摘要
为了提高谷物联合收获机自动测产系统的测产精度,在研究了谷物联合收获机田间工作状态和升运器速度变异的基础上,通过分析谷物的运动学原理及其对冲量传感器作用的力学原理,建立了电压/升运器速度产量模型。为进一步消除收获机作业时的噪声干扰,在测产原始数据预处理时,先采用回归差分法降低振动噪声,然后采用双阈值滤波以及阈值取代法、前值取代法2种插值方法以消除差分电压中的奇异值,结果显示前值取代法效果较佳。此外,还提出了升运器速度归一化方法和冲量电压标准化方法以消除量纲影响并简化计算。田间测产试验结果表明,提出的电压/升运器速度模型比传统的质量-电压模型更能准确表征谷物运动实际情况,测产精度高,验证均方根误差为2.03%。
Based on the research of working condition and speed variation of elevator on the grain combine harvester,the kinematics and physical mechanics of grain movement on the momentum sensor were analyzed,and then a model of the mass flow sensor named voltage / elevator rotating speed model was put forward to improve the estimation accuracy of grain yield,so that the relationship between grain mass flow and impact force was described deeply by the proposed model. In the harvesting experiments,there existed many vibration noises. In order to eliminate those noises and ensure the accuracy of yield prediction,the impact voltage signals needed to be further processed. Firstly,a regression difference method was used to reduce vibration interference. Secondly,a dual threshold de-noising method and two kinds of interpolation methods were proposed to eliminate the singular value of the difference voltages.One of the interpolation methods was threshold replacement and the other was former-value replacement.The former-value replacement method could relatively keep the real yield data,and its estimation error was smaller. Furthermore,the rotating speed was normalized to eliminate the influence of dimension andsimplify calculation. Consequently,a standardized processing method of impulse voltage was put forward.The verification results showed that the proposed voltage / elevator rotating speed had a high accuracy and the RMSE was 2. 03%,which could really characterize the actual situation of grain movement.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2015年第7期91-96,共6页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家高技术研究发展计划(863计划)资助项目(2012AA101901)
农业部引进国际先进农业科学技术计划(948计划)资助项目(2011-G32)
关键词
联合收获机
测产系统
产量模型
数据预处理
速度变异
Combine harvester Yield monitor Yield model Data pre-process Rotational speed variation
