摘要
智慧粮食最大的任务就是要确保粮食安全。粮仓储粮环境实时监测是粮食安全保障的重要技术。本研究针对气象因素对储粮环境的影响,建立了仓储粮堆温度参数和气象因素的数学关系,证明了基于气象数据进行粮堆温度预测的可行性。提出了基于气象8因素的储藏粮堆表层(粮面以下50 cm处)平均温度预测模型,利用气象的气温、气压、相对湿度、0 cm地面温度、日照时间、降水量、蒸发量、风速多个因素展开构建储藏粮堆表层平均温度估计。针对多元回归预测问题,提出了线性最小二乘回归和支持向量机(SVM)不同核函数回归的方法,设计了具体的建模算法。结果表明,仓储粮堆传感器获取的实际值与预测结果的趋势一致,均方根误差均小于5.3,证明了基于气象数据的仓储粮堆表层平均温度预测模型与方法的有效性,为加快实现智慧粮食提供了参考。
A key goal of smart grain initiatives is to significantly improve the capacity of grain storage.Real-time monitoring of grain storage environment in granaries is an important technology to ensure food security.Given the impact of meteorological factors on the environment of grain storage,in this paper,we set up the mathematical relationship between storage environment parameters and meteorological factors and proved the feasibility of forecasting grain temperature based on meteorological data.The average temperature prediction model of the surface layer of the storage grain pile(50 cm below the grain surface)based on the meteorological 8 factors is proposed,referring to the air temperature,air pressure,relative humidity,0 cm ground temperature,sunshine duration,precipitation of the meteorological.As for the multivariate regression prediction problem,we proposed a linear least squares regression and different kernels of support vector machine(SVM)regression method,and design a specific modeling algorithm.The results showed that the actual value obtained by the storage grain reactor sensor is consistent with the trend of the prediction results,confirming that the effectiveness of the model and method for predicting the average surface temperature of the storage grain pile based on meteorological data,which lays a theoretical foundation for accelerating the realization of smart food.
作者
段珊珊
杨卫东
肖乐
张元
Duan Shanshan;Yang Weidong;Xiao Le;Zhang Yuan(College of Information Science and Engineering,Henan University of Technology,Zhengzhou 450001;Key Laboratory of Ministry of Education of Grain Information Processing and Control,Henan University of Technology,Zhengzhou 450001)
出处
《中国粮油学报》
EI
CAS
CSCD
北大核心
2020年第2期152-158,共7页
Journal of the Chinese Cereals and Oils Association
基金
中原学者(172101510005)
国家重点研发计划(2017 YFD0401001)
河南省自然科学基金(152102210068)
粮食信息处理与控制教育部重点实验室开放基金(KFJJ-2017-108)
粮食信息处理与控制重点实验室开放基金(KFJJ-2018-105)
关键词
气象因素
粮堆表层平均温度
线性最小二乘
SVM核函数
预测
smeteorological factors
average surface temperature of grain bunker
linear least squares
SVM kernel function
prediction
