基于离散粒子群算法的测试选择  

Test Selection Based on Binary Particle Swarm Optimization

作　　者：蒋荣华[1] 王厚军[1] 龙兵[1] 

机构地区：[1]电子科技大学自动化学院,成都610054

出　　处：《电子测量与仪器学报》2008年第2期11-15,共5页Journal of Electronic Measurement and Instrumentation

基　　金：国防基础科研项目(编号:A1420061264);电子科大博士平台建设项目(编号:05BS00701)

摘　　要：本文将离散粒子群算法(BPSO)首次应用于测试选择,结合测试选择自身的特点,重新定义了粒子及其速度;通过定义带有测试选择指标的适应度函数对粒子进行优化,并根据其容易陷入局部最优的特点,引入了线性惯性权重因子;同时本文将故障发生的概率作为评价测试集优劣的一个重要指标,具有重要的应用价值。文中的实例验证了该算法的有效性,利用该算法不仅可以获得较高的故障检测率、故障隔离率及较小的测试矢量集,而且还可以很快地找到发生概率大的故障。This paper firstly applies particle swarm optimization algorithm to test selection. Based on characteristics of test selection, it redefines particles and velocities of BPSO （Binary Particle Swarm Optimization）. It optimizes the particles using fitness function that includes the indexes of test selection and introduces linearly decreasing weight to BPSO according to its characteristic of getting into local optimal easily. It also evaluates the test with the probability of the occurrence of faults, which has high values in practical application. Experimental results show that the proposed algorithm can not only achieve higher fault detection rate, isolation rate and more compact test sets when compared with other similar test selection algorithms, but also detect the faults with higher probability quickly.

关 键 词：离散粒子群算法 线性惯性权重 测试选择 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]