When applying the game-theoretic approach to find the optimal strategy for industrial cyber-physical systems defenders, most existing work assumes both the system states (e.g., for a power grid system, the system state captures which buses are compromised) and the attacker’s instant actions are observable and can be used to make the decision for the defender’s next move. Also, the reward and expected utilities are calculated based on the most likely system state and attack action. However, there is uncertainty in determining the system state and attack actions as the attack unfolds in the system in practice. This work shows that such an approximation is non-optimal in determining the defense strategy. Instead, we propose a framework that models the uncertainty in the system state and attack action. We derive the defender’s optimal strategy under such uncertainty by calculating the expected utilities across different action spaces and redefining the immediate reward within the deep learning algorithm based on the expected utilities and our estimation of the probabilistic distribution of the system state and attack action, ultimately employing the agent system for learning and generating the optimal defense strategy. The simulation experiments are carried out based on the generic industrial cyber-physical system testbed and the numerical results confirmed that the proposed solution can improve the defender’s expected utilities by 38.8% compared to the state-of-the-art.

Load frequency control is critical for maintaining grid stability, particularly in modern power systems with wind power penetration and increasing exposure to denial-of-service attacks. This paper presents a data-driven dynamic event-triggered reinforcement learning framework for constrained H∞load frequency control in multi-area power systems. The control problem is formulated as a min–max optimization task, and a dynamic event-triggered strategy is designed to reduce computational and communication burdens. A neural network-based reinforcement learning framework is developed to approximate the near-optimal event-triggered control strategy without requiring explicit system dynamics. To further counteract the impact of frequency-based denial-of-service attacks, a dedicated attacks compensation mechanism is designed. Theoretical analysis proves input-to-state stability of the closed-loop system and guarantees convergence of the neural network parameters. Extensive simulation studies on multi-area power systems with wind power integration demonstrate that the proposed method ensures stable frequency regulation while effectively alleviating the transmission burdens and mitigating the adverse effects of cyberattacks.

Detecting power equipment using by neural network-based detector is the fundamental task of unmanned aerial vehicle-mediated distribution lines inspection. Suffering from the heavy annotation burden of existing fully-supervised methods, this article proposes a novel Random Rendering and Recovery (R3) framework to explore the unsupervised detection of distribution equipment. In the R3framework, the 3D model is employed to provide richer equipment prior instead of manual structures, which enables the detector's monocular 2D and 3D detection ability without relying on any other annotations or pretrained weights. In total 1 60 761 distribution images with 8 equipment categories are collected to verify the adaptability and effectiveness of the proposed method. The numerical results show that R3achieves a 31.3 AP in the unsupervised detection task, significantly exceeding general methods such as Cutler (4.7) or FreeSOLO (2.3). Besides, R3can also serve as an advanced pretraining method to improve existing fully-supervised methods with unlabeled images.

In data center networks, cross-host collective communications (CC) for LLM training often suffer from ECMP's hash-based randomness which funnels flows onto overlapping spine-leaf links, creating hotspots, rank stragglers, and degraded CC efficiency. A promising yet underexplored approach is to plan cross-host paths ahead during CC initialization. This leverages host-side steering, exploiting ECMP hash linearity via lightweight packet-header modification. Assigning cross-host paths to minimize link overlap and balance load is NP-complete for large-scale networks. To address this, we propose PathPlanner, a centralized service that heuristically selects near-optimal paths with minimal spine-leaf overlaps, generates multiple valid source ports via the host-side steering, and distributes them to workers. By cycling through these ports, each flow traverses the intended path without modifying software logics. High-fidelity SimAI simulations with realistic LLM workloads demonstrate that PathPlanner significantly reduces link overlap and straggler effects, cutting CC primitive flow completion times by up to 44.7 % and execution times by up to 21.35 %. In 32-rank Mixtral training, it shortens per-iteration runtimes by 1.6–2.1s, yielding estimated cumulative savings of 2.65-3.52 days over a complete training run.

The rise of Industry 4.0 exposes traditionally isolated Industrial Cyber-Physical Systems (ICPSs) to increasing network attacks, posing serious security threats and potential damage. Traffic anomaly detection is essential for identifying such attacks. Nevertheless, existing work faces a dilemma between high accuracy and real-time performance. In this paper, we resolve this dilemma through P4-IDet, a novel traffic anomaly detection framework based on programmable switches, achieving both high accuracy and real-time performance. P4-IDet first deploys a low-complexity detector in the data plane to stamp timestamps, extract traffic features, and perform line-rate preliminary detection. Only suspicious packets and their features are uploaded to a server for fine-grained analysis by a high-accuracy machine learning model. To further reduce the upload and accelerate detection, a Bayesian optimizer adaptively tunes detection rules based on differences between detection results of the switch and the server. Moreover, P4-IDet can be integrated with existing detection models to enhance accuracy and real-time performance. Finally, we implement the prototype on a Barefoot Tofino 2.0 switch using the P4 language and an x86 server, and validate it on a large-scale ICPS platform with real-world industrial systems. Experiments show 5.6–41.1% accuracy gains, a 28.93% reduction in machine learning model workload, and 8.31–25.90% improvements in real-time performance.

The reliable and safe operation of the Cyber-Physical Power System (CPPS) consisting of power generation and transmission highly depends on the security of the underlying communication infrastructure. The facilities of the CPPS are often geographically distributed and vulnerable to coordinated cyber-attacks. This demands proactive security management solutions considering the various security situation of CPPS facilities to protect them. This article presents a game theory-based dynamic decision-making solution for collaboratively securing critical facilities to minimize system performance degradation under cyber-attacks. We take the essential generation facility (i.e., power plant) as the research object to validate the solution. An analysis of the attack penetration process described by the Bayesian Attack Graph (BAG) is carried out to assess the security situation of each power plant. Then, a stochastic game model is developed to characterize the interaction of the attacker and the defender considering the varying situation of every power plant. A novel reinforcement learning algorithm is presented to solve the Nash equilibrium and obtain the dynamic optimal security strategies that defend the most important power plants. The proposed solution is extensively evaluated through a range of experiments based on the IEEE 57-bus test system, and the numerical results demonstrated the effectiveness of the proposed solution.

