一种测试数据自动生成算法的研究  

Research on an Automatic Generation Algorithm of Test Data

作　　者：苗晓旭 曾佩杰[1] MIAO Xiao-xu;ZENG Pei-jie(China Helicopter Research and Development Institute,Jingdezhen 333001,China)

机构地区：[1]中国直升机设计研究所,江西景德镇333001

出　　处：《测控技术》2018年第B11期36-38,43,共4页Measurement & Control Technology

摘　　要：测试数据自动生成是测试自动化技术实现的关键环节。为了提高测试自动化的效率，设计测试数据自动生成模型，提出一种基于单亲遗传算法的测试数据自动生成算法。该算法以三复本竞争选择算子较好地保持了种群的多样性，采用全新的遗传算子，基因换位、基因移位、基因倒位、基因突变，所有的遗传操作均在一条染色体上进行，不仅可以保证新一代个体具有成为可行解的基本特性，又可以提高对解空间的搜索能力。通过三角形判别和折半查找程序，改进的算法与传统遗传算法进行了对比分析。实验结果表明，该算法相较传统遗传算法具有搜索效率高、不易陷入局部最优解等特点，在测试数据自动生成问题中具有较强的可行性和高效性。Automatic generation of test data is a key link in the realization of test automation technology. In or- der to improve the efficiency of test automation, a test data automatic generation system model was designed and a test data automatic generation algorithm based on the IPGA （improved parthenon-genetic algorithm） was ~ The algorithm maintained the diversity of the population with a three-copy competition selection oper- ator. It used new genetic operators, all genetic operations such as gene transposition, gene translocation, gene inversion, gene mutation were performed on one chromosome, which not only ensured that a new generation of individuals has the basic characteristics of being a feasible solution, but also can improve the search ability of the solution space. Through the triangle program and binary search program, the IPGA algorithm was compared with the GA algorithm. Experimental results show that the IPGA algorithm has high search ability and out of lo- cal optimal solution compared with the GA algorithm in the test data generation.

关 键 词：软件测试 单亲遗传算法 测试数据自动生成 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]