Abstract

Topology optimization is a promising technique to define holes and links in the material design space to reduce mass and inertial effect and enhance energy efficiency. In this work, topology optimization of the upper and forearm of SCORBOT ER-4U manipulator is performed considering dynamic loading conditions. The optimized topology corresponding to a loading condition may not be sufficient for other conditions because of changes in applied load directions. Here a methodology is applied to consolidate multiple optimized topologies at different angle values in working range. The superimposition technique is used to combine all the density elements of material domain. Later, highly sensitive elements of the material are refined using normalization followed by re-penalization approaches. It also helps to reduce volume and grayscale elements in the material domain. Due to this process, sharp corners, irregularities, and tiny holes persist in synthesized topology, which leads to the poor performance of the manipulator. Therefore, image morphology-based processing techniques are applied to eliminate the geometric complexities and improves the performance of a robot. The synthesized topology obtained using this methodology shows a 15–36.4% reduction in deflection and Von-Mises stress. The SCORBOT educational manipulator is considered to illustrate the proposed method. The complete manipulator is designed in SOLIDWORKS, and a motion study is conducted to generate the dynamic analysis data concerning the angular position. The manipulator-link volume fraction is reduced by 50%, which improves energy efficiency by 43.2% with better performance values.

摘要

拓扑优化是一种很有前途的技术，可以在材料设计空间中定义孔和链接，以减少质量和惯性效应并提高能源效率。在这项工作中，考虑动态负载条件，对 SCORBOT ER-4U 机械臂的上臂和前臂进行了拓扑优化。由于施加的负载方向的变化，对应于负载条件的优化拓扑可能不足以满足其他条件。这里应用一种方法来整合工作范围内不同角度值的多个优化拓扑。叠加技术用于组合材料域的所有密度元素。随后，使用归一化和重新惩罚方法对材料的高度敏感元素进行细化。它还有助于减少材料域中的体积和灰度元素。由于这个过程，合成拓扑中仍然存在尖角、不规则和小孔，导致机械手的性能较差。因此，应用基于图像形态学的处理技术来消除几何复杂性并提高机器人的性能。使用此方法获得的合成拓扑显示偏转和 Von-Mises 应力降低了 15–36.4%。SCORBOT 教育机械手被用来说明所提出的方法。完整的机械臂在 SOLIDWORKS 中设计，并进行运动研究以生成有关角位置的动态分析数据。机械手连杆体积分数减少50%，能源效率提高43%。