In data center networks (DCNs) with high-density computing power, the path selection managed by the equal-cost multi-path (ECMP) hashing function often causes path overlap and overuse of switch ports, undermining performance. While recent approaches allow hosts to influence egress port selection via single-bit changes in packet headers, they offer limited port coverage and are restricted to determining a single hop’s egress port. To address this, we propose a host-based path selector (HPS) that enables source hosts to precisely select paths via multiple hops by making targeted multi-bit header changes. HPS is based on two key principles: (a) exploiting the predictable relationship between the relative changes to hash values to select switch egress ports effectively, and (b) applying the criteria for ensuring modified headers accurately direct packets along the desired path. HPS iteratively adjusts packet headers at each switch along the path, ensuring efficient path selection with polynomial time and constant space complexity, making it scalable for large networks. We evaluated HPS in two and three-layer DCN topologies, testing up to 1,000 paths. The results show that HPS enables precise path selection, greatly reducing path overlap compared to traditional ECMP and state-of-the-art RePaC port influence method, resulting in substantial performance improvements.

The remarkable development of the Industrial Internet of Things (IIoT) has undoubtedly elevated industrial operations to a more intelligence and efficiency level, yet it has also introduced a range of security challenges. The widespread of intelligent IoT devices has greatly expanded the attack surface for cyber-attacks. Additionally, the cloud-based centralized management architecture of traditional IIoT is susceptible to single-point-of-failure, which exacerbates the security risks. Nowadays, the secure and decentralized nature of blockchain has been considered a promising solution to address the security and privacy challenges in IIoT. This article proposes a blockchain-based operational architecture for IIoT (SecureArchi- IIoT) to enhance security and privacy in IIoT operations. Under this architecture, a set of smart contracts are designed to provide operational functionalities that are suitable for actual industrial demands. An operational control policy is designed to realize precise and effective management of the operation permissions with distinct granularity. Furthermore, a reputation-based behavioral punishment mechanism is developed to enhance the security performance of the proposed architecture. The prototype of the proposed architecture is implemented in a private IIoT environment to demonstrate its feasibility and effectiveness. Experimental results confirm that the proposed architecture outperforms the traditional architecture in aspects of security and privacy and maintains acceptable real-time performance.

In today’s power transmission grids, Internet-of-Things networks employ long-haul optical wires for regular sensor data dissemination to a server. Ensuring resilience against link faults is paramount to observe the grid states accurately via a process known as state estimation (SE). The accuracy is achieved by minimizing the end-to-end failure rate in packet delivery (EEFR). Current approaches focus on hop-by-hop retransmission control with in-path caching. Notably, the disruption-resilient transport protocol (DRTP) stands out for achieving the lowest EEFR. DRTP employs robust hop-by-hop retransmission and a recursive collaboration process guided by arrival timeouts. However, challenges arise in maintaining recursiveness with timeouts, leading to increased EEFR due to cache mismatch. These intensify when a hop triggers arrival timeouts, spawning retransmission instances in an unexpected sequence, which can experience an unprotected parallel race condition. To address this, we propose RSDD, a resilient mechanism for sensor data dissemination for implementing DRTP in the correct and fully verified manner. RSDD orchestrates concurrent retransmission instances, ensuring exclusive execution for the same lost packet, precisely scheduled based on timeouts. We evaluated the performance of RSDD in a simulated network that combines SE and a grid, using ndnSIM, MATPOWER, and RTDS. The results validate RSDD as a correct DRTP implementation, highlighting its exclusiveness and quality-of-service performance. RSDD achieves an EEFR of 2.44% and an average end-to-end packet delivery time (EEDT) of 2.7 ms during full path disruption with a 20% link loss rate in packets. Moreover, RSDD excels in enabling SE to maintain the grid observability and accuracy.

The increasing proliferation of electric vehicles (EVs) and renewable energy sources (RESs) poses challenges to the operation of coupled power and transportation networks due to their uncertainties, where EV users' routing and charging behaviors are subjected to their complex decision-making rationality. To economically manage numerous fast charging stations with onsite RESs, this article focuses on the optimal day-ahead bidding and intraday scheduling strategies of a fast charging station aggregator (FCSA) to maximize its profit in the electricity market. Traffic simulation based on boundedly rational dynamic user equilibrium is presented to model charging demand under bounded rationality of EV users. To efficiently manipulate large-scale EVs, a group charging scheduling framework is proposed to reduce decision variables. Uncertainties in electricity prices, RES generation, traffic demand, and user rationality are addressed by stochastic programming. Case studies have validated the effectiveness of the proposed method in reducing the FCSA's operational costs and RES curtailment.

The integrated energy system (IES) is considered to be an efficient paradigm for low-carbon energy provision. The accurate modeling of inherent operational uncertainties in IES is of paramount importance for its optimal planning and energy management. This article develops a generation and forecasting approach of IES operational scenarios based on Wasserstein generative adversarial network (WGAN) for the characterization of IES operational uncertainties. The proposed solution can efficiently generate high-quality IES operational scenarios consisting of multivariate uncertain variables without explicit statistical assumptions. In addition, the well-trained WGAN is integrated into a constrained optimization problem to realize scenario forecasting in compliance with certain information (i.e., the past observation and point forecasting) for a specific coming period, and it is not restricted by the forecast look-ahead horizons. The proposed solution is assessed through a range of qualitative and quantitative evaluations, and the numerical results validate its effectiveness.

