Search

Decarbonization of the European energy networks is critical to meet Commission targets in the coming decades. The presented study aims to contribute to this by analysing one of the proposed solutions: power-to-gas. A technoeconomic model is created for the purposes of evaluating specific projects on their feasibility in terms of local constraints and opportunities, using a current project as a template for model generation and analysing different possible configurations in 8 operational scenarios. Five metrics were used for scenario analysis: levelized cost of methane, minimum selling price, operational hours, hydrogen tank size and capital cost. The results from the analysis indicate that, in terms of the stated project, synthetic natural gas production and grid injection along with on-site mobility applications provide the best economical result. However, selling prices of synthetic natural gas obtained are one magnitude higher than current natural gas prices, indicating government support is required for further development. Future projections of electrolyser efficiency and equipment capital costs will greatly reduce production costs, giving promise for feasible business cases in the coming years.

International audience; Nowadays, large amounts of renewable energy have been installed around the world, and building a zero carbon emission energy supply system has been put on the agenda. Then, how to size the zero carbon emission energy system to achieve cost-effective is an essential problem. In this paper, we build a 100% renewable energy based generating station and microgrid clusters to supply the load demands. Three objectives are considered, namely, minimizing the total cost, minimizing the total exchanged energy, and maximizing the installed PV panels. Genetic algorithm is adopted to solve the problem. The simulation results show that: 1) the hydrogen storage operates as the core device to build the 100% renewable energy based power supply system; 2) when the investment cost is decreasing, the total exchanged energy is increasing, and the installed PV panels are decreasing; when the PV panels are decreasing, the total exchanged energy is decreasing; 3) the volume of the hydrogen tanks in generating station is larger than that in microgrids, because it needs to cover the demands from microgrids.

In Press, Corrected Proof; International audience; As more and more activities are moving online, the need for datacenters is increasing which raises their energy consumption and the resulting greenhouse gas emissions. The supply of datacenters with renewable energy is often seen as the main solution to this nexus. However, multiple challenges are posed by their integration. Solutions to these challenges, mainly in the form of flexibility and energy efficiency, require multidisciplinary skills. As a consequence, this article aims to review scientific efforts in several complementary disciplines: electrical engineering, computer science, control system engineering and operations research. It covers datacenter architecture, infrastructure sizing and resource management on both information and energy sides. Conclusions on current progresses are drawn and challenges that remain to be addressed are listed as well.

Tropical climate is characterized by hot temperatures throughout the year. In areas subject to this climate, air conditioning represents an important share of total energy consumption. In some tropical islands, there is no electric grid, in these cases, electricity is often provided by diesel generators. In this study, in order to decarbonize electricity and cooling production and to improve autonomy in a standalone application, a microgrid producing combined cooling and electrical power was proposed. The presented system was composed of photovoltaic panels, a battery, an electrolyzer, a hydrogen tank, a fuel cell, power converters, a heat pump, electrical loads, and an adsorption cooling system. Electricity production and storage were provided by photovoltaic panels and a hydrogen storage system, respectively, while cooling production and storage were achieved using a heat pump and an adsorption cooling system, respectively. The standalone application presented was a single house located in Tahiti, French Polynesia. In this paper, the system as a whole is presented. Then, the interaction between each element is described, and a model of the system is presented. Thirdly, the energy and power management required in order to meet electrical and thermal needs are presented. Then, the results of the control strategy are presented. The results showed that the adsorption cooling system provided 53% of the cooling demand. The use of the adsorption cooling system reduced the needed photovoltaic panel area, the use of the electrolyzer, and the use of the fuel cell by more than 60%, and reduced energy losses by 7% (compared to a classic heat pump) for air conditioning.

International audience; The integration of Demand-Side Management (DSM) in the planning of Isolated/Islanded Microgrids (IMGs) can potentially reduce total costs and customer payments or increase renewable energy utilization. Despite these benefits, there is a paucity in literature exploring how DSM affects the planning and operation of IMGs. The present work compares the effects of five different strategies of DSM in the planning of IMGs to fulfill the gaps found in the literature. The present work embodies a Disciplined Convex Stochastic Programming formulation that integrates the planning and operation of IMGs using three optimization levels. The first level finds the capacities of the energy sources of the IMG. The second and third levels use a rolling horizon for setting the day-ahead prices or the stimulus of the DSM and the day-ahead optimal dispatch strategy of the IMG, respectively. A case study shows that the Day-Ahead Dynamic Pricing DSM and the Incentive-Based Pricing DSM reduce the total costs and the Levelized Cost of Energy of the project more than the other DSMs. In contrast, the Time of Use DSM reduces the payments of the customers and increases the delivered energy more than the other DSMs

