摘要
本文以沈阳市M水厂的月供水量预测为背景,阐明了预处理建模数据序列的必要性,并提出一种新的供水量预测方法,即一次累加-指数平滑法,在简介了一次累加-指数平滑模型和自适应过滤模型的基本原理之后,进行了实例计算与分析.研究结果表明,一次累加-指数平滑法和自适应过滤法可分别作为月供水量的短期和长期预测模型,两者的预测误差均不超过0.69万吨.
Taking the water supply prediction of M water treatment plant at Shenyang as background,the essentiality of pre-processing the data used to determine the parameters of the forecasting model is clarified in this paper. Thus, a new prediction method of water supply quantity-simple accumulation exponent smoothing is put,forward. The fundamentals of both the simple aceumulatio. exponent smoothing and self-adaptive filtering and their calculating methods are briefly introduced. Moreover,two kinds of forecasting are calculated and analysed with the data of M water treatment plant. The research result shows that the simple accumulation exponent smoothing and self-adaptive filtering can be used as the short-term and. long-term prediction models of water supply quantity respectively and that none of the maximum values of the two forecasting errors exceed 0. 69 ten-thousand tons.
关键词
城市
供水量
预测模型
供水系统
优化调度
urban water supply,forecasting,self-adaptive filtering .simple accumulation exponent smoothing