摘要
基于组合电路测试生成的Hopfield神经网络模型,讨论分析了利用混沌神经网络的全局搜索能力进行测试生成的有效算法和基于遗传算法的自适应测试生成。基于混沌神经网络的算法利用混沌所表现出的遍历性与内随机性进行全局搜索;而遗传算法不同于传统的方法,它不需要故障传播、回退等过程,并具有并行计算的能力。计算机仿真结果表明了这两种测试生成算法的可行性与高效性。
Based on a Hopfield neural network model for combi- national circuit test generation, a test generation algorithm with global searching ability of chaotic neural networks and a selfadaptive test generation based on genetic algorithm is analyzed. By means of the inherent stochastic and ergodic property of chaotic searching. The genetic algorithm is radically different from the conventional methods, and it doesn＇t need the processes of propagation and backtracks. The experimental results confirm the feasibility and effectiveness of the algorithm.
出处
《微计算机信息》
北大核心
2005年第08X期125-126,92,共3页
Control & Automation
关键词
神经网络
混沌搜索
遗传算法
测试生成
neural network, chaotic searching, genetic algorithm, test generation
