Intelligent energy management system and method for hydrogen-light complementary micro-grid in railway park

Abstract

The invention provides an intelligent energy management system and method for a hydrogen-light complementary micro-grid in a railway park, and the system comprises a data sensing layer which collects dynamic operation data in real time; the decision control layer is in communication connection with the data sensing layer, generates an optimization scheduling control instruction according to the dynamic operation data and a set decision control mechanism, and comprises a day-ahead economic scheduling mechanism, a real-time rolling optimization mechanism and a second-level dynamic adjustment mechanism; and the coordinated execution layer comprises a photovoltaic subsystem, a storage battery unit, an electrolytic cell, a fuel cell and a controller corresponding to the hydrogen storage module, and is used for responding to the optimal scheduling control instruction and performing accurate execution to realize dynamic distribution of power and energy balance. By adopting the scheme, the problems of single scheduling power, poor switching scheduling timeliness and insufficient economy in the prior art can be solved, a hydrogen-electricity hybrid energy storage state machine and multi-mode switching logic are designed, and intelligent dynamic energy distribution is realized; and a rolling optimization algorithm is introduced, so that challenges caused by uncertainty of photovoltaic power generation are effectively coped with.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent management and control of electric energy, and particularly relates to a smart energy management system and method for a hydrogen-light complementary micro-grid in a railway park, which can be effectively applied to micro-grid energy management and intelligent control; the system realizes efficient conversion and utilization of energy through multi-energy cooperation and intelligent scheduling, and is particularly suitable for typical scenarios such as railway parks with high requirements for power supply continuity, complex load fluctuations, and the need to balance economy and environmental protection. BACKGROUND

[0002] As a typical high-energy consumption subject, the load type of the railway park includes repair workshops, high-power tools, rail transit scheduling equipment, office lighting, and railway signal and communication systems. Such a park load has the characteristics of strong time-varying, large peak-valley difference, and extremely high power supply reliability requirements.

[0003] In recent years, with the promotion of the low-carbon optimization development goal, more and more railway parks have begun to build distributed photovoltaic systems to reduce electricity costs and achieve green and low-carbon operation. However, photovoltaic power generation has intermittency and random fluctuation, and its output is significantly affected by weather conditions. There is almost no power output during cloudy, rainy, snowy or night periods. If this fluctuation is not effectively regulated, it will have an adverse effect on the voltage stability, frequency fluctuation and power supply reliability of the railway park power grid.

[0004] To cope with the uncertainty of renewable energy output, the introduction of hydrogen energy storage systems (including electrolytic hydrogen production devices, hydrogen storage tanks and fuel cell power generation units) is considered by some researchers as an effective long-term energy storage and energy balancing scheme. However, the present inventors have found that in actual engineering, there are significant differences in dynamic response and coupling complexity between the hydrogen energy system and photovoltaic, energy storage batteries, and the power grid: the battery has a fast response speed (seconds), suitable for short-term power balancing; the fuel cell has a slower response (minutes), suitable for continuous power output; the electrolytic tank needs to consider the start-stop cost and efficiency curve, and frequent start-stop will reduce the service life. At present, most railway park micro-grids still use control strategies based on fixed thresholds or simple rules, which are difficult to balance economic operation, energy balance and power supply reliability at the same time. Especially when the power grid fluctuates, the load mutates or the time-of-use electricity price changes significantly, the existing system often cannot achieve dynamic optimization scheduling.

[0005] The information disclosed in the background section of this application is only intended to deepen the understanding of the general background of the application, and should not be regarded as acknowledging or implying in any form that this information constitutes prior art known to those skilled in the art. SUMMARY

[0006] To solve the above problems, the present application provides a kind of for railway park hydrogen light complementary micro-grid intelligent energy management system, using the scheme can overcome the prior art scheduling power single, switching scheduling time efficiency is poor, the problem of insufficient economy, hydrogen-electricity hybrid energy storage state machine and multi-modal switching logic are designed, intelligent dynamic energy distribution is realized;And introduce the rolling optimization algorithm based on model predictive control (MPC), effectively deal with the challenge brought by photovoltaic power generation uncertainty.The data perception layer in the system real-time acquisition and photovoltaic subsystem, hydrogen energy subsystem, battery module and grid related dynamic operation data;Decision control layer is connected with data perception layer, generates optimization scheduling control instruction according to dynamic operation data according to the set decision control mechanism, including day-ahead economic dispatching mechanism, real-time rolling optimization mechanism and second-level dynamic adjustment mechanism;Coordination execution layer, including the controller corresponding to photovoltaic subsystem, battery unit, electrolytic cell, fuel cell and hydrogen storage module, for responding optimization scheduling control instruction and accurate execution, realize the dynamic distribution and energy balance of power.Optimally, in one embodiment, the system includes.

[0007] Data perception layer, configured to real-time acquisition and photovoltaic subsystem, hydrogen energy subsystem, battery module and grid related dynamic operation data; Decision control layer, which is connected with the data perception layer, is configured to generate optimization scheduling control instruction according to the dynamic operation data according to a specific decision control mechanism, and the decision control mechanism includes day-ahead economic dispatching mechanism, real-time rolling optimization mechanism and second-level dynamic adjustment mechanism; Coordination execution layer, including photovoltaic inverter controller, energy storage converter controller, electrolytic cell controller, fuel cell controller and hydrogen storage controller, configured to respond to optimization scheduling control instruction and accurate execution, realize the dynamic distribution and energy balance of power.

[0008] In one embodiment, the data perception layer acquires dynamic operation data in real time through multiple types of sensors and communication networks, and the dynamic operation data includes photovoltaic power, demand load power, grid power supply state information, real-time energy storage state and hydrogen energy operation data.

[0009] Further, in one embodiment, the data perception layer is connected with the hydrogen system controller of the hydrogen energy subsystem to obtain hydrogen energy operation data, including hydrogen storage tank pressure, hydrogen storage state, hydrogen fuel cell state and electrolytic cell state data.

[0010] Preferably, in one embodiment, the decision control layer is configured to generate daily optimization scheduling operation plan according to day-ahead economic dispatching mechanism according to the following logic: The decision control layer is based on high-precision photovoltaic output and load prediction data, and takes the optimal economy of all-day operation as the primary goal, adopts a mixed integer linear programming algorithm, and formulates a 24-hour operation plan as an optimized dispatching control reference information.

[0011] Optionally, in an embodiment, the decision control layer is configured to generate optimized dispatching control instructions according to a real-time rolling optimization mechanism according to the following logic: The model predictive control framework is used to analyze the operation deviation of the target time step according to the set time period, the power distribution information of the future time step is configured according to the operation deviation, and the corresponding configuration instructions of the photovoltaic subsystem, the hydrogen energy subsystem, the battery module and the power grid in the matching operation mode are generated in combination with the daily optimized dispatching operation plan.

[0012] Further, in an embodiment, the decision control layer generates optimized dispatching control instructions according to a second-level dynamic adjustment mechanism according to the following logic: The instantaneous change of the DC bus voltage and the state of charge of the energy storage battery are integrated to generate battery charging and discharging power control instructions, so as to quickly adjust the charging and discharging power of the battery, make up for the slow response of the fuel cell, maintain real-time power balance.

[0013] In an embodiment, the decision control layer is further configured to judge the adaptive operation mode in real time according to the dynamic operation data obtained by the data perception layer, and generate mode execution instructions; the operation mode includes a photovoltaic dominant hydrogen production mode, a fuel cell grid-connected power supply mode, an off-grid island operation mode, a valley electricity hydrogen production mode and a peak electricity generation mode.

[0014] Optionally, in an embodiment, the decision control layer is configured to judge the adaptive operation mode according to the following logic: When the photovoltaic power is greater than the demand load data, the state of charge of the battery is further judged, if the state of charge of the battery is greater than the set upper limit index, the photovoltaic dominant hydrogen production mode is judged as the adaptive operation mode; the electrolyzer module is started, and the excess photovoltaic power is used to produce hydrogen by electrolyzing water; if the state of charge of the battery is less than or equal to the set upper limit index, the photovoltaic excess power is preferentially used to charge the battery, and the photovoltaic dominant hydrogen production mode is not immediately entered.

