Electromechanically coupled beam models for stacked dielectric elastomer actuators

这项工作的总体目标是为介电弹性体作动器提供计算成本低且准确的动态光束模拟模型，用于最优控制问题。叠层介电弹性体致动器在受到刺激时，在长度方向上的收缩与人体肌肉的行为相似。它们适用于逐点施加力。因此，介电弹性体可以实现复杂、高效和无噪音的系统驱动。然而，弹性作动器的使用也伴随着新的控制挑战。先进的控制策略需要避免不必要的振荡，使系统尽可能快地进入稳定状态，并尽可能紧密地遵循规定的轨迹。最优控制理论用于避免电介质执行器固有的弹性振荡，并获得最优控制轨迹。由于求解最优控制问题的计算成本受到模型自由度数量的显著影响，因此简化且针对特定问题的执行器模型优于成本密集型的通用有限元模型。在项目的前两年，开发并测试了一种经济高效的光束致动器模型，该模型涵盖了堆叠介质致动器的应用。在仿真时间和精度方面，将该模型与现有的基于有限元的仿真框架以及现有的集总模型进行了比较。此外，束致动器模型举例耦合到一个多体系统，以研究介电弹性体致动系统的行为。在前两年的项目延续中，执行器模型和驱动结构之间的耦合以结构化和模块化的方式制定，允许任意复杂的场景，并使用经济高效的光束模型来解决介电弹性体驱动系统的最优控制问题。数值算例显示了将仿真框架应用于简单的生物力学系统，如肘关节，以研究电介质弹性体替代骨骼肌的要求。计划进行小型实验，以解决参数识别和小型仿真实例的验证问题。

The overall objective of this work is to provide computationally cost efficient but also accurate dynamic beam simulation models for dielectric elastomer actuators to be used in optimal control problems. Stacked dielectric elastomer actuators bear analogy to the behaviour of human muscles in terms of contracting in length direction when stimulated. They are suitable for point-by-point application of a force. Therefore, dielectric elastomers allow for a sophisticated, efficient and noiseless actuation of systems. However, the use of elastic actuators is also accompanied by new control challenges. Advanced control strategies need to avoid undesirable oscillations, bring the system as quickly as possible into its steady state and follow prescribed trajectories as closely as possible. Optimal control theory is used to avoid oscillations that are inherent with the elastic nature of dielectric actuators and to obtain optimised control trajectories. As the computational cost for solving optimal control problems is significantly affected by the number of model degrees of freedom, reduced and problem specific actuator models are superior to general but cost-intensive finite element models. In the first two years of the project, a cost-efficient beam actuator model that covers applications with stacked dielectric actuators is developed and tested. The cost-efficient dielectric actuator model is compared to an existing finite element based simulation framework as well as to an existing lumped model regarding simulation time and accuracy. Moreover, the beam actuator model is exemplarily coupled to a multibody system to investigate the behaviour of dielectric elastomer actuated systems. In the continuation of the project after the first two years, the coupling between the actuator model and an actuated structure is formulated in a structured and modular way that allows for arbitrarily complex scenarios and the cost-efficient beam model is used to solve optimal control problems of dielectric elastomer actuated systems. Numerical examples show the application of the simulation framework to simple biomechanical systems, such as an elbow joint, in order to investigate the requirements on dielectric elastomers to replace skeletal muscles. Small experiments addressing the identification of parameters and the validation of small simulation examples are planned.