A real-time scheduling and compensation system and method for hydrogen production systems from renewable energy sources

Abstract

This invention belongs to the field of scheduling and control technology for renewable energy hydrogen production systems. It proposes a real-time scheduling compensation system and method for renewable energy hydrogen production systems. The main scheme is as follows: First, it acquires the electrolyzer cluster scheduling plan generated by the optimized scheduling model used by the scheduling platform, the real-time grid price signal, the real-time operating status of the electrolyzer cluster, and the real-time output of renewable energy. Second, it compares the real-time output of renewable energy with the minimum operating power threshold of the electrolyzers, and, combined with the real-time grid price signal and the real-time operating status of the electrolyzer cluster, generates compensation decision instructions based on the trust scheduling principle. Then, it sends the compensation decision instructions to the power control system of the electrolyzer cluster, and the scheduling control system performs real-time correction on the periodically generated electrolyzer cluster scheduling plan according to the compensation decision instructions. This invention can significantly improve hydrogen production, wind energy utilization, and overall operating efficiency.

Description

Technical Field

[0001] This invention relates to the field of scheduling and control technology for renewable energy hydrogen production systems, and particularly to a real-time scheduling and compensation system and method for renewable energy hydrogen production systems. Background Technology

[0002] With the rapid growth of installed capacity of renewable energy sources such as wind and solar power, the intermittency and volatility of their output have brought significant challenges to the operation of power systems. Combining renewable energy with water electrolysis for hydrogen production is an important technological path for absorbing excess renewable energy and producing green hydrogen. Against this backdrop, the optimal scheduling of renewable energy hydrogen production systems, especially multi-electrolyzer cluster systems, has attracted widespread attention.

[0003] To improve system economy, optimal scheduling strategies based on mathematical programming models have been extensively studied. These strategies typically aim to maximize economic benefits, comprehensively considering multiple factors such as real-time grid electricity prices, renewable energy power forecasting, electrolyzer state transition constraints, and load balancing. They generate scheduling plans ranging from minutes to hours through rolling calculations using optimization models. However, these advanced optimal scheduling strategies face two inherent technical bottlenecks when transitioning from theoretical research to engineering applications: First, the generated scheduling plan for future periods needs to reference the output of the power prediction model. However, predictions inevitably contain errors, especially in the low-power range, where the bias and noise of the prediction model are more significant. The scheduling plan is an "optimal" decision based on the predicted value. When the actual renewable energy output deviates from the predicted value, the original plan is no longer the optimal decision, and may even lead to frequent start-ups and shutdowns of electrolyzers and the system operating in an inefficient range, thereby damaging the system's economy and equipment lifespan.

[0004] Second, optimization algorithms suffer from computational latency, making them difficult to apply directly to real-time control. Optimization scheduling models are complex to solve and computationally time-consuming, with scheduling plan generation cycles typically on the order of minutes (e.g., 15-30 minutes). However, the power control system of an electrolytic cell cluster requires response times on the order of seconds or even milliseconds. This significant gap between computational speed and real-time control requirements means that optimization scheduling models cannot be directly used as real-time controllers, preventing their excellent global optimization capabilities from being fully utilized in engineering practice.

[0005] To address these issues, traditional engineering solutions typically employ simple rule-based control (RBC) strategies for real-time control. While these strategies offer fast response times and high stability, their fixed decision-making logic makes them ill-suited to fluctuating grid prices and complex system states. They fail to achieve the global economic optimization sought in optimal scheduling, representing a compromise that sacrifices economic efficiency for reliability.

