| With the development of science and technology, the area of people’s research is constantly expanding. We find the information in the real world is not only the certain information, but also a lot of uncertain information, so it is very important to research uncertain information. Causality diagram is developed on the basis of reliability of online. which is a kind of reasoning method based on probability theory, it has a unique advantage in fault diagnosis. When we use causality diagram for diagnosis, we usually for a particular event in the known evidence of the posterior probability, that is the first cut sets of the event, the final cut sets, non intersection cut sets and the posterior probability. But in the real world, it is difficult to get exact probability of the event because of the environment and human factors, in order to find fault cause and improve the system of weak links, in this paper, we introduce the binary decision diagram(BDD)and T-S model to the causality diagram, the main innovation is reflected in the following two parts.The first part, at first, we should find the minimum cuts and sort of event causality diagram, so we can get the causality diagram BDD, based on the study of BDD, we can get minimum cuts and sort of event causality diagram and find the influence of basic events on causality diagram. In the actual operation we can reduce the diagnosis space and analyze fault diagnosis rapidly. At last, we prove the feasibility and applicability of the method according to the example.The second part, the T-S model is applied to the causality diagram, it essence is to use fuzzy number to replace the failure probability of event, T-S fuzzy door instead of causality diagram logic gate. We can get the fuzzy probability of event fault according to the given rules and T-S algorithm, so we can analyze the most important factor and improve the weak link of the system. Through examples, this method is effective and feasible in causality diagram. |
PDF Full Text Request