Neural network based optimization for cascade filling process of on-board hydrogen tank

Compressed hydrogen storage is widely used in hydrogen fuel cell vehicles (HFCVs). Cascade filling systems can provide different pressure levels associated with various source tanks allowing for a variable mass flow rate. To meet refueling performance objectives, safe and fast filling processes must be available to HFCVs. The main objective of this paper is to establish an optimization methodology to determine the initial thermodynamic conditions of the filling system that leads to the lowest final temperature of hydrogen in the on-board storage tank with minimal energy consumption. First, a zero-dimensional lumped parameter model is established. This simplified model, implemented in Matlab/Simulink, is then used to simulate the flow of hydrogen from cascade pressure tanks to an on-board hydrogen storage tank. A neural network is then trained with model calculation results and experimental data for multi-objective optimization. It is found to have good prediction, allowing the determination of optimal filling parameters. The study shows that a cascade filling system can well refuel the on-board storage tank with constant average pressure ramp rate (APRR). Furthermore, a strong pre-cooling system can effectively lower the final temperature at a cost of larger energy consumption. By using the proposed neural network, for charging times less than 183s, the optimization procedure predicts that the inlet temperature is 259.99–266.58 K, which can effectively reduce energy consumption by about 2.5%.     

Refuelling scenarios of a light urban fuel cell vehicle with metal hydride hydrogen storage. Comparison with compressed hydrogen storage counterpart

The high price of hydrogen fuel in the fuel cell vehicle refuelling market is highly dependent on the one hand from the production costs of hydrogen and on the other from the capital cost of a hydrogen refuelling station's components to support a safe and adequate refuelling process of contemporary fuel cell vehicles. The hydrogen storage technology dominated in the vehicle sector is currently based on high-pressure compressed hydrogen tanks to extend as much as possible the driving range of the vehicles. However, this technology mandates the use of large hydrogen compression and cooling systems as part of the refuelling infrastructure that consequently increase the final cost of the fuel. This study investigated the prospects of lowering the refuelling cost of small urban hydrogen vehicles through the utilisation of metal hydride hydrogen storage. The results showed that for low compression hydrogen storage, metal hydride storage is in favour in terms of the dispensed hydrogen fuel price, while its weight is highly comparable to the one of a compressed hydrogen tank. The final refuelling cost from the consumer's perspective however was found to be higher than the compressed gas due to the increased hydrogen quantity required to be stored in fully empty metal hydride tanks to meet the same demand.     

Effect of cascade storage system topology on the cooling energy consumption in fueling stations for hydrogen vehicles

One of the main obstacles of the diffusion of fuel cell electric vehicles (FCEV) is the refueling system. The new stations follow the refueling protocol from the Society of Automotive Engineers where the way to reach the target pressure is not explained. This work analyzes the thermodynamics of a hydrogen fueling station in order to study the effects of the cascade storage system topology on the energy consumption for the cooling facility. It is found that the energy consumption for cooling increases, expanding the total volume of the cascade storage system. Comparing the optimal and the worst volume configurations of the cascade storage tanks at different ambient temperatures, the energy saving is approximately 12% when the average ambient temperature is 20 °C and around 20% when the average ambient temperature is 30 °C. The energy consumption for cooling is significantly influenced by the topology of the cascade storage system and it is particularly relevant in the case of low daily-dispensed amount of hydrogen.     

Optimization of the overall energy consumption in cascade fueling stations for hydrogen vehicles

Hydrogen fueling stations are emerging around and in larger cities in Europe and United States together with a number of hydrogen vehicles. The most stations comply with the refueling protocol made by society of automotive engineers and they use a cascade fueling system on-site for filling the vehicles. The cascade system at the station has to be refueled as the tank sizes are limited by the high pressures. The process of filling a vehicle and afterward bringing the tanks in refueling station back to same pressures, are called a complete refueling cycle. This study analyzes power consumption of refueling stations as a function of number of tanks, volume of the tanks and the pressure in the tanks. This is done for a complete refueling cycle. It is found that the energy consumption decreases with the number of tanks approaching an exponential function. The compressor accounts for app. 50% of the energy consumption. Going from one tank to three tanks gives an energy saving of app. 30%. Adding more than four tanks the energy saving per extra added tank is less than 4%. The optimal numbers of tanks in the cascade system are three or four.     

