This paper introduces a novel cyberattack-resilient model designed for the optimal operation management of electrolysis-based hydrogen refueling stations (eHRSs) integrated with electric power systems. The optimization model aims to coordinate the scheduling of eHRSs to concurrently support both the transportation sector and the electric power utility. This includes fulfilling the hydrogen demand of electric mobility systems (e-Mobility) and enhancing the resilience of the electric grid by following ancillary service signals issued by the grid operator. Adaptive feasibility areas (FAs) are estimated using the operating parameters of the integrated transportation and power system to identify potential cyberattacks. A framework is developed wherein dispersed eHRSs are managed by an eHRS chain aggregator. The operating parameters of eHRSs are communicated between the individual stations and the eHRS chain aggregator. Additionally, the eHRS chain aggregator interfaces with the electric power utility operator to address the utility’s requirements. Various scenarios are modeled to assess the technical and financial impacts of cyberattacks on eHRS. The proposed model is employed to detect false data injection attacks and mitigate the adverse effects of cyberattacks on the integrated transportation and power system. Simulation studies are conducted to evaluate the effectiveness and practicality of the proposed model. The performance of the FA-based method is compared with traditional deep neural network models and data-driven methods, demonstrating 13.5% and 8.68% improvements, respectively in detection accuracy. In addition, the proposed model achieves a 19% reduction in training time.

Advanced stealthy cyberattacks are capable of infiltrating the cybersecurity layers of power grids and alter their operating conditions, resulting in adverse effects on the system performance. Detecting such Stealthy Infiltrated Cyberattacks (SICA) at the earliest opportunity becomes crucial in order to enable power system operators to implement appropriate corrective measures. To that end, this paper proposes the addition of a new cybersecurity layer for SICA after they have broken through existing cyberattack prevention layers. The paper develops the Feasibility Area (FA) as a classifier mechanism to detect SICA in the collected data of Power System State Variables (PSSV). The proposed detection layer consists of two computational stages. The first stage involves estimating the FA parameters through a historical window of data over a specified period of time, which is then inputted to the second stage. In the second stage, the position of each PSSV with respect to the estimated FA is assessed and utilized by the SICA detection mechanism to identify broken through attacks. A flag vector is created indicating the location of each PSSV with respect to the defined FA. The location of each PSSV and its pattern represented in the flag vector are utilized to identify the existence of SICA. Various SICA detection mechanisms using mathematical techniques and the Pattern Recognition Neural Network (PRNN) have been applied. The numerical results from the evaluation of the proposed FA approach demonstrate a promising performance in detecting the SICA using the proposed method.

The utilization of hydrogen-powered vehicles as an alternative to fossil-fueled ones has been identified as a promising solution to reduce greenhouse gas emissions. To promote the deployment of hydrogen-powered vehicles, however, it is imperative that hydrogen refueling stations with on-site electricity-produced hydrogen (eHRSs) are economically utilized for concurrent services. Therefore, this article proposes a new estimation-based method for the optimal scheduling of eHRS for the provision of congestion management services to the distribution system. The method allows for real-time adaptive management of the congestion level in the feeder as decided by the distribution system operator. A new correlation-based method is proposed for online estimation of the amount of congestion and determination of the eHRSs setpoints to relieve congestion on the distribution feeders. This article also develops a congestion management pricing mechanism for the compensation of eHRSs that serve the distribution grid. Numerical results indicate how distribution feeder congestion can be adaptively managed, while the financial profile of the eHRSs is improved via joint application and provision of concurrent services to the transportation sector and grid operator.

