Experimental Study on Dynamic Simulation for Biofouling Resistance Prediction by Least Squares Support Vector Machine

In consideration of influencing factors of biofouling in the cooling water system, a new method of dynamic analog test of cooling water biofouling resistance for shell-tube heat exchanger was presented. The test conditions are as follows: the tube material and process are identical, the water temperature(30 ̊C) and flow rate (0.4m.s-1) both are constant. According to the degree of influence of pH, conductivity, total number of bacteria, dissolved oxygen, TN, NH3-N on the fouling resistance, this paper selected these water quality factors as the input variable, fouling resistance as the output variable, found a prediction model of cooling water fouling resistance based on Least Squares Support Vector Machine(LS-SVM). Data from another running period was used to verify model prediction accuracy. The mean square error of prediction model is 3.768%; the BP neural network is 9.223%. The results show that the LS-SVM model is pithily, and it has better extensive capability than traditional methods. The new method can be viewed as a new approach to advance the development of cooling water treatment technology and improve the prediction accuracy of the biofouling resistance.