摘要
棉花是我国重要的经济作物与棉纺织业发展的主要原材料之一,是我国经济发展的支柱产业。在棉花种植过程中,农田措施、气象环境等都会对棉花生产产生影响。对棉花生长因子进行分析,建立棉花预测模型,预测我国棉花产量,对于指导棉花生产和促进我国经济发展具有重要意义。为此,针对传统BP神经网络在预测中存在测试精度低、鲁棒性差等问题,利用遗传算法(Genetic Algorithm, GA)对BP神经网络模型进行优化,构建GA-BP神经网络模型;同时,基于湖北省2011-2021年棉花播种面积、气象因子、自然灾害和棉花产量,构建BP神经网络、GA-BP神经网络模型,对湖北地区棉花产量进行预测。研究结果表明:GA-BP神经网络模型精度明显高于BP神经网络模型,R2达到0.991。因此,通过GA-BP预测能够更加科学、合理地进行棉花产量预测,对棉花生产及管理措施的调整具有重要的指导意义。
Cotton is an important cash crop and one of the main raw materials for the development of cotton textile industry in China.In the process of cotton cultivation,farmland measures.Meteorological environment,etc.can have an impact on cotton production.Therefore,it is important to analyze cotton growth factors and establish a cotton prediction model to predict cotton production in China for guiding cotton production and promoting China's economic development.For the problems of low testing accuracy and poor robustness of traditional BP neural network in prediction,this study uses Genetic Algorithm(GA)to optimize the BP neural network model and construct GA-BP neural network model.Based on the cotton sowing area,meteorological factors,natural disasters and cotton yield in Hubei province from 2011-2021,BP neural network and GA-BP neural network models were constructed to predict cotton yield in Hubei province.The results showed that the accuracy of GA-BP neural network model was significantly higher than that of BP neural network model,and the R2 reached 0.96.Therefore,the GA-BP prediction can make cotton yield prediction more scientifically and reasonably,which has important guiding significance for cotton production and management measures adjustment.
作者
董宁
赵丙秀
王俊杰
Dong Ning;Zhao Bingxiu;Wang Junjie(Wuhan Vocational College of Software and Engineering,Wuhan 430205,China;Hainan Lance Technology Co.Ltd.,Haikou 570216,China)
出处
《农机化研究》
北大核心
2024年第12期39-43,共5页
Journal of Agricultural Mechanization Research
基金
教育部高校学生司第一期供需对接就业育人项目(20220105995)。
关键词
棉花
产量预测
遗传算法
BP神经网络
全局寻优
cotton
yield prediction
genetic algorithm
BP neural network
global optimization search