Ensuring the cost-effective operation of an unbalanced islanded microgrid (UBIMG) hinges on achieving a proportional power sharing relative to the capacity of the connected distributed energy resource units (DERs). However, inherent characteristics of UBIMG, such as heterogeneous line impedance and unbalanced loads, inevitably result in mismatching the reactive power-sharing (RPS) among the droop-controlled DERs. As a solution, this article introduces an advanced control scheme that combines unsymmetrical per-phase droop control with unsymmetrical per-phase virtual impedance, referred to as unsymmetrical per-phase droop-virtual impedance control (USPDVIC), to enhance the RPS among DERs within the UBIMG. To determine the settings of the proposed control scheme, this study formulates a multiobjective optimization approach to minimize the average generation costs and mismatching in the per-phase RPS within the UBIMG across a set of operating states simultaneously. The performance of the proposed USPDVIC is comprehensively evaluated within a parallel architecture UBIMG and a radial UBIMG-based IEEE 13-bus, IEEE 34-bus, and IEEE 123-bus benchmark systems under various states of operation. These states include changes in loading conditions, plug-and-play of DERs, and system reconfiguration and partitioning. The results, along with comparisons to existing literature, provide solid evidence for the effectiveness of the proposed control scheme in improving the per-phase RPS among the parallel-connected and dispersed DERs within UBIMGs.

Electrochemical modeling is commonly used to model the characteristics of proton exchange membrane (PEM) electrolyzer cells where all losses caused during the electrolysis process are taken into account. The model has a nonlinear relationship between current density and voltage (J–V), with five model parameters that are subjected to change depending on the physical properties and chemical conditions of the PEM electrolyzer. In this article, a novel analytical approach based on the least square error method is proposed to estimate the model parameters and characterize the electrochemical behavior of the PEM electrolyzer under various operating conditions. The accuracy and validity of the proposed approach are tested under different case studies at various operating temperatures, output pressures, hydrogen production rates, and sizes of the dataset. Also, the relationship between the estimated parameters and the operating conditions of the PEM electrolyzer is explored. Finally, the superiority of the proposed approach is demonstrated by comparison to numerical and heuristic optimization parameter identification methods.

Deployment of electric vehicles (EVs) in a fleet system to deal with environmental issues has been at the center of attention over the past several years. While the battery of each EV offers small storage, hundreds of EVs collectively can offer large energy storage to serve a power grid. This article develops a model for a central controller in a fleet system that allows adaptive utilization of EV batteries distributed energy for concurrent services to the transportation and power grid. The optimization model integrates various slack variables and control parameters for managing real-time fare prices, adaptive energy, and reserve margin allocation, interaction with the grid operator, and meeting the fleet target revenue. The proposed model incorporates EV driver's input into the scheduling process to allow the driver to flexibly manage their battery capacities based on their availability and assessment of the transportation services demand. A dynamic pricing mechanism is developed for real-time calculation of fare rates to allow the EV fleet optimization problem to achieve a daily revenue target while limiting fare prices in a competitive market. Numerical results indicate that the model can manage several EVs for various services while enhancing the fleet financial metrics.

This article proposes a new mathematical formulation for optimal operation scheduling for a remote photovoltaic (PV) farm and distributed electric vehicle charging stations (D-EVCSs), which are owned by a private entity. The proposed model is formulated to maximize the profit of the D-EVCSs private investor through optimal electric vehicle (EV) charging coordination and pricing mechanism. The proposed pricing mechanism aims to achieve an expected revenue by the private investor, while guaranteeing a low charging price for EVs to ensure the EV owners’ satisfaction. For the EV supply application, D-EVCS integrates a battery storage system and rooftop PV. However, the rooftop PV installation is constrained to the limited footprint of the EVCS. For this reason, the investment in a remote PV farm and its impact on the expected revenue and the EVs charging price is introduced in this article. The power generated by the remote PV farm can be transacted to the D-EVCSs using a power purchase agreement through the utility grid infrastructure. In such a transaction, the electricity service and distribution fees are paid to the utility grid for overseeing the PV farm power transaction. The proposed model also considers the opportunity for D-EVCSs to participate in the provision of operating reserve and demand response ancillary services. Different case studies are evaluated for the purpose of validating the effectiveness of the proposed model.