[0006] Therefore, a prominent technical contradiction has long existed in the field of renewable energy hydrogen production: if the site scheduling strategy adopts a rule-based approach, it cannot achieve global optimization; if an optimization-based scheduling scheme is adopted, it lacks real-time control capabilities. How to bridge the gap between optimization scheduling models and real-time engineering applications, enabling the system to enjoy both the economic advantages of optimization scheduling and the real-time reliability of rule-based control, has become a critical technical challenge that urgently needs to be solved. Summary of the Invention

[0007] The purpose of this invention is to provide a real-time scheduling compensation system and method for renewable energy hydrogen production systems. This system can achieve real-time closed-loop correction of scheduling plans without complex online calculations, effectively bridging the gap between the economic efficiency of optimized scheduling and the reliability of real-time control. It significantly improves the system's hydrogen production, wind energy utilization rate, and overall operating efficiency. Therefore, it can solve the technical problem that optimized scheduling strategies cannot be directly used for real-time control due to prediction errors and calculation delays.

[0008] The technical solution adopted by this invention to solve its technical problem is as follows: On one hand, the present invention provides a real-time scheduling and compensation system for a renewable energy hydrogen production system, comprising: The data acquisition and input module is used to acquire the electrolyzer cluster scheduling plan, real-time grid electricity price signal, real-time operating status of the electrolyzer cluster, and real-time output of renewable energy generated by the optimized scheduling model used by the scheduling platform, and output them to the rule compensation decision module. The rule-based compensation decision module stores the trust scheduling principle, which is used to compare the real-time output of the renewable energy with the minimum operating power threshold of the electrolyzer, and combine the real-time electricity price signal of the power grid with the real-time operating status of the electrolyzer cluster to generate a compensation decision instruction based on the trust scheduling principle and output it to the control instruction output module. The control command output module is used to send compensation decision commands to the power control system of the electrolytic cell cluster. The scheduling control system corrects the periodically generated electrolytic cell cluster scheduling plan in real time according to the compensation decision commands.

[0009] As a further optimization, when the rule-based compensation decision module compares the real-time output of the renewable energy source with the minimum operating power threshold of the electrolyzer, it includes: When the real-time output of the renewable energy is lower than the minimum operating power threshold of the electrolyzer, the system determines whether the current situation is economically viable based on the electrolyzer cluster scheduling plan of the scheduling platform. If the scenario is one with economic benefits, a compensation decision instruction is output to maintain or execute the electrolytic cell cluster scheduling plan of the scheduling platform. If the scenario is not economically viable, a compensation decision instruction is output to switch the affected electrolyzer to hot standby or shutdown state based on the real-time output of the renewable energy source.

[0010] As a further optimization, the scenario with economic benefits refers to the operating condition that is expected to generate economic benefits, which is calculated by the optimized scheduling model based on the real-time electricity price of the power grid, the predicted value of renewable energy power, and the operating status information of the electrolyzer cluster.

[0011] As a further optimization, maintaining or executing the electrolyzer cluster scheduling plan of the scheduling platform includes: controlling the system to purchase electricity from the grid to maintain the operation or start-up process of the load when the real-time output of renewable energy is lower than the required power.