Effects of pressure levels in three-cascade storage system on the overall energy consumption in the hydrogen refueling station

Studies show that compared with the one-buffer system, the cascade storage system has lower energy consumption in high-pressure hydrogen refueling stations. In the present study, practical dynamic models of the whole hydrogen refueling process are established to evaluate the energy consumption. Accordingly, the filling performance of the three-cascade storage system and single tank storage system are analyzed. Moreover, the impact of the three pressure levels and the charging sequence of the three tanks on the energy consumption are investigated. The obtained results show that changing from one buffer to three tanks gives a total energy saving of approximate 34%. For the three-cascade storage system, the total energy consumption increases approximately linearly with the increase of the pressure of the high-pressure tank. Whereas it shows concave curve shape trends with the increase of low-pressure level and the medium-pressure level. Furthermore, the charging sequence from the low-pressure buffer to the high one decreases the total operation energy consumption to a value slightly lower than the adverse charge sequence.     

Single-tank storage versus multi-tank cascade system in hydrogen refueling stations for fuel cell buses

Many countries in Europe are investing in fuel cell bus technology with the expected mobilization of more than 1200 buses across Europe in the following years. The scaling-up will make indispensable a more effective design and management of hydrogen refueling stations to improve the refueling phase in terms of refueling time and dispensed quantity while containing the investment and operation costs. In the present study, a previously developed dynamic lumped model of a hydrogen refueling process, developed in MATLAB, is used to analyze tank-to-tank fuel cell buses (30–40 kg_H2 at 350 bar) refueling operations comparing a single-tank storage with a multi-tank cascade system. The new-built Aalborg (DK) hydrogen refueling station serves as a case study for the cascade design. In general, a cascading refueling approach from multiple storage tanks at different pressure levels provides the opportunity for a more optimized management of the station storage, reducing the pressure differential between the refueling and refueled tanks throughout the whole refueling process, thus reducing compression energy. This study demonstrates the validity of these aspects for heavy-duty applications through the technical evaluation of the refueling time, gas heating, compression energy consumption and hydrogen utilization, filling the literature gap on cascade versus single tank refueling comparison. Furthermore, a simplified calculation of the capital and operating expenditures is conducted, denoting the cost-effectiveness of the cascade configuration under study. Finally, the effect of different pressure switching points between the storage tanks is investigated, showing that a lower medium pressure usage reduces the compression energy consumption and increases the station flexibility.     

The simulation and analysis of leakage and explosion at a renewable hydrogen refuelling station

The number of hydrogen refuelling stations (HRSs) is steadily growing worldwide. In China, the first renewable hydrogen refuelling station has been built in Dalian for nearly 3 years. FLACS software based on computational fluid dynamics approach is used in this paper for simulation and analysis on the leakage and explosion of hydrogen storage system in this renewable hydrogen refuelling station. The effects of wind speed, leakage direction and wind direction on the consequences of the accident are analyzed. The harmful area, lethal area, the farthest harmful distance and the longest lethal distance in explosion accident of different accident scenarios are calculated. Harmful areas after explosion of different equipments in hydrogen storage system are compared. The results show that leakage accident of the 90 MPa hydrogen storage tank cause the greatest harm in hydrogen explosion. The farthest harmful distance caused by explosion is 35.7 m and the farthest lethal distance is 18.8 m in case of the same direction of wind and leakage. Moreover, it is recommended that the hydrogen tube trailer should not be parked in the hydrogen refuelling station when the amount of hydrogen is sufficient.     

全球加氢站产业、技术及标准进展综述

以美国、欧洲、日本、中国的加氢站作为考察对象,就产业投资运营、主要设备商、相关标准进行梳理,结果表明：在加氢站投资及运营管理方面,美国、欧洲、日本具有较为成熟的经验;在氢增压、储存、加注等技术方面,德国、日本保持领先地位。在诸多氢气和液氢的技术领域,美国保持领先地位;在一些细分技术领域,英国、法国、挪威、俄罗斯具有优势;在加氢站标准方面,美国、日本具有较为完善的标准体系。中国已解决70 MPa氢气增压、加注、储存等领域的部分技术难题,但与其他国家相比,离子液式氢压缩机、液氢泵、液氢储罐、液氢加氢枪等产品的研发仍需持续推进。     

Optimisation of the fuelling of hydrogen vehicles using cascade systems and ejectors

The present study investigates the replacement of expansion valves, used in the cascade system of hydrogen fuelling stations, by a series of ejectors. The major advantage of using ejectors is to recover part of the kinetic energy lost during the expansion of a high-pressure primary flow, in order to entrain a lower pressure secondary flow; thus resulting in a more efficient fuelling.Firstly, a quasi-steady 1-D simulation model of the ejector was calibrated using computational fluid dynamics in terms of the main geometry and pressure conditions.Secondly, the quasi-steady 1-D model of the ejector was used in a dynamic model of the hydrogen fuelling station, in order to investigate the influence of its geometry on the transient fuelling performances. Different fuelling scenarios were explored with varying number of buffer tanks in the cascade system of the fuelling station, and different initial pressures in the vehicle's tank. The results show that the replacement of the expansion valve by an ejector may reduce the energy consumption for hydrogen compression by up to 6.5% using two buffer tanks in the cascade system. On the other hand, increasing the number of buffer tanks reduces the energy savings as the driving pressure ratio decreases.     

