Bayesian Network Model Test Configuration Method based on Genetic and Binary Discrete Particle Swarm Combination Algorithm
Lu Han,Xianjun Shi,Taoyu Wang
DOI: https://doi.org/10.1088/1742-6596/1642/1/012007
2020-09-01
Journal of Physics: Conference Series
Abstract:Abstract Test configuration is an important part in the process of testability design. Most of the existing test configuration methods are based on multi-signal flow diagram model and adopt genetic algorithm or particle swarm optimization (pso). However, there are many problems in this solution. First of all, the multi-signal flow diagram model has poor ability to express uncertain information, low model accuracy, large deviation in the calculation of testability indicators, and the model does not have the ability to learn and update. Secondly, the efficiency of genetic algorithm is low and the computation time is long, while the binary discrete particle swarm optimization algorithm is easy to fall into the local optimal. To solve the above two problems, a test configuration method based on the hybrid algorithm of genetic -binary discrete particle swarm optimization is proposed. This method can combine the global search ability of genetic algorithm with the optimal speed of binary particle swarm optimization, and use the bayesian network model to calculate more accurate testability indexes. It is proved that the algorithm can make full use of the high-precision information provided by the model, and the calculation speed is fast.And it is not easy to fall into the local optimal solution.