Application Of SVM-FuzzyES In Fault Diagnosis Of Wastewater Treatment Technics

Nowadays, the existing fault diagnosis theories and technologies are facing the difficult challenge that is brought by the complexities of process control, so studying the fault technology of complex process has an important and practical meaning for ensuring the safe and stable running and improving economy of the large flow control industry. Hence, taking the wastewater treatment on activated sludge process as research subjects, this dissertation researches the application of intelligent fault diagnosis technology.Firstly, an framework of the intelligent integrated fault diagnosis modeling, which consisted by "fault detecting fault diagnosis", is proposed and constructed based on fuzzy logic, support vector machine and expert system: to get the fault area, the support vector machine algorithms are used in fault detecting of wastewater treatment processes; to get the exactly fault position, cause and countermeasure, a fuzzy expert system is used fault diagnosis of wastewater treatment on activated sludge process.Next, with libsvm tool, some algorithms are concentrated and used in classifying operation state of wastewater treatment processes. The numerical experiments show the high classification accuracy and these algorithms are effective.After that, by analyzing the fault types and parameters of operation state for wastewater treatment on activated sludge process, a fuzzy expert system for fault diagnosis is designed in detail. The fuzzy reasoning of this fuzzy expert system is implemented based on Fuzzy Jess, and applies to fault diagnosis of activated sludge process. The test shows that the expert system diagnoses fault with the level of wastewater treatment expert. And the ways to improve its performance are also presented.Through the application of support vector machine and fuzzy expert system in fault detecting and diagnosis for wastewater treatment, the dissertation proves the validity of the intelligent integrated fault diagnosis modeling and the correction of application technology route, tools and methods. Meanwhile, the dissertation proposes one intellectualized solution blue print in the field of intellectualized fault diagnosis of wastewater treatment, which is provided with self-adaptation, reasoning to solve complicated problem that can be only solved by expert, and the blue print is provided with innovation.