基于灰信息挖掘的视情维修决策方法研究

针对列装新设备缺乏状态劣化信息和维修阈值难以进行视情维修决策的问题，基于新旧设备故障率变化趋势的一致性，定义改进性系数表示基本故障率的变化不同，得到新型设备的故障率函数，结合回归支持向量机拟合新型设备风险度函数曲线，得到新型设备的维修阈值，进而建立新型设备的视情维修决策模型，最后以船用柴油机监测数据进行了实例验证。结果表明，该方法能有效挖掘旧型柴油机的历史数据信息，充分利用新旧柴油机之间的内在联系，为灰信息条件下新设备的视情维修决策提供了新途径。

Study on Condition Based Maintenance Decision Method Based on Gray Information

For new equipment, Condition Based Maintenance decision model is difficult to establish in the case of lacking historical failure information and maintenance threshold. An improved coefficient was defined which indicates the difference of a basic failure rate based on the trends consistency of old and new equipment failure rate, and the failure rate function was obtained. Then the new equipment risk function curve was fit with Support Vector Machine regression, further, the maintenance threshold and Condition Based Maintenance decision model of new equipment were established. Finally, the monitoring data of marine diesel engine were analyzed as an example using the method mentioned above. The result shows that approached method can effectively tap the old equipment historical data information and make full use of the intrinsic relation between the old and new equipment. It provides a new way for condition based maintenance under gray information.