Study On The Method Of Water Quality Control In Water Works Based On Support Vector Machine SVM

The safety of residential water supply is an important issue related to the national economy and people's livelihood.And it is the most important part that the quality of factory water reaches the standard safely.At present,the factory water quality control usually adopts the extensive and relatively fixed conventional mode under artificial control,and lacks the technical measures for the interaction between the water quality of the source and the process operation,resulting in the increase of energy consumption and the instability of the water quality of the factory water.This model will consume a large amount of water purification materials and electricity.That not only reduced the economic benefits of water plants,but also caused additional burden to the environment.With the rapid development of monitoring technology,there are more and more process operation parameters and water quality data in water plants.It is feasible to use big data analysis and mining methods to guide the timely adjustment and optimization of water purification process,which is of great significance for water plants to minimize water quality risk and reduce energy consumption.Based on the research of data mining technology and the collection of operating rules and operating parameters of water plants,this project aimed at the stability of effluent water quality of water plants.Under the known inflow conditions,the optimal dosage and timing of dosage were realized to ensure the safe water supply under different raw water states.According to the research status of water quality prediction and control at home and abroad,the precision and accuracy of the quality prediction model established by SVM are higher,and it is more suitable for the multi-element and time-delay water quality control prediction model.The main contents of this project include:(1)The model of water quality control system based on support vector machine was constructed.Through the analysis of water quality control process,it was concluded that the key factors of water quality control,which optimized the on-off control of dosage and dosing time.Then the input variables of the prediction model are obtained by analyzing the influencing factors of water quality control.Then,the theoretical model of support vector machine was analyzed to establish the mathematical model of water quality control in this project.(2)The quality online testing system of finished water was constructed.The basic principle and hardware and software environment of the inspection system for the water quality control system of the factory were discussed.Firstly,the basic principle of water quality control system was introduced,the parameter setting of water quality control monitoring system was analyzed,and a numerical acquisition system was established to realize the real-time acquisition and storage of operation parameters of water plants,and the hardware environment of the detection system was built.(3)The water quality control method based on support vector machine were applied and verified.Taking a water plant in Zhejiang Province as the research object,the water quality control system was constructed,and data acquisition was carried out.The model was trained by using the real data in the process of water purification,and different water quality data are randomly selected to test the model.Taking the main pipe pressure,source water flow rate,source water turbidity,residual chlorine of source water,source water PH,turbidity of filtered water,residual chlorine of filtered water and PH of filtered water as variables,the dosage control model was established by using MATLAB simulation tool.The model was validated by simulation and application results.The results showed that the PAC automatic dosing model based on LS-SVM support vector machine was reliable for field water quality control.In this paper,the effect of dosing on the quality of filtered water was considered to have a certain time difference.The model was very practical.The model was validated by simulation and application results.The results showes that the PAC automatic dosing model based on LS-SVM support vector machine was reliable for field water quality control.Applying the research results of this subject to on-line monitoring and intelligent decision-making control of water-making process and dosing in Waterworks can realize accurate,fast and convenient real-time control of water quality on site,and the turbidity value of outgoing water was lower and the water quality was better.It can guarantee the quality of outgoing water and the safety of water supply under different urban raw water conditions.