With the continuous development of Web service composition, many services with the same or similar functions but different non-functional attributes have emerged. How to meet the user's demand for quality of service (QoS) is a challenging topic when selecting services. Aiming at the problem of service composition optimization, a genetic algorithm for Web service selection based on QoS perception is proposed. When this method is used to find the optimal composition of a service, in order to improve the efficiency of the solution, several candidate services are selected for each abstract service, which greatly reduces the size of the solution space. Then, the genetic algorithm is used to solve this nonlinear programming problem. The experimental results show that this method has better execution efficiency and fitness, and the path is templated to reduce the number of service groups The chromosome coding method is used to solve the problem of multi-path selection. Through the simulation experiment, Compared with the traditional algorithm, when the genetic algebra is more than 300 and the number of candidate services is more than 60, the proposed algorithm converges faster and the fitness of the optimal solution is higher.

随着Web服务组合的不断发展，出现了许多功能相同或相似但非功能属性不同的服务。如何满足用户对服务质量（QoS）的需求是选择服务时的一个具有挑战性的课题。针对服务组合优化问题，提出了一种基于QoS感知的Web服务选择遗传算法。当使用这种方法寻找服务的最优组合时，为了提高解的效率，每个抽象服务都选择了几个候选服务，大大减少了解空间的大小。然后，遗传算法被用来解决这个非线性规划问题。实验结果表明，该方法具有较好的执行效率和适应度，并对路径进行模板化，减少服务组数量，采用染色体编码方法解决多路径选择问题。通过仿真实验，与传统算法相比，当遗传代数大于300，候选服务数量大于60时，所提算法收敛速度更快，最优解的适应度更高。