Research On Sewage Abnormity Diagnose Intelligent Algorithm And Its Application

Sewage treatment is an important part of the socialist industrialization. Because of thecomplexity of water elements and treatment processes, the key point of wastewater plant is howto maintain the daily operation. However, with the application of modern equipment, thetraditional sewage fault diagnose shows its limitation. How to use effective method to solve thisproblem, then reduce the manpower costs as well as improve the capability of processing thewastewater would contributes the smooth running of wastewater plant.Recently, the intelligent algorithm based on computer technology has been widelyresearched and applied in traditional industries. Using intelligent algorithm to simulate dataanalysis and manipulation, the process of diagnose in wastewater treatment, and then design a setof automatic sewage treatment diagnosis system becomes an effective way to solve the problemand replace the manual operation.In this paper we adopt intelligent algorithms including expert system, data mining and roughset theory as the research theories. Based on these theories, we analyzed sewage data andmodeled the treatment process. Then we designed the online diagnose system to replace thetraditional diagnose method. The main research work is as follows:Firstly, focused on the sewage abnormity diagnose, we proposed and built the sewage expertsystem model based on certain factors. Expert system is an intelligent computer system tosimulate the diagnose process of human expert. We used the ES to model the sewage expertdiagnostic procedure, transferred the experts’ diagnose experience and operation procedures intocomputer recognition rules and then stored them into database. During this process, weintroduced certain factors to regulate the rule conflicts and then designed the sewage treatmentexpert system reasoning engine.Focused on the problem of knowledge acquisition, we made an intensive research andintroduced association rule mining algorithm. It is used to mining rules correlation with sewage abnormal state. In order to improve the efficiency of Apriori algorithm, we brought in theknowledge reduction based on rough set theory to optimize the mining method. Based on that, weremoved the irrelevant properties and use the K-means method to disperse the data. As the sametime, we applied the optimized algorithm on UCI data sets and evaluated the results.At last, according to the actual operation logic and treatment process in a sewage plant inJilin, we used the theory of ES and association rule mining, adopted SSH2web lightweightsoftware architecture to design a set of automatic online wastewater decision system. Thisindicated that our research theories have application values on sewage treatment.With the development of artificial intelligence, the combination of AI and traditionalindustries will be more and more closely. In this paper, we introduced machine learningalgorithms to get a good solution in increasing wastewater diagnose efficiency. Through thepractical application, we verified the feasibility of the method. Meanwhile, this paper couldprovide some reference in both intelligent engineering and intelligent sewage treatment.