Validation Study Of Hydraulic System Simulation Model

It is a popular technique method to analysis the hydralic system property on the softwareplatform by its simulation model. The hydraulic system simulation models are simplified andapproximate during modeling process, and the degree of approximation of the simulation modelto determine the credibility of the simulation results. Therefore, based on the analysis ofhydraulic test rig and its simulation model, it is significant to verify the validity of the simulationmodel of the hydraulic system.This paper was baseed on the major science and technology topics of Fujian Province,which was called “hydraulic drive technology and product R&D and industry "(itemnumber:2008HZ0002-1), combined with the comprehensive hydraulic test rig of our projectconstruction, taked the simulation model which was builded on EASY5software as the object ofstudy, and make the verification study of the validity of the simulation model. In the same inputconditions, verify the consistency or similar of the simulation model output and the output of thetest bed system, and then analyze the validity of the simulation model, the credibility of thesimulation results as well as differences.This thesis mainly developed to study include:Firstly, making a study of the data analysis and simulation model validation methodcorrelative theory. The premise of the simulation model validation is that output of the two typesof systems are real and effective, for this reason,using the PauTa rule to estimate the testing andsimulation data, and eliminate of outliers; Using the change-times check method to test thestationarity of the dynamic data and smooth the non-stationary data through remove the meantrend; Using the K-S test method to determine the data distribution form, and obtain the twotypes system output data are meet normal distribution; At the same time, the simulationverification method are classificaed and summarized, and combining with the objest of study,select the validation method.Secondly, making the validation study of the steady-state performance of the two systems inthe same experimental conditions. Verification study is conduncted separately from twoaspects:non-parameter verification and parameter verification. Non-parametric test adoptrank-sum testing method and run-length testing method to complete the verification, which aresimply calculated and commonly used in engineering, and the two methods constitute a mutualcorroboration. The verification results indicate that the steady state of the simulation modeloutput is valid. On the basis of inspection and verification of non-parametric tests, the mean value and variance parameters of the two systems are verified, and the simulation model is alsoavailable and credible.Finally, the dynamic performance’s consistency of the validation study of the system arecondunct in several typical experimental conditions. On the basis of analysis of two types of timeseries data, the qualitative validation analysis is realized in time domain by using the TIC’sinequality coefficient method and gray relational analysis method respectively, and theverification results show that dynamic performance of the two systems is consistent in the timedomain; The classical spectrum analysis methods is further applicated for frequency domainvalidation, and the results show that simulation model and actual system have certaindiscrepancy, and then given cause analysis of the differences.In this paper, the two systems are processed a comprehensive verification, from steady stateto dynamic, and from the time domain to the frequency domain analysis, and more profoundunderstanding of the two types of systems by verifying. At the same time, the analysis resultsprovide a reference for further research.