基于IABC-RBF神经网络的地下水埋深预测模型

为了验证基于改进人工蜂群算法的径向基函数（RBF）神经网络模型在地下水埋深预测中的可行性和优越性，在基本人工蜂群算法中引入高斯变异算子，优化初始蜜源位置，设计建立基于改进人工蜂群算法的RBF神经网络模型（IABC-RBF）. 通过输入泾惠渠灌区的年降雨量、年渠首引水量、年田间灌溉用水量、年地下水开采量和前一年的地下水埋深共5个相关影响因子的数据，对地下水埋深进行预测，与实测的地下水埋深数据进行比较，误差很小. 与RBF神经网络模型和基于基本人工蜂群算法训练的RBF神经网络模型（ABC-RBF）的预测结果进行比较，结果表明，基于改进人工蜂群算法的RBF神经网络模型收敛速度更快，预测结果误差最小，精度最高.

Groundwater depth prediction model based on IABC-RBF neural network

The Gaussian mutation operator was introduced into the basic artificial bee colony algorithm, and the initial nectar location was optimized in order to verify the feasibility and superiority of radial basis function (RBF) neural network model based on improved artificial bee colony algorithm in groundwater depth prediction. A RBF neural network model was designed based on improved artificial bee colony algorithm (IABC-RBF). The groundwater depth was predicted by inputting the annual rainfall, the amount of annual water intake in the first ditch, the amount of annual field irrigating, the annual groundwater exploitation and the groundwater depth in the previous year of Jinghui irrigation district. The error is very small compared with the measured groundwater depth data. The model was compared with the RBF neural network model and the RBF neural network model based on the basic artificial bee colony algorithm (ABC-RBF). Results show that the RBF neural network model based on the improved artificial bee colony algorithm has faster convergence speed, the least error of the prediction result and the highest accuracy.