摘要
为探寻降水量预测的适用模型,根据河南省4个地区气象观测站1953—2013年的地面气象资料,基于BP神经网络算法和遗传算法,建立多种模型,对降水量进行预测。结果表明:将遗传算法与BP算法有机融合,大大提高了模型预测精度。遗传算法优化后的GA-BP神经网络在预测误差与精度等方面都比BP神经网络更好。但从局部看,遗传算法优化后的BP神经网络还不是很理想。
In order to explore suitable models for precipitation prediction,a variety of models are established based on BP neural network and genetic algorithm according to the surface meteorological data from 4 meteorological stations in Henan Province from 1953 to 2013.The results show that the organic integration of genetic algorithm and BP algorithm has greatly improved the accuracy of model prediction.The GA-BP neural network optimized by genetic algorithm is better than BP neural network in prediction error and precision.But from the local point of view,the BP neural network optimized by genetic algorithm is not ideal.
作者
李建磊
付世豪
宋金繁
Li Jianlei;Fu Shihao;Song Jinfan(North China University of Water Resources and Electric Power,Zhengzhou 450046,China)
出处
《黑龙江科学》
2023年第8期27-30,33,共5页
Heilongjiang Science
基金
河南省青年骨干教师项目(2019GGJS100)
河南省高等学校重点科研项目计划基础研究专项(20zx003)
河南省自然科学基金(222300420579)。
