Large-scale Networked Systems Research Articles

WGAN-DL-IDS: An Efficient Framework for Intrusion Detection System Using WGAN, Random Forest, and Deep Learning Approaches

The rise in cyber security issues has caused significant harm to tech world and thus society in recent years. Intrusion detection systems (IDSs) are crucial for the detection and the mitigation of the increasing risk of cyber attacks. False and disregarded alarms are a common problem for traditional IDSs in high-bandwidth and large-scale network systems. While applying learning techniques to intrusion detection, researchers are facing challenges mainly due to the imbalanced training sets and the high dimensionality of datasets, resulting from the scarcity of attack data and longer training periods, respectively. Thus, this leads to reduced efficiency. In this research study, we propose a strategy for dealing with the problems of imbalanced datasets and high dimensionality in IDSs. In our efficient and novel framework, we integrate an oversampling strategy that uses Generative Adversarial Networks (GANs) to overcome the difficulties introduced by imbalanced datasets, and we use the Random Forest (RF) importance algorithm to select a subset of features that best represent the dataset to reduce the dimensionality of a training dataset. Then, we use three deep learning techniques, Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM), to classify the attacks. We implement and evaluate this proposed framework on the CICIDS2017 dataset. Experimental results show that our proposed framework outperforms state-of-the-art approaches, vastly improving DL model detection accuracy by 98% using CNN.

Distributed Unmanned Aerial Vehicle Cluster Testing Method Based on Deep Reinforcement Learning

In the process of the collaborative work of Unmanned Aerial Vehicle (UAV) clusters, the cluster communication node test is often carried out by a single-node test, which leads to poor topology and robustness of the overall network system, an imbalanced communication network load, and high complexity of the communication test, which seriously affects the diversified needs of current users and the efficiency of large-scale task processing. To solve this problem, a distributed method for UAV cluster testing, called UTDR (distributed UAV cluster Testing method by using Deep Reinforcement learning), based on the Deep Deterministic Policy Gradient (DDPG) is proposed in this work. The system management node is used to monitor the status of the UAV testing task execution node and bandwidth resources. By taking advantage of the method of continuous interaction between the agent and the environment, the future state of the node after processing the current task to be assigned is predicted and evaluated from the perspective of interpretability, so as to achieve the effectiveness and stability of the UAV cluster testing task collaborative execution. The experimental results show that our proposed method can ensure the stable operation of the UAV cluster, accurately predict the future state, and reduce the load degree and bandwidth resource consumption of the large-scale test task network system.

Optimizing energy efficiency in buildings’ cold water systems: A differential pressure control-based global approach

Buildings’ cold water systems have become increasingly complex, with independent equipment control leading to greater energy consumption and greenhouse gas emissions. To improve control efficiency and maximize energy savings, simulation and experimental studies of these systems are essential, although full-scale experimental setups are rare. This paper employs FloMASTER modeling to establish a large-scale practical pipeline network system for validating models, evaluating strategies, and optimizing energy efficiency. The modeling method is validated as accurate, with normalized mean bias error (NMBE) and coefficient of variation of the root mean square error (CVRMSE) values for nodal pressure and flow distribution meeting established evaluation criteria. A quantitative evaluation of constant differential pressure control strategies, considering hydraulic and thermal characteristics as well as energy consumption, reveals that the strategy based on intermediate loop users performs best. This strategy achieves an average hydraulic imbalance rate of only 5.05%, the quickest hydraulic stabilization and restoration time, and comparable secondary pump energy consumption across three control strategies. Furthermore, this simulation study proposes a global combined control strategy that enables more stable operation of the system, reduces chiller energy consumption by 4.66%, secondary pump energy consumption by 9.93%, and overall system energy consumption by approximately 5.38%. These findings suggest that this methodology can be applied to large-scale pipeline networks in complex cold water systems, yielding substantial energy savings.

