铁道学报
鐵道學報
철도학보
2014年
4期
48-53
,共6页
动车组列车制动系统%思维进化算法%Hammerstein模型%CRH2型动车组
動車組列車製動繫統%思維進化算法%Hammerstein模型%CRH2型動車組
동차조열차제동계통%사유진화산법%Hammerstein모형%CRH2형동차조
EMU braking system%mind evolutionary algorithm%Hammerstein model%EM U CRH2
针对动车组列车制动系统的非线性及其在ATO中的重要性,从控制和动力学角度提出动车组列车制动系统的Hammerstein模型。根据制动指令信号的流向介绍动车组列车制动系统的工作过程;分别考虑系统各环节,用经过曲线拟合得到的静态非线性函数描述动车组列车制动特性表,用延时环节描述制动指令信号传输和制动控制器动作的延时,用两个一阶线性环节分别描述制动力反馈调节过程和动车组列车减速度冲动缓解过程,提出动车组列车制动系统的 Hammerstein模型;并介绍了思维进化算法辨识模型参数的方法。最后以CRH2型动车组为仿真对象验证模型和参数辨识方法的有效性。
針對動車組列車製動繫統的非線性及其在ATO中的重要性,從控製和動力學角度提齣動車組列車製動繫統的Hammerstein模型。根據製動指令信號的流嚮介紹動車組列車製動繫統的工作過程;分彆攷慮繫統各環節,用經過麯線擬閤得到的靜態非線性函數描述動車組列車製動特性錶,用延時環節描述製動指令信號傳輸和製動控製器動作的延時,用兩箇一階線性環節分彆描述製動力反饋調節過程和動車組列車減速度遲動緩解過程,提齣動車組列車製動繫統的 Hammerstein模型;併介紹瞭思維進化算法辨識模型參數的方法。最後以CRH2型動車組為倣真對象驗證模型和參數辨識方法的有效性。
침대동차조열차제동계통적비선성급기재ATO중적중요성,종공제화동역학각도제출동차조열차제동계통적Hammerstein모형。근거제동지령신호적류향개소동차조열차제동계통적공작과정;분별고필계통각배절,용경과곡선의합득도적정태비선성함수묘술동차조열차제동특성표,용연시배절묘술제동지령신호전수화제동공제기동작적연시,용량개일계선성배절분별묘술제동력반궤조절과정화동차조열차감속도충동완해과정,제출동차조열차제동계통적 Hammerstein모형;병개소료사유진화산법변식모형삼수적방법。최후이CRH2형동차조위방진대상험증모형화삼수변식방법적유효성。
The Hammerstein model of EMU braking system was ectablished in view of nonlinearity and impor-tance of the braking system in ATO(automatic train operation) .It was a kinetic model and accorded with the control law .The working process of the EMU braking system was introduced according to the transport mech-anism of braking instructions .The static nonlinear function obtained by curve fitting was used to depict the braking characteristics table and the delay system was used to describe the delay characteristics of the braking instruction transmission and the braking controller operation . One first-order linear system represented the feedback process of breaking .The other first-order linear system described the remission process of EMU accel-eration impulses .The above three links of braking constituted the Hammerstein model .Then ,the way how to identify model parameters with the mind evolutionary algorithm (MEA) was introduced .Finally ,simulation by EM U CRH2 proved the effectiveness of the proposed model and parameter identification algorithm .