In order to realize the multi-objective optimization of hypoid gears,a neural network surrogate model was established to describe the relationship between Ease off modification parameters and transfer error,tooth root stress and meshing loss power.Firstly,the dynamics software MAST A was used to establish the hypoid gear drive axle model.Based on the sensitivity coefficient matrix,the machine tool modification parameters corresponding to the second-order Taylor expansion of tooth surface deviation were derived,and the modified gear model was established.Secondly,the transmission error,tooth root stress and meshing loss power of the modified gear model were calculated through MASTA's loading tooth surface contact analysis function,and the neural network agent model was finally established.NSGA-Ⅱ multi-objective optimization algorithm was used to optimize the surrogate model for comparative verification.The results show that the proposed multi-objective optimization method can effectively reduce transmission error,tooth root stress and meshing power loss of hypoid gears.
维普
万方数据