Fault diagnosis method of engine crankshaft position sensor

Abstract

A fault diagnosis method of an engine crankshaft position sensor is composed of a data acquisition system, a fuzzy processing system and a neural network system, and mainly diagnoses faults of Hall type and magnetic induction type crankshaft position sensors. According to the method, wave shape eigenvalues of fuzzified electronic signals of the crankshaft position sensor serve as input values of the neural network system, the fault diagnosis of the crankshaft position sensor is completed by a back propagation (BP) neural network in a multi-input and multi-output structure. The fault diagnosis method of the engine crankshaft position sensor has the advantages of being capable of diagnosing a plurality of faults of the crankshaft position sensor rapidly and accurately, providing technological help for automobile repair enterprises, and being widely used in the fault diagnosis of various kinds of automobile sensors.

Description

Technical field: [0001] The invention relates to a method for diagnosing faults of an engine crankshaft position sensor, which can diagnose faults of automobile sensors and related components through neural network technology. The technical fields include automotive electronic control technology, neural network technology, fuzzy mathematics, etc. Background technique: [0002] Every sensor in the car has its own standard working waveform. When the tested waveform is different from the standard waveform, it means that the component or its related parts have failed. Therefore, its faults can be diagnosed by the waveform characteristics of the sensor. At this stage, the research on sensor fault diagnosis is relatively extensive, but because the characteristics of sensor waveform extraction are different in each study, the types of faults that can be diagnosed are not the same. The crankshaft position sensor is one of the most critical sensors in the automobile engine, which d...

AI Technical Summary

Problems solved by technology

At this stage, the research on sensor fault diagnosis is relatively extensive, but since each study has different characteristics of sensor waveform extraction, the types of faults that can be diagnosed are not the same

The crankshaft position sensor is one of the most critical sensors in the automobile engine, which directly affects the starting and running conditions of the engine. However, there is no example of using neural network technology to diagnose the fault of the crankshaft position sensor at this stage.