MR damper identification using EHM-based feedforward neural network

This paper proposes a novel method for modeling magneto-rheological (MR) dampers. It uses elementary hysteresis model (EHM) with feedforward neural network (FNN) to capture hysteresis characteristics of MR damper, and another FNN to determine the current gain. These parts can be trained separately, thus reducing the size of the training dataset. The inputs of the proposed model include the velocity, acceleration, and current to estimate generated damping force. Unlike previous FNN models, this model does not require force sensor inputs. Simulation results show the high performance of proposed EHM-based FNN when compared to conventional methods like recurrent neural network (RNN).