基于多元混合时序分析及灰色理论的水华藻类生长阶段多因素预测方法

Water bloom multi-factor prediction in algae growth stages based on multidimensional time series analysis and grey theory

在水华防治工作中,水华预测一直都是1个难点,由于藻类水华整个生成过程各阶段的变化机理不尽相同,现有预测方法采用某1个模型对水华生成的全过程进行建模预测并不合理。因此本文针对水华生成过程中持续时间最长且变化规律明显的藻类生长阶段,采用多元时序分析技术,通过对藻类生长阶段中的多特征因素时序建模分析,以影响和表征藻类生长的特征因素作为模型输入和输出,考虑藻类生长阶段的趋势性、周期性、随机性变化规律,建立多元混合时序模型,并结合灰色理论思想,改进模型预测算法,从而提出基于多元混合时序分析及灰色理论的水华藻类生长阶段多因素预测方法,以实现对包含影响因素在内的特征因素变化特性的全面描述和预测以及对模型预测精度的进一步提高。采用本文方法及传统方法分别对江苏太湖水华特征因素监测数据进行建模预测,结果表明,基于本文方法的水华特征因素预测结果与实测结果更相符、预测平均误差绝对值更小。

In water bloom prevention and control, water bloom prediction is always a difficult problem. For the change mechanism difference in each stage of the algal bloom formation process, modeling and prediction the whole process of algal bloom formation by a certain model as the existing prediction methods is not reasonable. Hence, this paper proposes a new water bloom prediction method based on multiple characteristic factors time series analysis and grey theory in algae growth stages. In this paper, multiple characteristic factors time series is described by multidimensional time series hybrid model, which describes not only deterministic variation trend but also the integrated effect of multiple characteristic factors along with periodicity and random effect of environmental variables. A new prediction method is put forward by grey theory based on multidimensional time series hybrid model to solve the problem that existing bloom prediction is not accurate enough. Comparing the proposed method with other traditional methods, such as one-dimensional time series analysis or time series analysis without grey theory, it has been found that prediction based on multidimensional time series hybrid model and grey theory is useful and accurate for establishing multiple characteristic factors time series of water bloom.