地表水源水质预测方法研究

Forecasting methods of surface water quality

对时间序列、灰色系统、人工神经网络等水质预测方法及其特点进行了分析、探讨,举实例说明各种预测方法的特点,据此将预测方法分为适用于短期预测的统计模型法和长短期预测均适用的非统计模型法;同时对解读水质预测模型内在水质变化机理的研究新动向进行了分析讨论,提出了按照预测的实际情况和要求具体选择预测方法的流程.

This paper analyses and discusses such forecasting methods of water qualities as time series methods, grey system methods, artificial neural network methods at home and abroad. Characteristics of all kinds of forecasting mehtods are commented by examples, and forecasting methods are divided into two parts accrding to above analysis. One is the statistic model method fit for short-term forecasting, and the other is the non-statistic model fit for short-term and long-term forecasting. The paper also outlines the up-to-date research direction of extracting the variation mechanism of water qualities from water quality model. A flow chart of slecting forecast methods is developed according to the actual conditions and demands.