融合Cox回归与维纳过程的设备状态评估方法

Equipment condition assessment method integrating Cox regression and Wiener process

摘要:

针对机场自助行李托运设备运行状态评估存在数据利用率低和预测精度不高的问题,提出一种融合Cox回归与维纳过程的设备状态评估方法。首先基于事件型数据构建多风险因素影响下的设备状态突变模型,又基于关键子系统的状态型数据构建设备状态渐变模型,提出了复合退化指标并基于维纳过程建立设备性能退化模型,得到设备整体的健康状态值并提出相应维修决策。利用商用模块化航空推进系统仿真数据集和自助行李托运设备运行监测数据对所提方法进行实验验证,结果表明融合Cox回归与维纳过程的设备状态评估方法提高了设备数据利用率和预测精度。

Abstract:

Aiming at the problems of low data utilization and low prediction accuracy in the evaluation of the operation status of self-service baggage check-in equipment at airports, a method of equipment status evaluation combining Cox regression and Wiener process is proposed. First, build a sudden change model of equipment status under the influence of multiple risk factors based on event data, and build a gradual change model of equipment status based on the status data of key subsystems. A composite degradation index is proposed and the equipment performance degradation model is established based on the Wiener process to obtain equipment. The overall health status value and propose corresponding maintenance decisions. The experimental verification of the proposed method is carried out using the simulation data set of Commercial Modular Aero-Propulsion System Simulation and the operation monitoring data of self-service baggage check-in equipment. The results show that the equipment condition assessment method fusing Cox regression and Wiener process improves equipment data utilization and prediction accuracy.