Probability of Extreme Learning Machine Applications in Wastewater Treatment Operating Mode Recognition

Abstract

Sewage treatment process of central plains water instantaneous changeable and sewage treatment system resistance and load ability, often leads to the effluent quality can't satisfy the design index. Effective identification of sewage operation condition changes on the optimization of sewage treatment and safe system operation is very important. In this paper, by using the theory of extreme learning machine set up sewage treatment process of instantaneous variables and operating conditions, the prediction model according to the average output value, respectively establish sample function, determine the conditional probability density function. According to the principle of Bayesian computation of posterior probability and finally determine the sample classification. The model was validated using test data of desulfurization wastewater waste gas washing. The results show that, the model predicted results can meet the engineering requirements.