用非线性模型对循环水pH值动态预报控制

Dynamic Forecasting Control of pH in Circulating Water by a Nonlinear Model

在用线性时间序列模型对循环冷却水pH值动态变化规律研究的基础上,作者对水质各种参数进行了研究,为了提高预报精度,讨论了几种典型的非线性时间序列模型,并提出一种复合自回归时间序列模型,其平均相对预报误差可达2％左右。

With a linear time sequence model, a study on the rule of dynamic variation of pH value in circulating cooling water is presented. Furthermore, based on above work, various parameters describing water quality are explored. In order to improve forecast accuracy, a complex time sequence model with autoregression is proposed from several typical nonlinear time sequence models. It results in about 2% of average relative forecasting error.