基于遗传二进制粒子群混合算法的测试点决策研究  

An Approach for Decision of Test points based on GABPSO

作　　者：庞威[1] 吕晓峰[2] 姚成柱[3] 马羚[1] 

机构地区：[1]海军航空工程学院研究生管理大队 [2]海军航空工程学院兵器科学与技术系 [3]91206部队

出　　处：《计算机测量与控制》2014年第1期149-151,158,共4页Computer Measurement &Control

摘　　要：测试点选择是测试性分析和设计的前提。通过对测试问题的分析,建立了数学模型和构造了衡量测试集优劣程度的启发式函数,提出了一种遗传-二进制粒子群混合算法求解满足测试性指标要求的最小完备子集。应用实例表明,该算法能够有效的克服单一算法陷入局部最优和早熟收敛等不足,提高了搜索效率,能够有效快速的获得全局最优解。Test points selecting is the primary problem of the analysis and design for a testing system. A mathematical model is founded and a heuristic function is structured which is used to measure the degree of the test sets. Then a hybrid algorithm based on GA and BPSO is proposed to solve the minimum complete test set that satisfies the testability requirement. Experiments application shows that the hybrid algorithm not only overcome the local optimization and premature convergence, but also improves the searching efficiency and then achieve the global optimal solution.

关 键 词：测试点选择 GABPSO算法 启发函数 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]