基于数据挖掘的静变电源故障预测研究

静变电源是机场桥载设备最重要的组成部件之一,应用非常广泛,但与此同时,其产生的高频故障会造成设备利用率低、修复率时长和重大经济损失等问题,在基于桥载设备的安监系统上,设计了静变电源安监信息采集节点,通过数据挖掘软件,建立了静变电源故障预测模型.经过比较Apriori算法和其他典型数据挖掘算法的性能,结合在线数据库测试结果表明,得到了不同的典型算法在预测静变电源故障上显现的特点,通过选择最优算法高效地预测了静变电源的未来状态,实现了对静变电源的实时故障预测,进而为解决故障提供了方向和目标,最终达到了降低经济损失最大化的目的,具有很深的实际意义.

Study of Fault Prediction on Static Power Based on Data Mining

Static power is one of the most important part of the bridge born equipment on airport.By structuring the model of fault prediction and the node of safety monitoring information collecting with the data mining software,frequent fault appearing on static power can be effectively avoided.The frequent fault will usually arise a series of problems,including low utilization rate,long repairing rate time and severely economic losses and so on.By comparing Apriori algorithm with other typical algorithms on the basis of online database,it can be concluded that the obvious characteristics emerge in predicting the future state of static power.Moreover,the optimal algorithm that can effectively predict the future state of static power and make the real time error prediction come ture.Furthermore,the model can provide error solution with direction and goal.Finally,the purpose that economic losses can be reduced to maximization is achieved,which will have considerably practical significance.