[0015] Optionally, in an embodiment, the decision control layer is configured to judge the adaptive operation mode according to the following logic: When the photovoltaic power is less than or equal to the demand load data, it is further judged whether it belongs to the peak electricity price period, if it is in the peak electricity price period, the fuel cell grid-connected power supply mode is determined as the adaptive operation mode, the fuel cell is started to generate power to make up for the power shortage, otherwise, the external power grid is used to purchase power to make up for the power shortage.

[0016] Furthermore, in one embodiment, the decision control layer monitors different distribution ranges of the battery's state of charge in real time, and makes energy storage control commands according to the following logic to achieve hierarchical energy storage operation scheduling: When the battery charge drops to the emergency charging threshold, the system first determines whether the photovoltaic charging power is sufficient. If so, the battery is charged based on the surplus photovoltaic power. If not, the system further determines whether the fuel cell is effective and starts the fuel cell to supply power to the load and charge the battery. When the battery charge is within the normal floating threshold, the photovoltaic system and the grid provide priority power supply, and the battery smooths out short-term fluctuations. When the battery charge meets the hydrogen production start-up threshold, the electrolyzer is automatically started to convert excess electrical energy into hydrogen energy for storage.

[0017] In one embodiment, the decision control layer is further configured to set an importance level for each load and set different weight indicators for each level. Based on the weight indicators, the loads with higher importance are selected for priority power supply, while the power supply to other loads is selectively suspended.

[0018] Based on the application aspects of the system described in any one or more of the above embodiments, the present invention also provides a smart energy management method for a hydrogen-solar hybrid microgrid in a railway park, which is applied to the system described in any one or more of the above embodiments.

[0019] Based on other aspects of the methods described in the above embodiments, the present invention also provides a storage medium storing program code capable of implementing the methods described in the above embodiments.

[0020] Compared with the closest existing technology, the present invention has at least the following beneficial effects: This invention provides a smart energy management system and method for a hydrogen-solar hybrid microgrid in a railway park. The system includes a data sensing layer that collects dynamic operational data related to the photovoltaic subsystem, hydrogen energy subsystem, battery modules, and the power grid in real time; a decision control layer that generates optimized scheduling control commands based on the dynamic operational data according to a set decision control mechanism, including a day-ahead economic scheduling mechanism, a real-time rolling optimization mechanism, and a second-level dynamic adjustment mechanism; and a coordination and execution layer that responds to and precisely executes the optimized scheduling control commands to achieve dynamic power allocation and energy balance. This scheme proposes a multi-timescale hierarchical optimization framework based on the load characteristics of railway parks, combining day-ahead economic dispatch and real-time robust control; it designs a hydrogen-electric hybrid energy storage state machine and multi-modal switching logic to achieve intelligent dynamic energy allocation; and it introduces a rolling optimization algorithm based on model predictive control (MPC), with a built-in multi-timescale optimization algorithm to generate optimal power allocation instructions based on information from the data sensing layer, comprehensively coordinating the power allocation among photovoltaic, energy storage, electrolyzer, and fuel cell; achieving economically optimal operation under different electricity prices, weather, and load conditions; maintaining uninterrupted power supply to critical loads in the event of sudden power outages or grid disturbances; and fully leveraging the complementary characteristics of "hydrogen-photovoltaic-storage" through intelligent control algorithms to improve the overall operating efficiency and stability of the system.

[0021] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0022] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0023] Figure 1 This is a schematic diagram of the overall architecture of a smart energy management system for a hydrogen-solar hybrid microgrid in a railway park, provided by an embodiment of the present invention. Figure 2 This is a schematic diagram of the hierarchical decision control principle of the smart energy management system for a hydrogen-solar hybrid microgrid in a railway park, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating the optimized control effect of the smart energy management system for a hydrogen-solar hybrid microgrid in a railway park, as provided in this embodiment of the invention, compared to existing technologies. Figure 4 This is a schematic diagram of the hydrogen-electric hybrid energy storage state machine and multi-mode switching logic of the intelligent energy management system for a hydrogen-solar complementary microgrid in a railway park, provided in an embodiment of the present invention. Detailed Implementation

[0025] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples. Those skilled in the art will then fully understand how the present invention uses technical means to solve technical problems and achieve technical effects, and will be able to implement the present invention specifically based on the above-described implementation process. It should be noted that, as long as there is no conflict, the various embodiments and features of the present invention can be combined with each other, and the resulting technical solutions are all within the protection scope of the present invention.

[0026] Although the flowchart describes the operations as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. The order of the operations can be rearranged. A process can terminate when its operation is complete, but it may also have additional steps not included in the diagram. A process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0027] Computer equipment includes user equipment and network equipment. User equipment or clients include, but are not limited to, computers, smartphones, and PDAs (Personal Digital Assistants); network equipment includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Computer equipment can operate independently to implement this invention, or it can connect to a network and implement this invention through interaction with other computer equipment within the network. The network in which the computer equipment resides includes, but is not limited to, the Internet, wide area networks (WANs), metropolitan area networks (MANs), local area networks (LANs), and VPN networks.

[0028] The terms “first,” “second,” etc., may be used herein to describe various units, but these units should not be limited by these terms; they are used merely to distinguish one unit from another. The term “and / or” as used herein includes any and all combinations of one or more of the associated listed items. When a unit is referred to as “connected” or “coupled” to another unit, it may be directly connected or coupled to said other unit, or there may be intermediate units present.

[0029] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms “a” and “an” as used herein are also intended to include the plural. It should also be understood that the terms “comprising” and / or “including” as used herein specify the presence of the stated features, integers, steps, operations, units, and / or components, without excluding the presence or addition of one or more other features, integers, steps, operations, units, components, and / or combinations thereof.

[0030] Railway industrial parks, as typical high-energy-consuming entities, have load types including maintenance workshops, high-power machinery, rail transit dispatching equipment, office lighting, and railway signaling and communication systems. These park loads are characterized by strong time-varying characteristics, large peak-to-valley differences, and extremely high requirements for power supply reliability.

[0031] In recent years, with the advancement of low-carbon and optimized development goals, more and more railway parks have begun to build distributed photovoltaic systems to reduce electricity costs and achieve green and low-carbon operation. However, photovoltaic power generation is intermittent and subject to random fluctuations, and its output is significantly affected by weather conditions, with almost no power output during cloudy, rainy, snowy, or nighttime periods. If this volatility is not effectively regulated, it will adversely affect the voltage stability, frequency fluctuations, and power supply reliability of the railway park's power grid.

[0032] To address the uncertainty of renewable energy output, the introduction of hydrogen energy storage systems (including hydrogen electrolysis production devices, hydrogen storage tanks, and fuel cell power generation units) is considered by some researchers as an effective long-term energy storage and energy balance solution. Hydrogen is produced and stored through water electrolysis when needed, and then used to generate electricity through fuel cells, achieving an "electricity-hydrogen-electricity" energy cycle. This mechanism not only smooths out photovoltaic fluctuations but also enables cross-period energy utilization, improving the overall efficiency of the system.

[0033] However, after analysis, the researchers of this invention found that there are significant differences in dynamic response and coupling complexity between hydrogen energy systems and photovoltaics, energy storage batteries, and power grids in actual engineering: batteries have a fast response speed (seconds) and are suitable for short-term power balance; fuel cells have a slower response speed (minutes) and are suitable for continuous power output; the operation of electrolyzers requires consideration of start-up and shutdown costs and efficiency curves, and frequent start-ups and shutdowns will reduce lifespan.

[0034] Currently, most railway park microgrids still employ control strategies based on fixed thresholds or simple rules, making it difficult to simultaneously achieve economical operation, energy balance, and power supply reliability. Especially during periods of grid fluctuation, sudden load changes, or significant time-of-use pricing variations, existing systems often fail to achieve dynamic optimization scheduling.