Optimal design of wind-powered hydrogen refuelling station for some selected cities of South Africa

The transport sector is considered as one of the sectors producing high carbon emissions worldwide due to the use of fossil fuels. Hydrogen is a non-toxic energy carrier that could serve as a good alternative to fossil fuels. The use of hydrogen vehicles could help reduce carbon emissions thereby cutting down on greenhouse gases and environmental pollution. This could largely be achieved when hydrogen is produced from renewable energy sources and is easily accessible through a widespread network of hydrogen refuelling stations. In this study, the techno-economic assessment was performed for a wind-powered hydrogen refuelling station in seven cities of South Africa. The aim is to determine the optimum configuration of a hydrogen refuelling station powered by wind energy resources for each of the cities as well as to determine their economic viability and carbon emission reduction capability. The stations were designed to cater for 25 hydrogen vehicles every day, each with a 5 kg tank capacity. The results show that a wind-powered hydrogen refuelling station is viable in South Africa with the cost of hydrogen production ranging from 6.34 $/kg to 8.97 $/kg. These costs are competitive when compared to other costs of hydrogen production around the world. The cities located in the coastal region of South Africa are more promising for siting wind powered-hydrogen refuelling station compared to the cities located on the mainland. The hydrogen refuelling stations could reduce the CO₂ and CO emissions by 73.95 tons and 0.133 tons per annum, respectively.     

Thermal management of hydrogen refuelling station housing on an annual level

A housing insulation of hydrogen refuelling station is vital from the aspect of safe operation of equipment in an environment that is installed. To secure hydrogen supply during the whole year, this work brings the solution for both cooling and heating insulation equipment inside of hydrogen refuelling station installed in Croatia, Europe. This hydrogen refuelling station was designed as an autonomous photovoltaic-hydrogen system. In the interest of improving its energy efficiency, an optimal thermal management strategy was proposed. To select the best technological solution for thermal management design which will maintain optimal temperature range inside the housing in cold and warm months, a detailed analysis of the system components thermodynamic parameters was performed. Optimal operating temperatures were established to be 25 °C in summer and 16 °C in winter, considering components working specifications. Insulation, type of cooling units, and heaters have been selected according to the HRN EN 12831 and VDI 2078 standards, while the regime of the heating and cooling system has been selected based on the station's indoor air temperature. The annual required heating and cooling energy were calculated according to HRN EN ISO 13790 standard, amounting to 1135.55 kW h and 1219.55 kW h, respectively. Annual energy share obtained from solar power plant used for the heating and cooling system resulted in 5%. The calculated thermal management system load turned out to be 1.437 kW.     

Model predictive control of an on-site green hydrogen production and refuelling station

The expected increase of hydrogen fuel cell vehicles has motivated the emergence of a significant number of studies on Hydrogen Refuelling Stations (HRS). Some of the main HRS topics are sizing, location, design optimization, and optimal operation. On-site green HRS, where hydrogen is produced locally from green renewable energy sources, have received special attention due to their contribution to decarbonization. This kind of HRS are complex systems whose hydraulic and electric linked topologies include renewable energy sources, electrolyzers, buffer hydrogen tanks, compressors and batteries, among other components. This paper develops a linear model of a real on-site green HRS that is set to be built in Zaragoza, Spain. This plant can produce hydrogen either from solar energy or from the utility grid and is designed for three different types of services: light-duty and heavy-duty fuel cell vehicles and gas containers. In the literature, there is a lack of online control solutions developed for HRS, even more in the form of optimal online control. Hence, for the HRS operation, a Model Predictive Controller (MPC) is designed to solve a weighted multi-objective online optimization problem taking into account the plant dynamics and constraints as well as the disturbances prediction. Performance is analysed throughout 210 individual month-long simulations and the effect of the multi-objective weighting, prediction horizon, and hydrogen selling price is discussed. With the simulation results, this work shows the suitability of MPC for HRS control and its significant economic advantage compared to the rule-based control solution. In all simulations, the MPC operation fulfils all required services. Moreover, results show that a seven-day prediction horizon can improve profits by 57% relative to a one-day prediction horizon; that the battery is under-sized; or that the MPC operation strategy is more resolutive for low hydrogen selling prices.     

