Hydrogen fuel engine fault diagnosis system based on SOM neural network and method thereof

Abstract

The invention belongs to the technical field of hydrogen fuel engine maintenance, and particularly relates to a hydrogen fuel engine fault diagnosis system based on an SOM neural network and a methodthereof. The method comprises the steps that fault learning samples containing the fault data of various working conditions of a hydrogen fuel engine are collected; the fault learning samples are usedas training data and transmitted to the SOM neural network for training learning; the distribution characteristics and topology of the input data are trained, and the trained output result is used ascalibration data for determining the fault type; the parameter data, which are received by a monitoring module, of various working conditions are used as test data, and the test data are transmittedto the SOM neural network for testing; and according to the calibration data, the Euclidean distance is used to determine the fault type of the hydrogen fuel engine of the tested output result. According to the invention, the SOM neural network is used to adaptively change the network parameters and structure in a self-organized manner to effectively and accurately diagnose the fault type of the hydrogen fuel engine, which is convenient for later timely maintenance, ensures the reliability of hydrogen fuel engine equipment, and greatly prolongs the service life thereof.

Description

technical field [0001] The invention belongs to the technical field of hydrogen fuel engine maintenance, and particularly relates to a hydrogen fuel engine fault diagnosis system and method based on a SOM neural network, which combines the SOM neural network to perform fault diagnosis during the operation of the hydrogen fuel engine to ensure the safety of the hydrogen fuel engine equipment. reliability. Background technique [0002] The direct result of scientific and technological progress is the improvement of the economy, and economic development has promoted the entire automobile industry to make great strides forward. In recent years, the number of automobiles has increased sharply. In 2015, the number of civilian vehicles in China has reached 162.8445 million, a year-on-year increase of 12%, and the direct consequence of this trend is environmental pollution. Undoubtedly, the emergence of automobiles has made people's lives completely different, but this convenience ...

AI Technical Summary

Problems solved by technology

However, when its working process is improperly organized, the possibility of abnormal combustion in a hydrogen engine is greater than that of a gasoline engine, and its output power is lower than that of a gasoline engine. In addition, the ignition energy required by hydrogen fuel is extremely low. When the intake pipe is opened and the fuel enters the cylinder, Pre-ignition may be caused by residual waste in the cylinder or ignited by hot spots, and this process may continue to advance, which will cause the flame to enter the intake pipe when the intake valve is not closed, resulting in abnormal combustion such as backfiring and combustion fluctuations.

This abnormal combustion will not only affect the basic use of the engine, but will even cause serious consequences such as stalling, stopping, and failing to work normally.

Images