State Key Laboratory of Hydrology-Water resources and Hydraulic Engineering, Nanjing Hydraulic Research Institute, Nanjing, 210029, China.
Faculty of Civil Engineering Osijek, University J.J. Strossmayer in Osijek, Osijek, Croatia.
Environ Sci Pollut Res Int. 2019 Apr;26(12):12622-12630. doi: 10.1007/s11356-019-04716-y. Epub 2019 Mar 20.
River water temperature (RWT) forecasting is important for the management of stream ecology. In this paper, a new method based on coupling of wavelet transformation (WT) and artificial intelligence (AI) techniques, including multilayer perceptron neural network (MLPNN) and adaptive neural-fuzzy inference system (ANFIS) for RWT prediction is proposed. The performances of the hybrid models are compared with regular MLPNN and ANFIS models and multiple linear regression (MLR) models for RWT forecasting in two river stations in the Drava River, Croatia. Model performance was evaluated using the coefficient of correlation (R), the Willmott index of agreement (d), the root mean squared error (RMSE), and the mean absolute error (MAE). Results indicate that the combination of WT and AI models (WTMLPNN and WTANFIS) yield better models than the conventional forecasting models for RWT simulation for both regular periods and heatwave events. The MLPNN and ANFIS models outperform the MLR models for RWT simulation for the studied river stations. RMSE values of WTMLPNN2 and WTANFIS2 models range from 1.127 to 1.286 °C, and 1.216 to 1.491 °C for the Botovo and Donji Miholjac stations respectively. Additionally, modeling results further confirm the importance of the day of year (DOY) on the thermal dynamics of the river. The results of this study indicate the potential of coupling of WT and MLPNN, ANFIS models in forecasting RWT.
河流水温（RWT）预测对于溪流生态管理至关重要。在本文中，提出了一种基于小波变换（WT）和人工智能（AI）技术耦合的新方法，包括多层感知机神经网络（MLPNN）和自适应神经模糊推理系统（ANFIS），用于 RWT 预测。在克罗地亚德拉瓦河的两个河站，将混合模型的性能与常规 MLPNN 和 ANFIS 模型以及 RWT 预测的多元线性回归（MLR）模型进行了比较。使用相关系数（R）、Willmott 一致性指数（d）、均方根误差（RMSE）和平均绝对误差（MAE）评估模型性能。结果表明，WT 和 AI 模型（WTMLPNN 和 WTANFIS）的组合比常规预测模型在 RWT 模拟中表现更好，无论是在常规时期还是热浪事件中。对于研究河站的 RWT 模拟，MLPNN 和 ANFIS 模型优于 MLR 模型。WTMLPNN2 和 WTANFIS2 模型的 RMSE 值分别在 1.127 到 1.286°C 和 1.216 到 1.491°C 之间。此外，建模结果进一步证实了日历年（DOY）对河流热动力学的重要性。本研究结果表明，WT 和 MLPNN、ANFIS 模型耦合在 RWT 预测中的潜力。