Network administrators usually realize network functions in data plane programs. They employ the network-wide program deployment that decomposes input programs into match-action tables (MATs) while deploying each MAT on a specific switch. Since MATs may be deployed on different switches, existing solutions propose the inter-switch coordination that uses the per-packet header space to deliver crucial packet processing information among switches. However, such coordination incurs non-trivial per-packet byte overhead, leading to end-to-end performance degradation. We propose, a framework that aims to minimize the per-packet byte overhead. The key idea is to formulate network-wide program deployment as a mixed-integer programming (MIP) problem with the objective of minimizing the per-packet byte overhead. Also, offers a greedy-based heuristic that solves the problem in a near-optimal and timely manner. We have implemented on Tofino switches. Compared to existing frameworks, decreases the per-packet byte overhead by 156 bytes while preserving end-to-end performance in terms of flow completion time and goodput.

Programmable switches allow administrators to customize packet processing behaviors in data plane programs. However, existing solutions for program deployment fail to achieve resource efficiency and high packet processing performance. In this paper, we propose SPEED, a system that provides resource-efficient and high-performance deployment for data plane programs. For resource efficiency, SPEED merges input data plane programs by reducing program redundancy. Then it abstracts the substrate network into an one big switch (OBS), and deploys the merged program on the OBS while minimizing resource usage. For high performance, SPEED searches for the performance-optimal mapping between the OBS and the substrate network with respect to network-wide constraints. It also maintains program logic among different switches via inter-device packet scheduling. We have implemented SPEED on a Barefoot Tofino switch. The evaluation indicates that SPEED achieves resource-efficient and high-performance deployment for real data plane programs.

The popularized named data network (NDN) needs a fast reroute (FRR) to improve resilience in dealing with the norm of link faults. The resilience of the existing disjoint arborescences and paths based approaches is restricted in reroute options, due to the limited number of disjoint end-to-end routes available. Such restrictions can be weakened by exploiting of the disjoint subpaths (DSP) between two intermediate nodes in the primary path (PP), as what DSP can link together is not only the two ends. In this paper, we propose a novel NDN based traffic-aware fast reroute (TA-FRR) mechanism that exploits a new resilient subgraph (RSG) route involving multiple DSPs. TA-FRR proactively constructs an RSG with scalable computational complexity to maximize the resilience and to avoid a too high reroute stretch. In the RSG, TA-FRR efficiently computes the shortest detour path for the disconnected components actively identified according to packet arrival timeouts. We evaluated the performance of TA-FRR in the ndnSIM simulator using synthsized Internet topologies with different degrees of link faults. The results show that TA-FRR can efficiently recover serious link faults in PP with a scalable RSG construction and outperforms the existing approaches with a significantly higher resilience and the maximum stretch similar to the arborescences approach.

Substation communication network (SCN) provides real-time, high-speed, and reliable data transmissions for the advanced monitoring and control functionalities, which are facing increasing cyberspace threats and attacks. Efficient threat perception and cyber situational awareness are essential to enhance secure and reliable SCN operations. This article explores multiscale SCN traffic pattern characteristics with holistic network traffic, separated network traffic for included devices (especially IoT devices) and separated network traffic of certain types of protocol. The proposed online traffic-oriented SCN traffic anomaly detection and cyber situational awareness models are designed for the network anomalies and cyber-attacks that could cause network traffic pattern variations. We leverage a fractional autoregressive integration moving average (FARIMA)-based dynamic threshold model to detect abnormal traffic patterns without sophisticated computations or deep packet inspection. The SCN real-time operation conditions are timely quantified through the statistical methods with the alliance of SCN topology and protocols. The cyber situational awareness model is further carried out to evaluate the most affected protocol and security risks of various devices in SCN using Grubbs’ test. The experiment results are carried out based on a real 110-kV intelligent power substation. The numerical results confirm the comparative low mean square error (MSE) and low complexity of the online traffic characterization when forecasting holistic network traffic and separated network traffics. Furthermore, the timely and quantified cybersecurity risk analysis is conducted based on the SCN traffic with varying scales to detect cyberspace threats and identify the high-risk SCN devices and the most affected protocol.

Given the huge installed capacity of photovoltaic (PV) worldwide, the traditional defect detection system for PV plants is infeasible, especially for large-scale plants. In this article, unmanned aerial vehicles (UAVs) mounted with several sensors and a computer in the cloud are used cooperatively to establish an Internet of Things-based cloud-edge computing infrastructure, which can automatically detect defects with low latency, low cost, and high accuracy. The pretrained models trained in the cloud server are embedded into the processor in UAVs to implement online detection. Specifically, given the characteristic of defects in electroluminescence images, a two-stage algorithm is proposed to identify the defects with high performance. In the first stage, cells, the basic unit in the PV module, are extracted using an encoder–decoder network. Then, a vision-based incremental defect classification algorithm is proposed for defect detection that integrates deep learning with prior knowledge to maximize computing efficiency. The performance of the proposed system is evaluated through extensive experiments.

