Topology Identification of General Dynamical Network with Distributed Time Delays

This paper addresses the problem of identifying the graph structure of a dynamical network using measured input/output data. This problem is known as topology identification and has received considerable attention in recent literature. Most existing literature focuses on topology identification for networks with node dynamics modeled by single integrators or single-input single-output (SISO) systems. The goal of the current paper is to identify the topology of a more general class of heterogeneous networks, in which the dynamics of the nodes are modeled by general (possibly distinct) linear systems. Our two main contributions are the following. First, we establish conditions for topological identifiability, i.e., conditions under which the network topology can be uniquely reconstructed from measured data. We also specialize our results to homogeneous networks of SISO systems and we will see that such networks have quite particular identifiability properties. Secondly, we develop a topology identification method that reconstructs the network topology from input/output data. The solution of a generalized Sylvester equation will play an important role in our identification scheme.

This paper aims at investigating the topology identification problem of complex dynamical networks with varying node dynamics parameters and fixed inner coupling matrices. In particular, by employing the unified chaotic system as node dynamics, this work further explores the influence of continuously changing node dynamics parameters on topology identification of complex dynamical networks with different coupling strengths. Results show that for sufficiently small or large coupling strengths, the performance of topology identification is not affected by the change of node parameters. Specifically, for small enough coupling strengths, the topological structure can be completely identified regardless of the change of node parameters, while for sufficiently large coupling strengths, the connectivity (presence and absence of connections) cannot be successfully identified. Furthermore, for certain coupling strengths, with the increase of node dynamics parameters, the topology identification varies from completely unidentifiable to partially or event completely identifiable. Therefore, the synchronization-based topology identification depends on node dynamics. Even for the same node dynamical model, different parameters can have a significant impact on identification results. Furthermore, for networks consisting of chaotic oscillators defining node dynamics, small coupling strengths are conducive to topology identification. A broader conclusion is that projective synchronization, rather than just complete synchronization, is an obstacle to the network topology identification. The findings in this paper will add to our understanding of conditions for identifying topologies of complex networks.

Distribution network (DN) topology identification (TI) is the basis of distribution network state estimation. However, the connection of high-penetration renewable energy makes TI of DN more challenging. Thus, a TI method of active distribution network (ADN) based on one-dimensional convolutional neural network is proposed in this manuscript. Based on the sensitivity of node voltage to topology changing of DN, the characteristics of nodes are analyzed to select the key nodes to place the distribution network phasor measurement unit (DPMU), which can save investment and reduce the redundancy of model training. Several tests are carried out with the modified IEEE-33 bus DN with photovoltaic (PV) units. The results show that the proposed distribution network topology identification method can realize the high accuracy TI in ADN under limited DPMU measurement.

Topology identification of biological neural networks has received great attention in recent years. But most existing researches ignored the importance of linear independence condition. Once network occurs inner synchronization, such a condition would inevitably break, which could lead to the failure of topology identification. However, synchronous firing between neurons is unavoidable in biological neural network. This phenomenon makes it difficult to identify the topology of neural networks. In this paper, we proposed a new method to identify Hindmarsh-Rose Neural Networks. By this method, the original neural network can implement inner desynchronization and the topology observers can successfully identify the unknown topology. It is worth mentioning that the our method can ensure that the linear independence condition naturally holds. This means that topology identification will not meet with failure. Finally, the effectiveness of the proposed method is successfully validated by numerical simulations.

A Lyapunov functional for a neutral system with a distributed time delay

One of the important problems that exist today in distribution networks is the lack of accurate and reliable information about its topology and equivalent circuit. The article presents a new approach to identification of the distribution network topology. The approach is based on active distribution network management. It is possible thanks to the development and applying of a mobile laboratory built on high-precision measuring equipment with an accuracy class of 0.05. The mobile laboratory is built on a three-level architecture, which includes: intelligent electronic devices; Supervisory Control And Data Acquisition server; application software that interacts with the user. It is shown that due to active load control, it becomes possible to identify the network topology using a simple recursive algorithm based on the analysis of voltage changes. The description of methodology and algorithm of network topology identification is resulted. The method for identifying the topology of 0.4 and 6 (10) kV networks consists of two stages: identification of the phases of each load; network topology identification independently for each phase. Considering 10kV feeder, the efficiency of this method was proved with the help of mathematical modelling. A detailed description of each stage of the algorithm is presented.

We study the permanence, extinction, and global asymptotic stability for a nonautonomous malaria transmission model with distributed time delay. We establish some sufficient conditions on the permanence and extinction of the disease by using inequality analytical techniques. By a Lyapunov functional method, we also obtain some sufficient conditions for global asymptotic stability of this model. A numerical analysis is given to explain the analytical findings.

The automatic identification of the topology of power networks is important for the data-driven and situation-aware operation of power grids. Traditional methods of topology identification lack a data-tolerant mechanism, and the accuracy of their performance in terms of identification is thus affected by the quality of data. Topology identification is related to the link prediction problem. The graph neural network can be used to predict the state of unlabeled nodes (lines) through training on features of labeled nodes (lines) with fault tolerance. Inspired by the characteristics of the graph neural network, we applied it to topology identification in this study. We propose a method to identify the topology of a power network based on a knowledge graph and the graph neural network. Traditional knowledge graphs can quickly mine possible connections between entities and generate graph structure data, but in the case of errors or informational conflicts in the data, they cannot accurately determine whether the relationships between the entities really exist. The graph neural network can use data mining to determine whether a connection obtained between entities is based on their eigenvalues, and has a fault tolerance mechanism to adapt to errors and informational conflicts in the graph data, but needs the graph data as database. The combination of the knowledge graph and the graph neural network can compensate for the deficiency of the single knowledge graph method. We tested the proposed method by using the IEEE 118-bus system and a provincial network system. The results showed that our approach is feasible and highly fault tolerant. It can accurately identify network topology even in the presence of conflicting and missing measurement-related information.

