Design And Development Of Fault Diagnosis Platform For High Pressure Common Rail Fuel System

With the requirements of the "double carbon" goal,the performance requirements for energy conservation and emission reduction of ship power are also constantly improving.The high-pressure common rail fuel system has become the mainstream technology of marine diesel engine fuel system due to its superior energy saving and emission reduction performance.However,its complex structure and harsh working environment of high frequency and pressure make it the subsystem with the highest failure rate of diesel engines.In order to improve the safety and reliability of marine common rail diesel engines,domestic and foreign experts and scholars have made some achievements in the fault diagnosis research of the Common rail fuel system,but there are still some problems that need to be solved urgently,mainly including: 1.The design and development of the fuel system diagnosis platform is less,and the diagnostic accuracy and efficiency of some existing diagnostic systems are low;2.Foreign collection equipment is expensive and has a long purchase cycle,resulting in high research and development costs and reduced practicality of diagnostic platforms.In response to the above issues,the main research content of this article is as follows:In this paper,the failure mode and effects analysis method are used to sort out the typical fault types of the Common rail fuel system,and two fault types are selected for fault simulation.Based on the analysis of the research status of fault diagnosis of fuel system at home and abroad,the system architecture of the fault diagnosis platform of Common rail fuel system is designed,the diesel engine test platform and the Common rail fuel system test platform are built,and the fault source signal acquisition experiment is carried out on the platform,providing data support for the effectiveness verification of the diagnosis method and diagnosis platform.In order to independently develop and design a signal acquisition system and replace expensive and timeconsuming acquisition equipment from abroad,a signal acquisition system was designed and developed based on the powerful signal processing ability and extremely high computational speed of Digital Signal Processor(DSP).Hardware signal processing circuits and adaptive digital filters were also designed in the acquisition system,Real time filtering and acquisition of pressure wave signals from high-pressure oil pipes have been achieved.In order to ensure the reliability and accuracy of the diagnostic results of the fault diagnosis platform,this paper conducts the following research on the intelligent diagnostic algorithms embedded in the diagnosis platform: In response to the problem of poor signal denoising effect caused by mode mixing defects in the Complementary Ensemble Empirical Mode Decomposition(CEEMD)algorithm,an improved CEEMD adaptive threshold denoising algorithm is proposed,Through simulation and experimental data analysis,it is shown that the improved CEEMD algorithm improves the signal-to-noise ratio of the signal by 9.58 d B and 1.45 d B,respectively,compared to commonly used noise reduction algorithms.In response to the impact of the accuracy of complex signal feature extraction on fault recognition accuracy,this article cites the Composite Multi-scale Weighted Permutation Entropy(CMWPE)method for feature extraction of denoised signals.By comparing it with commonly used feature extraction methods,Compared to the Multi-scale Weighted Permutation Entropy and Hierarchical Weighted Permutation Entropy algorithms,the fault recognition accuracy based on CMWPE algorithm has been improved by 5.0% and 6.67%,respectively.In the final step of the fault diagnosis process-fault pattern recognition,this paper introduces the particle swarm optimization Least Squares Support Vector Machines(LSSVM)method to improve the efficiency of fault diagnosis and recognition accuracy.Compared with the cross validation LSSVM parameter algorithm and the standard LSSVM algorithm,the accuracy of fault recognition is increased by 1.67% and 6.67% respectively,and the false alarm rate is the lowest.Finally,this paper completes the design of a signal monitoring system and a fault diagnosis system based on LabVIEW software,and embeds the intelligent diagnosis algorithm studied in this paper into the diagnosis system.On the basis of integrating system hardware and software,a DSP acquisition system was used to conduct signal acquisition experiments under different fault states of the fuel system on a diesel engine test platform.The results were compared with the data collected by the NI acquisition card.The experimental results showed that the fault source signals obtained by the acquisition system designed based on DSP in this paper were real and effective,and could replace foreign acquisition equipment.The function of the fault diagnosis system is tested using the fault data of the Common rail fuel system test bench.The diagnosis results show that the fault diagnosis accuracy of the diagnosis system reaches 98.551%,which meets the actual diagnosis requirements.