Prediction-Based Control of Large-Scale Networked Control Systems

Prediction-Based Control of Large-Scale Networked Control Systems

Extinction Chains Reveal Intermediate Phases Between the Safety and Collapse in Mutualistic Ecosystems

Ecosystems are undergoing unprecedented persistent deterioration due to unsustainable anthropogenic human activities, such as overfishing and deforestation, and the effects of such damage on ecological stability are uncertain. Despite recent advances in experimental and theoretical studies on regime shifts and tipping points, theoretical tools for understanding the extinction chain, which is the sequence of species extinctions resulting from overexploitation, are still lacking, especially for large-scale nonlinear networked systems. In this study, we developed a mathematical tool to predict regime shifts and extinction chains in ecosystems under multiple exploitation situations and verified it in 26 real-world mutualistic networks of various sizes and densities. We discovered five phases during the exploitation process: safe, partial extinction, bistable, tristable, and collapse, which enabled the optimal design of restoration strategies for degraded or collapsed systems. We validated our approach using a 20-year dataset from an eelgrass restoration project. Counterintuitively, we also found a specific region in the diagram spanning exploitation rates and competition intensities, where exploiting more species helps increase biodiversity. Our computational tool provides insights into harvesting, fishing, exploitation, or deforestation plans while conserving or restoring the biodiversity of mutualistic ecosystems.

Large-scale cultural heritage conservation and utilization based on cultural ecology corridors: a case study of the Dongjiang-Hanjiang River Basin in Guangdong, China

In the field of world heritage conservation, there has been broad consensus on carrying out heritage conservation research on the basis of spatial integration and interregional and international cooperation. However, there are still many deficiencies in the integration of culture with the environment, regional economic and social development, and the regional, holistic and multimodal conservation and utilization of cultural heritage sites. In China, the Dongjiang-Hanjiang River Basin is a representative area of substantial cultural and ecological value for both Guangdong Province and the whole country. This paper uses the morphological spatial pattern analysis and the minimum cumulative resistance model to integrate cultural ecology sources and establish a cross-regional and large-scale cultural ecology network system that includes 1 main corridor, 22 important corridors and 17 secondary corridors. In addition, based on identified cultural landscape nodes and cultural ecology services, the economy of the cultural ecology corridor could be developed with large-scale co-construction, co-governance and shared working mechanisms to overcome administrative limits and realize the conservation and utilization of multimodal and large-scale heritage sites. This approach has strong theoretical and practical significance for innovative methods in cultural ecology research, as well as for new content in the research of Lingnan culture, ecosystem restoration, and the economic and social development of towns and villages. This article supplements unilateral studies of regional culture and ecology and demonstrates an in-depth application of cultural ecology theory.

Distributed state estimation for leak detection in water supply networks

In this paper, we focus on observer-based approaches of leak detection and isolation in pipeline networks, which has the potential to be extended to large-scale water supply network systems, while in previous studies only single pipeline scenario is considered. Due to several disadvantages of implementing centralized state estimation for large-scale systems, the distributed Kalman filter algorithm has been applied in this paper. The observers are designed on the basis of a set of nonlinear fluid models of pipelines, pumps and other hydraulic accessories, with uncertainties in leak position and magnitude. The approach assumes only flow and pressure measurements at the ends of each duct. Simulation results with a water transmission line are presented to assess the efficiency of the proposed distributed state estimation method while comparing with its centralized counterpart.

Fault detection and diagnosis using the dynamic network framework

A local model-based method for fault detection and diagnosis (FDD) in large-scale interconnected network systems is introduced, using models in a dynamic network framework. To this end, model validation methods are developed for validating single modules in a dynamic network, which are generalized from the classical auto- and cross-correlation tests for open- and closed-loop systems. Invalidation of the model can indicate the detection of a fault in the system. A fault diagnosis algorithm is developed that includes fault isolation and optimal placement of external excitation signals. Numerical illustrations demonstrate the method’s capability to detect a fault in a local module and isolate it within the entire network system.

Predictor-Based Load Frequency Control for Large-Scale Networked Control Power Systems

Predictor-Based Load Frequency Control for Large-Scale Networked Control Power Systems

Controllability scores for selecting control nodes of large-scale network systems

Controllability scores for selecting control nodes of large-scale network systems

Quadratic stability of uncertain large-scale networked systems

In this paper, the quadratic stability of continuous-time uncertain large-scale networked systems is considered. The systems are composed of many subsystems distributed in different locations in space and interconnected according to a certain network topology. A computationally efficient sufficient condition for the quadratic stability of large-scale networked systems is established, utilising effectively the block diagonal structure of the system parameter matrix and the sparseness of the subsystem connection matrix (SCM). Furthermore, a stability condition is presented, which is entirely contingent on the parameters of each subsystem. The simulation results show that the obtained conditions have obvious advantages in analysing the quadratic stability of large-scale networked systems.

