Optimal design methodology of magneto-rheological (MR) rotary brake for robotic ankle systems considering temperature effects

摘要

Rotary magneto-rheological (MR) brake applied in robotic ankle systems often works at higher temperature which will degrade the torque due to the harsh environment and self-heating. To ensure the brake satisfy the requirements about the torque and compensation ability, an optimal methodology is proposed. In the design process, an evaluation coefficient is proposed to assess whether the brake has the compensation ability. The maximum torque, dynamic range and mass are set as the optimal objectives. The multi-objective functions are solved by the nondominated sorting genetic algorithm II. To describe the nonlinear relationships between the torque and design variables, a mathematical prediction model is constructed based on the finite element method. The simulation results show that this design optimal methodology is efficient to improve the performance and the compensation ability of rotary MR brake.