[0035] Specifically, based on practical application, the following shortcomings of the existing technology can be identified: 1. The multi-energy coordinated dispatch capability is insufficient and it is difficult to adapt to the "hydrogen-photovoltaic-storage-grid" coupled system. Existing technologies have not established a systematic coordinated mechanism for hydrogen energy, photovoltaics, batteries and the power grid, which is mainly reflected in the following aspects.

[0036] 1) Ignoring differences in equipment response: Batteries (second-level response), fuel cells (minute-level response), and electrolyzers (frequent start-stop needs to be avoided) have different dynamic characteristics, but existing strategies mostly use uniform threshold control without targeted power allocation, resulting in a disconnect between the functions of "long-term energy storage (hydrogen energy) and short-term regulation (battery)". 2) Lack of energy cycle design: The cross-time characteristics of the "electricity-hydrogen-electricity" cycle are not fully utilized. The hydrogen energy and photovoltaic system are simply connected in parallel. It is impossible to convert the surplus photovoltaic power into hydrogen energy for storage, and it is also difficult to efficiently supplement the energy through fuel cells when the photovoltaic power is insufficient, resulting in energy waste.

[0037] 2. The control strategy is too simplistic and cannot take into account multiple objectives for optimization, which is mainly reflected in the following aspects.

[0038] Existing technologies mostly rely on fixed thresholds or simple rule control, which cannot balance the multiple needs of railway parks in terms of "economy, reliability and environmental protection". Lack of economy: The scheduling strategy is not combined with time-of-use electricity price signals, which makes it impossible to achieve the optimized coordination of electricity price for "valley electricity to produce hydrogen and peak electricity to generate electricity". It also does not consider the cost of equipment operation loss and energy storage life decay, resulting in high operating costs. Low reliability priority: No priority weights were set for "critical loads (train control, signaling system, dispatching communication)" in the railway park. In the event of sudden load changes or power grid disturbances, power supply to critical loads may be interrupted, which does not meet the high reliability requirements of railways.

[0039] 3. Incomplete time scale coverage and poor dynamic stability: Existing technologies lack multi-time scale coordinated control capabilities and cannot cope with photovoltaic fluctuations and load changes, mainly manifested in the following aspects.

[0040] Lack of long-term planning: The 24-hour operation plan was not formulated based on the day-ahead photovoltaic output and load forecast, making it impossible to optimize the start-up, shutdown and reference power of the electrolyzer and fuel cell in advance, resulting in chaotic power distribution during real-time operation; Insufficient short-term response: No second-level dynamic adjustment mechanism was designed. When photovoltaic power fluctuates drastically (such as the power change rate exceeding 20% ​​of rated power / minute within 2 minutes) or the bus voltage is abnormal, the traditional PID control alone cannot quickly smooth out the fluctuations. The bus voltage deviation often exceeds 5%, affecting power quality.

[0041] 4. Weak off-grid and on-grid switching capability and insufficient power supply continuity. Existing technologies cannot guarantee the continuous power supply of railway parks during grid failures or mode switching, mainly in the following aspects.

[0042] 1) Slow switching response: The on-grid / off-grid switching time is long, and it cannot quickly switch to off-grid mode when the grid fails. In addition, it lacks the temporary support design of "UPS + battery", which can easily lead to power outages of critical loads. 2) Simple switching logic: The coordinated power supply rules of "photovoltaic-battery-fuel cell" in the off-grid state have not been established. When running off-grid, the system may collapse due to loss of power balance.

[0043] 5. Poor adaptability and lack of design for railway park load characteristics. Existing technologies do not take into account the characteristics of railway park loads, such as "strong time-varying nature, large peak-valley difference, and high reliability requirements", which are mainly reflected in the following aspects.

[0044] 1) Lack of load adaptation: The system does not take into account the power consumption patterns of different types of loads such as railway park maintenance workshops, high-power machinery, and dispatching equipment, and does not adopt a general control strategy, which cannot cope with the problem of large load peak-valley differences; 2) Data Interaction Disconnection: The Energy Management System (EMS) has not established a communication interface with the railway integrated monitoring system, making it unable to obtain load switching plans. This results in a disconnect between dispatch instructions and actual load demand, further reducing power supply reliability and energy utilization efficiency.

[0045] Based on the optimization requirements of the existing technology, the purpose of this invention is to provide a smart energy management system with multi-timescale coordination capabilities, which can achieve the following improvement goals in the complex operating environment of railway parks.

[0046] (1) Comprehensively coordinate the power distribution among photovoltaic, energy storage, electrolyzer and fuel cell; (2) Achieve optimal economic operation under different electricity prices, weather and load conditions; (3) Maintain uninterrupted power supply to critical loads in the event of sudden power outages or grid disturbances; (4) By using intelligent control algorithms, the complementary characteristics of "hydrogen-photovoltaic-storage" can be fully utilized to improve the overall operating efficiency and stability of the system.

[0047] This invention proposes a smart energy management system for hydrogen-solar hybrid microgrids in railway parks, capable of achieving multiple optimization objectives and adapting to the specific needs of railway parks. The system comprises a data sensing layer, a decision control layer, and a coordination execution layer. It proposes a multi-timescale hierarchical optimization framework based on the load characteristics of railway parks, combining day-ahead economic dispatch and real-time robust control. A hydrogen-electric hybrid energy storage state machine and multi-mode switching logic are designed to achieve intelligent dynamic energy allocation. Furthermore, a rolling optimization algorithm based on model predictive control (MPC) is introduced to effectively address the challenges posed by the uncertainties of photovoltaic power generation.

[0048] The structural components, connection methods, and functional principles of the system according to embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Although the logical order of each operation is shown in the description of the system's structural operation, in some cases, the operations shown or described may be performed in a different order than that shown here.

[0049] Example 1 Figure 1 This diagram illustrates the structure of a smart energy management system for a hydrogen-solar hybrid microgrid in a railway park, as provided in Embodiment 1 of the present invention. (Refer to...) Figure 1 It can be seen that the system includes: The data sensing layer is configured to collect dynamic operating data related to the photovoltaic subsystem, hydrogen energy subsystem, battery module, and power grid in real time. The decision control layer, which is communicatively connected to the data perception layer, is configured to generate optimized scheduling control instructions based on the dynamic operating data according to a specific decision control mechanism. The decision control mechanism includes a day-ahead economic scheduling mechanism, a real-time rolling optimization mechanism, and a second-level dynamic adjustment mechanism. The coordination and execution layer, consisting of photovoltaic inverter controllers, energy storage converter controllers, electrolyzer controllers, and fuel cell controllers, is configured to respond to and precisely execute optimized scheduling control commands to achieve dynamic power allocation and energy balance.

[0050] This invention sets up a data sensing layer to collect dynamic operating data in real time through multiple types of sensors and communication networks. The dynamic operating data includes photovoltaic power, demand load power, real-time electricity price information, real-time energy storage status, and hydrogen energy operation data.

[0051] The data sensing layer uses a power sensor to obtain the photovoltaic power (P_pv) of the photovoltaic subsystem; in an optional embodiment, the power sensor is located at the photovoltaic array outlet of the photovoltaic subsystem.

[0052] The data perception layer is connected to the railway integrated monitoring system to obtain the demand load data (P_load) of the railway park and clarify the overall dynamic load demand of the park.

[0053] The data sensing layer also includes grid-connected energy meters, which are installed at the interface between the railway park microgrid and the external power grid. These meters are key monitoring nodes for system energy management and are mainly used to monitor the power supply status information of the external power grid in real time. The power supply status information includes energy flow direction, current, power, voltage, and frequency information.

[0054] Based on the monitored power grid power supply status information, it can determine in real time whether the park is supplying power to or drawing power from the power grid, thus providing core data for the decision-making and control layer; at the same time, it is also used for power grid fault identification, operation mode switching and economic dispatch optimization calculation, and is an important sensor for realizing intelligent switching between grid connection and off-grid.

[0055] The data sensing layer is connected to the battery management system (BMS) of the battery unit to obtain the state of charge (SOC_batt) of the energy storage battery in real time.