(Q, S, R)-Dissipativity Analysis of Large-Scale Networked Systems

This brief investigates the (Q,S,R)-dissipativity of large-scale networked systems in the discrete-time case. The networked system consists of a large number of subsystems, each of which can have different dynamic characteristics, and the connections among subsystems are arbitrary. When the number of subsystems is large, the traditional analysis method based on lumped description may have computational difficulties. To solve this problem, some computationally attractive conditions are derived, which essentially depend on the block-diagonal structure of the system parameter matrices and the sparseness of the subsystem connection matrix. Numerical simulations illustrate that compared with the existing results, the conditions suggested in this brief have higher computational efficiency in the dissipative analysis of large-scale networked systems.

A Review of Optimal Design for Large-Scale Micro-Irrigation Pipe Network Systems

Micro-irrigation pipe network systems are commonly utilized for water transmission and distribution in agricultural irrigation. They effectively transport and distribute water to crops, aiming to achieve water and energy conservation, increased yield, and improved quality. This paper presents a model for the scaled micro-irrigation pipeline network system and provides a comprehensive review of the fundamental concepts and practical applications of optimization techniques in the field of pipeline network design. This paper is divided into four main sections: Firstly, it covers the background and theoretical foundations of optimal design for scaled micro-irrigation pipeline network systems. Secondly, the paper presents an optimal design model specifically tailored for scaled micro-irrigation pipeline networks. And then, it discusses various optimization solution techniques employed for addressing the design challenges of scaled micro-irrigation pipeline networks, along with real-world case studies. Finally, this paper concludes with an outlook on the ongoing research and development efforts in the field of scaled micro-irrigation pipeline network systems. In addition, this paper establishes a fundamental model for optimizing pipeline networks, to achieve minimum safe operation and total cost reduction. It considers constraints such as pipeline pressure-bearing capacity, maximum flow rate, and diameter. The decision-making variables include pipeline diameter, length, internal roughness, node pressure, future demand, and valve placement. Additionally, this paper provides an extensive overview of deterministic methods and heuristic algorithms utilized in the optimal design of micro-irrigation pipeline networks. Finally, this paper presents future research directions for pipeline network optimization and explores the potential for algorithmic improvements, integration of machine learning techniques, and wider adoption of EPANET 2.0 software. These endeavors aim to lay a strong foundation for effectively solving complex and challenging optimization problems in micro-irrigation pipeline network systems in the future.

Mode-Dependent Scalable Control for Large-Scale Networked Systems

A novel mode-dependent scalable control strategy is proposed for large-scale networked systems, which are composed of several dynamically interconnected subsystems. First, to remove the restrictions on the coupling structure of these subsystems, an arbitrary topology is carefully constructed. Then, some sufficient conditions under the subsystem level are derived, which uses each subsystem’s knowledge of its own dynamics and a limited amount of information about the coupled subsystems. And, a set of flexible controller gains are solved, guaranteeing all closed-loop signals to be globally exponentially stable. Finally, simulation results on a coupled microgrid validate the effectiveness of the proposed approach.

Distributed Design of Glocal Controllers via Hierarchical Model Decomposition

This paper proposes a distributed design method of controllers having a glocal (global/local) information structure for large-scale network systems. Distributed design, independent design of all subcontrollers that constitute a structured controller, facilitates scalable controller synthesis. While existing distributed design methods confine attention to the decentralized or distributed information structures, this study addresses distributed design of glocal-structured controllers. Glocal control exploits the nature that network system's behavior can typically be represented as a superposition of spatially local fluctuations and global interarea oscillations by incorporating a global coordinating subcontroller with local decentralized subcontrollers. The key idea to distributed design of glocal controllers is to represent the original network system as a hierarchical cascaded system composed of reduced-order models representing the global and local dynamics, referred to as hierarchical model decomposition. Distributed design is achieved by independently designing and implementing subcontrollers for the reduced-order models while preserving the cascade structure. This paper provides a condition for existence of the hierarchical model decomposition, a specific representation of the hierarchical system, a clustering method appropriate for the proposed approach, and a robust extension. Numerical examples of a power grid evidence the practical relevance of the proposed method.

