论文部分内容阅读
引入共轭梯度算法对传统WANN模型求解搜索进行优化,从而提高传统WANN模型计算效率,改善模型求解精度。本文将改进的WANN模型用于新疆某区域中长期水文预报中,并结合区域实测水文数据,对比分析改进的WANN模型在中长期水文预报中的适用性以及预报精度。研究结果表明:相比于传统WANN模型,改进的WANN模型可提高中长期水文预报的精度,模型在中长期降水量和水量的预报误差分别减少8.8%和3.6%,收敛度分别提高0.18和0.17。研究成果对于地区中长期水文预报方法提供参考价值。
Conjugate gradient algorithm is introduced to optimize the traditional WANN model search, so as to improve the computational efficiency of the traditional WANN model and improve the accuracy of the model. In this paper, the improved WANN model is applied to mid- and long-term hydrological forecasting in a certain area of Xinjiang. In combination with the measured hydrological data in the region, the applicability and accuracy of the improved WANN model in mid- and long-term hydrological forecasting are compared and analyzed. The results show that compared with the traditional WANN model, the improved WANN model can improve the accuracy of medium- and long-term hydrological forecasting. The prediction error of the model in mid- and long-term precipitation and water decreases by 8.8% and 3.6% respectively, and the convergence increases by 0.18 and 0.17 . The research results provide reference value for the medium- and long-term hydrological forecasting methods.