Holo-balancing method of flexible rotors based on multi-objective differential search algorithm

The rotor is a critical component of rotating machinery, often experiencing high amplitude vibrations caused by unbalance. Therefore, achieving efficient and accurate dynamic balancing of rotors is essential for effective monitoring and maintenance . Traditional optimization algorithms for dynamic balancing have several limitations. For instance, the vibration data obtained from a single sensor cannot adequately represent the overall vibration of the system, and the balancing objectives are often too narrow. This paper addresses the dynamic balancing of flexible rotors using multi-objective intelligent optimization. Leveraging the comprehensive information extraction capabilities of the holospectrum, three metrics are selected to evaluate the balancing state of the rotor: the average value, the maximum value, and the extreme difference of residual vibration. An optimal evaluation function for multiple targets is introduced to resolve the coordination issues among these three metrics. To enhance the efficiency and accuracy of dynamic balancing, an improved differential search algorithm, E2-S-DSA, is proposed. Balancing experiments demonstrate that this method achieves a vibration reduction ratio of 85.3%, further verifying the reliability and practical value of the proposed approach.