Abstract: In order to solve the Shortest Path（SP） problem in network optimization，this paper establishes a model with constraints for SP problem，and then explores the framework and key steps of Tabu Search（TS） for solving it.This method with advantages of intelligent computation has strong optimization ability and brief structure.It can also handle the problem constraints easily.In the end of this paper，the related algorithm is tested and compared in real example.The results indicate that the algorithm which has a good performance in convergence speed can obtain optimal solution set under certain constraints.It’s more suitable for poor network conditions.Therefore，the algorithm is feasible and effective.

Key words: shortest path, tabu search, network optimization, constraint, intelligent computation

摘要： 针对网络优化算法中的最短路径（Shortest Path，SP）问题，建立了有约束条件的SP问题模型，并探讨了使用禁忌搜索（Tabu Search，TS）算法对其求解的算法框架及关键步骤。该求解方法寻优能力强，结构简明，能方便处理问题约束，具有智能计算方法的优点。最后，通过实例进行测试和比较，证明算法收敛速度快，并能够获得满足约束条件的优解集合，能适应较差网络条件下的多条路径选择，算法是可行和有效的。

关键词: 最短路径, 禁忌搜索, 网络优化, 约束, 智能计算