Development of efficient hydrogen refueling station by process optimization and control

Development of efficient hydrogen refueling station (HRS) is highly desirable to reduce the hydrogen cost and hence the life cycle expense of fuel cell vehicles (FCVs), which is hindering the large scale application of hydrogen mobility. In this work, we demonstrate the optimization of gaseous HRS process and control method to perform fast and efficient refueling, with reduced energy consumption and increased daily fueling capacity. The HRS was modeled with thermodynamics using a numerical integration method and the accuracy for hydrogen refueling simulation was confirmed by experimental data, showing only 2 °C of temperature rise deviation. The refueling protocols for heavy duty FCVs were first optimized, demonstrating an average fueling rate of 2 kg/min and pre-cooling demand of less than 7 kW for 35 MPa type III tanks. Fast refueling of type IV tanks results in more significant temperature rise, and the required pre-cooling temperature is lowered by 20 K to achieve comparable fueling rate. The station process was also optimized to improve the daily fueling capacity. It is revealed that the hydrogen storage amount is cost-effective to be 25–30% that of the nominal daily refueling capacity, to enhance the refueling performance at peak time and minimize the start and stop cycles of compressor. A novel control method for cascade replenishment was developed by switching among the three banks in the order of decreased pressure, and results show that the daily refueling capacity of HRS is increased by 5%. Therefore, the refueling and station process optimization is effective to promote the efficiency of gaseous HRS.     

Design of a solar hydrogen refuelling station following the development of the first Croatian fuel cell powered bicycle to boost hydrogen urban mobility

Hydrogen refuelling stations are important for achieving sustainable hydrogen economy in low carbon transport and fuel cell electric vehicles. The solution presented in this paper provides us with a technology for producing carbon dioxide free hydrogen, which is an approach that goes beyond the existing large-scale hydrogen production technologies that use fossil fuel reforming. Hence, the main goal of this work was to design a hydrogen refuelling station to secure the autonomy of a hydrogen powered bicycle. The bicycle hydrogen system is equipped with a proton exchange membrane fuel cell stack of 300 W, a DC/DC converter, and a metal hydride storage tank of 350 NL of hydrogen. The hydrogen power system was made of readily available commercial components. The hydrogen station was designed as an off-grid system in which the installed proton exchange membrane electrolyzer is supplied with electric energy by direct conversion using photovoltaic cells. With the hydrogen flow rate of 2000 cc min⁻¹ the hydrogen station is expected to supply at least 5 bicycles to be used in 20 km long city tourist routes.     

Exact algorithms for incremental deployment of hydrogen refuelling stations

Given the large investments required to establish hydrogen refuelling stations (HRSs) and the difficulty in forecasting the sales of fuel cell electric vehicles, incremental HRS deployment offers an efficient method of establishing hydrogen infrastructure with a sufficient load factor and low financial risk. Considering that some HRSs are already in use, this study assumed that the optimal location of a new HRS maximises its distance from existing HRSs and minimises its distance from customer demand points. Accordingly, a multi-objective location model and efficient exact solution methods were proposed to determine the optimal location of one or two new HRSs. As a case study, the solution methods were applied to supply hydrogen to an increasing captive fleet of taxis in a large metropolis such as Paris with fixed demand points. The methods can be widely applied to effectively install one or two HRSs incrementally.     

Experimental study of powder bed behavior of sodium alanate in a lab-scale H2 storage tank with flow-through mode

Chemical hydrogen storage in complex hydrides offers the potential of high gravimetric storage densities compared to intermetallic hydrides, and is therefore a promising technology for mobile applications. The main challenge for mobile application is still the required high refuelling rate of the hydrogen storage tanks. Since hydrogen is bonded by an exothermal chemical reaction in complex hydrides, appropriate storage tanks require high heat transfer rates for the cooling of the tank. Hydride tanks that are state of the art rely on an indirect cooling and are additionally equipped with e.g. finns, foams, etc. to improve the heat transfer rate. For the present study, an improved laboratory tank, which allows for indirect as well as direct cooling by excess H₂ gas (flow-through mode), has been designed and built. This laboratory tank is filled with 87 g of NaAlH₄ (doped with 2 mol% CeCl₃) and equipped with 8 thermocouples as well as two pressure sensors. Experimental results presented in this paper show a significant influence of the cooling by gaseous excess H₂ on the flow-directional temperature profiles at the part of the reaction bed close to the inlet. Considering the overall conversion, this influence is rather small due to the low heat capacity flux (ρc_p)_H2. Furthermore, it is shown that changes in material properties, attributed to the effects of heat and mass transport as well as intrinsic reaction kinetics, can be measured and assessed by the temperature and pressure sensors. After about 10 complete charging and discharging cycles, the initial permeability K of the bed has decreased by 50% to 1.6·10⁻¹² m². In the same time, the initial thermal conductivity has increased by a factor of 1.3 to values reported in literature (0.67 Wm⁻¹ K⁻¹) and remains constant during further cycles. Additionally, it is observed that the reaction rate of the second absorption step improves, even after a total of 36 cycles.     

