Vibration suppression device that has variable moment of inertia using magneto-rheological fluid (Seismic test during switching series inertia mass by reinforcement learning)

In previous paper the authors successfully developed a vibration suppression device (VSD) that has variable moment of inertia using Magneto-Rheological (MR) fluid. The MR fluid is filled into flywheel. Ferrite particles are clustered when magnetic field applied to flywheel by electromagnets, series mass VSD will be switchable. this seismic tests during switching reinforcement learning focused in order get more reduction while random excitations. Firstly, agents trained Deep Deterministic Policy Gradient (DDPG) algorithm one-degree-of-freedom system with VSD, which switchable mass, excited under earthquakes. Secondly, improved for having 8 analyzed FEM, resisting force characteristics and response performance investigated. Then, confirm control some earthquakes, one degree-of-freedom DDPG carried out shaking table, responses simulated numerically. From experimental results, peaks acceleration displacement decrease about 1/2 case DDPG, also cut 20 % mass. Finally, validity confirmed.