Research on Rotary Kiln Rotation Center Offset Fault Identification Based on ISBOA-VMD

To address the difficulty of extracting thermal bending failure and centerline horizontal displacement fault feature signals when judging the operating status of cement rotary kilns, we propose a method for extracting fault features based on improved secretary bird optimization algorithm (ISBOA) and variational modal decomposition (VMD). First, a strategy of randomly consuming prey with inertial weights is proposed to enhance the randomness of search results and avoid local optima. Then, the whale algorithm’s encirclement strategy is introduced into the secretary bird’s camouflage strategy to coordinate the capabilities of local search and global exploration. Finally, ISBOA demonstrated superior performance to other optimization algorithms in VMD parameter selection, achieving a 75% improvement in convergence speed compared to pre-optimization. Through validation with experimental and simulation data, this method demonstrates good feasibility. By decomposing actual signals and comparing the mean energy of their characteristic signals, the severity of thermal bending faults in the cylinder and centerline horizontal displacement faults in cement rotary kilns is diagnosed. Verified against actual measurement results, the accuracy reached 96.7%.