[0012] As a further optimization, the trust scheduling principle executes the following steps in a loop with a second-level cycle: Obtain the latest scheduling plan issued by the optimized scheduling platform and verify the current position in the scheduling plan; For each electrolytic cell, if the electrolytic cell is in a stopped state, and the scheduling plan indicates that the electrolytic cell should be switched to a cold start process, then the electrolytic cell cold start process will be started; if the scheduling plan indicates that the electrolytic cell should be kept in a stopped state, then the electrolytic cell should be kept in a stopped state. If the electrolytic cell is in the cold start process, the scheduling plan will not be considered, and the electrolytic cell will be maintained to complete the cold start process. If the electrolyzer is in hot standby mode, consider whether the real-time output of renewable energy allocated to the electrolyzer exceeds the minimum power threshold. If it does, the electrolyzer is switched to hot start-up mode according to the dispatch plan. If the power allocated to the electrolyzer does not meet the minimum power threshold, but the dispatch plan still instructs the electrolyzer to switch to hot start-up mode, then it is switched to hot start-up mode according to the dispatch plan. If the dispatch plan instructs the electrolyzer to be switched to shutdown mode, determine whether the real-time renewable energy output can meet the electrolyzer's hot standby power consumption. If it can meet the electrolyzer's hot standby power consumption, then the electrolyzer is kept in hot standby mode instead of being shut down according to the dispatch plan. If it does not meet the requirements, then it is switched to shutdown mode according to the dispatch plan. If the dispatch plan instructs the electrolyzer to be kept in hot standby mode, then it is kept in hot standby mode according to the dispatch plan. If the electrolytic cell is in the hot start process, then maintain the hot start of the electrolytic cell; If the electrolyzer is in operation, consider whether the real-time output of renewable energy allocated to the electrolyzer exceeds the minimum power threshold. If it does, maintain the electrolyzer's operation according to the dispatch plan. If the dispatch plan does not issue a dispatch instruction to maintain operation when the real-time output of renewable energy meets the electrolyzer's operating requirements (i.e., if the power allocated to the electrolyzer does not meet the minimum power threshold and the dispatch plan indicates that it should be switched to shutdown), consider whether the renewable energy output meets the electrolyzer's hot standby power consumption. If it does, prioritize switching to hot standby; otherwise, switch to shutdown according to the dispatch plan. If the dispatch plan indicates that it should be switched to hot standby, then switch the electrolyzer to hot standby according to the dispatch plan. If the dispatch plan indicates that the electrolyzer should be maintained in operation when renewable energy power is insufficient, determine whether the current grid electricity price is during off-peak hours. If not, switch the electrolyzer to hot standby; if it is during off-peak hours, maintain the electrolyzer's operation according to the dispatch plan.

[0013] As a further optimization, the optimized scheduling model is a mixed-integer linear programming, dynamic programming, or stochastic programming model; The optimized scheduling model adopts a rolling solution method.

[0014] As a further optimization, the electrolytic cell cluster scheduling plan includes the start / stop status of each electrolytic unit, the target power command, and the planned switching sequence.

[0015] On the other hand, the present invention also provides a real-time scheduling compensation method for a renewable energy hydrogen production system, applied to the aforementioned real-time scheduling compensation system for a renewable energy hydrogen production system, comprising the following steps: The system acquires the electrolyzer cluster scheduling plan generated by the optimized scheduling model used by the scheduling platform, the real-time electricity price signal of the power grid, the real-time operating status of the electrolyzer cluster, and the real-time output of renewable energy. The real-time output of the renewable energy source is compared with the minimum operating power threshold of the electrolyzer, and compensation decision instructions are generated based on the trust scheduling principle, taking into account the real-time electricity price signal of the power grid and the real-time operating status of the electrolyzer cluster. The compensation decision command is sent to the power control system of the electrolytic cell cluster, and the scheduling control system corrects the periodically generated electrolytic cell cluster scheduling plan in real time according to the compensation decision command.

[0016] As a further optimization, the optimized scheduling model generates scheduling plans over a period of 5 to 30 minutes, and the compensation decision instruction issuance period is in the second range.

[0017] As a further optimization, the method is triggered to execute at the end of each rolling optimization cycle.

[0018] The beneficial effects of this invention are as follows: This invention acquires optimized scheduling plans and real-time operation data through a data acquisition module, and performs rapid evaluation and correction of the plan instructions based on the trust scheduling principle through a rule compensation decision module. Finally, the compensation decision instructions are sent to the execution unit through a control instruction output module, thereby achieving real-time closed-loop correction of the scheduling plan without the need for complex online calculations, effectively bridging the gap between the economy of optimized scheduling and the reliability of real-time control. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system composition structure of a real-time scheduling and compensation system for a renewable energy hydrogen production system according to Embodiment 1 of the present invention; Figure 2 This is a system topology diagram of a wind power hydrogen production scenario in Embodiment 1 of the present invention; Figure 3 The wind field data is the actual measured historical data for the whole year in Embodiment 1 of this invention; Figure 4 This is a flowchart of the rule compensation decision module based on the trust scheduling principle in Embodiment 1 of the present invention. Figure 5 This is a flowchart of a real-time scheduling and compensation method for a renewable energy hydrogen production system according to Embodiment 2 of the present invention; Figure 6 The classification results are for eight typical days in Embodiment 3 of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Example 1

