Study On Method And Theory Based On Intellectualized Diagnosis Of Gasoline Engines Using Dynamic Causality Diagram

With the increasing improvement of the automatic manufacture and the rapid development of the modern science and technology,The human society has entered into a information age. As people are confronting the information explosion, speedy transferring and various complexities, it becomes more difficult and important to get these information, even fully process and manage them. In the traditional field of industrial fault diagnosis, it only depends on very limited skills sometimes. But as the data relation turns out to be more complex than ever, the capacity of data needs to be increased, so we introduce intelligence into computers. The high growing of computer offers the material basis for developing of artificial intelligence, that's the reason why people adopt artificial intelligence widely to solve the industrial faults.This paper mainly deals with the research of the probability knowledge expressing method and reasoning computation principle about Dynamic Causality Diagram brought forward by Pro. Qingzhang. After introducing the basic concept of fault diagnosis and the shortages of actual fault diagnosis methods, this pager conducts comprehensive research aiming at the reasoning algorism of causality diagram, which has single value or multiple value. The main theory work of this paper is as follows:1. The introduction of conventional reasoning algorism of dynamic causality diagram, which has single value.2. The discussion of the difficulty that the dynamic causality diagram of multiple value cannot strictly conform to the probability theory and the reasoning result could be incorrect using it to analyze the actual problem.3. Presenting a causality infection distribution-reasoning algorithm to dynamic diagram, which has multiple value.4. In order to improve the reasoning algorithm in multiple value causality diagram, which could not deal with the fuzzy case, a fuzzy reasoning algorithm was presented. It extended the definition of the multiple value causality diagram with fuzzy. The fuzzy mapping relation between every event variable and every reader variable was made. So, the fuzzy knowledge can be expressed with membership functions. This algorithm can deal with fuzzy causality diagram, which has multiple value, and effectively express the fuzzy uncertain knowledge in practice.After the introduction of the reasoning procedures about dynamic causality network,this paper concludes the present difficulties of it, then presents some achievements we have gotten aiming at these difficulties and brings forward the future development directions.In order to apply the theory of Dynamic Causality Diagram into practice, this paper also deals with the research that using dynamic causality diagram to analyze and diagnose the faults of gasoline engines. The diagnosis platform has been constructed, and the main fault mode of gasoline engines has been concluded into the corresponding nodes of dynamic causality diagram. The diagnosis result coincides with the practice, the speed of diagnose and the result are satisfied. In the end this paper summarizes the research work.