International audience

Proton exchange membrane (PEM) fuel cell hasbeen widely used in diverse applications, especially in automotivefield. However, durability and cost are two principle barriers totake PEM fuel cell into wide commercial use. The discipline ofprognostics and health management (PHM) tends to help againstthis problem. PHM aims at deploying predictive maintenance andanticipating degradation mitigation strategies for PEM fuel cell.This paper focuses on the prognostics of PHM and an adapteddata-driven fuel cell prognostics approach based on multiplicativefeature decomposition and echo state network is proposed to predict the fuel cell degradation behaviour under dynamicoperation conditions. Experimental data is used to verify theeffectiveness of the proposed algorithm.

Multi-energy-supply microgrids covering different types of demands are expected to play an important role in smart grids. Local generation, energy storage systems, and renewable energy sources can form load service entities, which can provide ancillary services to the utility grid and to consumers. On the other hand, microgrids can also sell energy to load service entities or the utility grid to obtain profits. But how the load service entities can decide the electricity selling price to multiple microgrids, and how the microgrids can decide the electricity selling price to load service entities are problems. In this paper, we present a price decision method for multiple microgrids considering demand response. Mixed integer linear programming is used to control the operation of each microgrid, and is also used to operate the load service entities. A genetic algorithm is used to search for the best price for each microgrid and the load service entities. The combined method is deployed in a decentralized way, namely, each microgrid runs its own operation problem. The simulation results show that the new searched price works better than a basic time of use price, which can reduce the operation cost of the whole system. The searching method is compared with a method based on the Cournot model. At last, a large system is tested, in which 4 load service entities, 16 microgrids and an IEEE30-node network are considered. In order to reduce the searching time, a neural network model is presented to estimate the operation of the whole system. Based on the neural network model, the prices are obtained, and the results show that the prices based on neural network model are better than with the time of use price.

Renewable energy-based multi-energy supply microgrids are expected to play an important role in smart cities. How to schedule such microgrids in grid-connected mode and dispatch power among sources inside the microgrids is a problem. Moreover, as electric vehicles are becoming more and more common, the charging of large numbers of vehicles is also a challenge for the utility grid. In this paper, we build a temporal-spatial electric vehicle charging demand model, which includes three parts: trip plans, duration of stay, and search for the shortest path based on the Dijkstra algorithm. Then, we build a grid-connected gas/electricity/heat microgrid and present a coordinated scheduling method for this microgrid. A day-ahead scheduling method is used to decide the role of the microgrid (i.e., operate as a load or as a generator from the point of view of the utility), a real-time rolling-horizon dispatching algorithm is used to respond to the forecasting errors and at the same time implement the real-time actual power exchange between the microgrid and the main grid. The problem is formulated as a mixed integer linear programming problem. The temporal-spatial electric vehicle charging demands model is simulated based on a 81-node transportation network, while the energy supply network is a combined IEEE-30, gas-20 and heat-14 network. The simulation results show the effectiveness of this coordinated scheduling method.

Multi-energy supply systems are expected to play an important role in smart grids. Today's energy supply systems are large nodes networks, and different types of energy are needed at each node to satisfy the different energy demands. These different types of energy can then be converted to each other through specific devices. How to decide the ratings of these devices at each node to make the system cost-effective is addressed in this paper. The focus is set on a gas/electricity/heat hybrid network. A hydrogen storage system (fuel cell, electrolyzer, and tanks) is used as electricity storage system, a combined heat and power device is used to produce heat and electric power, etc. A mixed integer linear programming algorithm is used to determine the optimal operation schedule of the system, where the goal is to minimize shed load. A genetic algorithm is also used to search for the best size of each component, where the goal is to minimize the total investment costs. In order to resist to contingency events, betweenness centrality (describing the relative importance of each node) is then used to find the worst case under contingency events. This worst case scenario is used to research about the influence of contingencies on the sizing results. At last, two cases (modified 13-node network and IEEE 30 + Gas 20 + Heat 14 nodes system) are tested using the proposed sizing method. The results show that the renewable energy location, investment cost of components, and the structure of the whole system influence the sizing results. When the installed capacity of photovoltaic panels is reduced by 50%, the capacity of the electrolyzer decreases by 3%, the capacity for the hydrogen tanks increases by 2%; when the investment cost of the fuel cell and electrolyzer decreases by 50%, the capacity of photovoltaic increases by 14%, the electrolyzer increases by 13%, and hydrogen tanks increase by 2%. After considering the worst case contingency event, for case I, the capacity of photovoltaic and fuel cell increase by 12% and 11%, and the electrolyzer increases by 34%; for case II, the capacity of photovoltaic and fuel cell increase by 8% and 11%, and the electrolyzer increases by 57%.

