基于惯性权重优化的测试用例生成方法  

Testing cases generation based on improved inertial weight

作　　者：李昭 陈雨杭 沈金伟 聂旭清 LI Zhao;CHEN Yu-hang;SHEN Jin-wei;NIE Xu-qing(College of Mathematics and Computer,Guangdong Ocean University,Zhanjiang 524088,China;College of Computer and Information Technology,China Three Gorges University,Yichang 443002,China;R&D Center,Guangdong Bangxin Data Technology Limited Company,Guangzhou 510000,China)

机构地区：[1]广东海洋大学数学与计算机学院,广东湛江524088 [2]三峡大学计算机与信息学院,湖北宜昌443002 [3]广东邦鑫数据科技股份有限公司研发中心,广东广州510000

出　　处：《计算机工程与设计》2023年第6期1743-1748,共6页Computer Engineering and Design

基　　金：广东省海洋经济发展专项基金项目(GDNRC[2020]052);湛江市科技攻关计划基金项目(2021B01021);广东海洋大学重大科研基金项目(GDOU2017052605);广东海洋大学科研启动经费基金项目(R20079)。

摘　　要：针对组合测试测试用例生成环节中局部最优的问题,提出一种基于惯性权重优化的测试用例生成方法。结合粒子群算法的优点,以生成最小规模覆盖表为目标,改进适应值函数,设计一种惯性权重微分递减策略;在此基础上,更新调优后的惯性权重值,使用测试用例演化策略迭代生成测试用例集,提升生成更小规模覆盖表的能力。实验结果表明,该方法在高覆盖需求下,覆盖表最小生成规模和平均生成规模优于其它方法。Aiming at the problem of local optimization in test case generation,a test case generation method based on inertia weight optimization was presented.To generate the minimum coverage table,the fitness function was improved,and a differential decreasing strategy of inertia weight was designed.On this basis,the optimized inertia weight value was updated,and the test case evolution strategy was used to iteratively generate the test case set,so as to improve the ability of generating smaller cove-rage tables.Experimental results show that the proposed method outperforms other methods in terms of minimum generation scale and average generation scale of coverage table under high coverage requirements.

关 键 词：软件测试 测试用例生成 惯性权重递减 适应值函数 测试用例集 覆盖表生成 最小规模覆盖表 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]