[0056] The data perception layer communicates and interacts with the coordination and execution layer to realize the acquisition of monitoring data and the transmission of control commands in real time.

[0057] In an optional embodiment, the data sensing layer is communicatively connected to the hydrogen system controller of the hydrogen energy subsystem to acquire hydrogen energy operation data, including hydrogen storage tank pressure, hydrogen storage status (Fill%), hydrogen fuel cell status, and electrolyzer status data.

[0058] The hydrogen energy subsystem includes an electrolyzer module, a fuel cell module, and a hydrogen storage system; while the hydrogen system controller is the upper-level management controller of the entire hydrogen energy subsystem, used for monitoring and sensing; for example, it is used to summarize various operational data of the electrolyzer, fuel cell, and hydrogen storage system.

[0059] Accordingly, each of the electrolyzer module, fuel cell module, and hydrogen storage system is equipped with a controller to control the electrolyzer module, fuel cell module, and hydrogen storage system to perform functional actions according to required instructions. For example, the electrolyzer controller is an equipment-level control unit used for execution and control, and is responsible for the specific execution of power regulation and protection of the electrolyzer.

[0060] In this embodiment of the invention, the decision control layer serves as the core control unit of the system, and constructs a multimodal operation logic in conjunction with railway load characteristics. It adopts a hierarchical collaborative optimization control strategy of "day-ahead economic dispatch + real-time rolling optimization + second-level dynamic adjustment", the core of which is the energy management controller (EMS), which has a built-in multi-time-scale optimization algorithm to generate the optimal power allocation command based on the information from the data perception layer.

[0061] The decision control layer generates optimized scheduling control instructions based on the data obtained by the data perception layer according to a specific decision control mechanism, which includes a day-ahead economic scheduling mechanism, a real-time rolling optimization mechanism, and a second-level dynamic adjustment mechanism.

[0062] The decision control layer is configured to generate a daily optimized scheduling operation plan based on the day-ahead economic scheduling mechanism, as reference information for optimized scheduling control, according to the following logic: The decision-making and control layer, based on high-precision photovoltaic output and load forecast data, prioritizes optimal economic efficiency throughout the day. It employs a mixed-integer linear programming (MILP) algorithm to formulate a 24-hour operational plan, serving as reference information for daily optimized scheduling instructions. This plan is then used as the basis for power allocation boundary conditions and operational modes in subsequent real-time rolling optimization (MPC) and second-level dynamic adjustment. This embodiment of the invention generates a macro-level energy flow plan for the park over the next 24 hours using the MILP algorithm. The result is a strategic allocation of energy levels such as hydrogen production, electricity purchase, and planned fuel cell power generation, serving as reference information for optimized scheduling and control.

[0063] In practical applications, during the process of generating optimized scheduling control instructions based on the day-ahead economic scheduling mechanism, at 0:00 every day, the decision control layer calls the weather forecast and the park's production plan, runs the MILP algorithm based on the day-ahead scheduling mechanism, and generates the plan curve for the next 24 hours.

[0064] In the current economic dispatch mechanism, high-precision photovoltaic power output forecasts are generated by combining real-time photovoltaic power data with weather forecasts (irradiance, cloud cover, temperature, etc.) through a short-term photovoltaic forecasting model.

[0065] The load forecast data is obtained based on a short-term load forecast model constructed from historical load records of the railway park, the next day's production plan, and time characteristics.

[0066] In practical applications, high-precision photovoltaic power output forecast data and load forecast data are used as input variables for MILP optimization to construct the power balance constraint for the next 24 hours. The production plan is a hard target that must be achieved. Combined with economic indicators such as time-of-use electricity price, electricity purchase cost, and operating costs of electrolyzers and fuel cells, an objective function is formed. By solving the problem, the total operating cost of the system is minimized.

[0067] In this embodiment of the invention, the specific start-up and shutdown status of the electrolyzer and fuel cell, and the real-time reference power point are determined by the EMS of the decision control layer based on multimodal operation logic, daily optimized scheduling operation plan, MPC rolling optimization mechanism, and the device's own controller combined with dynamic SOC, grid status, actual photovoltaic output and other data during operation.

[0068] When the decision control layer triggers the fuel cell operating mode (including grid-connected power supply mode or off-grid islanded mode), it comprehensively determines the reference power point of the fuel cell based on the current load power gap, time-of-use electricity price, battery state of charge, and the remaining hydrogen in the hydrogen storage tank.

[0069] Preferably, the calculation process of the fuel cell reference power includes: first, determining the theoretical power demand based on the difference between the load power and photovoltaic output, the battery's allowable discharge power, and the grid's allowable power purchase power; then, weighting and correcting the theoretical power demand according to the electricity price level and the fuel cell efficiency range; and finally, smoothly adjusting it under the constraints of the fuel cell's allowable power operating range and ramp-up rate to avoid large power fluctuations, thereby balancing system economy and fuel cell operating life.

[0070] The decision control layer is configured to generate optimized scheduling control instructions based on a real-time rolling optimization mechanism according to the following logic: The model predictive control (MPC) framework is used to analyze the operational deviation at the target time step according to the set time period. Based on the operational deviation, the power allocation information for the future time step is continuously optimized and configured. Combined with the daily optimized scheduling operation plan, configuration instructions for the photovoltaic subsystem, hydrogen energy subsystem, battery module and power grid corresponding to the matching operation mode are generated.

[0071] The real-time rolling optimization mechanism (MPC) is mainly used to perform power coordination optimization of key controllable units in the "photovoltaic-energy storage-hydrogen energy" microgrid system. Its core control objects include devices that can quantify and adjust the output, such as batteries (energy storage PCS), fuel cells, and electrolyzers. MPC predicts future power deviations in a rolling manner and corrects the target power of each unit in real time, determining the start-up and shutdown status and reference power point of electrolyzers and fuel cells at each time period to maintain bus stability and overall energy balance.

[0072] First, ultra-short-term forecasting methods are used to obtain forecast data for the target time step. In practical applications, this refers to time-series forecasts for photovoltaic power output, park load, battery SOC evolution, and power balance / bus voltage trends for multiple future time steps. The obtained forecast data is used as input to the MPC control model in the form of a multi-step time series to calculate the optimal power allocation for future time periods.

[0073] Furthermore, by combining the actual operating data at the current target time step, the deviation between the actual operating data and the predicted data is analyzed; Based on operational deviations, the power allocation information for future time steps is continuously optimized and configured, such as... Figure 2 As shown.

[0074] In the real-time rolling optimization mechanism of this invention, both the predicted data and the actual operating data are time-series data that change over time. Their "deviation" is not characterized by a similarity index, but rather by a time-step quantized error. Specifically, the system calculates error values ​​in each optimization cycle in the following forms: power deviation between predicted and measured photovoltaic power, load deviation between predicted and actual load, and voltage deviation between predicted and measured bus voltage. These deviations form an error sequence by subtracting the actual value from the predicted value, thus constituting a time-series vector reflecting the system's prediction error state. This error sequence is directly input into the Model Predictive Control (MPC) framework and used as the basis for power allocation optimization in the next time period. Through this time-step error approach, the system can accurately identify deviations in the prediction model and correct scheduling commands in subsequent optimization steps, enabling the control strategy to have real-time adaptive capabilities.

[0075] In the process of rolling optimization of power allocation information for future time steps based on operational deviations, MPC re-solves the optimal power allocation for several future time steps (such as four sub-periods within one hour) based on the planned power set in the previous cycle and the currently detected error sequence.

[0076] The correction logic is as follows: when there is a deviation between the prediction and the actual situation, the optimization algorithm automatically adjusts the target power of each unit by minimizing objective functions such as power error, voltage deviation and power change rate, so that the output of the battery, fuel cell and electrolyzer is corrected in the direction of compensating for the deviation in the next cycle, thereby achieving stable support for the bus voltage and early response to power fluctuations.

[0077] The optimal power commands (such as battery charging and discharging power, electrolyzer operating power, fuel cell power generation power, etc.) corrected by the real-time rolling optimization mechanism are issued to the controllers of each device in the coordination execution layer for execution, so as to realize dynamic optimization scheduling with continuous rolling and self-updating.

