基于混沌神经网络行星齿轮传动的模糊可靠性优化设计

The fuzzy reliability optimum design of the planetary gear based on chaos neural network

以渐开线行星齿轮传动机构为例,在考虑了机械传动各设计参数的模糊性和随机性的基础上,在运用神经网络求解优化方面问题的同时,引进全局耦合模型GCM(Globally Coupled Mode)混沌神经网络,通过混沌遍历,可使神经网络在整个相空间进行搜索,从而避免网络在运行过程中陷入局部极小值。利用混沌神经网络能达到系统的稳定平衡点和能够提高优化速度和精度等特点,探讨了基于混沌神经网的行星齿轮传动机构模糊可靠性的优化设计,实例进行实验,结果显示混沌神经网络的优化获得了全局最优解。表现出了令人满意的结果。

Introducing GCM （Globally Coupled Mode） chaotic neural network to improve the optimizing procedure, neural network is used to resolve the optimization problem. By chaotiet ravelling, network can do search in all phase space and so avoid trapping in local minimum. Make use of the Chaos neural networks to achieve the stability of the equilibrium point and optimization to improve the speed and accuracy characteristics Discussion The Fuzzy reliability optimum design of the planetary gear drive based on Chaos neural network a living example of design showed that the Chaos neural network optimization of the overall optimal solution and validated then gaind a good result.