基于ARMA模型对乡镇污水厂进水量的预测研究

The Prediction Research of Town Sewage Plant in Water Based on ARMA Model

乡镇污水的来源较多(包含工业污水、生活污水和自然降水量),其时间序列不平稳,受干扰严重,现有的预测模型对乡镇污水的进水量预测精确度不高。针对此问题,本文提出基于卡尔曼滤波的污水预测模型,通过滤波去除干扰,并结合自适应机制建立了ARMA模型的滚动预测,对进水量进行预测的方法。同时将本文提出的预测模型用于某市污水处理厂进水量的预测。通过工业案例分析,本预测模型的预测平均绝对误差小于5%,预测值与实测值基本一致,为进一步进行污水处理厂的曝气量优化控制提供了前提。

There are many sources of municipal sewage, including industrial wastewater, sewage and natural precipi- tation, and its time series is not stable, serious interference, the existing prediction model of municipal sewage wa- ter quantity prediction accuracy is not high. To solve this problem, this paper proposes a short - term prediction model method for short - term prediction of water inflow, based on Calman filter, removing the interference through the filter, and setting up a rolling prediction of ARMA model combined with the adaptive mechanism. At the same time, the prediction model proposed in this paper is used to predict the influent of Heyuan wastewater treatment plant. Through the industrial case analysis, the average absolute error of the prediction model is less than 5 % , and the predicted value is basically the same as the measured value, which provides the premise for the further optimi- zation of the aeration capacity of the sewage treatment plant.