Software-Defined Event-Triggering Control for Large-Scale Networked Systems Subject to Stochastic Cyberattacks

In large-scale networked control systems (NCSs), an important issue is how to guarantee the systems performance under the limited network bandwidth and stochastic cyber attacks. Centralized event-triggering mechanisms (ETMs) are now regarded as a desirable solution to ease the bandwidth pressure, but the application in large-scale NCSs is constrained by overall management complexity. In this paper, we will firstly propose a software defined centralized ETM to cost-efficiently conduct event-triggering decision based on global system states so as to assure the system transmission performance. Then, by taking deception attacks, which can hardly be detected and pose serious threats to NCSs, into account, we study secure control problem over a large-scale NCS with the presented centralized ETM. The considered deception attacks compromise controller-to-actuator channels, and the specific behaviors of the attacks on different channels are depicted by different Bernoulli processes. To solve the problem, a formal model of the envisioned large-scale system is established, the sufficient conditions that achieving the uniformly ultimately bounded (UUB) stability of the formulated system are analyzed, and then the controller gains are designed accordingly. The effectiveness of the proposed approach is finally validated by an illustrative example.

Robust H∞ Performance Analysis of Uncertain Large-Scale Networked Systems

This paper considers the robust H∞ performance problem of continuous-time uncertain large-scale networked systems (LSNSs). The systems consist of numerous arbitrarily connected subsystems, each of which has different dynamics. Currently it is computationally difficult to manage systems with the existing lumped analysis method; therefore, exploiting the structural properties of the systems, sufficient conditions are derived for robust H∞ performance. Based on these results, an analysis condition that depends only on the parameters of each single subsystem is further obtained. Some numerical simulations are proposed to verify the validity and superiority of the developed conditions.

SympGAN: A systematic knowledge integration system for symptom–gene associations network

Phenotypes (i.e., symptoms and clinical signs) are essential for clinical diagnosis and research related to symptom science and precision health. As clinical observational manifestations of a disease, symptoms are clinically significant because they act as direct causes for patients to seek medical care and the primary indicators for clinicians to provide diagnosis/treatments. However, a comprehensive phenotypic knowledge base and high-quality symptom–gene associations are lacking. Therefore, a thorough understanding of the relationships between symptoms and other entities is urgently needed to support scientific research and clinical health care. In this paper, we constructed a systematic, large-scale, and high-quality symptom-gene associations network system named SympGAN (accessible at http://www.sympgan.org/). We provide access to the database with millions of associations between symptoms, genes, diseases, and drugs, as well as the system for users to search, analyze, knowledge inference, and present data visualization. We utilize state-of-the-art machine learning and deep learning algorithms as the backbone to form the final dataset. In addition, we utilize the RoBERTa-PubMed neural network for name entity recognition to assist in data screening. The knowledge graph is adopted to organize the relationships between different entities. We adopt ConvE, TuckER, and HypER methods for knowledge completion experiments to validate the quality of final knowledge graph triples. Based on the results, we provide online automatic knowledge inference interfaces. The system, SympGAN, has promising value for disease diagnosis, decision support in health care, precision health, and scientific research, as researchers and practitioners can easily access information about symptoms, diseases, targets, gene ontology, and drugs.

The Structure Fault Tolerance of Alternating Group Networks

The alternating group network [Formula: see text] can be used to model the topology structure of a large-scale parallel network system. In this work, the structure fault tolerance of alternating group networks based on the star and path structures is investigated. For a graph [Formula: see text] and its a connected subgraph [Formula: see text], the [Formula: see text]-structure connectivity [Formula: see text] (resp. [Formula: see text]-substructure connectivity [Formula: see text]) of [Formula: see text] is the cardinality of a minimum family [Formula: see text] whose every element is isomorphic to [Formula: see text] (resp. isomorphic to a subgraph of [Formula: see text]) such that [Formula: see text] is disconnected. Specifically, we determine [Formula: see text] and [Formula: see text] for [Formula: see text].

Near-Optimal Distributed Linear-Quadratic Regulator for Networked Systems

This paper studies the trade-off between the degree of decentralization and the performance of a distributed controller in a linear-quadratic control setting. We study a system of interconnected agents over a graph and a distributed controller, called -distributed control, which lets the agents make control decisions based on the state information within distance on the underlying graph. This controller can tune its degree of decentralization using the parameter and thus allows a characterization of the relationship between decentralization and performance. We show that under mild assumptions, including stabilizability, detectability, and a subexponentially growing graph condition, the performance difference between -distributed control and centralized optimal control becomes exponentially small in . This result reveals that distributed control can achieve near-optimal performance with a moderate degree of decentralization, and thus it is an effective controller architecture for large-scale networked systems.