This paper incorporates a new perspective into P2P energy trading coordination schemes for EVs by considering Quality of Service (QoS) management. QoS could be utilized as a control metric to facilitate resilient and reliable transactions according to user preferences. To that end, this paper proposes a novel decentralized QoS-based system for P2P energy trading among EV energy providers and consumers. The system utilizes smart contracts to carry out the matching between EVs and monitor the delivery of a QoS-based P2P contract without the presence of a third party. Two QoS-based mechanisms are proposed to match trading EVs in this system. The proposed mechanisms are designed to match single-consumer to multiple-providers and multiple-consumers to multiple-providers based on consumers’ and providers’ QoS requirements and offers, respectively. A fuzzy-based approach with minimum and intelligible input is introduced to determine the weight values of each QoS attribute. Further, a penalty mechanism is developed to discourage dishonest requests/offers and ensure that trading parties stick to their contractual obligations. Numerical simulations are conducted to validate the effectiveness of the proposed QoS-based mechanisms.

This article proposes an optimal zone clustering algorithm of islanded microgrids (IMG) based on supply adequacy taking into account the dynamic performance of distributed state estimation units. The IMG is partitioned into several localized, yet coupled zones, where each zone is responsible for its local state estimate and performs data fusion to reach consensus for shared state variables between zones. The technique proposes a novel algorithm to optimally define the placement of the virtual boundaries of the zones by minimizing the potential power transfer between adjacent zones. The proposed algorithm adopts the distributed particle filter (DPF) technique for the state estimation process. The proposed algorithm also has the ability to come up with one optimal configuration considering different events and scenarios that might occur in the IMG. Monte Carlo simulations demonstrate the efficacy of the proposed technique in the presence of severely corrupted measurements and state values as well as displaying tolerance to major load changes within the IMG. The DPF shows similar performance when compared to its centralized implementation while also providing computational savings by a factor of the number of zones.

Integration of power and gas systems has been recently proposed as a portfolio solution to deal with the sporadic availability of renewables and enhance the flexibility of power systems. In an integrated system, where critical operating information and control signals of both systems need to be communicated, the risk of cyber attack is intensified. In this article, we present a new model for the integration of power and gas systems using power-to-gas (PtG) and gas-fired generation (GfG) facilities. We demonstrate how the operation of the integrated system can be adversely impacted during cyber attacks that may not be detected using traditional methods. We propose two new detection schemes for false data injection attacks against the input and output signals of the PtG/GfG facility scheduler. In the first scheme, a supervised machine-learning technique, based on the convolutional neural network and wavelet transforms, is adopted to detect attacks on the information received by the facility scheduler. In the second scheme, a hybrid neural network is developed, based on an unsupervised learning technique, that requires no labeled training information to detect attacks on the output control signals issued by the scheduler. In both schemes, information acquired from local sensors and deterministic estimation methods is utilized for signal evaluation. The proposed schemes are incorporated into the facilities' scheduler to create a cyber-attack resilient scheduling model in an integrated power and gas grid. The efficacy and feasibility of the proposed model are evaluated via numerical studies using the IEEE30-bus power system integrated with the Belgian gas grid as the test bed using historical operating parameters.

This paper proposes a novel and generic model to calculate the Electric Bus Energy Consumption (EBEC) without the need for a high-resolution speed profile data. The proposed model generates a set of speed profiles using the basic information of the bus trip: trip time, trip length, and distances between successive bus stops. The generated speed profiles could accurately reflect the various traffic conditions and speed behaviors of real-world situations. Roadway Level of Service (LoS) is incorporated in the proposed model to simulate different traffic conditions. Further, a stochastic model for the bus speed profile is adopted to simulate the probability of the bus to stop at each on-route designated stop. The generated speed profiles are then inputted to an accurate EBEC model that considers the route topography, auxiliary loads (lighting, sound, and radio systems) and the impact of the weather conditions. The operation of the heat, ventilation and air conditioning system (HVAC) is also incorporated in the model using the thermal mass balance principle. Using the proposed model, the characteristics of EBEC on a given route can be evaluated through generating a set of speed profiles for the studied route. The proposed model provides transit network planners with a useful tool to appropriately design electric-based transit networks when there is a lack or unavailability of real-time and high resolution data.

