Considering the influence of the uncertainty of design variables on the optimization design, the robust optimization design theory was used to build a robust model for the powertrain mounting system of automobiles. In this model, decoupling of energy distribution was taken as the target, the stiffness parameters of the mounting was taken as a design variable, and the mean and standard deviation of the target results were considered. Aiming at the problem that the particle swarm algorithm was easy to fall into a local optimal solution, a hybrid particle swarm algorithm was adopted to optimize the stiffness of the mounting of the powertrain mounting system, and the Monte Carlo method was used to analyze the optimized results to examine the influence of the variety of design values on the objective function. The results show that the method can improve the robustness of the mounting system effectively.

Constrained layer damping, as one type of the surface damping methods, is usually applied to solve the vibration and noise problems of metal plate structures. The loss factor and storage modulus of the damping material are usually used to evaluate the damping performance of the material. But this method can not be employed to evaluate the vibration and noise control performance of the constrained layer damping effectively and directly. In this paper, a new method is introduced to solve this problem. In this method, sinusoidal scanning is adopted and the vibration of a simply supported beam is excited by the quasiwhite noise random vibration. Through the analysis of the response, the vibration and noise control effect of the constrained layer damping is expressed.

The common faults of squirrelcage induction motor and the importance of fault diagnosis are introduced. The electrical and mechanical features of signals are extracted by the modified wavelet packet algorithm. A fault diagnosis method based on wavelet packet analysis of frequencyband energy is proposed. The experiment shows that with this method all the broken bars faults of the rotor can be detected rapidly with high accuracy. This method has high feasibility and extensive applicability.