[0022] See Figure 1 This embodiment provides a real-time scheduling and compensation system for a renewable energy hydrogen production system, wherein the system may consist of the following parts: The data acquisition and input module is used to acquire the electrolyzer cluster scheduling plan, real-time grid electricity price signal, real-time operating status of the electrolyzer cluster, and real-time output of renewable energy generated by the optimized scheduling model used by the scheduling platform, and output them to the rule compensation decision module. The rule-based compensation decision module stores the trust scheduling principle, which is used to compare the real-time output of the renewable energy with the minimum operating power threshold of the electrolyzer, and combine the real-time electricity price signal of the power grid with the real-time operating status of the electrolyzer cluster to generate a compensation decision instruction based on the trust scheduling principle and output it to the control instruction output module. The control command output module is used to send compensation decision commands to the power control system of the electrolytic cell cluster. The scheduling control system corrects the periodically generated electrolytic cell cluster scheduling plan in real time according to the compensation decision commands.

[0023] See Figure 2 This embodiment uses a grid-connected hydrogen production system in Northeast China, equipped with a 4MW wind turbine and four 1MW alkaline electrolyzers (ALK), as an application scenario. The wind power data are historical measured data from the wind farm, as shown in the attached figure. Figure 3 The implementation process of the present invention will be described in detail below.

[0024] This embodiment of a real-time scheduling compensation system for a renewable energy hydrogen production system can be deployed on an industrial computer or server located at the scheduling layer, and communicate with the optimized scheduling model platform and the underlying electrolyzer cluster power control system via a data bus (such as OPC UA, Modbus TCP, etc.).

[0025] It should be noted that the real-time scheduling compensation system for a renewable energy hydrogen production system in this embodiment mainly consists of three modules: a data acquisition and input module, a rule-based compensation decision module, and a control command output module. Specifically, the data acquisition and input module acts as a data interface, continuously acquiring and updating the following real-time data at a frequency of once per second: Electrolyzer Cluster Scheduling Plan: The scheduling plan for the next 4 hours is obtained from the upper-level optimization scheduling platform (in this embodiment, a mixed-integer linear programming model is used as the optimization calculation method for the scheduling platform, with a rolling solution in 15-minute intervals). This plan exists in the form of a data table, which includes the preset start / stop status of the four electrolyzers in each 15-minute interval (e.g., ALK1: On, ALK2: Standby, ALK3: Off, ALK4: Cold Start), target power command (e.g., ALK1: 0.8MW), and planned switching sequence. The electrolyzer cluster scheduling plan can provide sufficient judgment basis for rule compensation decisions.

[0026] Real-time electricity price signal from the power grid: Local time-of-use electricity price information is obtained from the power grid dispatch system or market trading platform. The specific price cycle is shown in Table 1. The data acquisition and input module provides the electricity price signal at the current moment to the rule compensation decision module.

[0027] Table 1. Local Industrial Electricity Price Peak-Valley Mechanism , Real-time operating status of the electrolytic cell cluster: The current status (On, Standby, Cold start, Off) and actual input power (e.g., ALK1: 0.75MW) of each electrolytic cell are read in real time from the underlying PLC control system.

[0028] Real-time output of renewable energy: The actual total output of the wind farm (unit: MW) is obtained in real time from the wind farm SCADA system.

[0029] For the rule-based compensation decision module, when comparing the real-time output of the renewable energy source with the minimum operating power threshold of the electrolyzer, it may include: When the real-time output of the renewable energy is lower than the minimum operating power threshold of the electrolyzer, the system determines whether the current situation is economically viable based on the electrolyzer cluster scheduling plan of the scheduling platform. If the scenario is one with economic benefits, a compensation decision instruction is output to maintain or execute the electrolytic cell cluster scheduling plan of the scheduling platform. If the scenario is not economically viable, a compensation decision instruction is output to switch the affected electrolyzers to hot standby or shutdown status based on the real-time output of the renewable energy source. Maintaining or executing the electrolyzer cluster scheduling plan of the scheduling platform includes: when the real-time output of the renewable energy source is lower than the required power, controlling the system to purchase electricity from the grid to maintain the operation or start-up of the load.