The proliferation of hydrogen fueling stations as a critical infrastructure is necessary for the successful materialization of hydrogen-powered vehicles. Such fueling stations can, in part, utilize the renewable/inexpensive electricity, which would otherwise be curtailed, to generate and store hydrogen. The stored hydrogen can later be used to serve the transportation sector and straightforwardly yield profit for the operator of the stations. The available energy in the storage stations, however, would not be utilized effectively during offpeak hydrogen demand by the transportation sector. While hydrogen fueling stations are primarily contemplated as the suppliers to hydrogen vehicles, this paper shows how the storage capacity in each station can be exploited to provide operating reserve (OR) to an electricity market. To that end, this paper proposes a new supervisory-based model for the optimal scheduling of distributed hydrogen storage stations for 1) energy supply to hydrogen-powered vehicles; and 2) OR provision to an electricity market. As such, the economic feasibility of the investment in such stations would be further intensified due to extra financial settlements for the stations via joint applications. This paper, then, unveils a model that brings about more opportunities for the deployment of hydrogen fueling stations, thereby further inspiring the private investment in such an area by private sectors. The efficacy and feasibility of the proposed model are validated using numerical illustration conducted on a test system.

The existing distributed gas-fired generation (GfG) units have (partially) connected the power and gas distribution grids. In addition, as the emerging technology for conversion of renewable/surplus power to synthetic natural gas, i.e., power-to-gas (PtG) materializes, the foundation for a fully integrated power and gas distribution grid is more likely to be set. This could, in turn, bring about new opportunities for exploiting the natural gas distribution grid for mitigation of the existing and imminent issues in power distribution systems. To that end, this paper unveils a new model for optimal joint scheduling of PtG and GfG units in a power-gas embedded grid. The PtG-GfG facility is operated for arbitrage and loss alleviation as a regulation service to the power distribution system. The mathematical formulation of a new method for estimation of the loss reduction in the integrated grid is developed and embedded into the optimization problem. The efficacy and feasibility of the model is numerically validated on a test system. The results indicate that the proposed model can reliably estimate the loss reduction percentage and accordingly determine the scheduling setpoints to achieve the determined loss reduction. It is demonstrated that the model increases the profitability of investment in PtG-GfG facilities via extra financial settlements for the facility operator.

This paper proposes a new online-calibrated time series based model with the application to the day-ahead natural gas demand (GD) forecasting. A double-stage parallel process is developed for creating the forecasting model. The two stages include analysis of the temperature-independent and temperature-dependent components of the GD. The former stage is executed by online processing of the historical GD information considering the intertemporal variation of the GD. The latter stage, however, is conducted by exploiting the features of the GD information correlated with the ambient temperature. The forecast of the temperature is incorporated into the GD forecasting model through a correlation-based function. The model can generate the day-ahead GD forecast with both the hourly and intrahourly resolutions without compromising the forecast accuracy. The model is calibrated online using the historical GD and temperature information to achieve a higher forecast accuracy. The practical challenges associated with the industrial application of the model are also discussed. The application of the proposed model is numerically examined using real-world GD and temperature data, and the results are comprehensively studied. The outcomes reveal the efficacy and feasibility of the proposed model under various cases.

Recent developments in microgrids place strict constraints on the underlying state estimation technology, including the need for a dynamic and distributed approach. Since the problem is reminiscent of classical information fusion [2], the paper explores the application of a fusion-based reduced order, distributed unscented particle filter (FR/DUPF) for dynamic state estimation in microgrids. By partitioning the nonlinear microgrid into a network of nsublocalized and dynamically coupled systems, the FR/DUPF provides computational savings of a factor of nsubover its centralized version. Monte Carlo simulations verify its accuracy by confirming that estimates from the FR/DUPF and centralized filter evolve close to the ground truth.