基于智能优化算法的测试用例集约简

Test Case Suite Reduction Based on an Intelligent Optimization Algorithm

现代智能电表程序的更新迭换速度不断加快,程序经常发生变化,对智能电表的测试提出了更高的要求,如何提高回归测试的效率,已成为当前一个研究焦点。测试开销的降低是提高回归测试效率最直接的表现,而测试用例集约简是降低测试开销的必要手段之一。以萤火虫算法为基础,文章提出了一种应用于智能电表测试用例集约简问题上的优化算法。首先,该算法用二元优化集覆盖问题来代表智能电表程序的测试用例集约简问题;然后,改进传统的萤火虫算法,通过细胞自动机分类模型,将萤火虫的位置转换为0/1矩阵;最后,结合贪心算法,求解得到最优的测试用例子集。通过对不同的智能电表程序和不同规模测试用例集进行仿真,实验结果显示萤火虫算法相比同类算法在相同测试次数上获得测试开销最优解和最优子集的比例都取得了更好的效果。

The update and replacement of modern smart meter programs is constantly accelerating,and the programs often change. Higher requirements are put forward for the testing of smart meters. It has become a hot issue about how to improve the efficiency of regression testing. The most direct manifestation of improving the efficiency of regression testing is the reduction of test cost. One of the necessary means to reduce test cost is the reduction of test case suite. Based on the firefly algorithm,the article proposes an optimization algorithm applied to the problem of smart meter test case reduction. First,the binary optimization set coverage problem is used to represent the problem of reducing the test case suite of smart meter programs;then,the firefly algorithm is improved to convert the position of the firefly into a 0/1 matrix through the cellular automata classification model;finally,the greedy algorithm is combined toobtainthe optimal suite of test cases. Through the simulation of different smart meter programs and test case suite of different scales,the experimental results show that,the proportion of the firefly algorithm to obtain the optimal solution and the optimal subset of the test cost for the same number of tests is better compared with similar algorithms.