Analysis And Optimization Of Powertrain Mounting System Considering Uncertainty

Powertrain mounting system,consisting by a powertrain and some elastic components that connecting the powertrain with vehicle body or sub-frame,is one of the decisive system that influence the NVH(Noise,Vibration and Harshness)performance of a vehicle.Optimization design for the powertrain mounting system is one effective approach to damp the vibration and noise transmitting from powertrain to vehicle interior.Traditional analysis methods for the powertrain mounting system are usually based on deterministic model,which supposes all system parameters have deterministic values.Nevertheless,due to the unavoidable material aging during the production and operation of mounts,as well as the errors in manufacture,assembly and measurement,uncertainties widely exist in the powertrain mounting system.Generally speaking,single uncertain parameter has little impact on system,but if many uncertain parameters are coupled,the design results of powertrain mounting system based on deterministic model are likely to be unreliable,because of the great deviation resulting from the great uncertainties.In order to solve the problem above,with the aim of improving the NVH performance of the powertrain mounting system with uncertainties,this dissertation discusses analysis models,numerical solutions and optimization processes for the powertrain mounting system with uncertainties.The main works are shown as follows:(1)By constructing a dynamic model of six degrees of freedom powertrain mounting system,the natural frequencies and decoupling ratios are calculated to evaluate the NVH performance of the powertrain mounting system.Plus,based on the trifilar torsional pendulum measurement,the center of gravity and the inertia tensor of the powertrain mounting system are acquired,thereby assembling the mass matrix,which are the basis parameters of the uncertain analysis and optimization in later chapters.(2)Chebyshev-Vertex method is developed to design the powertrain mounting system involved interval uncertainties.In the proposed method,the interval optimization model for the powertrain mounting system are build: the appropriate bounds of natural frequencies and decoupling ratios are taken to create optimization constrains,while the weighting sum of the lower bounds of decoupling ratios are used to construct the optimization objective.The numerical example verifies the effectiveness and accuracy of the proposed method.In addition,by means of the interval optimization,the powertrain mounting system can not only meet the requirements completely,but also have a much more reliable results with better NVH performance,compared with deterministic optimization.(3)For the case that the powertrain mounting system have both random parameters and interval parameters,two methods are provided to calculate the responses of the system.One method is hybrid random-interval perturbation-central difference method(HRIP-CDM),which is used to calculate the bounds of the means and the variances.The other is hybrid intervalrandom perturbation-central difference method(HIRP-CDM),which can solve the means and the variances of the bounds.Numerical examples indicate the both methods can solve the hybrid random and interval model of powertrain mounting system accurately and effectively,by comparison with Monte-Carlo method.Furthermore,the optimization results verify the proposed methods are capable of improving the NVH performance of the system.Based on the vibration decouple theory,uncertainty analysis theory,and optimization theory,this dissertation presents two uncertain models for powertrain mounting system systematically.In addition,three analysis and optimization approaches are concluded with corresponding numerical examples.In engineering practice,the above methods can guide the design of powertrain mounting system with uncertainty.