Vibration and noise reduction of the drive motor bearings using RBMO and PPO

To improve the ride comfort of new energy vehicles, a vibration and noise reduction method for the drive motor bearings of new energy vehicles using red-billed blue magpie optimizer (RBMO) and proximal policy optimization (PPO) is proposed. Combining the Tent map with the Sine map, the Tent-Sine map is designed to initialize the positions of red-billed blue magpies, make the distribution of individuals in the population more uniform, and enhance the randomness and traversal of population initialization. The nonlinear cosine function calculation formula of the equilibrium population coefficient is introduced. Thus, the equilibrium population coefficient stays at a higher value in the early stage, enabling the algorithm to search over a larger range, explore more areas, and enhance the algorithm's global search capabilities. Moreover, in the later stage, it converges to smaller values earlier and conducts precise searches for the possible regions where the optimal solution may exist in order to improve the local convergence speed and search accuracy. PPO is introduced into RBMO, which enables RBMO to make autonomous decisions on foraging manners based on experience and rewards, which weakens the sensitivity of the initial parameter settings, solves the problem of being stuck in local optima, and accelerates the convergence speed of the algorithm. Simulation analyses demonstrate that the optimized bearings of the new energy vehicle drive motor using this method meet the requirements of high speed, low vibration, and low noise. Experimental investigations demonstrate that the proposed method effectively suppresses the vibration and noise generated by the drive-motor bearings of new-energy vehicles and reduces their peak values. The noise signal of the optimized bearing exhibits a smoother profile without noticeable whine noise.