摘要
为解决传统粮仓储粮温度预测模型的预测精度低、输入变量多的问题,提出一种基于主成分分析(PCA)、贝叶斯算法和XGBoost的组合预测模型。通过主成分分析法提取粮仓中影响粮温的主要因素,降低模型输入向量维数,利用贝叶斯算法对XGBoost超参数进行组合优化,以获取最优参数组合,建立粮温预测模型。结果表明,该模型预测误差小、精度高,可为粮仓的温度调控管理提供决策依据。
In order to solve the problems of low prediction accuracy and many input variables of the traditional grain storage temperature prediction model,a combined prediction model based on principal component analysis(PCA),bayesian algorithm and XGBoost was proposed.The main factors affecting grain temperature in granaries were extracted by principal component analysis,and the dimension of model input vector was reduced.The XGBoost hyper-parameters were combined and optimized by bayesian algorithm to obtain the optimal parameter combination,and the grain temperature prediction model was established.The results showed that the model has small prediction error and high precision,which could provide decision-making basis for the temperature control and management of granaries.
作者
郭利进
王永旭
GUO Li-jin;WANG Yong-xu(School of Electrical Engineering and Automation,Tianjin Polytechnic University,Tianjin 300387,China)
出处
《粮食与油脂》
北大核心
2022年第11期78-82,共5页
Cereals & Oils
