Study on Hot Spring Quality pH Value Forecast Based on Regression Supporting Vector Machine
CHEN Wufen;ZHANG Qianhua;HUANG Zhengwu;HUANG Li;LIN Nianwang;Pearl River Water Resources Commission of the Ministry of Water Resources;
Using Enping Jinjiang Hot Spring dynamic monitoring system engineering to acquire data that has temperature, dissolved oxygen, conductivity, and turbidity as the impact factor. Then Regression Support Vector Machine pH value of water quality prediction model is built. Through Jinjiang Hot Spring 18 No. of study, modeling and forecasting analysis of the spa water quality pH value is proceeded. Simulation results show that the pH value forecast mode training set based on SVR is 0. 854,and the pH value test set decision coefficient based on SVR is 0. 897. The average relative error of test set is 1. 419%. The model provides reference for water quality evaluation, the water environment monitoring and management. At the same time, it has some practical value.
【CateGory Index】： P314.1