SVR WATER QUALITY PREDICTION MODEL BASED ON GA OPTIMIZATION

Aiming at the fact that the BOD₅ in wastewater can not easily measured directly, a support vector machine regression water quality prediction model based on genetic algorithm optimization parameters was proposed. Using machine learning method, BOD₅ was determined by establishing a mathematical relationship model between COD and other parameters in wastewater. The genetic algorithm was used to optimize the key parameters in the SVR, which solved the problem of parameter selection of the traditional SVR prediction model. Experiments were carried out with the influent wastewater from the Beijing Wastewater Treatment Plant as the research object. The results showed that the average error and root mean square error of the results predicted by GA-SVR method were reduced to 0.009443 and 16.88 mg/L, respectively. Compared with BP neural network and the SVR,the result of GA-SVR was with reasonable advantage.