Study On Automatic Test Data Generation Based On Particle Swarm Optimization Algorithm

Meta-heuristic search algorithm in the field of intelligent computation solves different optimization problems in software engineering areas,especially in software test data automatic generation.Automatic software structure test data generation is important to enhance the efficiency of test,cut test cost and guarantee software quality.Particle swarm optimization algorithm as a random search algorithm,as a new intelligent optimization technology,because of its global search ability strongly,implement easily,understand easily,the advantages of high precision,is widely used.According to the basic particle swarm optimization algorithm that easily falls into the local optimal solution and the problem of low convergence rate and low search accuracy,this thesis proposes reduced adaptive particle swarm optimization algorithm(RAPSO),a new particle swarm optimization with mix topological structure(MPSO)and adaptive chaos particle swarm optimization algorithm(ACPSO)for generating software structural test data automatically.The contents of this thesis are shown as follows:(1)This thesis reduces the evolution equations of particle swarm,adjustment scheme based on inertia weight is proposed,inertia weight is directly acted on the particle position,and employs the summation of branch-function as fitness function.The RAPSO discards the particle velocity and reduces the PSO from the second order to the first order difference equation.The evolutionary process is only controlled by the variables of the particles position.Triangle discrimination programs are used to experiment.The experimental results show that this approach can effectively improve the efficiency of generating test data automatically.(2)This thesis presents a new particle swarm optimization with mix topological structure.Based on the analysis of the different neighborhood topology structure effect on the performance of particle optimization algorithm,MPSO based on the combination of global optimization and local optimization.In each generation,by observing the feedback information of diversity of the population,the particle speed update method is selected by global topology model or local topology model.Theexperimental results show that MPSO maintains the swarm diversity,avoids falling into local optimization,and improves the convergence speed.(3)This thesis proposes adaptive chaos particle swarm optimization algorithm(ACPSO)for generating software structural test data automatically.Firstly,by means of ergodicity and randomicity of chaos,the initial population position is generated by using appropriately chaotic mapping,so that these particles can be distributed uniformly over the solution space.Secondly,adaptive inertia weight scheme is adopted to improve the convergence rate and balance local search ability and global search ability.Furthermore,by the diversity of the population judging the local convergence,chaos perturbation strategy is utilized to avoid the premature convergence.The experiments show that this approach can effectively improve the efficiency of generating test data automatically and avoid premature convergence problem.