改进遗传算法在饲料配方设计中的应用

针对现有饲料配方软件的局限性，以白绒山羊饲料配方为例，提出一种基于改进遗传算法的求解方法。与标准遗传算法相比，采用随机联赛选择替代轮盘赌模型，优化了选择策略；采用不同的随机数交叉和高斯变异，改进了交叉算子与变异算子。测试结果表明，该方法具有良好的运算效率，拓宽了搜索空间，提高了对重点区域的搜索能力，降低了成本，为复杂问题的优化提供了一种新的思路。

Application of Improved Genetic Algorithm in Design of Feed Formula

According to the malpractices of many feeding formula software,a new method of feed formula design for white cashmere goats based on improved genetic algorithm was proposed.Compared with the standard genetic algorithm,it adopted real coding and introduces the problem areas of elicitation information,increased the search capabilities;Replaces the traditional roulette wheel model with random league matches in choosing strategy policy,using different random crossover and gauss mutation,improved crossover operator and mutation operator.The test results showed that the method has high computing efficiency,broadens the search space,reduces the cost and so on,and was very effective in solving design problems of feed formulation.Due to the improved genetic algorithm did well in maintaining the population diversity,resolving the long degree non-linear problem with more factors,so it could provided a new way for optimization of complex issues.