Critical industrial infrastructures are currently facing increasing cyberspace threats in their underlying information and communication systems. The advanced monitoring, control, and management functionalities of the industrial systems firmly rely on the reliable and secure operations of the industrial control system (ICS) network. This article characterizes the ICS network traffic and presents a scalable and efficient solution for real-time ICS network traffic anomaly detection, considering various forms of ICS anomaly events. The events due to the cyberattacks, malicious operating behaviors, and network anomalies can be effectively detected without sophisticated computational requirements and retrieval of communication protocols. The proposed hybrid statistical-machine learning model integrates a seasonal autoregressive integration moving average (SARIMA)-based dynamic threshold model and a long short-term memory (LSTM) model to jointly identify the abnormal traffic patterns with low false omission rates. The proposed solution is extensively evaluated at a realistic ICS cyber–physical system (CPS) testbed, and the numerical results confirm its high detection accuracy and low computational complexity. Note to Practitioners—This article was motivated by the challenge of real-time anomaly detection in industrial cyber–physical systems (CPSs). The existing industrial control system (ICS) network anomaly detection solutions are generally carried out based on a single model based on the historian database and cannot dynamically classify the abnormal conditions in a real-time fashion. A novel hybrid statistical-machine learning model is developed that integrates a seasonal autoregressive integration moving average (SARIMA)-based dynamic threshold model and a long short-term memory (LSTM) model to jointly identify the anomalous events through traffic pattern analysis. The proposed anomaly detection solution can efficiently provide accurate detection for cyberattacks, malicious operating behaviors, and network anomalies while meeting the real-time requirements of ICS networks. The proposed solution can be deployed in the realistic ICS CPSs, e.g., the power generation system, gas pipeline systems, and urban railway transportation systems. The preliminary numerical results obtained from the ICS-CPS testbed suggested that it can provide high detection accuracy with low computational complexity and, hence, can be adopted with minimal deployment hurdles.

In network function virtualization (NFV), network functions (NFs) are chained as a service function chain (SFC) to enhance NF management with low cost and high flexibility. Recent NFV solutions indicate that the packet processing performance of SFCs can be significantly improved by offloading NFs to programmable switches. However, such offloading requires a deep understanding of heterogeneous NF properties (e.g., NF resource consumption and NF performance behaviors) to achieve the maximum SFC performance. Unfortunately, none of existing solutions provide automatic analysis of these NF properties. Thus, network administrators have to manually examine the source codes of NFs and profile various NF properties by hand, which is extremely time-consuming and laborious. In this article, we propose LightNF, a novel system that simplifies NF offloading in programmable networks. LightNF automatically dissects comprehensive NF properties by means of code analysis and performance profiling while eliminating manual efforts. It then leverages its analysis results of NF properties in its SFC placement so as to make the performance-optimal offloading decisions. We have implemented LightNF on Tofino-based hardware programmable switches. We perform extensive experiments to evaluate LightNF with a real-world testbed and large-scale simulation. Our experiments show that LightNF outperforms existing solutions with an orders-of-magnitude reduction in per-packet processing latency and 9.5× improvement in SFC throughput.

The increasing penetration of distributed renewable generation has introduced significant uncertainties and randomness to the power distribution network operation. Accurate and timely awareness of the network operation is of paramount importance to ensure system safety and reliability and is considered nontrivial and costly as substantial network reinforcement with advanced measurement devices is generally required. Also, the existing state estimation methods, e.g., weighted least square, may not converge in the presence of incomplete and inaccurate measurements. This article proposes a spatio-temporal estimation generative adversarial network (ST-EGAN) consisting of feature extraction, information completion, data reconstruction, and fake data discrimination to generate high-resolution pseudo-measurements to promote the accuracy and robustness of state estimation. The task of high-resolution power distribution network state estimation is carried out based on the mixed dataset of multiple time-scale measurements obtained from supervisory control and data acquisition and phasor measurement units. The proposed solution is extensively assessed using the IEEE 33-bus test network compared with the existing solutions for a range of scenarios with different resolutions and noise intensities. The numerical results demonstrated that the proposed ST-EGAN can reduce the mean rmse by 4.78% compared to interpolation algorithms, and reduce the rmse by 0.14% and 0.21% compared with deep convolutional generative adversarial networks and super-resolution convolutional networks, respectively, in the presence of noises with different intensities. The proposed method can be generalized to cases with different topological structures and measurement assembly conditions.

The energy consumption of mobile network infrastructures has been witnessed to rise significantly in recent years due to the massive growth of mobile traffic triggered by the increasing adoption of 5G networks and massive applications of the Internet of Things (IoT). The increase in network traffic not only demands network densification but also increases the operational expenditure (OPEX) of the mobile network operators as well as the environmental and sustainability concerns. To address such a challenge, green energy technology has received increasing attention. However, keeping the density of small cell base stations (SCBSs) and matching the dynamics of available energy variations to the user service request arrivals are non-trivial that need further investigation. This paper proposes to utilize the microgeneration of renewable energy (RE) infrastructure with traffic-aware load offloading integrated with advanced sleep mode (ASM) operation. The centralized renewable energy microgeneration approach is adopted to power the high density of SCBSs installed at a dispersed geographical location. The stochastic modeling-based traffic offloading can offload the macro base station (MBS) users to SCBSs depending on the traffic intensity. The advanced sleep mode stochastic model is presented to gradually manage the underutilized SCBSs into sleep modes based on the traffic load. The proposed scheme is developed to reduce the on-grid energy consumption whilst meeting the quality of service (QoS) requirement in terms of blocking probability and reactivation delay. The numerical results demonstrated that the proposed solution can achieve a significant amount of energy-saving depending upon the renewable energy availability and MBS traffic offloading with ASMs policies.

Programmable switches empower stateful packet processing, in which incoming packets continuously update states in the data plane, while applications in the control plane read and write states. However, since the data plane and control plane are separated, a consistent view of states in both planes is required for stateful packet processing. Existing approaches suffer from either high latency or low accuracy. In this paper, we propose ApproSync, a framework that offers approximate state synchronization with low latency and high accuracy. To achieve low latency, ApproSync directly transfers states between switch ASICs and the control plane by bypassing switch operating systems. To achieve high accuracy, ApproSync utilizes the resources in the switch ASIC to realize rate control in state synchronization, such that it avoids potential state loss. It also bounds the divergence between the states in the data plane and that in the control plane under limited link capacity. We prototype ApproSync on Barefoot Tofino switches. The experimental results indicate that compared to existing approaches, ApproSync achieves order-of-magnitude latency reduction while maintaining high accuracy of state synchronization. Also, our experiments demonstrate that ApproSync provides significant latency benefits to existing network management applications and well preserves high application-level accuracy.

