基于灰色模型和模糊神经网络的综合水质预测模型研究

水质状态变化趋势预测研究对水资源管理和维护具有重要的现实意义。提出了一种将灰色模型和模糊神经网络相结合的水质预测模型。首先基于改进的灰色模型预测出水体中各理化因子在未来一段时间内的指标变化,然后采用T-S模糊神经网络对各单因子的预测值进行数据融合,构建水质变化综合趋势预测模型,预测出下一时间段的水质整体状态指标。实验表明,这种方式用来预测湖泊水质变化趋势具有可行性;与BP网络模型相比,基于T-S模糊神经网络系统的模型具有预测精度高、模型系统稳定等优越性。

Comprehensive prediction model of water quality based on Grey Model and Fuzzy Neural Network

Trend prediction for the changes of water quality plays an important role in the field of water management and maintenances. In this paper, a predictive model for water quality prediction system combined Grey model and fuzzy neural network model was proposed. First, the change trends in a period of time in the future for various physical and chemical factors in water environment were predicted based on improved Grey model. Then T-S fuzzy neural network was adopted to fuse the prediction value of various single factors, and the prediction model of changing trends of water quality was structured to predict the whole status index of water quality in the next time. The experimental results show that this approach can be used to predict the changing trends of water quality. Compared with the BP network model, T-S fuzzy neural network model has the advantages with higher prediction accuracy and more stable model system.