CAJ | 학술논문

针对传统粒子群优化算法生成测试数据容易产生早熟收敛而陷入局部最优的问题,提出一种基于改进粒子群算法的组合测试数据生成算法。该算法在粒子群算法的基础上引入一种惯性权重自适应调整策略,根据粒子的适应度不同采用不同的惯性权重,从而有效的平衡算法的全局和局部搜索能力,增加种群的多样性并提高算法的搜索效率。仿真实验表明该算法与传统粒子群算法相比,所需迭代次数减少,生成组合测试数据速度快。
침대전통입자군우화산법생성측시수거용역산생조숙수렴이함입국부최우적문제,제출일충기우개진입자군산법적조합측시수거생성산법。해산법재입자군산법적기출상인입일충관성권중자괄응조정책략,근거입자적괄응도불동채용불동적관성권중,종이유효적평형산법적전국화국부수색능력,증가충군적다양성병제고산법적수색효솔。방진실험표명해산법여전통입자군산법상비,소수질대차수감소,생성조합측시수거속도쾌。
To solve the problem of the traditional Particle Swarm Optimization （PSO） algorithm＇s premature convergence and local optimum, an improved PSO is presented for test data generation in combinatorial testing. Based on the traditional PSO, inertial weight adaptive adjust- ment strategy has been used. In the new algorithm, particles have different inertia weight with different fitness. These strategies improve the PSO algorithm at the aspects of diversity and the balance of exploration and exploitation. Simulation results show that the improved algorithm obviously reduces the number of iterations, improves the speed of combinatorial test data generation.