液压系统仿真模型验证研究

发布时间：2018-05-17 03:44

本文选题：验证研究 + 仿真模型　；参考：《集美大学》2012年硕士论文

【摘要】：在软件平台上建立实际液压系统的仿真模型进而对其性能进行仿真分析是当今流行的技术方法。液压系统仿真模型的建立过程中模型都是经过一定的简化和近似，仿真模型的近似程度决定了仿真结果的可信度。因此，在对液压试验台及其仿真模型分析的基础上，对液压系统仿真模型的有效性进行验证显得很有意义。本文以福建省科技重大专项专题“液压传动技术与产品的研发及产业化”（2008HZ0002-1）为背景，结合项目建设的液压系统综合试验平台，以采用EASY5软件搭建的试验平台液压系统仿真模型为研究对象，对仿真模型的有效性进行验证研究。在相同的输入条件下，对仿真模型输出和试验台系统输出的一致性或相似性的验证研究，进而分析仿真模型的有效性、仿真结果的可信性以及两系统的差异。本论文主要开展的研究包括：首先，对数据分析处理及仿真模型验证方法的相关理论进行研究。仿真模型验证的前提是两类系统的输出真实有效，，为此，采用拉依达准则对试验测试及仿真数据进行异常值判断与剔除；采用轮次检验法进行动态数据平稳性检验并对非平稳数据进行去除均值趋势项的平稳化处理；采用K-S检验法进行数据分布形式的判定，得两类系统的输出数据符合正态分布；同时，对仿真验证方法进行了分类归纳总结，结合研究对象的特点选择本文的验证方法。其次，在相同试验条件下对两系统的稳态性能进行验证研究。分别从非参数检验和参数检验两个方面进行验证，非参数检验中采用计算简便、工程中常用的秩和检验法及游程检验法完成验证，并构成相互佐证，验证结果表明仿真模型的稳态输出是有效的。在完成非参数检验验证的基础上实现两类系统均值、方差参数的检验验证，同样可得仿真模型是可信的。最后，就系统在几种典型实验条件下的动态性能进行一致性验证研究。在对两类数据时序分析处理的基础上，分别采用TIC不等式系数法及灰色关联分析法进行时域的定性验证分析，验证结果表明时域中两系统的动态性能是一致的；进一步应用经典谱分析方法进行频域的检验验证，验证结果可得频域中仿真模型与实际系统有一定差异，同时给出差异性原因分析。本文对两系统从稳态到动态、从时域到频域进行了全面的验证分析，通过验证更加深刻认识两类系统，同时也为进一步的研究提供一定的借鉴。
[Abstract]:It is a popular technical method to establish the simulation model of the actual hydraulic system on the software platform and to simulate and analyze its performance. The simulation model of hydraulic system is simplified and approximated, and the degree of approximation determines the credibility of simulation results. Therefore, on the basis of the analysis of the hydraulic test bench and its simulation model, it is very meaningful to verify the validity of the hydraulic system simulation model. This paper takes "hydraulic transmission technology and product R & D and industrialization" as the background, combined with the hydraulic system comprehensive test platform built by the project. Taking the hydraulic system simulation model of the test platform built by EASY5 software as the research object, the validity of the simulation model is verified. Under the same input conditions, the consistency or similarity between the simulation model output and the test-bed system output is verified, and then the validity of the simulation model, the credibility of the simulation results and the differences between the two systems are analyzed. The main research in this paper includes: Firstly, the theory of data analysis and simulation model verification is studied. The premise of simulation model verification is that the output of the two kinds of systems is true and effective. The stationary test of dynamic data and the stationary processing of removing the mean value trend term from the non-stationary data are carried out by the method of cycle test, and the data distribution form is judged by K-S test method, and the output data of the two kinds of systems accord with the normal distribution, and the results show that the output data of the two kinds of systems are in accordance with the normal distribution. At the same time, the simulation verification methods are classified and summarized, and combined with the characteristics of the research object, the verification method of this paper is selected. Secondly, the steady-state performance of the two systems is verified under the same experimental conditions. The non-parametric test and the parameter test are used to verify, the non-parametric test is easy to calculate, the rank sum test and the run-length test are used in engineering to complete the verification, and the mutual evidence is formed. The results show that the steady-state output of the simulation model is effective. On the basis of completing the verification of non-parametric test, the simulation model is also credible when the mean and variance parameters of two kinds of systems are tested and verified. Finally, the consistency verification of the dynamic performance of the system under several typical experimental conditions is carried out. On the basis of the time series analysis of two kinds of data, the TIC inequality coefficient method and the grey correlation analysis method are used to analyze the time domain qualitatively, and the results show that the dynamic performance of the two systems in the time domain is the same. Furthermore, the classical spectrum analysis method is used to verify the frequency domain. The results show that the simulation model in frequency domain is different from the actual system, and the reasons for the difference are also given. In this paper, the two systems from steady state to dynamic, from time domain to frequency domain are verified and analyzed. Through the verification, the two kinds of systems are understood more deeply, and some references are provided for further research.
【学位授予单位】：集美大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP391.9;TH137

【相似文献】