基于DS证据理论的道砟清筛机作业工况识别研究

Research on Working Condition Identification of Ballast Cleaning Machine Based on DS Evidence Theory

针对大型道砟清筛机作业工况的识别问题,提出一种基于DS证据理论的识别方法和判定流程。首先,采集清筛机各工作装置压力传感器的信号,提取特征参数并构建传感器信号特征库;然后,通过预设的判定流程快速实现对空载工况的识别;最后,运用DS证据理论,采用分布式结构进行多传感器信息融合,并结合分类决策规则实现作业工况的识别判定。经清筛机现场试验验证表明,该方法实现了对清筛机作业工况的准确识别,具有较高的实际应用价值。

Aiming at the identification problem of working conditions of large ballast cleaning machine,this paperproposed the identify method and judgment process based on DS evidence theory. Firstly,the characteristic parameterswere extracted from the signals collected by the pressure sensor of each working device of cleaning machine,and thesensor signal feature database was constructed. Then,through the preset judgment process,the identification of no-loadcondition can be realized quickly. Finally,using DS evidence theory,multi-sensor information fusion was carried outwith distributed structure,and the working conditions were identifiedand combined with the classification decisionrules. The field test shows that this method can accurately identify the working conditions of the ballast cleaning machineand has high application value.