[0078] The issued instructions are generated in real time by the EMS of the decision control layer, including immediate instructions such as power setpoints, start / stop commands and operating modes, and are transmitted to each unit controller for execution via communication links (such as Modbus-TCP, CAN).

[0079] In practical applications, the decision control layer operates on a 15-minute cycle, based on the Model Predictive Control (MPC) framework. It continuously optimizes the power allocation for several future time steps based on the deviation between ultra-short-term forecasts and actual operating conditions, with the main objectives of smoothing power fluctuations and maintaining bus stability.

[0080] Every 15 minutes, the real-time rolling optimization layer is activated. Based on the latest measured data, it optimizes the next hour (four time steps) and calculates more accurate power commands to be sent to each unit. Comparison data of the control performance of the Model Predictive Control (MPC) framework is referenced. Figure 3 .

[0081] For example, simulating a scenario where a cloud layer rapidly passes by, the photovoltaic power drops sharply from 1.5MW to 0.7MW within 2 minutes. Using traditional PID control, the bus voltage experiences a significant drop (>5%). However, using the MPC control of this invention, the algorithm predicts the power change trend in advance, instructing the fuel cell to increase output before the power decreases and instructing the battery to provide instantaneous support, thus suppressing the microgrid bus voltage fluctuation within ±3%.

[0082] The decision control layer generates optimized scheduling control instructions based on a second-level dynamic adjustment mechanism according to the following logic: By combining the instantaneous changes in DC bus voltage and the state of charge of the energy storage battery, a battery charging and discharging power control command is generated to quickly adjust the charging and discharging power of the battery, making up for the slow response of the fuel cell and maintaining real-time power balance.

[0083] The second-level scheduling mechanism operates continuously. Once it detects that the bus voltage fluctuation exceeds the threshold, it immediately adjusts the output of the energy storage PCS to provide support.

[0084] The researchers of this invention recognized that existing technologies typically employ simple PID or fixed threshold logic, failing to distinguish between fast-responding devices (batteries) and slow-responding devices (fuel cells, electrolyzers, and power grids), resulting in an inability to quickly stabilize the bus voltage during sudden photovoltaic fluctuations. This invention, however, uses a multi-timescale control architecture to explicitly delegate the "second-level" control to the battery, leveraging its second-level dynamic response capability to support the bus, thereby achieving a faster and more stable regulation effect than existing technologies.

[0085] Therefore, the rapid regulation of this invention does indeed prioritize the rapid charging and discharging of the battery, rather than simultaneously mobilizing photovoltaic, grid, or fuel cell energy sources—which are adjusted by the upper layer (15 minutes / MPC or day-ahead scheduling) on ​​a slower timescale.

[0086] In a preferred embodiment, the researchers of this invention considered that the battery charging and discharging power control command is adaptively controlled according to the instantaneous changes in the state of charge of the energy storage battery and the DC bus voltage under any circumstances, which can ensure the mobility and effectiveness of battery control.

[0087] Therefore, in the process of generating optimized scheduling control commands based on the second-level dynamic adjustment mechanism, the present invention does not adopt the logic of "immediately calling the battery as long as there is a voltage deviation". Instead, it first checks whether the battery SOC is in the allowable fast discharge / charge range, and then decides the adjustment range and direction.

[0088] For example, if the SOC is close to the upper or lower limit, it will limit the instantaneous power regulation capability of the battery, and the upper-level scheduling (fuel cells, grid power purchase, etc.) will compensate for the energy on a slower time scale.

[0089] Therefore, the second-level dynamic adjustment mechanism of this invention uses the bus voltage deviation as the trigger condition and the battery SOC as the safety constraint to achieve fast and sustainable power stability control.

[0090] In the railway park microgrid of this invention, there are multiple power supply terminals that can effectively supply power to the load. In practical applications, the priority order for power supply to the load is set as follows: photovoltaic power supply, grid power supply, fuel cell power supply, and battery discharge power supply. Among them, since the battery capacity is limited and the purpose is only to optimize regulation rather than to stabilize power supply or consumption, the power supply priority is usually photovoltaic power supply first, and then grid power supply and fuel cell power supply are supplemented by considering cost (electricity price cost).

[0091] The system described in this invention supports multiple operating modes and can dynamically and quickly switch between them through the decision control layer. The decision control layer is also configured to determine the appropriate operating mode in real time based on the dynamic operating data obtained by the data perception layer and generate mode execution instructions.

[0092] The operating modes of the system in this embodiment include at least photovoltaic-dominated hydrogen production mode, fuel cell grid-connected power supply mode, off-grid islanded operation mode, off-peak electricity hydrogen production mode, and peak electricity power generation mode.

[0093] The decision control layer determines the appropriate operating mode based on the following logic: When the photovoltaic power P_pv > the demand load data P_load, the state of charge (SOC) of the battery (energy storage battery) is further determined. If the SOC of the battery (energy storage battery) > the set upper limit (e.g., 80%), The photovoltaic-dominated hydrogen production mode is determined to be the suitable operating mode. The electrolyzer module is started, and the surplus photovoltaic power is used to power the electrolyzer module, and the excess photovoltaic power is used for water electrolysis to produce hydrogen. When there is sufficient sunlight and surplus electricity, water electrolysis to produce hydrogen can be effectively realized.

[0094] When the photovoltaic power is greater than the load and the battery SOC is lower than the upper limit, priority should be given to charging. That is, when P_pv>P_load, but the battery SOC has not yet reached the hydrogen production start-up threshold (e.g., 80%), the system should prioritize using the surplus photovoltaic power to charge the battery instead of immediately entering the photovoltaic-dominated hydrogen production mode.

[0095] Based on the above logic, the photovoltaic surplus will only be used for hydrogen production when the battery charge is already in the high SOC range and there is a surplus of energy storage. This will better maintain the stability of energy storage and the regulation capability of the microgrid.

[0096] Specifically, when the system determines whether to enter the photovoltaic-dominated hydrogen production mode, it should simultaneously check the remaining hydrogen storage space of the hydrogen storage tank. In an optional embodiment, it can determine whether the preset hydrogen storage safety threshold has been reached based on the hydrogen storage tank pressure or the hydrogen storage status Fill% to avoid an unsafe or ineffective operating state where "the hydrogen tank is full but the electrolyzer continues to produce hydrogen".

[0097] For example, if monitoring and analysis determine that a certain parameter in the hydrogen storage tank pressure or hydrogen storage status (Fill%) has reached the corresponding safety threshold, then the hydrogen storage status is determined to be at the hydrogen storage safety threshold. At this point, even if the photovoltaic-dominated hydrogen production mode is met, the hydrogen production operation will not be further switched on.

[0098] When the photovoltaic power P_pv < the demand load data P_load, it is further determined whether the current period is a peak electricity price period. If it is a peak electricity price period, the fuel cell grid-connected power supply mode is determined to be the adaptive operation mode, and the fuel cell power generation is started to make up for the power deficit. Otherwise, the power deficit is made up by purchasing electricity from the external grid. The electricity price period is divided into peak electricity price period, parity electricity price period, and off-peak electricity price period according to the local electricity price distribution time.

[0099] For example, at noon on a certain day, the photovoltaic power generation is high (P_pv = 2.5MW), while the load is relatively light (P_load = 1.2MW). It is determined that the photovoltaic power P_pv >= the demand load data P_load. Further analysis reveals that the battery SOC has reached 85%, and the battery (energy storage battery) SOC exceeds the set upper limit (e.g., 80%). The EMS determines that the conditions for "photovoltaic-dominated hydrogen production mode" are met. If the analysis determines that the hydrogen storage state has not reached the hydrogen storage safety threshold, the electrolyzer is automatically started, and 1.0MW of excess power from the photovoltaic subsystem is used for hydrogen production. Figure 4 As shown.

[0100] When a grid failure occurs, the off-grid islanding mode is determined to be the appropriate operating mode, and the system switches to off-grid mode, with photovoltaic, storage battery and fuel cell working together to ensure power supply to critical loads.

