Study on water quality prediction model of centralized drinking water wellhead

SVM and BP water quality prediction models were established following the basic theories for support vector machine (SVM) and artificial neural network (BP). RS-SVM and RS-BP water quality prediction models were established combining the complementary nature of rough set (RS), support vector machine (SVM), and artificial neural network (BP). These models were applied to the Gucheng Lake centralized drinking water wellhead. The results show that the water quality prediction model based on SVM has higher accuracy than the water quality prediction model based on BP predicted, so the RS-SVM prediction model is the best to predict water quality.