Learn More
This brief deals with the problem of stability analysis for a class of recurrent neural networks (RNNs) with a time-varying delay in a range. Both delay-independent and delay-dependent conditions are derived. For the former, an augmented Lyapunov functional is constructed and the derivative of the state is retained. Since the obtained criterion realizes the(More)
This paper is concerned with the estimation of the domain of attraction for a class of linear continuous-time systems subject to both interval time-varying delay and actuator saturation. A new type of delay-range-dependent condition is firstly proposed using the free-weighting matrix technique to derive a tighter upper bound on the derivative of a(More)
To solve the problem of gradient descent (GD) method which has low accuracy and easily falling into local optimum, the radial basis function (RBF) based on immune algorithm system (IAS-RBF) is proposed. In this method, each antibody is a RBF neural network and the optimal affinity is calculated by immune algorithm system (IAS) to get the best antibody, then(More)
Many artificial networks in the real world have the power-law degree distribution. This feature causes a lot of difficulties in network design, such as robustness, reliability and other performance issues. In this paper, the congestion problem in power-law communication networks is focused. Since nodes with very large degree usually become the bottleneck in(More)
Echo state network (ESN) is a powerful tool for nonlinear system modeling. However, the random setting of structure (mainly the reservoir) may degrade its estimation accuracy. To create the optimal reservoir for a given task, a novel ESN design method based on differential evolution algorithm is proposed. Firstly, the weight matrix of reservoir is(More)
In the wastewater treatment processes (WWTPs), the oxygen transfer is a nonlinear and large time-delay process, which makes the dissolved oxygen (DO) concentration difficult to control. In this paper, a novel kind of control method based on fuzzy neural network (FNN) is proposed for controlling DO concentration. The parameters of the neural network were(More)
An abnormal solution might occur during the learning process of echo state network if the least singular value of reservoir state matrix is very close zero. To solve this problem, an echo state network based on Levenberg-Marquardt (LM-ESN) algorithm replacing linear regression for output weights is proposed and a new damping term is given. In the proposed(More)