22 January 2025 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
Phase-change memory (PCM) is a novel type of nonvolatile memory and offers low power consumption, high integration, and significant plasticity, making it suitable for neural synapses. In this article, we investigate the global exponential stabilization (GES) of phase-change inertial neural networks (PCINNs) with discrete and distributed time-varying delays. Initially, a piecewise equation is established to model the electrical conductivity of PCM. Based on this, we use PCM to simulate neural synapses, and a class of PCINNs with discrete and distributed time-varying delays is formulated. A continuous state feedback controller is designed to obtain the ${\boldsymbol {\rho} {\textrm {th}}({\rho \ge 1})}$ moment GES conditions of PCINNs in the Filippov sense by using differential inclusion theory, comparison strategies, and inequality techniques. Additionally, the global exponential stability conditions of phase-change Hopfield neural networks are obtained, expressed in the form of an M-matrix. Finally, three simulation examples are provided to verify the effectiveness of the theoretical results.
27 July 2021 in International Journal of Control, Automation, and Systems
15 October 2020 in Neural Computing and Applications
9 January 2018 in Complexity
The dynamics behaviors of Kaldor–Kalecki business cycle model with diffusion effect and time delay under the Neumann boundary conditions are investigated. First the conditions of time‐independent and time‐dependent stability are investigated. Then, we find that the time delay can give rise to the Hopf bifurcation when the time delay passes a critical value. Moreover, the normal form of Hopf bifurcations is obtained by using the center manifold theorem and normal form theory of the partial differential equation, which can determine the bifurcation direction and the stability of the periodic solutions. Finally, numerical results not only validate the obtained theorems, but also show that the diffusion coefficients play a key role in the spatial pattern. With the diffusion coefficients increasing, different patterns appear.
28 April 2021 in Neural Processing Letters
10 December 2020 in IEEE Access
This paper studies the data-driven optimal synchronization problem for complex networks (CNs) with unknown dynamics. By using pre-compensation technology, a compensator and a controller are proposed. Then, an augmented error system is constructed, which can circumvent the requirement of system dynamics. It is revealed that the the optimal synchronization control of CNs works as the optimal regulation of the augmented system with a performance function. A novel policy iteration (PI) algorithm is given to ensure that the iterative performance function can converge to the optimal value which is the solution of the coupled Hamilton-Jacobi-Bellman equation (HJB), which means that the optimal regulation of the augmented system can be solved and the synchronization can be achieved. Based on this, a novel data-driven control scheme is proposed, which is composed of parts: compensator, controller and critic network. The iterative performance is generated by critic network. The compensator is used to construct the control parameter by using performance and the controller is used to construct control input by using control parameter. Both compensator and critic network are implemented by neural networks (NNs) and only depend on the process sampling data. Finally, we use robot network as an example to verify the effectiveness of proposed control scheme.