基于双源激振的果品自适应振动采收机理及装置研究

Vibratory fruit harvesting has been extensive researched. The traditional trunk shaker was limited application due to its inefficiency and high damage to the tree. To address this problem, this project puts forward a method for self-adaptive vibratory fruit harvesting under dual-source excitations, and develops a vibratory harvesting prototype for testing. After reconstructing the 3D motion status of the sub-system, including branch, peduncle and fruit, using high-speed stereo vision, the sub-system’s motion law can be analyzed. Thus, the detachment force causing the fruit to fall off can be estimated, and the mechanism of fruit detachment under vibration can be found eventually. Employing the transfer matrix method, a relatively efficient and accurate dynamics of the whole fruit tree was established, and then was modified by the experimental measurements of tree’s dynamic characteristics. To maximize the accelerations of the whole tree under the same power exciting, the multi-objective optimization model was established on the premise of high efficiency and low damage, and the optimized operating parameters for vibratory harvesting were calculated. Finally, a self-adaptive vibratory fruit harvester with double vibration heads was been developed, and then was applied to a test of vibratory fruit harvesting with the optimized operating parameters. The objective of this research will enhance the level of theory in the vibratory fruit harvesting, and provide some valuable technology for development and application of intelligent vibratory fruit harvester.

机械式果品振动采收已得到广泛的研究，但传统的树干振摇式果品振动采收方式由于其效率低、对树的高损伤率等缺点限制了它的应用。针对这个问题，本项目提出了一种双源激振主枝的果品自适应振动采收方法，研制了相应的振动采收装置。利用高速立体视觉技术重建果枝-果柄-果实子系统的三维运动状态，分析它们的运动规律，估计引起果实脱落的作用力，揭示果实的振动脱落机理。利用传递矩阵法建立相对高效、准确的果树系统动力学模型，并结合果树动力学特性测试试验进行修正。以等功率激振下树枝产生的最大加速度为目标，在满足高效、低损伤的前提下建立多目标优化模型，求解理想的振动采收工作参数。研制双振动头的果实自适应振动采收装置，并把优化后的振动工作参数应用到振动采收试验中。本项目的研究将提升我国在果品振动采收研究领域的理论水平，为智能化果品振动采收装置的研制和应用提供参考和技术储备。

目前，果品振动采收已开展了大量的研究，但由于果树个体的差异性，采用单一的激振频率对果树实施振动采收普遍存在效率低、损伤果树根部等问题。为此，本项目围绕提高果品振动采收的效率和采净率，研制高效的果品振动采收装置，研究果品自适应振动采收方法。为提高果树三维模型的精度，采用十字结构光对果树进行三维重建，相应地提出了鲁棒的基于方向灰度重心法的激光光条中心提取方法。研制了以空间连杆振摇机构为核心部件的三维激振采收装置，并借助遗传算法对结构尺寸进行优化，田间试验表明用该振动采收装置能更有效地使果树各分枝产生受迫振动。为满足自适应振动采收的需要，研制了一套智能振动采收系统，并提出了一种基于快速扫频在线测量果树分枝固有频率的方法，开展了相应的田间山核桃树试验。结果表明用提取的频率作为激振频率能使果枝产生明显的加速度响应，因此可以实现果品的自适应振动采收。最后，研究基于双源激振的果品振动采收方式，试验结果显示它能使不同分枝产生的加速度响应更均衡，可能实现更高的采收效率和采净率。果品自适应振动采收方式开辟果品振动采收的新思路，将有力推动我国果品振动采收的产业化。