浙江农业学报 ›› 2016, Vol. 28 ›› Issue (11): 1941-1946.DOI: 10.3969/j.issn.1004-1524.2016.11.21

• 生物系统工程 • 上一篇    下一篇

西瓜生长期的振动预测方法研究

汪迪松1, 2, 李臻峰1, 2, *, 浦宏杰1, 2, 陆勇1, 2   

  1. 1.江南大学 机械工程学院,江苏 无锡 214000;
    2.江苏省食品先进制造装备技术重点实验室,江苏 无锡 214000
  • 收稿日期:2016-04-01 出版日期:2016-11-15 发布日期:2016-11-16
  • 通讯作者: 李臻峰,E-mail:1078709105@qq.com
  • 作者简介:汪迪松(1991—),男,湖北黄冈人,硕士研究生,研究方向为食品无损检测。E-mail:sunnywds@126.com
  • 基金资助:
    国家自然科学基金(515082290); 江苏省产学研联合创新资金(BY2014023-32); 江苏省食品先进制造装备技术重点实验室开放课题(FM-201406)

Study on forecast method for watermelon growth period based on vibration

WANG Di-song1, 2, LI Zhen-feng11, 2, *, PU Hong-jie1, 2, LU Yong1, 2   

  1. 1.School of Mechanical Engineering, Jiangnan University, Wuxi 214000, China;
    2. Jiangsu Key Laboratory of Advanced Food Manufacturing Equipment and Technology, Wuxi 214000, China
  • Received:2016-04-01 Online:2016-11-15 Published:2016-11-16

摘要: 通过自行设计的敲击振动检测系统对西瓜进行敲击振动检测,得到西瓜的频响函数,分别针对频响函数的第一响应频率f1、第二响应频率f2和第三响应频率f3建立预测西瓜生长期的数学模型,模型的确定系数R2较高。且选取了f1f2f3三个响应频率建立了预测西瓜生长期的多元非线性回归模型,模型的确定系数R2可达0.830,均方根误差RMSE降低到2.095。该模型的建立为在线无损检测西瓜的生长期提供了理论参考。

关键词: 振动检测, 西瓜, 生长期, 多元非线性回归

Abstract: The study was conducted by knocking on watermelon for its’ vibration test with self-designed vibration test system, from which the frequency response function(FRF) of watermelon was obtained. The prediction mathematical model of watermelon’s growth period was established by using the first response frequency(f1), the second response frequency(f2) and the third response frequency(f3) of FRF, respectively. The determination coefficients R2 of the three models were high. All the f1, f2 and f3 response frequency were selected to establish multivariate nonlinear regression model which could predict watermelon’s growth period. The determination coefficient R2 of the model could reach 0.830, the root mean square error RMSE decreased to 2.095. The establishment of the model provides a theoretical reference for the on-line nondestructive testing of watermelon’s growth period.

Key words: vibration testing, watermelon, growth period, multivariate nonlinear regression

中图分类号: