Digital twin-driven focal modulation-based convolutional network for intelligent fault diagnosis

•An efficient digital twin-enabled physical model is constructed, incorporating dynamic modeling of the physical entity bearing to simulate its operating conditions.•A novel lightweight architecture, Focal Modulation-based lightweight CNN (FM-LCN), is proposed for fault recognition of the simulated mechanical dataset.•Experimental results indicate that the developed approach exhibits high accuracy on the simulated bearing dataset compared with state-of-the-art models.