中文摘要：

将服务选择问题建模为带QoS约束的非线性最优化问题，并提出了一种参数自适应的改进遗传算法（IPAGA）．构造了基于双曲正切函数的非线性参数变换函数，当迭代次数或种群多样性程度增加时，使遗传算法的交叉和变异概率相应地非线性递减，以保证算法的全局收敛性和收敛速度．实验结果表明：算法能够快速搜索出全局近似最优解，具有很高的有效性和可行性．

英文摘要：

The service selection problem was modeled as a problem of nonlinear optimization with QoS （quality of service） constraints. Then, an improved parameter adaptive genetic algorithm （IPAGA） was proposed. A nonlinear parameter transforming function based on the hyperbolic tangent function was constructed, which made the crossover probability and the mutation probability decrease nonlin- early with the increasing of iterations and population diversity. Thereby, the convergence speed and the global convergence were ensured. The experimental results show that an approximate optimal re- sult can be searched out quickly. The efficiency and feasibility of our approach are demonstrated in the experimental evaluation.