基于组合电路测试生成的离散Hopfield神经网络模型,将混沌搜索与Hopfield网络的梯度算法相结合,利用混沌搜索的内随机性及遍历性来克服梯度算法易于陷于局部极小的缺点,形成一种具有全局搜索能力的测试生成有效算法。该算法综合了随机性和确定性算法的优点,其性能优于一般的随机性算法。实验结果验证了该测试生成算法的有效性。 Based on the discrete Hopfield neural network model generated by the combinational circuit test, the chaos search is combined with the gradient algorithm of Hopfield network. By utilizing the internal randomness and ergodicity of chaotic search, the gradient algorithm is easily overcome by the shortcoming of local minima. Testing of global search capabilities generates efficient algorithms. The algorithm combines the advantages of stochastic and deterministic algorithms, and its performance is superior to the general stochastic algorithm. Experimental results verify the effectiveness of the test generation algorithm.