Modelling and control of hydrogen and energy flows in a network of green hydrogen refuelling stations powered by mixed renewable energy systems

The planning of a hydrogen infrastructure with production facilities, distribution chains, and refuelling stations is a hard task. Difficulties may rise essentially in the choice of the optimal configurations. An innovative design of hydrogen network has been proposed in this paper. It consists of a network of green hydrogen refuelling stations (GHRSs) and several production nodes. The proposed model has been formulated as a mathematical programming, where the main decisions are the selection of GHRSs that are powered by the production nodes based on distance and population density criteria, as well the energy and hydrogen flows exchanged among the system components from the production nodes to the demand points. The approaches and methodologies developed can be taken as a support to decision makers, stakeholders and local authorities in the implementation of new hydrogen infrastructures. Optimal configurations have been reported taking into account the presence of an additional hydrogen industrial market demand and a connection with the electrical network. The main challenge that has been treated within the paper is the technical feasibility of the hydrogen supply chain, that is mainly driven by uncertain, but clean solar and wind energy resources. Using a Northern Italian case study, the clean hydrogen produced can be technically considered feasible to supply a network of hydrogen refuelling stations. Results show that the demands are satisfied for each time period and for the market penetration scenarios adopted.     

Reinforcement learning based energy management systems and hydrogen refuelling stations for fuel cell electric vehicles: An overview

This paper examines the current state of the art of hydrogen refuelling stations-based production and storage systems for fuel cell hybrid electric vehicles (FCHEV). Nowadays, the emissions are increasing rapidly due to the usage of fossil fuels and the demand for hydrogen refuelling stations (HRS) is emerging to replace the conventional vehicles with FCHEVs. Hence, the availability of HRS and its economic aspects are discussed. In addition, a comprehensive study is presented on the energy storage systems such as batteries, supercapacitors and fuel cells which play a major role in the FCHEVs. An energy management system (EMS) is essential to meet the load requirement with effective utilisation of power sources with various optimizing techniques. A detailed comparative analysis is presented on the merits of Reinforcement learning (RL) for the FCHEVs. The significant challenges are discussed in depth with potential solutions for future work.     

Off-grid test results of a solar-powered hydrogen refuelling station for fuel cell powered Unmanned Aerial Vehicles

Fuel cell (FC) propulsion for small (MTOW < 25 kg) Unmanned Aerial Vehicles (UAVs) provides a route for lower capital cost, environmentally friendlier and low noise operation. Most FC-based UAVs tested to date rely on compressed gas cylinders delivered to the point of use and used to refill the UAV hydrogen tanks on-site or chemical hydride systems to produce hydrogen on-board. An attractive alternative option is to produce hydrogen on-site from an off-grid renewable source according to the UAV fuel demand. A prototype off-grid solar-based hydrogen refuelling station for UAVs was developed for that purpose by Boeing Research & Technology Europe. A test program was carried out to evaluate the dynamic response of the hydrogen UAV refuelling system operating in an off-grid manner (disconnected from the AC grid). The system comprises a concentrated photovoltaic (CPV) array, an alkaline electrolyser, a low pressure hydrogen buffer tank and the required power electronics. The electrolyser was connected to the CPV source in an off-grid manner. The results from the off-grid tests are presented in this paper.     

Optimization of hydrogen vehicle refueling via dynamic simulation

A dynamic model has been developed to analyze and optimize the thermodynamics and design of hydrogen refueling stations. The model is based on Dymola software and incorporates discrete components. Two refueling station designs were simulated and compared. The modeling results indicate that pressure loss in the vehicle's storage system is one of the main factors determining the mass flow and peak cooling requirements of the refueling process. The design of the refueling station does not influence the refueling of the vehicle when the requirements of the technical information report J2601 from Society of Automotive Engineers are met. However, by using multiple pressure stages in the tanks at the refueling station (instead of a single high-pressure tank), the total energy demand for cooling can be reduced by 12%, and the compressor power consumption can be reduced by 17%. The time between refueling is reduced by 5%, and the total amount of stored hydrogen at high pressure is reduced by 20%.