[0101] In a preferred embodiment, the decision control layer dynamically determines whether the power grid is in a fault state or a normal state based on the power grid power supply status information, according to the following logic: If the voltage drop or frequency deviation exceeds the set threshold and persists for a certain period of time, it can be determined that the external power grid has malfunctioned or is abnormal. Based on the direction of current and power detected by the electricity meter at the grid connection point, if the power drops to zero instantly, fluctuates violently, or has an abnormal direction, a grid instability fault is confirmed.

[0102] In practical applications, a continuous interruption of the synchronization signal with the upper-level dispatching or grid-connected protection equipment can also serve as an auxiliary criterion. By combining these monitoring signals, the data sensing layer can identify grid faults in a very short time and send the trigger conditions for off-grid islanding switching to the EMS, enabling the microgrid to ensure uninterrupted operation of critical loads.

[0103] In the microgrid system described in this invention, grid faults are determined by the decision control layer through the comprehensive abnormal characteristics of the electricity meter and voltage / frequency monitoring signals at the grid connection point. When any of the following situations occurs at the grid connection point: voltage drop or disappearance, frequency deviation exceeding the limit, power sudden change to zero or abnormal direction, especially when the grid connection point voltage is continuously lower than the threshold (e.g., 0.9 pu), the grid frequency deviates from the rated value, or the electricity meter data is interrupted, the EMS identifies it as an external grid fault and immediately triggers system disconnection, switching to off-grid islanding operation mode. In various modes, photovoltaic, storage battery and fuel cell work together to ensure power supply to critical loads.

[0104] During the continuous process, the changes in photovoltaic power and the state of charge of the battery are monitored in real time. When the photovoltaic power P_pv is greater than the demand load data P_load, but the state of charge of the battery is lower than the set upper limit, the excess photovoltaic power is used to charge the battery first. When the photovoltaic power P_pv is less than or equal to the demand load data P_load, the electrolyzer hydrogen production is stopped. Further determine whether it is in off-grid operation mode. If not, further determine whether it is during peak electricity price period. If it is not during peak electricity price period, supplement the photovoltaic power difference based on grid power supply.

[0105] During peak electricity price periods, fuel cells are used to supplement the photovoltaic power deficit. If the photovoltaic power drops to zero in the evening, it is determined that the peak electricity price period has begun, and the EMS automatically switches to the "fuel cell and microgrid grid-connected power supply mode" and starts the fuel cell to operate at 0.8MW.

[0106] During off-peak electricity prices at night, power is supplied by both fuel cells and the grid. During periods of high electricity prices, fuel cells are used to supply the entire power supply as much as possible, with the grid supplying the power beyond that capacity. During periods of low electricity prices, the grid supplies the entire power supply. After economic and parameter assessments, it is decided whether to consume grid electricity to power the water electrolysis.

[0107] In addition, during off-peak electricity price periods at night, the off-peak electricity hydrogen production mode is determined as the adaptive operation mode, and hydrogen production is carried out by purchasing electricity from the grid and operating the electrolyzer under the premise that the hydrogen storage status has not reached the hydrogen storage safety threshold. During peak electricity price periods, the peak power generation mode is determined as the appropriate operating mode; for example, fuel cell power generation can be used during peak daytime electricity price periods to achieve arbitrage.

[0108] Real-time monitoring of different distribution ranges of battery state of charge, and decision-making of energy storage control commands according to the following logic to achieve hierarchical energy storage operation scheduling: When the battery charge SOC_batt drops to the emergency charging threshold, the system first checks whether the photovoltaic charging power is sufficient. If so, the battery is charged based on the surplus photovoltaic power. If not, the system further checks whether the fuel cell is effective and starts the fuel cell to supply power to the load and charge the battery.

[0109] When the battery's SOC_batt is within the normal floating threshold, the photovoltaic system and the power grid prioritize power supply to the microgrid load, and the battery smooths out short-term fluctuations. During the process, the battery is charged through surplus photovoltaic power, fuel cells, or the power grid according to the actual situation.

[0110] When the battery's SOC_batt meets the hydrogen production start-up threshold, the electrolyzer is automatically started to convert excess electrical energy into hydrogen energy for storage.

[0111] In practical applications, three levels of battery charge index are set according to requirements, corresponding to emergency charging, normal floating and hydrogen production start-up requirements respectively. In an optional embodiment, three key thresholds of battery SOC are set, including the emergency charging threshold of 20%, the normal floating range of 20%-80%, and the hydrogen production start-up threshold of 80%.

[0112] The system, designed for the load characteristics of railway parks, adds load priority weights to the optimization objectives to prioritize ensuring the power supply reliability of critical loads.

[0113] The decision control layer is also configured to set importance levels for each load and set different weight indicators for each level. For example, in the off-grid state, if the power supply cannot fully meet the load requirements, the load with higher importance is selected to be given priority power supply according to the weight indicators from high to low, and the power supply to other loads is selectively suspended.

[0114] The system, designed for the load characteristics of railway parks, adds load priority weights to the optimization objectives. For example, key loads such as train control, signaling systems, and dispatch communication can be set as the highest-weighted loads.

[0115] The Energy Management Controller (EMS) establishes a communication interface with the railway integrated monitoring system to receive load switching plans.

[0116] The load switching plan refers to the load start-up and shutdown arrangements issued by the railway integrated monitoring system, including information such as the planned loads to be put online, the planned loads to be taken offline, the rated power and criticality level of each load.

[0117] The decision control layer is also configured to update the overall load list of the park in a timely manner based on the load switching plan data, and to dynamically update the priority and weight parameters of the load.

[0118] After obtaining the load switching plan, the decision control layer will reconstruct the classification results of critical and non-critical loads and input the updated load data into the day-ahead scheduling and rolling optimization algorithm, so that the power allocation strategy can adapt to the actual operation requirements in real time.

[0119] The microgrid in this embodiment of the invention is based on optimized scheduling control commands generated by the decision control layer, which are then executed by the coordination execution layer, thereby implementing the optimized scheduling mechanism of the railway park microgrid. The coordination execution layer consists of a photovoltaic inverter controller, an energy storage converter (PCS) controller, an electrolyzer controller, and a fuel cell controller, configured to respond to and precisely execute optimized scheduling control commands to achieve dynamic power allocation and energy balance.

[0120] The photovoltaic inverter controller and the energy storage converter (PCS) controller receive and execute the optimized scheduling control instructions for the photovoltaic subsystem and battery unit from the decision control layer, so as to realize the refined coordinated control of the photovoltaic subsystem and battery unit.

[0121] In an optional embodiment, the electrolyzer controller is an electrolyzer DC / DC controller; the fuel cell controller is a fuel cell DC / DC controller; both are used to receive and execute optimized scheduling control commands matched by the decision control layer, so as to realize fine-grained coordinated control of the power of the electrolyzer unit and the fuel cell unit.

[0122] The intelligent energy management system for hydrogen-solar hybrid microgrids in railway parks provided by this invention collects real-time system operation data and future prediction information; based on a multi-objective optimization algorithm, it obtains the optimal power allocation instruction set, with the objective function simultaneously considering economy, environmental protection, and reliability to ensure the rationality and efficiency of energy dispatch; the power instructions are decomposed and issued to each unit controller for execution, and the control instructions are corrected in real time through closed-loop feedback to ensure the stable operation of the system under varying loads and energy supply conditions.

[0123] In real-time control, the method introduces DC-side virtual capacitor / voltage droop and damping control, and the energy storage PCS and fuel cell DC / DC work together to achieve rapid support and buffering of bus voltage, thereby improving the DC-side voltage stability of the microgrid.

[0124] The system and method described in this invention have been demonstrated at a railway hub station, achieving significant economic and technical benefits and showing promise for large-scale application.

[0125] The intelligent energy management system for hydrogen-solar hybrid microgrids in railway parks provided by this invention has at least the following advantages when put into application.

