lay the groundwork for future research on the engine powerlessness fault remote on-line diagnosis method.