【摘要】：摩擦模型參數(shù)辨識在機械工程設計、控制、系統(tǒng)優(yōu)化等方面均發(fā)揮重要作用,針對無法通過恒速法獲取摩擦力的機械系統(tǒng)(譬如汽車轉向系統(tǒng)),本文以Stribeck摩擦模型為原型提出一種辨識其摩擦參數(shù)的方法。首先,基于能量法則將參數(shù)辨識問題轉化為參數(shù)優(yōu)化問題,并根據(jù)優(yōu)化目標的結構特點,分別設計了最小二乘法和遺傳算法兩種辨識算法。其次,利用Matlab/Simulink搭建含Stribeck摩擦模塊的單自由度和兩自由度運動仿真系統(tǒng),并針對最小二乘法和遺傳算法的優(yōu)化條件設計了運動方案,仿真結果驗證了能量法則的正確性與所提出參數(shù)辨識方法的有效性。再次,以循環(huán)球轉向?qū)嶒炁_架為應用對象,建立包含Stribeck摩擦模型的系統(tǒng)動力學方程,并結合本文所設計的參數(shù)辨識方法確定該轉向系統(tǒng)摩擦實驗方案。最后,設計了具體的轉向系統(tǒng)實驗方案并依次實施,對實驗結果采用遺傳算法逐個辨識出系統(tǒng)各摩擦模塊的摩擦參數(shù),通過相對誤差分析和系統(tǒng)綜合分析論證了參數(shù)辨識結果準確可信,并驗證了本文所提出的方法對汽車轉向系統(tǒng)摩擦參數(shù)辨識切實可行。
[Abstract]:Friction model parameter identification plays an important role in mechanical engineering design, control, system optimization, etc. For mechanical system (such as automobile steering system) which can not obtain friction force by constant speed method, this paper presents a method to identify friction parameters based on Stribeck friction model. Firstly, the parameter identification problem is transformed into the parameter optimization problem based on the energy rule, and two identification algorithms, least square method and genetic algorithm, are designed according to the structural characteristics of the optimization objective. Secondly, the motion simulation system of single degree of freedom and two degrees of freedom with Stribeck friction module is built by using Matlab/Simulink, and the motion scheme is designed according to the optimization conditions of least square method and genetic algorithm. Simulation results verify the correctness of the energy rule and the validity of the proposed parameter identification method. Thirdly, the system dynamic equation including Stribeck friction model is established, and the friction experimental scheme of the steering system is determined by using the parameter identification method designed in this paper. Finally, the specific experiment scheme of steering system is designed and implemented in turn. The friction parameters of each friction module of the system are identified one by genetic algorithm. Through the relative error analysis and the system comprehensive analysis, it is proved that the parameter identification results are accurate and credible, and it is proved that the method proposed in this paper is feasible for automobile steering system friction parameter identification.
【學位授予單位】：廣西大學
【學位級別】：碩士
【學位授予年份】：2015
【分類號】：TH117.1

【參考文獻】

相關期刊論文 前1條

1 徐升;葛文杰;張輝;;雙萬向節(jié)系統(tǒng)不等速性因素研究[J];機械傳動;2014年04期

相關碩士學位論文 前1條

1 李紅根;萬向節(jié)參數(shù)化建模及運動特性的研究[D];西南交通大學;2014年

，

本文編號：2200887