The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
This paper studies the consensus problem in second-order multi-agent systems. For the first time, an event-triggered impulsive consensus control scheme is proposed to solve this problem. Under the event-triggered impulsive consensus control scheme, a distributed event-condition is defined in advance for each agent and the impulsive control is taken only when such condition satisfies, and no any actions...
The Network Function Placement (NFP) problem involves placing Virtual Network Functions (VNFs) in a network in order to meet the Service Function Chain (SFC) requirements of the flows through the network. Simultaneously, the usage of network resources by the VNF instances must be optimized. Prior work primarily treated this as a constraint satisfaction problem, using linear programming to find optimal...
Analysis of network traffic behavior and modeling to predict, for network management and security early warning has a very important significance. An improved FOA-ESN method using opposition-based learning (OBL) mechanism for the network traffic prediction with multiple steps is proposed in this paper. Firstly, reconstructing the phase space of the original network flow time series, and then building...
This paper considers the depth control problem of autonomous underwater vehicles (AUVs) in discrete time. A neural-network-based deterministic policy gradient (NNDPG) controller is proposed by combining the deterministic policy gradient theorem with neural networks. Two networks, evaluation network and policy network, are designed to respectively approximate the long-term cost function and policy...
This paper deals with the fixed-time consensus for multi-agent systems with input delay. First, finite-time stability for single system with input delay is given with the aid of Artstein model reduction method. Then, two finite-time nonlinear consensus protocols are constructed for multi-agent systems with input delay using the Lyapunov functionals. In particular, the settling-time of the second consensus...
Based on the concept of persistent excitation (PE), a deterministic learning algorithm is proposed for neural network (NN)-based identification of nonlinear systems recently. This paper investigates the quantitative relationship between the PE levels (including the level of excitation), the architectures of NNs and the convergence properties of deterministic learning, which is motivated by a practical...
ADP is an effective optimal method. However, the optimality depends on its network structure and training algorithm. This paper adopts RBF neural network to realize its critic and action networks after a detailed analysis on ADP. The LSM method is introduced as training algorithm, and a novel basis function is defined, which achieves global optimization and online control. The validity is verified...
Based on L1 adaptive control theory, a novel block backstepping control for a class of uncertain multiple-input-multiple-output nonlinear system is proposed. The matched system parametric uncertainty and unmatched general uncertainty including modeling error and external disturbance are considered in the design. The L1 adaptive control is integrated with block backstepping to improve the transient...
In the proposed work, we presented an Artificial Neural Network approach to predict the stock market indices. We outlined the design of the Neural Network model with its salient features and customizable parameters. A number of the activation functions are implemented along with the options for the cross validation sets. We finally test our algorithm on the Nifty stock index dataset where we predict...
Multilayer Perceptron (MLP) and Radial Basis Function (RBF) were proposed in this paper as two models of non-linear ANN based equalization techniques in order to optimize processing performance, tracking, and minimize error of the channel effects and the ascending noise with 16 QAM Modulation, this work will be referenced with one of the most used linear adaptive equalization; Recursive least squares...
This paper builds upon the Koopman spectral analysis tools to develop a method for assessment of the performance of a class of first-order nonlinear consensus networks. This class of networks is defined over an interconnected graph with state-dependent weights that are nonlinear functions of the state of the network. The mean energy of the output of the system with respect to random initial conditions...
For a single-phase distribution network with constant-power, constant-current, and constant-impedance loads (ZIP loads), sufficient conditions are presented that explicitly define a region where a unique load-flow solution exists. The Z-Bus method is shown to be a contraction mapping iteration, which upon initialization within this region, is guaranteed to converge to the unique load-flow solution...
This paper considers the identification problem of nonlinear systems based on single-hidden-layer neural networks (SHLNNs) and Lyapunov theory. A nonlinearly parameterized neural model, whose weights are adjusted by robust adaptive laws, which are designed via Lyapunov theory, is proposed for ensuring the convergence of the residual state error to an arbitrary neighborhood of zero. In addition, a...
This paper deals with the problem of assigning tasks to a set of nodes communicating in a connected graph topology to satisfy the following requirements: assigning all the tasks to the agents; assigning to each agent no more than M tasks; minimizing the maximum total load of each agent. A gossip-based algorithm is presented: starting from an unfeasible solution, at each iteration a node solves a Local-Integer...
This paper considers distributed convex optimization problems over a multi-agent network, with each agent possessing a dynamic objective function. The agents aim to collectively track the minimum of the sum of locally known time-varying convex functions by exchanging information between the neighbors. We focus on scenarios when the communication among the agents is described by a directed network...
In this paper, we develop a projection neural network to solve the convex quadratic programming problem in support vector machine (SVM) learning. Then, we obtain a unique global solution for the proposed neural network. Furthermore, we prove that this network is completely stable and finite-time convergence. To present the feasibility and efficiency of the proposed neural network for solving the SVM...
The main result presented in this paper (whose proof can be found in [1]) is that the fraction of agents (YNk(t)) at state k ∊ X := {1, …, K} associated with an interacting particle system over an appropriate dynamical communication network converges weakly to the solution of a differential equation. The vector macroprocess (YN(t)) = (YN1(t),…, YNk(t)) is not Markov since its evolution depends not...
A new technique for the training of ANNs is presented. The time-domain vibration signals of rolling bearings with different fault conditions are preprocessed using differential evolution method, then further being trained by Levenberg Marquardt method. The processed data are applied as input vectors to artificial neural networks (ANNs) for rolling bearing fault classification. The hybrid training...
The nonnegative least mean square (NNLMS) algorithm has the advantages of simplicity and ease of implementation, but it has a slow convergence rate in sparse nonnegative system identification and its robustness is not strong in an impulsive interference environment. To solve these problems, an lo-norm NNLMS (lo-NNLMS) algorithm is presented by using an lo-norm optimization. Then, an lo-norm nonnegative...
Both boosting and deep stacking sequentially train their units taking into account the outputs of the previously trained learners. This parallelism suggests that it exists the possibility of getting some advantages by combining these techniques, i.e., emphasis and injection, in appropiate manners. In this paper, we propose a first mode for such a combination by simultaneously applying a general and...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.