基于量子行为粒子群优化的软件可靠性模型参数估计  

Estimating parameters of software reliability models based on quantum-behaved particle swarm optimization

作　　者：陈晓[1] 江建慧[1] 

机构地区：[1]同济大学软件学院,上海201804

出　　处：《高技术通讯》2014年第5期479-485,共7页Chinese High Technology Letters

基　　金：863计划(2007AA01Z142)资助项目

摘　　要：研究了准确估计软件可靠性模型参数这一难题,在分析现有软件可靠性模型参数估计方法的基础上,提出了一种基于量子行为粒子群优化(QPSO)算法的软件可靠性模型参数估计方法。针对该方法,选取了7个数据样本和三个软件可靠性模型(G-O模型、M-O模型和Weibull模型)进行了仿真实验,并且与粒子群算法、蚁群算法仿真实验进行对比。实验结果表明,基于QPSO算法的软件可靠性模型参数估计方法与粒子群算法、蚁群算法相比,求解精度高,误差较小,模型适应性强。The problem of precise estimation of the parameters of most software reliability models was studied, and a new approach for parameter estimating for software reliability models based on the quantum-behaved particle swarm optimization （QPSO） algorithm was proposed on the basis of the analysis of the existing methods for software reliability models＇ parameter estimation. To verify the performance of the proposed method, three software reliability models of G-O, M-O and Weibull were compared according to the classification of software reliabiliy models and seven data samples were adopted to conduct the simulation experiment, and the experimental results were compared with the particle swarm optimization （PSO） algorithm and the ant colony algorithm. The comparison results demonstrate that the proposed method has the higher precision, minor error and stronger adaptability.

关 键 词：量子行为粒子群优化(QPSO) 软件可靠性模型 参数估计 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] O212.1[自动化与计算机技术—控制科学与工程]