基于改进遗传算法的行星齿轮优化设计

针对行星齿轮的设计参数是由一些混合离散变量组成的特点，通过距离测度改进遗传算子，引入菱形思维，得到了具有实值编码技术的改进型自适应遗传算法．将齿数、模数、齿宽及行星轮个数作为设计变量，以机构的体积最小为目标函数，建立了有别于现有行星轮系的优化设计数学模型．并利用改进遗传算法对模型进行了优化，较常规设计体积减少了31．64％，较SGA减少了17．8％，具有一定的实际意义．

Planetary gear optimization based on improved genetic algorithm

The planetary gear＇ s design parameters are composed of a number of hybrid discrete variables. By using distance measurement to improve operators and introducing lozenge thinking, an improved genetic algorithm（GA） that adopts a real code method is obtained. The design variables are composed of the number of teeth, modulus, tooth width and the number of planetary wheels. Taking the smallest volume as the objective function, the mathematic model is built up. A mixed GA is introduced into the fundamental genetic algorithm to carry out the optimization. Through a planetary gear design calculation, the volume reduced 31.64% comparing with the normal design and 17.8% comparing with SGA.