基于灰狼优化算法的灌区流量预测模型研究

Research on irrigation area discharge prediction model based on grey wolf optimization algorithm

快速准确的灌区流量预测对于水资源分配有着重要依据。现有的流量预测模型一般采用传统方法或是简单的神经网络模型,采用上述方式存在操作复杂或预测精度低等缺陷。因此,本文提出使用灰狼优化算法优化BP模型,设计出一种新的GWO-BP模型。为了验证模型的优越性,将该模型应用于甘肃省花海灌区的实测流量上,并与传统的经验公式法和BP神经网络模型进行对比。结果显示GWO-BP预测结果更加稳定可靠,满足灌区水资源计量的工程精度要求。

Fast and accurate irrigation area flow forecast is an important basis for water resource allocation. Existing traffic prediction models generally use traditional methods or simple neural network models, and the above methods have defects such as complicated operations or low prediction accuracy. Therefore, this paper proposes to use the gray wolf optimization algorithm to optimize the BP model and design a new GWO-BP model. In order to verify the superiority of the model, the model was applied to the measured flow in Huahai Irrigation District of Gansu Province, and compared with the traditional empirical formula method and BP model. The results show that the GWO-BP prediction results are more stable and reliable, and can meet the engineering accuracy requirements of water resources measurement in irrigation areas.