Recent efforts in network measurement place measurement tasks on programmable switches to measure high-speed traffic. These tasks extract flow data, i.e., events, from packets and send events to the control plane. However, the tasks may generate massive events in a short time. In this context, the links transferring events to the control plane and the control plane servers that handle events may be overloaded, i.e., control plane overload. None of existing solutions can eliminate control plane overload. In this paper, we propose MTP, a framework that eliminates control plane overload via careful measurement task placement. Our key idea is to allocate enough resources for each task during task placement to avoid control plane overload at runtime. For each task, MTP estimates its maximum possible rate of sending events to the control plane. Then its optimization framework addresses the resource restrictions of both switches and the control plane. The experiments on Tofino switches indicate that MTP outperforms existing solutions with higher accuracy in several use cases.

A massive number of small cell base stations are expected to be deployed in the 5G and beyond 5G mobile communication networks due to the exponential increase in mobile traffic. This will directly lead to not only a significant increase in energy consumption but also the overall operational cost and carbon footprint. An energy provision based on renewable energy generation to power these small cell base stations is considered a sustainable and promising solution to address this challenge. This paper exploits the cost-effective and low-carbon energy provision solution for individual small-cell mobile networks and presents two different potential frameworks, i.e., centralized and distributed energy provision, respectively. The former supplies nearby small cell base stations through a centralized renewable energy source with energy storage facilities. For the latter, small cell base stations can be supplied by utilizing local renewable energy and storage facilities. These two frameworks are assessed and compared in terms of renewable energy utilization and carbon emission reduction in the presence of time-varying traffic loads, small cell locations and renewable energy availabilities. In addition, we devise energy management for these configurations by incorporating a resource-on-demand strategy in the proposed framework. The numerical simulation results demonstrate that the proposed centralized renewable energy generation strategy for nearby small cells maximizes the cost and energy efficiencies of the network.

An industrial cyber–physical system (ICPS) tightly integrating both physical processes and information and communication technologies (ICTs) leads to increasing cyberspace threats and attacks for the critical electrical infrastructure. With the limited defense resources availability, the efficient threat perception and mitigation of potential impacts of cyber attacks are essential to enhance the ICPS operational security. This article proposes an optimal defense resource allocation solution to prioritize the ICPS asset protection based on the distributed network traffic anomaly detection. The traffic anomalies and attack paths can be timely detected simultaneously over multiple security zones of the electrical infrastructure through local computing devices. The defense resource allocation is formulated as a multiobjective optimization (MOO) problem considering the tradeoff among the asset vulnerability, cost, and criticality, and solved by the Pareto optimal solution generation approach. The proposed solution is extensively evaluated using a realistic electrical CPS (ECPS) testbed for a range of cyber-attack scenarios. The numerical results confirm the effectiveness of the proposed distributed anomaly detection model and defense resource allocation strategy for varying defense resource availabilities.

Sustainable energy management systems have been increasingly studied in recent years. Non-intrusive load monitoring (NILM), as a key component, estimates the power consumption of individual appliances from the main readings only. However, most NILM approaches are computationally expensive, and their generality is negatively affected by the data drift occurred when the models are used across domains. Besides, the threats of privacy violation will rise in the model transfer due to the possible leakage of the personal information of the users from the source domain. To address all these challenges, we designed a cost-efficient learning method using LightGBM for energy disaggregation. We also proposed a model-based transfer learning algorithm using feature importance analysis, which enhances the generalisation capability of tree-based ensemble models applied in different domains while protecting privacy. We conducted experiments with real-world data sets. The performance of our approach is superior to the state-of-the-art solutions.

The paradigm of the Internet of Things (IoT) and cloud-edge computing plays a significant role in future smart grids. The data-driven solution integrating the artificial intelligence functionalities brings novel methods to address the nontrivial task of economic dispatch in microgrids in the presence of uncertainties of renewable generations and loads. This article proposes a learning-based decision-making framework for the economic energy dispatch of an islanding microgrid based on the cloud-edge computing architecture. Cloud resources are utilized to solve the optimal dispatch decision sequences over historical operating patterns. It can be considered as a sample labeling process for the supervised training that can implement the complex mapping of input-output space through an advanced machine learning model. Then, the well-trained model can be adopted locally at edge computing devices keeping the long-term parameters unchanged for implement the real-time microgrid energy dispatch. The key benefit of the proposed solution is that it effectively avoids the prediction of multiple stochastic variables and the design of sophisticated regulation strategies or reward policy functions for real-time dispatch. The solution is extensively assessed through simulation experiments by the use of real data measurements for a set of operational scenarios and the numerical results validate the effectiveness and benefit of the proposed algorithmic solution.

Partial discharge (PD) is a widely adopted method for the internal insulation detection of electrical transformers. The impact of the refraction and diffraction of the ultrasonic signal in the oil, winding, and core in the transformer makes the localization of the PD source inside the transformer a nontrivial task. The refraction and diffraction factors can introduce additional complexity to the localization equations that can hardly be solved. This article proposes an algorithmic solution based on the time difference of arrival algorithm and semidefinite relaxation convex optimization to improve the PD source localization accuracy. The refraction and diffraction error, measurement error, and their relationship in the process of PD signal propagation are fully considered. The proposed solution is extensively assessed through simulation, testbed, and field experiments, and the results confirm that the proposed solution outperforms the Chan and particle swarm optimization algorithm with the localization error of about 0.1 m.

