摘要
针对目前北方日光大棚温度、湿度、光照强度等环境因子随天气变化而波动较大的问题,搭建了用改进遗传算法优化BP神经网络的预测模型(IAGA-BP),对大棚中的环境因子做出预测以便提前调控。传统BP神经网络存在易陷入局部最小化、预测精度不高、收敛时间长的问题,使用改进的遗传算法来对网络中的权、阈值进行迭代寻优,利用自适应变化的交叉与变异算子来对种群中的子代不断优化以达到最优值。通过选取大棚中的实际数据来进行仿真实验,对BP、GA-BP和IAGA-BP模型进行大棚环境预测分析,实验结果表明,IAGA-BP模型的预测精度最高、误差最小,拥有更好的稳定性。
Aiming at the problem that the environmental factors such as temperature,humidity and illumination intensity fluctuate greatly with the change of weather in northern solar greenhouse,an improved genetic algorithm(IAGA)was used to optimize the BP neural network prediction model to predict the environmental factors in the greenhouse in order to regulate and control them in advance.The traditional BP neural network has the problems of low prediction accuracy,long convergence time and easy to fall into local optimum.The improved genetic algorithm is used to iteratively optimize the thresholds and weights in the network,and the adaptive crossover operator and mutation operator are used to optimize the descendants of the population to achieve the optimal value.By selecting the actual data in the greenhouse for simulation experiment、BP、GA-BP and IAGA-BP models are used to predict the greenhouse environment.The experimental results show that the iaga-bp model has the highest prediction accuracy,the smallest error and better stability.
作者
李渊朴
王秀玲
Li Yuanpu;Wang Xiuling(College of Information Engineering,Inner Mongolia University of Technology,Hohhot 010080,China)
出处
《电子测量技术》
2020年第7期46-49,166,共5页
Electronic Measurement Technology
基金
内蒙古科技计划基金项目(2017030219)资助
关键词
大棚环境
遗传算法
BP神经网络
预测分析
优化
greenhouse environment
genetic algorithm
BP neural network
prediction analysis
optimization
