蹇红梅， 成新文， 曾  燕
（四川理工学院计算机学院，四川 自贡 643000）
摘  要：针对软件测试数据采用遗传算法和粒子群算法自动生成算法复杂和容易早熟等问题，提出一种动态调整收缩扩张因子的自适应量子粒子群算法（AQPSO）。该算法通过引入粒子进化度和聚合度，收缩扩张因子随粒子进化度因子和聚合度因子变化而变化，从而实现算法的动态自适应性，提高算法收敛速度和求解精度。软件测试数据自动生成实验验证了该算法的有效性和可行性。
关键词：量子粒子群；软件测试；测试数据生成；收缩扩张因子
中图分类号：TP206+.1；TP301.6；TP311.52；TP311.55     文献标志码：A    文章编号：1674-5124（2013）03-0100-04
Automatic generation of software test data based on adaptive QPSO algorithm
JIAN Hong-mei， CHENG Xin-wen， ZENG Yan
（School of Computer Science，Sichuan University of Science & Engineering，Zigong 643000，China）
Abstract: For the complexity and prematurity of the automatic software test data generation algorithm based on the genetic algorithm and the standard particle swarm optimization algorithm， an adaptive quantum-behaved particle swarm optimization （AQPSO） algorithm is presented to dynamically adjust the contraction expansion factro to overcome these shortcomings. By introducing the evolution degree and polymerization degree of the particle into this method， the contraction expansion factor keeps changing as the evolution dgree and polymerization dgree factors are changing， orderly the dynamical and adaptive algorithm is realize， which improves the convergence speed and precision the traditional algorithm. The experiment on automatic generation of software test data verified the validity and feasibility of the algorithm.
Key words: QPSO； software testing； test data generation； contraction expansion factor