Successful detection of false data injection attacks (FDIAs) and removal of state bias due to FDIAs are essential for ensuring secure power grids operation and control. This article first extends the approximate dc model of FDIA to a more general ac model that can handle both traditional and synchronized measurements. To automatically filter out the established FDIAs, we propose a state reconstruction scheme consisting of a contaminated state separation method, an enhanced bad data identification approach and a state recovery algorithm. In this scheme, a classifier is developed by aggregating a series of extreme learning machines (ELMs) to detect anomaly states caused by FDIAs. Gaussian random distribution and Latin hypercube sampling are adopted to initialize the input weights of base ELMs, which can provide more diversities to enhance the ensemble performance. Then, to identify the exact locations of the compromised measurements, a state forecasting-based bad data identification approach is proposed by exploiting the consistency between the forecasted and the received measurements. Finally, an effective state recovery algorithm applies quasi-Newton method and Armijo line search to address the possible system unobservable problem due to the removal of attacked measurements. Numerical tests on serval IEEE standard test systems verify the efficiency of the proposed FDIA model and state reconstruction scheme.

Internet of Things (IoT) technology provides the necessary foundation and support for smart city water management. To address the challenge of river pollution prevention and flood control requirements in the urban river system, this article proposes a data-driven model to carry out the optimal operation scheduling of water diversion and drainage pumping stations in the presence of the complex hydrometeorological constraints. The proposed solution in the model predictive control (MPC) framework first adopts the long short-term memory (LSTM) network through supervised learning from IoT data to simulate and predict the river flow dynamics and the water quality. Consequently, the optimal scheduling of controllable pumping stations to minimize the operational cost (e.g., the flocculant consumption) can be formulated as a stochastic optimization problem, while meeting the river flood control and water quality constraints. The particle swarm optimization (PSO) algorithm is further used to solve the above unit commitment (UC) optimization problem and obtain the optimal operational schedules of the water pumping units (e.g., startup time and working periods). The performance of the proposed optimal water pumping scheduling solution is evaluated through a field case study of the urban river diversion system and the numerical results clearly confirm its effectiveness and improved economic performance compared to the existing benchmark solution.

The power efficiency of photovoltaic (PV) modules is highly correlated with their health status. Under dynamically changing environments, PV defects could spontaneously form and develop into fatal faults during the daily operation of PV power plants. To facilitate defect detection with less human intervention, a nondestructive, contactless, and automatical visual inspection system with the help of unmanned aerial vehicles and edge computing is proposed in this article. During the processing of the incoming data stream, the system may collect some new, unknown, and unlabeled defects that have not been identified yet in the existing database. To distinguish them from the existing ones, a deep embedded restricted cluster algorithm is designed to identify the unknown and unlabeled PV module defects in an unsupervised manner. Limited by the resources of edge devices and the availability of images of PV defects for training, we developed an online solution combined with deep learning, data argumentation, and transfer learning to properly address the issues of running resource-hungry applications on edge devices and lack of training samples faced by the deep learning approaches used in the field. In addition, pointwise convolution layers are introduced into the network to reduce the parameters and the size of the model. With the reduction of the network depth of the deep convolutional neural network model and the features transferred from the learned defects, the resource consumption of our proposed approach is significantly reduced, and thus can be used on a wide range of edge devices to complete defect detection in a timely manner with high accuracy. The experimental results clearly demonstrate the practicality and effectiveness.

In the last two decades, the installation and production of photovoltaic (PV) plants increased widely. As the PV system operation time grew, more and more defect occurred on the PV modules. One of the most significant open issues in the PV sector is to find appropriate inspection methods to detect PV modules' failures. In this paper, two approaches are proposed to detect the hotspots in the infrared image of PV modules. The classical digital image processing technology mainly uses Hough line transformation and canny operator to detect hotspots. The deep learning model is based on Faster-RCNN and transfer learning, which perform better with more compute resources.

Cyborg insects have attracted great attention as the flight performance they have is incomparable by micro aerial vehicles and play a critical role in supporting extensive applications. Approaches to construct cyborg insects consist of two major issues: 1) the stimulating paradigm and 2) the control policy. At present, most cyborg insects are constructed based on invasive methods, requiring the implantation of electrodes into neural or muscle systems, which would harm the insects. As the control policy is basically manual control, the shortcomings of which lie in the requirement of excessive amount of experiments and focused attention. This paper presents the design and implementation of a noninvasive and much safer cyborg insect system based on visual stimulation. The tethered paradigm is adopted here and we look at controlling the flight behavior of bumblebees, especially the abdominal-waving behavior, in the context of a model-free reinforcement learning problem. The problem is formulated as a finite and deterministic Markov decision process, where the agent is designed to change the abdominal-waving behavior from the initial state to the target state. Sarsa with transformed reward function which can speed up the learning process is employed to learn the optimal control policy. Learned policies are compared to the stochastic one by evaluating the results of ten bumblebees, demonstrating that abdominal-waving state can be modulated to approximate the target state quickly with small deviation.

In electrical transmission grids, the redundant communication paths nondisjointly overlapping at links can be established between certain substations and the control center to guarantee reliable packet delivery under link failures. However, the generation of nondisjoint paths with multiplicative and concave constraints is with the NP-complete complexity and the failovers can lead to out-of-order packets. This paper presents an OpenFlow-based nondisjoint path aggregation mechanism to heuristically compute the constrained nondisjoint paths in a centralized fashion and reorganize the out-of-order packets at edge switches. The solution is evaluated through simulations of the IEEE 30-bus network scenario under 2% link failure rate and the result confirms its effectiveness: the packet delivery success rate is significantly improved in comparison with the current nondisjoint and disjoint path algorithms. TCP throughput is improved by 121.62% with the packet reordering. Also, the memory usage for the packet reordering buffer is reduced by 76.563% compared with the distributed cognitive packet network.