This paper concentrates on the research of the topology identification problem for coupled neural networks with multiple output couplings (CNNMOCs). Based on the designed adaptive state feedback controller and coupling weight adjustment scheme, a topology identification criterion is presented. Furthermore, the case that the parameter uncertainties occur in individual neural network is also discussed. Lastly, two numerical examples are provided to demonstrate the validity of the proposed criteria.

This paper deals with the proportional-integral control problem for a class of stochastic Markovian jum v-Krasovskii functional and free-weighting matrix method, the novel delay-dependent robust stabilization criterion for the stochastic Markovian jumping systems is formulated in terms of linear matrix inequality (LMI). When the LMI is feasible, an explicit expression of the desired proportional-integral controller is given. Designed controller, based on the obtained criterion, ensures asymptotically stable in the mean square sense of the resulting closed-loop system for all admissible uncertainties and time delay.

This paper investigates non-fragile exponential H∞ control problems for a class of uncertain nonlinear networked control systems (NCSs) with randomly occurring information, such as the controller gain fluctuation and the uncertain nonlinearity, and short time-varying delay via output feedback controller. Using the nominal point technique, the NCS is converted into a novel time-varying discrete time model with norm-bounded uncertain parameters for reducing the conservativeness. Based on linear matrix inequality framework and output feedback control strategy, design methods for general and optimal non-fragile exponential H∞ controllers are presented. Meanwhile, these control laws can still be applied to linear NCSs and general fragile control NCSs while introducing random variables. Finally, three examples verify the correctness of the presented scheme.

This paper presents a neotype control strategy for maximum power point tracking (MPPT) of doubly-fed induction generator based wind turbine (DFIG-WT). The strategy adopts proportional-integral (PI) based regulators and feedback linearization control (FLC) based regulators in the power and current control loops of DFIG, respectively. The FLC based regulator can fully decouple and linearize the nonlinear system and provide optimal control performance in a wide range operating conditions. However, since the FLC strategy might require some unaccessible state variables of the mechanical system, the PI based regulators are able to simplify the control implementation in the power loop. Simulation studies are undertaken in a single-machine system operating under ramp-change and random wind profiles, respectively. Simulation results reveal that the proposed strategy performs better than the conventional vector control (VC) scheme under varying operation conditions.

Many distribution system operations (e.g., state estimation, control, fault detection/localization) rely on the assumption that the underlying topology is accurately defined. In general, topology identification is a challenging problem in distribution systems as these systems are unobservable with a very limited number of available measurements. In this paper, we tackle this problem by designing a compressive sensing framework that jointly estimates the systems states and network topology via an integrated mixed integer nonlinear program (MINLP) formulation. Two reformulations of the original MINLP problems are investigated. Firstly, in order to remove the nonlinearity in the MINLP formulation, a mixed integer linear programming (MILP) problem that employs auxiliary variables is derived. Furthermore, to achieve a faster solution, convex relaxation of the original formulation is derived. Finally, using a Markovian model for topology changes, prior information about system topology is used to improve topology identification particularly when a limited amount of measurements is available. Simulation results on IEEE 37-bus test feeder and IEEE 123-bus test feeder illustrate the efficiency and scalability of the proposed approaches from both state estimation and topology identification point of view (even with 30% of available data).

In this paper, we propose a machine learning-based mobility robustness optimization algorithm to optimize handover parameters for seamless mobility under dynamic small-cell networks. Small cells can be arbitrarily deployed, portable, and turned on and off to fulfill wireless traffic demands or energy efficiency. As a result, the small-cell network topology dynamically varies challenging network optimization, especially handover optimization. Previous studies have only considered dynamics due to user mobility in a specific static network topology. To optimize handovers under dynamic network topologies, together with user mobility, we propose an algorithm consisting of two steps: topology adaptation and mobility adaptation. To adapt to a dynamic topology, the algorithm obtains prior knowledge, which presents a belief distribution of the optimal handover parameters, for the current network topology as coarse optimization. In the second step, the algorithm fine-tunes the handover parameters to adapt to user mobility based on reinforcement learning, which utilizes the knowledge obtained during the first step. Under a dynamic small-cell network, we showed that the proposed algorithm reduced adaptation time to 4.17% of the time needed by a comparative machine-based algorithm. Furthermore, the proposed algorithm improved the user satisfaction rate to 416.7% compared to the previous work.

Due to the constantly moving of the communication nodes in the combat environment, the network topology is highly dynamic and uncertain. However, combat missions require the real-time and accurate information transmission through the military network. Therefore, it is of great practical interest to design a dynamic routing algorithm to accommodate to the dynamic and uncertain network topology. We consider this important problem in this paper and make the following contributions. First, a semi-Markov prediction model is proposed to predict the uncertainty of the network topology. Second, a dynamic resource table is updated by the formation transition probability which can describe the dynamic of the network topology. Third, the Q-learning method is applied to find an optimal routing in the uncertain and dynamic network topology. Numerical experiments demonstrate that the proposed method can predict the network topology more accurate and obtain an optimal routing strategy with improved transmission efficiency and reduced transmission delay.KeywordsMilitary networksSemi-MarkovTopology predictionQ-learningDynamic routing