鉴于标准遗传算法有局部搜索能力差,求解精度不高等缺点,引入多元回归分析和模式搜索算法,利用最小二乘法求得种群寻优方向的信息,建立一种新的遗传算法,并由此指导种群中的每一个个体模式搜索,提高整个种群的寻优速度和效能,以非线性连续优化仿真为例,结果表明,该方法比遗传算法性能更好。该算法应用于优化化工过程,效果良好。 In view of the shortcomings of standard genetic algorithm, such as poor local search ability and poor solution precision, the multi-regression analysis and pattern search algorithm are introduced, and the least square method is used to obtain the information of population optimization direction and establish a new genetic algorithm. The search for each individual model in the population improves the optimization speed and efficiency of the whole population. Taking non-linear continuous optimization simulation as an example, the results show that the proposed method performs better than the genetic algorithm. The algorithm is applied to optimize the chemical process, the effect is good.