Microgrids are small-scale power systems with local generation, storage systems and load demands, that can operate connected to the main grid or islanded. In such systems, optimal components sizing is necessary to make the system secure and reliable, while minimizing costs. In this paper, a stand-alone microgrid considering electric power, cooling/heating and hydrogen consumption is built. A unit commitment algorithm, formulated as a mixed integer linear programming problem, is used to determine the best operation strategy for the system. A genetic algorithm is used to search for the best size of each component. The influence of three factors (operation strategy, accuracy of load and renewable generation forecasts, and degradation of fuel cell, electrolyzer and battery) on sizing results is discussed. A 1-h rolling horizon simulation is used to check the validity of the sizing results. A robust optimization method is also used to handle the uncertainties and evaluate their impact on results.

International audience; This paper proposes an advanced aging-tolerant control for a fuel cell/super-capacitor hybrid system applied to a commercial vehicle. The controller is designed with the Interconnection and Damping Assignment-Passivity-Based Control (IDA-PBC) method to solve the converters coordination problem, where the state-of-charge of the super-capacitors and all current limitations are considered into the non-linear controller. The aging of the fuel cell is estimated in real-time by an extended Kalman filter and is integrated in the controller in order to preserve the stability of the whole system. Finally, a hardware-in-the-loop platform based on an INTEL/ALTERA FPGA is designed in order to validate the real-time operation of the algorithms for a specific case study with a fuel cell vehicle.

International audience; Demand response (DR) and renewable energy sources have opened new avenues for end-users to lower their energy expenses via energy management systems. Aggregators facilitate the participation of end-users by acting on their behalf and interacting with bulk electricity markets. In this paper, an energy management algorithm is presented to investigate the impact of distributed photovoltaic (PV) and central energy storage system (ESS) assets on the economic performance of an energy aggregator in the residential sector. To enable DR, the aggregator provides a competitive incentive price to end-users, and centrally optimizes the central ESS assets and schedule of committed customer elastic loads. Thus, customers reduce their daily electricity bill while the aggregator decreases the aggregated peak consumption and earns profits as a return for providing DR services. The scope of this paper pertains to the economic impact of distributed PV and central ESS assets on aggregator profits and customer savings resulting from DR, including ESS degradation. Simulation results showed that the central ESS increases the income of the aggregator, while residential PV reduces the impact of DR.

International audience; This paper describes the design of a measurement system for supraharmonic emissions in the frequency range of 2 to 150 kHz, and analyzes the measurements obtained in real grid scenarios. The measurement system is first characterized in the laboratory. The Design of Experiment approach then uses an adequate number of experiments to identify the effects and interactions of factors responsible for supraharmonic emissions. For each of these experiments, the measured supraharmonic emissions are analyzed and quantified using the discrete Fourier transform. Next, this data is studied using the Analysis of Variance method, which enables identifying the critical factors that generate supraharmonic emissions in the network. The measurement and analysis results show the individual effects and interactions between these factors.

International audience; We present a new performance metric for co-optimization of dispatch and reserves in microgrids. A metric based on NERC criteria is used for each asset to quantity its reliability-based value. A system level metric is obtained through the individual performance metrics, dispatched capacity, and net dispatchable capacity available as reserve. Simulations are performed on a notional microgrid with a dispatchable and a non-dispatchable distributed energy resource to demonstrate the calculation of the metric.

