Study On The Expert System Of Fault Diagnosis For Seawater Reverse Osmosis

With the development of economy, the world is being faced with increasing water shortage disaster. And it has restrained the social development. By developing seawater desalination technology, it will have important meaning to alleviate the current shortage of water resources. As an advanced water treatment technology, reverse osmosis membrane separation technology has less space demand, lower energy consumption, small investment and high separation efficiency. And it is widely used in seawater desalination industry and brackish water desalination industry etc. The development of membrane desalination project is trending large, continuous and intelligent, so the traditional fault diagnosis technology will not be able to meet the growing demand of system reliability. Therefore there is important significance to explore an intelligent fault diagnosis system which is supported by advanced science and technology.Based on comparing these common fault diagnosis methods, the expert system based on fault tree analysis is selected as the main fault diagnosis method for reverse osmosis seawater desalination system. The design of system's framework and structure is identified through analyzing the function requirement of expert system. Based on the detailed analysis of the symptoms and fault cause of the key components and subsystems, the fault tree model was established. Meanwhile, it is used as the method of knowledge acquisition. Knowledge Base is established by adopting framework-rule mixed knowledge representation method. The Diagnosis system selects the method of Pros and cons of hybrid reasoning as the main control strategy of the inference engine. And the efficiency of the inference engine is enhanced by using the strategies of Heuristic search and Conflict resolution, which is related with Statistical Analysis of Failure Case History and Quantitative fault tree analysis.The expert system software is developed by the OOP language of C# and SQL Sever2005. It achieves the main functional modules and establishes knowledge base and integrated database. Then Fault diagnosis expert system is tested through inputting simulated fault, and the result is ideal.