The optimization of the energy consumption of Industrial Robots (IRs) has been widely investigated. Unfortunately, on the field, the prediction and optimization strategies of IRs energy consumption still lack robustness and accuracy, due to the elevated number of parameters involved and their sensitivity to environmental working conditions. The purpose of this paper is to present, and share with the research community, an extensive experimental campaign that can be useful to validate virtual prototypes computing the energy consumption of robotic cells. The test cell, comprising a high payload IR equipped with multiple sensors and different payloads, is firstly described. The testing procedures are then presented. Experimental results are analyzed providing novel qualitative and quantitative evaluations on the contribution and relevance of different power losses and system operating conditions, clearly depicting the nonlinear relation between the energy consumption and various freely programmable parameters, thus paving the way to optimization strategies. Finally, all the experimental tests data are provided in the form of an open research dataset, along with custom Matlab scripts for plotting graphs and maps presented in this paper. These tests, which are verifiable via the shared dataset, consider the overall measured IR energy consumption (as drawn from the electric network) and highlight that, in some industrially interesting case scenarios, optimization potentials for energy savings of more than 50% are possible.

工业机器人（IR）的能耗优化已得到广泛研究。不幸的是，在现场，由于涉及的参数数量增加以及它们对环境工作条件的敏感性，IR能耗的预测和优化策略仍然缺乏鲁棒性和准确性。本文的目的是介绍广泛的实验活动，并与研究社区共享，该活动可用于验证计算机器人细胞能耗的虚拟原型。首先描述包括具有多个传感器和不同有效载荷的高有效载荷IR的测试单元。然后介绍测试程序。分析了实验结果，对不同功率损耗和系统工作条件的贡献和相关性提供了新颖的定性和定量评估，清楚地描绘了能耗与各种自由可编程参数之间的非线性关系，从而为优化策略铺平了道路。最后，所有实验测试数据都以开放研究数据集的形式提供，以及自定义的Matlab脚本，用于绘制本文中介绍的图形和地图。这些测试可通过共享数据集进行验证，其中考虑了整体测量的IR能耗（从电网得出），并强调了在某些工业上有趣的情况下，可以实现50％以上的节能优化潜力。