[0126] 1. High economic efficiency: By optimizing time-of-use electricity pricing and leveraging the time-shifting characteristics of hydrogen storage, an economical operation mode of "hydrogen production from off-peak electricity and power generation from peak electricity" is achieved. Simulation verification shows that the system can reduce the park's annual comprehensive electricity costs by approximately 15% to 22% (depending on local electricity price differences and photovoltaic power output conditions).

[0127] 2. High Reliability: Layered control and multi-modal coordinated switching ensure that the system can complete rapid on-grid / off-grid switching within 2 seconds under different operating conditions (including grid disturbances, sudden weather changes, etc.). Critical loads are connected to the power supply bus of the grid-type energy storage GFM / uninterruptible power supply UPS to achieve uninterrupted operation; non-critical loads are allowed short-term voltage jumps during switching.

[0128] 3. High stability: Based on the rolling optimization of the MPC algorithm and the fast response of the battery, the DC bus voltage fluctuation of the system is stabilized within ±3%, effectively suppressing the impact of photovoltaic fluctuations on power quality.

[0129] 4. Intelligent Adaptability: The system can autonomously adjust its control strategy based on real-time operating data, achieving intelligent energy management with less human intervention and stronger autonomous decision-making, thereby improving the energy utilization rate and operational safety of the railway park.

[0130] The present invention will be further described below with reference to specific embodiments. The scope of the present invention is not limited to the embodiments, but is defined in the claims.

[0131] Comparative example: a traditional fixed threshold controlled hydrogen-photonic microgrid system for railway parks.

[0132] 1. System configuration: Energy unit: 1.5MW photovoltaic array + 500kWh battery + 500Nm³ / h electrolyzer + 500kW fuel cell.

[0133] (1). Control strategy: "Simple threshold control" is adopted, without multi-timescale optimization; When photovoltaic output > load + battery SOC ≥ 80%, 50% of surplus power is fixedly allocated for hydrogen production (no dynamic adjustment); in the event of a grid fault, non-critical loads are directly disconnected, and power is supplied solely through the battery (without fuel cell collaboration); it is not combined with time-of-use pricing (peak and valley pricing is uniformly calculated at 0.5 yuan / kWh, without the logic of "valley electricity for hydrogen production and peak electricity for power generation"); (2) Load adaptation: No priority is given to critical / non-critical loads, and all loads are connected to the same bus.

[0134] 2. Core operating data (simulating typical operating conditions in a railway park: average daily photovoltaic output of 8MWh, critical load of 2MWh / day, non-critical load of 3MWh / day, peak electricity price of RMB 1.2 / kWh, and off-peak electricity price of RMB 0.3 / kWh), and the evaluation index data for comparative application are shown in the table below.

[0135]

[0136] Implementation Case: Reference Figure 1 In this implementation, the data sensing layer includes a power sensor at the photovoltaic array outlet, a grid-connected energy meter, a battery management system (BMS), a hydrogen system controller (providing hydrogen storage pressure and fuel cell / electrolyzer status), and load information obtained from the railway integrated monitoring system. The decision control layer consists of an industrial control computer (equipped with the algorithm software of this invention) acting as an energy management controller (EMS). The coordination and execution layer comprises the local controllers of each unit (photovoltaic inverter, energy storage PCS, electrolyzer DC / DC converter, and fuel cell DC / DC converter), which communicate with the EMS via MODBUS-TCP or CAN bus.

[0137] In the intelligent energy management system for hydrogen-solar complementary microgrids in railway parks provided by the embodiments of the present invention, each module or unit structure can operate independently or in combination according to actual parameter sensing needs and dynamic decision-making sequence to achieve the corresponding technical effects.

[0138] Example 2 The above-described embodiments of the present invention have provided a detailed description of the system. Based on other aspects of the system described in any one or more of the above embodiments, the present invention also provides a smart energy management method for a hydrogen-solar complementary microgrid in a railway park. This method is applied to the smart energy management system for a hydrogen-solar complementary microgrid in a railway park described in any one or more of the above embodiments. Specific embodiments are given below for detailed description.

[0139] Specifically, the smart energy management method for hydrogen-solar hybrid microgrids in railway parks provided in this embodiment of the invention includes: The data sensing layer collects dynamic operating data related to the photovoltaic subsystem, hydrogen energy subsystem, battery module, and power grid in real time. The decision control layer generates optimized scheduling control instructions based on the dynamic operating data according to a specific decision control mechanism, which includes a day-ahead economic scheduling mechanism, a real-time rolling optimization mechanism, and a second-level dynamic adjustment mechanism. The coordinated execution layer responds to and precisely executes the optimized scheduling control commands to achieve dynamic power allocation and energy balance. The coordinated execution layer includes a photovoltaic inverter controller, an energy storage converter controller, an electrolyzer controller, a fuel cell controller, and a hydrogen storage controller.

[0140] The smart energy management method for hydrogen-solar hybrid microgrids in railway parks is applied to the smart energy management system for hydrogen-solar hybrid microgrids in railway parks described in this embodiment of the invention. The system includes: The data sensing layer is configured to collect dynamic operating data related to the photovoltaic subsystem, hydrogen energy subsystem, battery module, and power grid in real time. The decision control layer, which is communicatively connected to the data perception layer, is configured to generate optimized scheduling control instructions based on the dynamic operating data according to a specific decision control mechanism. The decision control mechanism includes a day-ahead economic scheduling mechanism, a real-time rolling optimization mechanism, and a second-level dynamic adjustment mechanism. The coordination and execution layer, including photovoltaic inverter controllers, energy storage converter controllers, electrolyzer controllers, fuel cell controllers, and hydrogen storage controllers, is configured to respond to and precisely execute optimized scheduling control commands to achieve dynamic power allocation and energy balance.

[0141] In one embodiment, the data sensing layer collects dynamic operating data in real time through multiple types of sensors and communication networks. The dynamic operating data includes photovoltaic power, demand load power, grid power supply status information, real-time energy storage status, and hydrogen energy operation data.

[0142] Furthermore, in one embodiment, the data sensing layer is communicatively connected to the hydrogen system controller of the hydrogen energy subsystem to acquire hydrogen energy operation data, including hydrogen storage tank pressure, hydrogen storage status, hydrogen fuel cell status, and electrolyzer status data.

[0143] Preferably, in one embodiment, the decision control layer is configured to generate a daily optimized scheduling operation plan according to the following logic based on the day-ahead economic scheduling mechanism: The decision control layer, based on high-precision photovoltaic power output and load forecast data, prioritizes optimal economic efficiency throughout the day and employs a mixed-integer linear programming algorithm to formulate a 24-hour cycle operation plan as reference information for optimized scheduling and control.

[0144] Optionally, in one embodiment, the decision control layer is configured to generate optimized scheduling control instructions according to a real-time rolling optimization mechanism based on the following logic: The model predictive control framework is used to analyze the operational deviation at the target time step according to the set time period. Based on the operational deviation, the power allocation information for the future time step is continuously optimized and configured. Combined with the daily optimized scheduling operation plan, configuration instructions are generated for the photovoltaic subsystem, hydrogen energy subsystem, battery module and grid corresponding to the matching operation mode.

[0145] Furthermore, in one embodiment, the decision control layer generates optimized scheduling control instructions based on a second-level dynamic adjustment mechanism according to the following logic: By combining the instantaneous changes in DC bus voltage and the state of charge of the energy storage battery, a battery charging and discharging power control command is generated to quickly adjust the charging and discharging power of the battery, making up for the slow response of the fuel cell and maintaining real-time power balance.

[0146] In one embodiment, the decision control layer is further configured to determine the appropriate operating mode in real time based on the dynamic operating data obtained by the data perception layer, and generate mode execution instructions; the operating modes include photovoltaic-dominated hydrogen production mode, fuel cell grid-connected power supply mode, off-grid islanded operation mode, valley electricity hydrogen production mode, and peak electricity power generation mode.

