Research On Hierarchical Directed Graph Model Fault Diagnosis Approach Based On Rough Set And Its Application

The thermal power unit faults have a large proportion of in a thermal system, which has a significant impact on the economical safety of the unit as a whole run. Because of the complex structure of the thermodynamic system and the complex system operating conditions, it is difficult to establish and perfect the knowledge base in the process of fault diagnosis, so the thermal system fault diagnosis research is of great significance.The combination graph theory and rough graph on the basis of the SDQ which make SDG model be easier to diagnosis, thereby improving the accuracy of the SDG fault diagnosis. Firstly ambiguity reasoning of the SDG will lead to many problems that the resolution is low, the slow diagnosis speed and poor real-time of large-scale systems, a hierarchical directed graph algorithm is proposed. This dissertation proposes a novel qualitative diagnosis approach based on hierarchical digraph integrating the goodness of the fault propagation graph and SDG (signed directed graph), the model has the better completeness and explanation facility of SDG, and can express the connection relation among all components. After that, four diagnosis methods for selecting optimal measuring points are developed which avoid inefficient testing in diagnosis process to increase diagnosis efficiency. Finally the hierarchical techniques, constraint propagation, intermediate optimization and the fault backward reasoning of SDG are combined to locate the fault of the system. The algorithm can distinguish unknown faults in the same fault mode. The method can avoid the possible "combinatorial explosion" problem. The conclusion is compared to the traditional SDG diagnostic results, which shows that the method is superior to the traditional method. The methodology is applied to a centrifugal pump and liquid level system for a thermal power plant. The diagnosis simulation result show the approach is valid.Secondly, concerning the characteristic that the information system obtained from doing faults analysis by the SDG is rough and inaccurate, this paper added the rough set mining the advantage of uncertain relationship to the SDG fault diagnosis process and proposed the hierarchical SDG fault diagnosis model based on the rough graph. The model consists of three parts of main contents including the component node rough correlation graph, rough fault graph generation algorithm and the fault maximum flow analysis algorithm. The example validation and model advantage analysis showed that the model can reflect the actual situation better than the traditional SDG fault diagnosis model to better reflect the actual situation, which diagnoses results received is more accurate. Finally, the hierarchical SDG fault diagnosis algorithm based on rough graph used to handle the different relationship levels fault of the thermodynamic system. The result shows that the model can reduce the complexity of dynamic characteristics of the thermal equipment which fault knowledge is difficult to obtain. And then the model can accurately identify gentle and early faults of the thermal system to greatly improve diagnostics efficiency of the thermal system, which promotes the progress of the thermal system fault diagnosis technology.The paper will introduce hierarchical theory and rough graph into the SDG fault diagnosis, which is proposed hierarchical SDG fault diagnosis algorithm based on rough graph. The algorithm in a large part reduces and even avoids the conflict probability of the SDG search reasoning algorithm to improve the reasoning resolution. The method is effective and has a good versatility, which has a very important role to improve the safety and economy of the unit operation, and which played a certain role in promoting the development of intelligent fault diagnosis.