Non destructive detection of Zea mays' critical periods: A parameter identification based approach

详细信息    查看全文

• 作者：M. Bonilla^aAuthor Vitae ; J.C. Martí ; nez-Garcí ; a^b ; ^{juancarlos_martinez-garcia@conciliencia.org}Author Vitae ; C. Antonio^cAuthor Vitae ; R. Arteaga^dAuthor Vitae
• 关键词：Cultured plant dynamics ; Maize production improvement ; Physical-based modeling ; Vector parameter estimation ; Model-based systems identification ; Filtering
• 刊名：Neurocomputing
• 出版年：2017
• 出版时间：12 April 2017
• 年：2017
• 卷：233
• 期：Complete
• 页码：23-33
• 全文大小：1474 K
• 卷排序：233

文摘

Because of its significance for the improvement of the efficiency in crop production, it is of fundamental importance to identify the plant critical periods (which characterize the sensitivity of the plant to the atmospheric elements). Since in the case of some plant development stages it is not possible to observe them by a simple sight, it is usual to destroy some sampling plants to get the required information. In this work, the crop growth process of the maize (Zea mays, sp) is analyzed for finding its critical phases by means of non destructive techniques. For this, at first the stem elongation process is analyzed by studying the dynamical behavior of an hypothetic unitary mass particle, concentrated in the stem appendix, traveling through an apparent potential well. And then, a parameter identification Non Destructive Detection Procedure is proposed, which is based on two concatenated Recursive Constrained Least–Square Algorithm with orthogonal projections. The critical periods are predicted until 3 weeks of anticipation.