论文部分内容阅读
针对机坪地面空调间歇故障引起的使用效能低、维修滞后等问题,提出了二次关联累加数组(AS)-Apriori与聚类K-means相结合的间歇故障预测方法,并基于此实现了延误维修预测.其中:AS-Apriori算法解决了Apriori频繁扫描事务库的低效问题,通过实时构造间歇故障数组并对其对应项累加求和;延误维修预测是为了估计出永久故障临界区以安排合理维修,可采用正态分布求出不同间歇故障变量的维修波及延误概率并进行依次累加而实现.验证表明,ASApriori提高了运行效率,且二次关联规则支持度提升了20.656个百分点,能更准确预测间歇故障,同时参照数据分析,预测的维修波及延误累加概率呈线性分布,即可预测性高的间歇故障更便于预先维护管理,减少永久故障的形成.
Aiming at the problems of low utilization efficiency and lag in maintenance caused by intermittent airfield apron faults, an intermittent fault prediction method based on quadratic associative additive array (AS) -Apriori and clustering K-means is proposed and delayed Maintenance prediction.Among them: AS-Apriori algorithm solves the problem of inefficient Apriori frequent scan transaction library by constructing intermittent fault array in real time and summing up its corresponding items; delay maintenance prediction is to estimate the permanent fault critical section to arrange the reasonable Maintenance can be achieved by using the normal distribution to find out the delay probability of maintenance-induced delays and accumulate them sequentially.The verification shows that ASApriori improves the operating efficiency and the support for secondary association rules is improved by 20.656 percentage points, which can be more accurate Prediction of intermittent faults, meanwhile, with reference to data analysis, predicted cumulative delays and linear accumulated probabilities are predictable. Intermittent faults with high predictability are easier to maintain and reduce the formation of permanent faults.