Hybrid system response model for condition monitoring of bearings under time-varying operating conditions

•A CM method originating from the perspective of system response is proposed with the influence of TVOCs eliminated.•HSSM is constructed under the combination of physically informed and data-driven ideas that models the transition of the bearing operating state and varying OCs.•The dual estimation of the operating state and the weights of the neural network is realized by dual extended Kalman filtering.•Experiments on accelerated fatigue degradation of bearings verify the effectiveness and superiority of the proposed method.