中文摘要：

神经网络是高度复杂的大规模的非线性动力学系统,具有丰富的动力学行为。神经网络被广泛应用于信号处理、图像处理和人工智能等问题中,所以其理论研究受到了很大的重视,并得到一系列好的结果。时滞现象是普遍存在于神经网络中的,而造成神经网络系统不稳定和震荡的根源往往是时滞。同时,时滞的种类有多种,不同的时滞会表现出不同的动力学特性,因此研究不同种类的时滞神经网络的稳定性问题具有非常实际的意义。另外由于所获系统信息的不全面或存在一定的信息偏差,系统常常被刻画的不准确,因此利用所获信息得到的系统模型必然存在不确定性。研究一类具有分布时滞的中立型不确定神经网络模型稳定性问题,应用Lyapunov函数理论,以及线性矩阵不等式的技巧,进一步研究了神经网络的稳定性问题,提出了神经网络渐近稳定性的判据,最后通过数值算例来验证所得结论的正确性和有效性。

英文摘要：

Delayed neural networks exhibiting rich dynamical behaviors are a class of complex large scale nonlinear dynamical system. Since neural network have a wide application on signal and image processing and artificial intelligence, its theoretical research has received considerable attention and achieved many perfect products. As we know, delay phenomenon often exists in the neural network systems, making the neural network systems unstable and shocked. At the same time, there are many types of time delays, and different time delays show different dynamic characteristics. So it is significantly important in theory and practice to qualitatively study the stability of the neural networks with many types of delays. On the other hand, the information of system we obtained is incomplete or declinational, system is often depicted inaccurate, so it is necessary to get the system has uncertainty. This paper discusses the stability problem of the uncertain neural networks model with distributed delays and neutral delay. By applying Lyapunov-Krasovskii functional and linear matrix inequality（LMI） technique, the sufficient condition on the stability for the considered system is obtained. The numerical example is given to demonstrate the effectiveness of the proposed method.