Research On Fault Diagnosis Of Hydraulic System Of A Large-caliber Artillery Loading Device

The artillery loading device is an important part of the artillery weapon system,which plays an important role in the process of projectile loading.Its performance will directly affect the speed,power,reliability and safety of the loading system.As one of the important subsystems of the artillery,its structure is complex,the working environment is diverse,and the working conditions are harsh.It is one of the subsystems with the most failures of the artillery.With the continuous improvement of modern weapon equipment intelligence,primitive manual fault judgment can no longer meet the actual combat requirements of modern weapon equipment.Therefore,it is of great significance for the study of intelligent fault diagnosis of large-caliber artillery.This paper takes a large-caliber artillery project as the background,uses co-simulation technology,combined with ant colony algorithm to optimize the BP neural network algorithm to carry out the fault intelligent diagnosis of the hydraulic system of the loading device.The main research contents of this article are:（1）Explain in detail the system composition,working principle and working process of the artillery loading device,establish the logical relationship between the components of the system,and analyze the common failures of the system and the generation mechanism of the failures through the failure mode and impact analysis（FMEA）.After analysis,common failure modes and failure causes were obtained,which served as the theoretical basis for subsequent research.（2）Based on the ADAMS and AMESim platforms,the dynamic model of the loading device and the hydraulic system model were established respectively,and the mechanicalhydraulic integrated simulation model was established through the corresponding interface module,and the simulation dynamic characteristics of the loading device were analyzed.The co-simulation model was verified through bench tests to ensure the accuracy of subsequent fault diagnosis data.（3）The fault injection method based on simulation is selected to study the fault simulation of the hydraulic system through fault injection technology.Through the establishment of a fault simulation model,the waveform characteristics of the parameters such as the pressure of the cylinder,the angular velocity and the angular displacement of the coordinating arm are obtained,the corresponding characteristic values are extracted,and the influence of the characteristic values is analyzed,and the corresponding fault samples are constructed.（4）Aiming at the defects of the BP neural network in the fault diagnosis of the hydraulic system of the loading device,a BP neural network algorithm optimized based on the ant colony algorithm is proposed.The fault classification and training of the hydraulic system are carried out through the classification and training of the fault samples.Part of the samples were used as test data to verify the accuracy of the algorithm and complete the fault diagnosis of the hydraulic system of the loading device.