International audience; Renewable energy based multi-energy supply microgrids not only can cover different types of demands (such as, electricity/heat/gas), but also can interconnect with different utility grid networks (electricity/heat/gas). When there are large numbers of grid-connected microgrids, how to operate these multiple microgrids in real-time is a problem. In this paper, day-ahead stochastic optimization scheduling and real-time sliding window model predictive control are used to control the operation of microgrids. In order to consider the influence of future prediction on the current optimal decision results, different prediction methods are adopted to predict the load demands and renewable energy output. For example, online learning Markov chain prediction, and support vector machine are used to predict the future values. As for comparison, robust prediction and bilevel optimization are adopted to describe the future prediction uncertainty. The real-time operation of microgrids aims to follow the day-ahead exchanged energy with utility grids, which can minimize the impact of the microgrid on the utility grids. The supply network is an IEEE30 + gas20+heat14 network. The results show that: 1) when the sliding window number is smaller, the total operation cost is larger, but the calculation time is smaller, the trade-off between sliding window numbers and calculation time should be considered; 2) the accuracy of the prediction impacts the 2-norm error of the operation cost, when we decrease by “1” unit of 2-norm prediction error of the whole system, the 2-norm operation cost will decrease by “0.15” unit; 3) from the view of the post-event analysis (total operation cost), for the Markov chain prediction method, the relative error is about 0.32%, is better than the support vector machine method; 4) in the robust cases, the larger the conservative value, the higher the stored hydrogen energy. At last, the results of real-time sliding window model predictive control problem are influenced by the future prediction methods and the window numbers.

International audience; This paper describes the design of the measurement system for supraharmonic emissions in the frequency range 2-150 kHz in real grid scenarios and factors considered for design. It also portrays the validation of the measurement system on a smart grid platform and the design of experiments performed on this platform. Various network combinations and factors considered for real grid measurements are described in the paper. Index Terms-Measurement system, network combinations, power quality, real grid measurements, supraharmonic emissions.

International audience; Multi-energy-supply microgrids covering different types of demands are expected to play an important role in smart grids. Local generation, energy storage systems, and renewable energy sources can form load service entities, which can provide ancillary services to the utility grid and to consumers. On the other hand, microgrids can also sell energy to load service entities or the utility grid to obtain profits. But how the load service entities can decide the electricity selling price to multiple microgrids, and how the microgrids can decide the electricity selling price to load service entities are problems. In this paper, we present a price decision method for multiple microgrids considering demand response. Mixed integer linear programming is used to control the operation of each microgrid, and is also used to operate the load service entities. A genetic algorithm is used to search for the best price for each microgrid and the load service entities. The combined method is deployed in a decentralized way, namely, each microgrid runs its own operation problem. The simulation results show that the new searched price works better than a basic time of use price, which can reduce the operation cost of the whole system. The searching method is compared with a method based on the Cournot model. At last, a large system is tested, in which 4 load service entities, 16 microgrids and an IEEE30-node network are considered. In order to reduce the searching time, a neural network model is presented to estimate the operation of the whole system. Based on the neural network model, the prices are obtained, and the results show that the prices based on neural network model are better than with the time of use price.

International audience; Microgrid (MG) clusters are small power systems that can help improve the resilience and flexibility of larger systems. In this work, we focus on designing an agent-based decentralized coordination method that healthy MGs to help unhealthy ones (e.g., in the case of insufficient generation). Coordination is expected to eliminate the power imbalance inside MGs, thus limiting the impact for customers and maximizing local generation. Decentralized coordination is especially useful as the resulting system is not subject to single points-of-failure. To ensure the feasibility of the power dispatch resulting from agents interaction, a distributed power flow calculation is used and enables verifying line capacity constraints. The operation of the system is tested with a system including 12 MGs and 33 MGs.

International audience; This paper introduces a day-ahead energy management algorithm for the coordination of smart homes with renewable energy sources and energy storage systems in neighborhood areas. The aim of this study is to establish a day-ahead decentralized coordination method with appliance scheduling and energy sharing (among smart homes) to minimize the electricity bills of consumers under dynamic pricing. The energy sharing algorithm focuses on increasing the utilization of renewable sources by controlling storage units. A multi-agent system is used to model entities (smart homes, aggregator and utility) as agents and the optimization problem is solved in a decentralized manner by home agents with a genetic algorithm. The performance of the coordination algorithm is evaluated annually with and without considering forecasting errors.

