|
|
|
| Fuzzy optimization design of engineering elevator mechanism applying genetic algorithm and neural networks |
| XI Ping-Yuan1, LI Gui-San1, HU Heng-Yin2, SHENTU Liu-Fang1 |
1.Department of Mechanical Engineering, Huaihai Institute of Technology, Lianyungang 222005, China;
2.Engineering Department, Continental Teves Corporation(LYG) Co., Ltd, Lianyungang 222005, China |
|
|
|
Abstract Concerning that some factors influencing design are difficult to display with definite value, this paper established the fuzzy optimization design model by taking the minimum of the volume of tooth corona of worm gear in elevator mechanism as the optimization objective. The method of secondclass comprehensive evaluation was adopted to obtain the optimal level cut set, and thus fuzzy optimization is transformed into usual optimization. Moreover, the neural networks algorithm is adapted to train feedforward networks so as to fit relative coefficient. Then the fitness function with penalty terms is established with penalty strategy. Genetic Algorithm Toolbox of Matlab are utilized to seek the optimal solution and then reach the aim of improve accuracy and search efficiency.
|
|
Published: 28 October 2005
|
应用遗传算法和神经网络的工程电梯传动机构模糊优化设计
考虑到影响设计的某些因素很难用确定数值表示,以工程电梯传动机构中蜗轮齿冠体积最小为优化目标,建立了模糊优化设计的数学模型。采用二级模糊综合评判法按最大隶属度原则求出最优水平截集，将模糊优化问题转化为普通优化问题。另外，通过神经网络方法得出网络权值和阈值以拟合待求系数，并采用加法形式的惩罚策略来构造带有惩罚项的适值函数，应用Matlab遗传算法工具箱寻求问题最优解，从而提高设计精度和搜索效率。
关键词：
变幅机构,
模糊优化设计,
遗传算法,
神经网络工具箱
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
