基于动态调整简化粒子群优化的组合测试用例生成方法  

Combinatorial Test Case Generation Method Based on Simplified Particle Swarm Optimization with Dynamic Adjustment

作　　者：包晓安[1] 鲍超 金瑜婷 陈春宇 钱俊彦[2] 张娜[1] BAO Xiao-an;BAO Chao;JIN Yu-ting;CHEN Chun-yu;QIAN Jun-yan;ZHANG Na(School of Information Science and Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China;Guangxi Key Laboratory of Trusted Software,Guilin University of Electronic Technology,Guilin,Guangxi 541004,China)

机构地区：[1]浙江理工大学信息学院,杭州310018 [2]桂林电子科技大学广西可信软件重点实验室,广西桂林541004

出　　处：《计算机科学》2018年第11期199-203,共5页Computer Science

基　　金：国家自然科学基金(61502430;61379036;61562015);广西自然科学重点基金(2015GXNSFDA139038);浙江理工大学521人才培养计划资助

摘　　要：优化的组合测试中的一个关键是生成的测试用例能够覆盖更多的组合,而粒子群算法在生成强组合覆盖用例方面有其独特的优势和能力。文中提出了一种基于动态调整简化粒子群优化的组合测试用例生成方法。该方法基于粒子群算法生成测试用例,结合混合的优先级one-test-at-a-time策略和基于动态调整的简化粒子群算法生成组合测试用例集,排除了速度因素对粒子优化过程的影响。定义了一个粒子收敛指标,以粒子群早熟收敛程度为依据来动态调整惯性权值,以防止粒子陷入局部最优和后期出现收敛速度慢的情况,从而提高粒子群算法所生成的覆盖表的覆盖组合能力。通过对比实验表明,基于动态调整的简化粒子群优化算法在用例规模和时间成本上具有一定的优势。One of the keys for the optimized combinatorial test is that the generated test case can cover more combinations,and the particle swarm algorithm has the distinctive advantage and capability in generating strong combinatorial coverage cases.This paper proposed a combinatorial test case generation method based on simplified particle swarm optimization based on dynamic adjustment.In this method,test case is generated based on particle swarm algorithm,and the mixed priority one-test-at-a-time strategy and simplified particle swarm optimization algorithm based on dynamic adjustment are combined to generate combinatorial test case set,excluding the influence of velocity factors on the process of particle optimization.Then,a particle convergence criterion is defined,and the inertia weight is dynamically adjusted based on the premature convergence degree ofparticle swarm,so as to prevent that the particles fall into the local optimum and its convergence is slow later,thus improving the capability of coverage combination of the coverage table gene-rated by the particle swarm algorithm.Experiments show that the simplified particle swarm optimization algorithm based on dynamic adjustment has certain advantages in the aspect of case scale and time cost.

关 键 词：简化粒子群算法 测试用例 惯性权值 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]