基于最小径集的系统故障概率分配方法研究

A new approach to the study of system failure probability distribution based on the minimum path set

为准确、快捷地解决在复杂系统中的可靠性分配问题,在基于最小割集的可靠性分配方法的基础上建立了基于最小径集的系统故障概率分配模型。该模型的分配思想不是对系统所有的单元都进行分配,而是选择最有可能导致系统发生故障的单元进行分配。最小径集分配法只需对最小径集中的基本事件直接进行分配计算,从而大幅度减少了中间计算工作量。它的另一个突出优点是能提供多个可选择的分配方案,因为每一个最小径集都代表1种分配方案。根据安全系统工程的优化思想,提出了不可靠度分配方案的选择原则:在基本事件数相同的最小径集中选择概率较大的最小径集;在概率接近的最小径集中选基本事件数较少的最小径集。对于最小径集中基本事件的分配,根据概率重要度来分配是合理的,因为重要度越大的基本事件对系统的影响就越大,分配的概率就应该越小。通过矿井火灾事故树的举例,进一步描述了基于最小径集的系统故障概率分配方法的定量计算过程。研究表明,基于最小径集的系统故障概率分配法是工作量最小的分配方法,还可以进行分配方案的比较选择。

The present paper takes its main interest in introducing our new approach to the study of system failure probability distribution based on the minimum path set. The distribution theory of this model is not to allocate all the units, but to choose the unit which is most likely to cause the system to get damaged. As is known, it is possible to work out the distribution of the probabilities of all the basic events directly by using the minimum path set, and in turn to reduce the calculation workload significantly. Another prominent advantage of using the minimum path set is to provide a certain options for the distribution scheme. According to the optimization idea of the system safety engineering, we have to give a selection principle of the failure probability distribution. If the number of the basic events in the path sets is same, it is also favorable to select the minimum path set which may have a larger failure probability. However, if the failure probability values for all the path sets prove to be approximately equal, it can also choose the minimum path set which has less number of basic events. For according to the probability importance to distribution, the distribution of the basic events in the minimum path set is reasonable. Since the basic event that has greater importance may have bigger impact upon the system, the probability distribution should be smaller. In order to further describe the quantitative calculation process of the failure probability distribution method, a good example to take is the fire hazard probability distribution in a mine. Thus, it can be seen that, with the help of this failure probability distribution method proposed in this paper, the allocation workload can be reduced to the least through comparison and choice of the distribution scheme. If so, the safety-production managers will be in a position to make a right and reasonable decision to control the system risk more effectively and trouble-savingly.