摘要
粮食的储备对于一个国家来说十分重要,是关乎民生的重要战略资源。储粮对于温度要求十分严格,机械通风是保持粮仓环境温度的重要方法。针对传统机械通风不能准确识别通风时机而导致的风机损耗和粮食损害问题,选用RBF神经网络建立温度预测模型,利用粒子群算法(Particle Swarm Optimization,PSO)及列文伯格-马夸尔特算法(Levenberg-Marquard,LM)优化网络的参数。仿真结果可以看出,用PSO算法和LM算法优化后的神经网络模型预测精度更加准确,与用PSO优化后的RBF神经网络进行比较后可知,PSO-LM-RBF算法预测误差小,模型更加稳定。
Grain reservation is of great importance to a country,which is an important strategic resource related to people’s livelihood.The temperature requirement for grain storage is very strict.Mechanical ventilation is an important method to maintain the temperature of granary.To address the problem of fan damage and grain deterioration caused by the failure of accurately identify the ventilation time by traditional mechanical ventilation,RBF neural network was used to establish a temperature prediction model.Particle Swarm Optimization(PSO)and Levenberg-Marquard(LM)algorithms were used to optimize parameters of the network.As simulation results suggested,the accuracy of neural network model optimized by both PSO algorithm and LM algorithm was more accurate.Compared to the RBF neural network optimized by PSO algorithm,the prediction error of PSO-LM-RBF algorithm was smaller.In addition,the PSO-LM-RBF model is more stable.
作者
黄琦兰
王涛
HUANG Qi-lan;WANG Tao(School of Control Science and Engineering,Tiangong University,Tianjin 300387,China)
出处
《保鲜与加工》
CAS
2022年第3期30-34,共5页
Storage and Process
关键词
机械通风
RBF神经网络
PSO算法
LM算法
温度预测
mechanical ventilation
RBF neural network
PSO algorithm
LM algorithm
temperature prediction
