Abstract: Aiming at the problem that it was difficult to realize efficient dynamic balancing of some rotating machinery rotors for less than once per revolution (OPR) signals or for decreasing the times of system startup and shutdown during trim balancing, a virtual dynamic balancing algorithm without OPR signal based on Particle Swarm Optimization (PSO) was proposed by studying the structure of propeller rotor system on test-bed and summarizing dynamic balancing methods without key phase signal. A software integrating dynamics algorithm and particle swarm optimization was designed, and the positions and value of counterweight were gained by evolutionary search technology of particle swarm optimization in three steps. First, the initial vibration amplitude was input into the algorithm software, and the initial imbalance mass was accurately calculated. Second, after trial weight was added on the rotor, the angle between the initial imbalance and trial weight was searched and optimized. Finally, the direction of the two weights was identified by performing FFT for vibration signal to gain balancing position and achieve balance. The results indicate that the synchronous vibration is suppressed to at least 95% with one the trial weight to achieve balance without once per revolution (OPR) signal.

Key words: virtual dynamic balancing; finite element analysis; particle swarm optimization; without OPR sensor; imbalance vibration