Weak fault diagnosis method of rolling bearing based on improved dung beetle optimisation feature mode decomposition

To accurately extract the weak fault features of rolling bearings under significant noise interference, this paper proposes a feature mode decomposition method based on an improved dung beetle optimization algorithm (DBOFMD). The method constructs a weighted impulse response energy ratio (WIRER) as the objective function by integrating the kurtosis of the square envelope spectrum (KSES) as a weighting factor with the impulse response energy ratio (IRER). This approach optimizes the key parameters of FMD, overcoming the reliance on empirical settings. The improved DBO algorithm efficiently searches for the optimal parameters, combining fast convergence with local and global search capabilities to avoid premature convergence. Validation using both simulated and experimental datasets demonstrates that DBOFMD outperforms methods such as POFMD, VMD, and CYCBD in extracting weak fault features, enabling effective diagnosis through envelope spectrum analysis.