论文部分内容阅读
提出了一个新的方法用来确定当设备发生退化故障征兆后的最优停车维修时间。优化的目标是使从出现故障征兆到停车检查这段时间内的费用率最小。影响该费用率的两个变量是故障率和故障特征量小于最大允许值的概率,通过假设从出现故障征兆到发生功能性故障的时间的概率分布为已知,可以得到故障率;分两步估计,第一步通过神经网络预测退化故障过程的退化量及增量,第二步对退化增量密度函数参数进行估计,这两步的执行都是实时动态进行的,最佳时间确定过程也采取动态决策方式。一个仿真例子表明了该方法的可行性。
A new method is proposed to determine the optimal maintenance time after a device has experienced signs of degradation failure. The goal of optimization is to minimize the rate of expense from the time a symptom of a failure occurs to the time of a parking check. The two variables that affect the cost rate are the probability that the failure rate and failure characteristic amount are less than the maximum allowable value, and the failure rate can be obtained by assuming that the probability distribution from failure symptom to occurrence of functional failure is known; in two steps It is estimated that the first step is to predict the degradation and increment of the degenerative fault process through the neural network and the second step to estimate the degenerated incremental density function parameters. The execution of these two steps is carried out in real time and dynamically, and the optimal time determination process Take dynamic decision-making. A simulation example shows the feasibility of this method.