This letter proposes an efficient algorithmic solution to diagnose multiple mechanical faults in wind turbine gearbox through source number estimation using an empirical mode decomposition (EMD) and singular value decomposition (SVD) joint approach, and source signal recovery based on short-time Fourier transform (STFT), fuzzy C-means clustering and l1 norm decomposition. The effectiveness of the solution is validated using real wind turbine measurements under multifault scenarios.

Social Internet-of-Things (SIoT) environment comprises not only smart devices but also the humans who interact with these IoT devices. The benefits of such system are overshadowed due to the cyber security issues. A novel approach is required to understand the security implication under such a dynamic environment while taking both the social and technical aspects into consideration. This paper addressed such challenges and proposed a 3-D security modeling platform that can capture and model the security requirements in the SIoT environment. The modeling process is graphical notation based and works as a security extension to the Business Process Model and Notation. Still, it utilizes the latest 3-D game technology; thus, the security extensions are generated through the third dimension. Consequently, the introduction of security extensions will not increase the complexity of the original SIoT scenario, while keeping all the key information on the same platform. Together with the proposed security ontology, these comprehensive security notations created a unique platform that aims at addressing the ever complicated security issues in the SIoT environment.

The integration of a massive number of plug-in electric vehicles (PEVs) into current power distribution networks brings direct challenges to network planning, control, and operation. To increase the PEV penetration level with minimal negative impact, the dynamical PEV travel behaviors and charging demand need to be better understood. This paper presents a Markov-based analytical approach for modeling PEV travel behaviors and charging demand. The travel behaviors of individual PEVs are expressed mathematically through Monte Carlo simulation considering two essential factors: temporal travel purposes and state of charge (SoC). Markov model and hidden Markov model (HMM) are adopted to explicitly formulate the probabilistic correlation between multiple PEV states and SoC ranges. This modeling approach provides an efficient and generic tool for analyzing PEV travel behaviors and charging demand based on available PEV statistics. The analytical model is further adopted in the impact assessment of two PEV normal charging scheduling strategies for a range of PEV penetration levels in an IEEE 53-bus test network with field data (network parameters and realistic PEV statistics). The results demonstrate the benefit of the proposed modeling approach in network analysis considering PEV integration.

Currently the false data injection (FDI) attack bring direct challenges in synchronized phase measurement unit (PMU) based network state estimation in wide-area measurement system, resulting in degraded system reliability and power supply security. This paper assesses the performance of state estimation in electric cyber-physical system paradigm considering the presence of FDI attacks. The adverse impact on network state estimation is evaluated through simulations for a range of FDI attack scenarios using IEEE 14-bus network model. In addition, an algorithmic solution is proposed to address the issue of additional PMU installation and placement with cyber security consideration and evaluated for a set of standard electric transmission networks (IEEE 14-bus, 30-bus, and 57-bus network). The numerical result confirms that the FDI attack can significantly degrade the state estimation and the cyber security can be improved by an appropriate placement of a limited number of additional PMUs.

This paper establishes a new framework for modeling electrical cyber-physical systems (ECPSs), integrating both power grids and communication networks. To model the communication network associated with a power transmission grid, we use a mesh network that considers the features of power transmission grids such as high-voltage levels, long-transmission distances, and equal importance of each node. Moreover, bidirectional links including data uploading channels and command downloading channels are assumed to connect every node in the communication network and a corresponding physical node in the transmission grid. Based on this model, the fragility of an ECPS is analyzed under various cyber attacks including denial-of-service (DoS) attacks, replay attacks, and false data injection attacks. Control strategies such as load shedding and relay protection are also verified using this model against these attacks.

Various solutions have been proposed to enable mobile users to access location-based services while preserving their location privacy. Some of these solutions are based on a centralized architecture with the participation of a trustworthy third party, whereas some other approaches are based on a mobile peer-to-peer (P2P) architecture. The former approaches suffer from the scalability problem when networks grow large, while the latter have to endure either low anonymization success rates or high communication overheads. To address these issues, this paper deals with an enhanced dual-active spatial cloaking algorithm (EDA) for preserving location privacy in mobile P2P networks. The proposed EDA allows mobile users to collect and actively disseminate their location information to other users. Moreover, to deal with the challenging characteristics of mobile P2P networks, e.g., constrained network resources and user mobility, EDA enables users (1) to perform a negotiation process to minimize the number of duplicate locations to be shared so as to significantly reduce the communication overhead among users, (2) to predict user locations based on the latest available information so as to eliminate the inaccuracy problem introduced by using some out-of-date locations, and (3) to use a latest-record-highest-priority (LRHP) strategy to reduce the probability of broadcasting fewer useful locations. Extensive simulations are conducted for a range of P2P network scenarios to evaluate the performance of EDA in comparison with the existing solutions. Experimental results demonstrate that the proposed EDA can improve the performance in terms of anonymity and service time with minimized communication overhead.

Recent research confirms that current slow central control based upon Supervisory Control and Data Acquisition (SCADA) systems is no longer sufficient to support power distribution grid with Distributed Generators (DGs). Distributed Network Operators (DNOs) need novel management mechanisms coupled with advanced communication infrastructure to meet the emerging technical challenges. This paper exploits the effectiveness of using a Low Earth Orbit (LEO) satellite network as the key component of “Power Utility Intranet” to support an active network management solution in the UK - AURA-NMS. Our investigation demonstrates encouraging result which suggests that a LEO network can be a viable communication solution for managing the future smart power distribution grids.

