基于改进花粉算法优化的神经网络负荷预测研究

为预测短期电力负荷，可采用神经网络方法预测，而神经网络复杂权重和阈值的参数调节为预测精度的提升增加了一定程度的困难。采用进化计算算法优化过后的神经网络拥有更为优良的结构，能够提高神经网络的预测精度，为了使求解神经网络结构的进化计算算法拥有更为优秀的搜索能力，可改进算法求解网络模型结构。对进化计算花粉算法的改进及改进效果进行了研究。

Research on load forecasting of neural network based on improved pollen algorithm optimization

In order to predict short-term power load，neural network method could be used to predict，and the parameter adjustment of the complex weight and threshold of neural network added a certain degree of difficulty to the improvement of prediction accuracy.The neural network optimized by the evolutionary computing algorithm had a better structure，which couldn improve the prediction accuracy of the neural network.In order to make the evolutionary computing algorithm for solving the neural network structure had better search capabilities，the algorithm could be improved to solve the network model structure.The improvement and improvement effect of the pollen algorithm of evolutionary computation were studied.