[0147] Optionally, in one embodiment, the decision control layer is configured to determine the appropriate operating mode according to the following logic: When the photovoltaic power exceeds the demand load data, the battery state of charge is further assessed. If the battery state of charge exceeds the set upper limit, the photovoltaic-dominated hydrogen production mode is determined to be the appropriate operating mode. The electrolyzer module is then started, and the surplus photovoltaic power is used to power the electrolyzer module, with the excess photovoltaic power used for water electrolysis to produce hydrogen. If the battery state of charge is less than or equal to the set upper limit, the surplus photovoltaic power is used to charge the battery first, and the photovoltaic-dominated hydrogen production mode is not immediately entered.

[0148] In an optional embodiment, the decision control layer is configured to determine the appropriate operating mode according to the following logic: When the photovoltaic power is less than or equal to the demand load data, further determine whether the current period is a peak electricity price period. If it is a peak electricity price period, determine that the fuel cell grid-connected power supply mode is the appropriate operating mode and start the fuel cell power generation to make up for the power deficit. Otherwise, use external grid power purchase to make up for the power deficit.

[0149] Furthermore, in one embodiment, the decision control layer monitors different distribution ranges of the battery's state of charge in real time, and makes energy storage control commands according to the following logic to achieve hierarchical energy storage operation scheduling: When the battery charge drops to the emergency charging threshold, the system first determines whether the photovoltaic charging power is sufficient. If so, the battery is charged based on the surplus photovoltaic power. If not, the system further determines whether the fuel cell is effective and starts the fuel cell to supply power to the load and charge the battery. When the battery charge is within the normal floating threshold, the photovoltaic system and the grid provide priority power supply, and the battery smooths out short-term fluctuations. When the battery charge meets the hydrogen production start-up threshold, the electrolyzer is automatically started to convert excess electrical energy into hydrogen energy for storage.

[0150] In one embodiment, the decision control layer is further configured to set an importance level for each load and set different weight indicators for each level. Based on the weight indicators, the loads with higher importance are selected for priority power supply, while the power supply to other loads is selectively suspended.

[0151] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0152] It should be noted that, in other embodiments of the present invention, the method can also be combined with one or more of the above embodiments to obtain a new smart energy management method for hydrogen-solar complementary microgrids in railway parks, so as to achieve efficient and economical dispatch of microgrids in railway parks.

[0153] Example 3 It should be noted that, based on the methods in any one or more embodiments of the present invention described above, the present invention also provides a storage medium storing program code that can implement the methods described in any one or more embodiments. When the program code is executed by the operating system, it can implement the smart energy management method for hydrogen-solar complementary microgrids in railway parks as described above.

[0154] It should be understood that the embodiments disclosed herein are not limited to the specific structures, processing steps, or materials disclosed herein, but should be extended to equivalent substitutions of these features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0155] The phrase "an embodiment" in the specification means that a specific feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0156] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A smart energy management system for a hydrogen-solar hybrid microgrid in a railway park, characterized in that, The system includes: The data sensing layer is configured to collect dynamic operating data related to the photovoltaic subsystem, hydrogen energy subsystem, battery module, and power grid in real time. The decision control layer, which is communicatively connected to the data perception layer, is configured to generate optimized scheduling control instructions based on the dynamic operating data according to a specific decision control mechanism. The decision control mechanism includes a day-ahead economic scheduling mechanism, a real-time rolling optimization mechanism, and a second-level dynamic adjustment mechanism. The coordination and execution layer, including photovoltaic inverter controllers, energy storage converter controllers, electrolyzer controllers, fuel cell controllers, and hydrogen storage controllers, is configured to respond to and precisely execute optimized scheduling control commands to achieve dynamic power allocation and energy balance.

2. The system according to claim 1, characterized in that, The data perception layer collects dynamic operation data in real time through multiple types of sensors and communication networks. The dynamic operation data includes photovoltaic power, demand load power, grid power supply status information, real-time energy storage status, and hydrogen energy operation data.

3. The system according to claim 1, characterized in that, The data sensing layer is connected to the hydrogen system controller of the hydrogen energy subsystem to acquire hydrogen energy operation data, including hydrogen storage tank pressure, hydrogen storage status, hydrogen fuel cell status, and electrolyzer status data.

4. The system according to claim 1, characterized in that, The decision control layer is configured to generate a daily optimized scheduling operation plan based on the day-ahead economic scheduling mechanism according to the following logic: The decision control layer, based on high-precision photovoltaic power output and load forecast data, prioritizes optimal economic efficiency throughout the day and employs a mixed-integer linear programming algorithm to formulate a 24-hour cycle operation plan as reference information for optimized scheduling and control.

5. The system according to claim 1, characterized in that, The decision control layer is configured to generate optimized scheduling control instructions based on a real-time rolling optimization mechanism according to the following logic: The model predictive control framework is used to analyze the operational deviation at the target time step according to the set time period. Based on the operational deviation, the power allocation information for the future time step is continuously optimized and configured. Combined with the daily optimized scheduling operation plan, configuration instructions are generated for the photovoltaic subsystem, hydrogen energy subsystem, battery module and grid corresponding to the matching operation mode.

6. The system according to claim 1, characterized in that, The decision control layer generates optimized scheduling control instructions based on a second-level dynamic adjustment mechanism according to the following logic: By combining the instantaneous changes in DC bus voltage and the state of charge of the energy storage battery, a battery charging and discharging power control command is generated to quickly adjust the charging and discharging power of the battery, making up for the slow response of the fuel cell and maintaining real-time power balance.

7. The system according to claim 1, characterized in that, The decision control layer is also configured to determine the appropriate operating mode in real time based on the dynamic operating data obtained by the data perception layer, and generate mode execution instructions; the operating modes include photovoltaic-dominated hydrogen production mode, fuel cell grid-connected power supply mode, off-grid islanded operation mode, valley electricity hydrogen production mode, and peak electricity power generation mode.

8. The system according to claim 1, characterized in that, The decision control layer is configured to adapt to the following operating mode based on the following logic: When the photovoltaic power exceeds the demand load data, the battery state of charge is further assessed. If the battery state of charge exceeds the set upper limit, the photovoltaic-dominated hydrogen production mode is determined to be the appropriate operating mode. The electrolyzer module is then started, and the surplus photovoltaic power is used to power the electrolyzer module, with the excess photovoltaic power used for water electrolysis to produce hydrogen. If the battery state of charge is less than or equal to the set upper limit, the surplus photovoltaic power is used to charge the battery first, and the photovoltaic-dominated hydrogen production mode is not immediately entered.

9. The system according to claim 1, characterized in that, The decision control layer is configured to adapt to the following operating mode based on the following logic: If photovoltaic power is less than or equal to demand load data, further determine whether the current period is a peak electricity price period; If the grid-connected power supply mode of the fuel cell is determined to be the appropriate operating mode during peak electricity price periods, the fuel cell power generation will be started to make up for the power deficit; otherwise, the power deficit will be made up by purchasing electricity from the external power grid.

10. The system according to claim 1, characterized in that, The decision control layer monitors different distribution ranges of the battery's state of charge in real time, and makes decisions on energy storage control commands according to the following logic to achieve hierarchical energy storage operation scheduling: When the battery charge drops to the emergency charging threshold, the system first determines whether the photovoltaic charging power is sufficient. If so, the battery is charged based on the surplus photovoltaic power. If not, the system further determines whether the fuel cell is effective and starts the fuel cell to supply power to the load and charge the battery. When the battery charge is within the normal floating threshold, the photovoltaic system and the grid provide priority power supply, and the battery smooths out short-term fluctuations. When the battery charge meets the hydrogen production start-up threshold, the electrolyzer is automatically started to convert excess electrical energy into hydrogen energy for storage.

11. The system according to claim 1, characterized in that, The decision control layer is also configured to set importance levels for each load and set different weight indicators for each level. Based on the weight indicators, the loads with higher importance are selected for priority power supply, while the power supply to other loads is selectively suspended.

12. A smart energy management method for hydrogen-solar hybrid microgrids in railway parks, characterized in that, The method is applied to the system described in any one of claims 1 to 11.

13. A storage medium, characterized in that, The storage medium stores program code capable of implementing the method as described in any one of claims 1 to 11.

Images