Multiple virtual networks sharing an underlying substrate network is considered a promising tool to diversify and reshape the future inter-networking paradigm. As a simple and straightforward approach, the static bandwidth allocation in virtual networks (VNs) can often be inefficient in practice, and the adaptive allocation scheme with a small time-scale may lead to the transient network and service instability. Due to the fact that the traffic patterns vary over time, the challenge still remains to meet the expected resources allocation whilst promote the network scalability and robustness. In this paper, based on the robust optimization theory we present a robust dynamic approach which periodically identifies bandwidth allocation to VNs to work reasonable well for a range of traffic patterns over a period of time, rather than certain traffic pattern instance. This problem is formulated as a robust optimization problem using path-flow model aiming to compute the minimum-cost bandwidth allocation. Through the primal decomposition, we present a distributed algorithm which consists of two components running in individual VNs and the substrate network respectively. The numerical result obtained from simulation experiments demonstrates the strength and the effectiveness of the proposed algorithm in terms of convergence and acceptance ratio.

Very often, network users expect to access services relevant to their locations, whilst preserve their privacy without disclose their exact locations. The well-known privacy preserving method is the spatial cloaking technique where exact user locations are blurred into a cloaked region to meet the privacy requirement, e.g. k-anonymity. Most of current solutions are designed with a centralized architecture in mind and rely on a third trustworthy party, i.e. a location anonymizing server (LAS). Unfortunately, these solutions cannot be directly applied to the mobile peer-to-peer (P2P) networks where no centralized servers are possible. In this paper, we present a dual-active spatial cloaking algorithm for mobile P2P networks. The key difference between the suggested algorithm and two existing algorithms, on-demand and proactive, is that: our algorithm allows peers not only actively collect but also actively disseminate location information to others. The three approaches are assessed through extensive simulation experiments for a range of P2P network scenarios. The experimental result shows that the dual-active approach uses the least anonymizing time and has the best anonymization success rate at the price of acceptable communicating cost.

Power distribution networks with distributed generators (DGs) can exhibit complex operational regimes which makes conventional management approaches no longer adequate. This paper looks into key communication infrastructure design aspects, and analyzes two representative evolution cases of Active Network Management (ANM) for distributed control. Relevant standard initiatives, communication protocols and technologies are introduced and underlying engineering challenges are highlighted. By analyzing two representative case networks (meshed and radial topologies) at different voltage levels (33 and 11 kV), this paper discusses the design considerations and presents performance results based on numerical simulations. This study focuses on the key role of the telecommunications provision when upgrading and deploying distributed control solutions, as part of future ANM systems.

As a key issue of building a virtual network (VN), the VN mapping problem can be addressed by various state-of-the-art algorithms. While these algorithms are efficient for the construction of unicast service-oriented VNs, they are generally not suitable for multicast cases. In this paper, we investigate the mapping problem in the context of virtual multicast service-oriented network subject to delay and delay variation constraints (VMNDDVC). We present a novel and efficient heuristic algorithm to tackle this problem based on a sliding window approach. The primary objective of this algorithm is in two-fold: to minimize the cost of VMNDDVC request mapping, and to achieve load balancing so as to increase the acceptance ratio of virtual multicast network (VMN) requests. The numerical results obtained from extensive simulation experiments demonstrate the effectiveness of the proposed approach and superiority than existing solutions in terms of VN mapping acceptance ratio, total revenue and cost in the long term.

Increasing complexity of the recent introduced quality of service (QoS) provisioning tools in packet-switched networks becomes a major obstacle for Internet service providers to practically deploy and manage these mechanisms to guarantee a range of transparent real-time service-level agreements. The authors revisit the existing Internet protocol (IP) QoS approaches and present a cost-effective solution to provide voice over Internet protocol (VoIP) service guarantees in converged IP networks by combining open shortest-path first traffic engineering with new insights for the configuration of standard active queue management random early detection mechanism. Extensive simulation studies are carried out, and the service quality is explicitly quantified using international telecommunication union E-model for a range of scenarios representing different types of network uncertainties. The direct representation of VoIP service quality well demonstrate that such a simple, coordinated approach, in keeping with the Internet paradigm, can achieve increased load for a given quality level and greater resilience under degraded network conditions.

A large and increasing number of distributed generators connected to the existing UK distribution networks brings enormous challenges to network operation and management. The regional network management system is suggested to tackle these industrial challenges through ubiquitous deployment of autonomous regional controllers with reliable and flexible communication among them. Such regional power network control requires a more advanced communication infrastructure than the existing SCADA system. In this paper, we present a conceptual communication infrastructure for regional control of power distribution networks in the context of an autonomous regional active network management system, and highlight the key communication system requirements. A case study of communication infrastructure supporting radial distribution network regional control is provided with the performance evaluation. Our study shows that TCP/IP based communication would need to be introduced into the communication infrastructure design to support regional network management system.

To satisfy the low delay, low jitter performance requirements of real-time traffic such as VoIP, DiffServ EF class using priority scheduling is normally recommended. This requires limits on admissible load, configured to meet the most stringent QoS in the real-time traffic mix. To handle multiple realtime service types with heterogeneous QoS, we propose a novel alternative: DiffServ AF classes with RED queue management. Recent queue theoretic advances have demonstrated RED's ability to control the delay distribution of inelastic traffic under congested conditions. This significantly reduces late delivery of packets at the cost of (probabilistically) dropping a greater proportion in the network. We investigated the load-quality trade-off in a DiffServ domain with both hop-based and link weight optimized OSPF routing. End-to-end delay, jitter and loss were measured for two AF classes carrying VoIP across all source-destination paths in order to compare the effects of tail- drop and RED queue management. We used the ITU-T E-model to express how these performance measures affect voice quality; results demonstrate that our novel AF configuration enables the network to carry more traffic for a given quality level, and to degrade more gracefully under severe congestion.