摘要
粮仓储粮重量自动检测是国家粮食安全的重要保障技术.本文针对粮堆散粒体特性,建立了粮仓储粮重量与粮仓底面和侧面压强的数学关系,证明了基于压力传感器进行粮仓数量在线检测的可行性.提出了基于内外圈两圈布置的压力传感器布置模型和基于多项式展开的粮仓储粮重量检测模型,利用内外圈传感器输出值均值的多项式展开构建粮仓储粮重量估计.针对实仓检测中内外圈传感器输出值均值存在较大波动的问题,提出了基于SVR的粮仓储粮重量检测模型,给出了SVR输入项序列的具体提取方法,设计了具体的建模算法.实验表明,实验粮仓检测模型建模与预测结果的误差小于±3%,证明了所提出的粮仓储粮重量检测模型与方法的有效性,可以满足国家粮仓储粮重量检测的要求.
Automatic detection of granary storage quantity is an important technology for national food security.In this paper,according to the characteristics of grain peaks,the mathematical relationship between granary storage weight and the bottom / side pressure of granaries is established;the feasibility of the online granary storage quantity detection based on pressure sensors is also demonstrated.Furthermore,a new granary storage weight detection model based on polynomial expansion is proposed by using pressure sensors arranged along the inner and outer rings.The polynomial expansion of the average value of the pressure sensors is used to evaluate the granary storage weight.As the average value fluctuates inpractical warehouse detection,a granary storage weight detection model based on SVR is proposed.The detailed extraction method for SVR input sequences is described,and specific modeling algorithm is designed.Practical storage weight detection results show that the detection accuracy of the proposed model is better than 97%,which demonstrates that the proposed granary storage weight detection model is effective and well meet the demand of national granary storage quantity detection.
作者
张德贤
张苗
张庆辉
张元
吕磊
ZHANG De-xian;ZHANG Miao;ZHANG Qing-hui;ZHANG Yuan;L Lei(School of Information Science and Engineering,Henan University of Technology,Zhengzhou,Henan 450001,China;Grain Information Processing and Control,Key Laboratory of Ministry of Education,Zhengzhou,Henan 450001,China)
出处
《电子学报》
EI
CAS
CSCD
北大核心
2018年第5期1179-1185,共7页
Acta Electronica Sinica
基金
国家高科技研究发展计划(863计划)(No.2012AA01608)
国家科技支撑计划(No.2013BAD17B04)
国家自然科学基金(No.U1404617)
粮食信息处理与控制教育部重点实验室开放基金(No.KFJJ2016102)
河南省高校科技创新团队(No.16IRTSTHN026)
关键词
储粮重量监测
压力传感器
检测模型
支持向量回归
检测精度
grain storage weight monitoring
pressure sensor
detection model
support vector regression（SVR）
detection accuracy