International audience; Microgrids are increasingly popular small-scale power systems with local generation, storage and loads. Interconnecting several microgrids together could provide a flexible structure for power distribution and enable large renewable and distributed penetration levels, as well as coordination between microgrids when disturbances occur. This paper proposes a control strategy for such connected microgrids. Although a microgrid aims to operate independently in normal conditions, several microgrids can support each other in case of contingencies or insufficient generation. Such coordination aims at reducing load shedding and generation curtailment, and is achieved by coordinating the output power of microgrids. Each microgrid is equipped with an agent to achieve self-control and to negotiate with other microgrids, for example to request power to its neighbors to support its loads. The Newton-Raphson and consensus methods are used to calculate the output power of each microgrid. The control strategy is validated using simulations on an IEEE 13-node test feeder.

International audience; To solve the converters coordination problem of a fuel cell/super capacitor hybrid power system, an advanced passivity-based control is proposed in this paper. The Interconnection and Damping Assignment Passivity Based Control (IDAPBC) is applied to design the controller. This nonlinear controller considers the state of charge of super capacitor and achieves the stability of the whole closed-loop system. Its feasibility for an electrical vehicle is proven by the simulation results and the comparison with the previous method.

International audience; Aggregators are market participants that bridge the gap between the bulk electricity market and the emerging active end-user (smart home) by efficiently scheduling or allocating resources to meet certain objectives in the electricity grid. The computational burden and processing time of such allocation problems increases with the number of resources. Using high performance computing and parallel processing techniques, the computational time for simulating the environment can be managed. In this paper, we present and compare three parallel processing techniques executed on a dedicated high performance computer for simulating a multi-day aggregator-based resource allocation problem in the Smart Grid.

International audience; Abstract—In this paper, a decentralized control algorithm is presented for coordinated energy sharing among smart homes in neighborhood areas using a game-theoretic approach and a multi-agent system (MAS). The aim of the study is to reduce theelectricity bill of end-users with dynamic pricing where price is associated to aggregated consumption. To reduce the cost of consumption, a control algorithm performs home appliance scheduling and battery control while enabling energy sharing among neighbors in the neighborhood. We assume that photovoltaic (PV) and battery systems are installed in smart homes and end-users are decision-makers willing to optimize the run time of electricity appliances and the control inputs of the battery.In particular, end-users aim to schedule controllable appliances and/or decide about battery charging during low price hours anddischarging during high price hours. The battery can be charged by three strategies: using local PV generation, from neighborhoodresidual generation and grid energy jointly or distinctly. In this study, a MAS is used for modeling entities (homes and aggregator) in the neighborhood as agents. The aggregator agent is the supervisor agent which determines the aggregated profile and dynamic price by communicating with home agents. Home agents are independent and selfish decision-makers which only focus on the maximization of their own welfare while achieving near-optimal performance at Nash equilibrium of a formulated non-cooperative coordination game. Results show that each smart home can benefit from this scheme, compared to a baseline (no control) scenario, as well as reduce the neighborhood total cost and peak load consumption.

International audience; Microgrids are small-scale power systems that include local generation and storage units to serve their loads. The effectiveness of such systems depends on both sizing and operations, that need to be efficient to minimize costs while ensuring reliable power delivery. In this paper, we build a stand-alone microgrid while considering not only electric power, but also cooling, heating, and hydrogen consumption. A unit commitment algorithm, formulated as a mixed integer linear programming problem, is used to determine the best operation strategy of the system. A genetic algorithm is used to search for the best size of each component. Results show the feasibility of the proposed sizing method.

International audience; As natural disasters seem to become more and more frequent in some parts of the world, power system resilience is a growing concern, especially due to the economic consequences of the induced blackouts. In this paper, we use a temporal-spatial model to describe natural disasters. Based on fragility curves, we use a Monte Carlo method to estimate the destructive dynamic of such disasters, and simulate a hurricane occuring in the IEEE 37-bus network. We then propose an optimization method to find the most suitable nodes to integrate distributed generators (DG). With the resulting DG additions, we select the lines to be hardened to further improve system resilience.

This paper proposes a model of proton exchange membrane (PEM) fuel cells for real-time applications, with a 2-dimensional model of mass flow in the channel layer and a 1-dimensional model of electrochemical and mass diffusion in the diffusion layer and polymer membrane. It is also shown how an optimized program based on finite difference and finite volume methods may be developed for embedded applications, and how the error, convergence and hardware properties of this program may be analyzed to determine whether such a program can be uploaded to a real-time controller. New ideas for increasing the real-time simulation speed are given and discussed. An experimental validation of the model is done. Results from the 2-dimensional real-time mass flow model are presented and discussed, as well as performance results of the real-time simulation model.