CAJ | 학술논문

针对高速列车安全性能跟踪监测点多、监测数据量大而导致的走行部故障识别难的问题，提出了一种基于模糊证据理论的多特征、多源信息融合的走行部故障识别方法。首先根据不同传感器信息的某类特征属于不同故障模式下的隶属度间的差异来计算传感器间的信息融合度，利用融合度来确定不同传感器在融合中的权重，从而得到同类特征不同传感器间信息融合后的隶属度；然后由融合后的隶属度转化为基本概率分配函数；最后用证据理论对不同特征间信息进行融合。实验结果表明，该方法能有效地识别列车正常、空气弹簧失气、抗蛇形减震器全拆、横向减震器全拆四种情况，同时在不同速度下均取得了满意的识别率，验证了该方法的有效性。
침대고속열차안전성능근종감측점다、감측수거량대이도치적주행부고장식별난적문제，제출료일충기우모호증거이론적다특정、다원신식융합적주행부고장식별방법。수선근거불동전감기신식적모류특정속우불동고장모식하적대속도간적차이래계산전감기간적신식융합도，이용융합도래학정불동전감기재융합중적권중，종이득도동류특정불동전감기간신식융합후적대속도；연후유융합후적대속도전화위기본개솔분배함수；최후용증거이론대불동특정간신식진행융합。실험결과표명，해방법능유효지식별열차정상、공기탄황실기、항사형감진기전탁、횡향감진기전탁사충정황，동시재불동속도하균취득료만의적식별솔，험증료해방법적유효성。
To solve the problem that it is difficult to identify faults of high-speed train running gears caused by multiple detection points and large amount of data,a running gear fault recognition method with multi-feature and multi-source information fusion was proposed based on the fuzzy evidence theory.Firstly,the information fusion levels of sensors were obtained by calculating the difference of membership degrees reflecting a certain feature of different sensors belonging to different fault modes.The fusion levels were used to determine the weights of different sensors during fusion,thus the membership degrees after information fusion between different sensors with similar characteristics were obtained.Then,the fused membership degrees were converted into the basic probability assignment functions.Finally,the information of different characteristics was fused on the basis of the evidence theory.Experimental results showed that the four cases including normal running,air loss of air spring,without anti-yaw shock absorber and without transverse shock absorber can be identified effectively with the proposed method,and the satisfactory fault identification rates are obtained at different speeds,all these facts verify the validity of the new method.