[0030] In this embodiment, the scenario with economic benefits refers to the operating condition that is expected to generate economic benefits, which is calculated by the optimized scheduling model based on the real-time electricity price of the power grid, the predicted value of renewable energy power, and the operating status information of the electrolyzer cluster.

[0031] It should be noted that in this embodiment, the optimized scheduling model adopts a rolling solution method to periodically adapt to changes in the system state. The optimized scheduling model is a mathematical programming model, such as, but not limited to, mixed integer linear programming (MILP), dynamic programming, or stochastic programming models.

[0032] Specifically, the rule-based compensation decision module has built-in features such as... Figure 4 The decision logic flowchart shown below executes the following steps in a loop with a second-level cycle: Step 1: Obtain the latest scheduling plan issued by the optimized scheduling platform and verify the current position in the scheduling plan.

[0033] Step 2: For each electrolytic cell, if the electrolytic cell is in a stopped state, and the scheduling plan indicates that the electrolytic cell should be switched to the cold start process, then the electrolytic cell cold start process will begin; if the scheduling plan indicates that the electrolytic cell should be kept in a stopped state, then the electrolytic cell should be kept in a stopped state.

[0034] Step 3: If the electrolytic cell is in the cold start process, the scheduling plan is not considered, and the electrolytic cell is maintained to complete the cold start process.

[0035] Step 4: If the electrolyzer is in hot standby mode, consider whether the real-time output of renewable energy allocated to the electrolyzer exceeds the minimum power threshold. If it does, refer to the dispatch plan to switch the electrolyzer into hot start-up mode. If the power allocated to the electrolyzer does not meet the minimum power threshold, but the dispatch plan still indicates to switch the electrolyzer into hot start-up mode, then switch it into hot start-up mode according to the dispatch plan. If the dispatch plan indicates to switch the electrolyzer into shutdown mode, it is necessary to determine whether the real-time renewable energy output can meet the electrolyzer's hot standby power consumption. If it can meet the electrolyzer's hot standby power consumption, then maintain the electrolyzer in hot standby mode instead of shutting it down according to the dispatch plan. This decision can maximize the utilization rate of renewable energy. If it does not meet the requirements, then switch it into shutdown mode according to the dispatch plan. If the dispatch plan indicates to maintain the electrolyzer in hot standby mode, then maintain the hot standby mode according to the dispatch plan.

[0036] Step 5: If the electrolytic cell is in the hot start process, then maintain the hot start of the electrolytic cell.

[0037] Step 6: If the electrolyzer is in operation, consider whether the real-time output of renewable energy allocated to the electrolyzer exceeds the minimum power threshold. If it does, maintain the electrolyzer's operation according to the dispatch plan. If the dispatch plan does not issue a dispatch instruction to maintain operation even when the real-time output of renewable energy meets the electrolyzer's operating requirements, refer to the subsequent procedures. The main reason may be that individual electrolyzers need to balance their operating time to reduce the inter-cell degradation difference. If the power allocated to the electrolyzer does not meet the minimum power threshold, and the dispatch plan instructs it to be switched to a shutdown state, consider whether the renewable energy output meets the electrolyzer's hot standby power consumption. If it does, it should be switched to hot standby first; otherwise, it should be switched to a shutdown state according to the dispatch plan. This can improve the utilization rate of renewable energy. If the dispatch plan instructs it to be switched to hot standby, then the electrolyzer should be switched to hot standby according to the dispatch plan. If the dispatch plan instructs to maintain the electrolyzer's operation when renewable energy power is insufficient, determine whether the current grid electricity price is during off-peak hours. If not, the electrolyzer should be switched to hot standby; if it is during off-peak hours, then the electrolyzer should be maintained according to the dispatch plan. The reason is that the optimization scheduling platform calculates the scheduling plan within a certain time period, and its calculation time is limited. It may actively purchase electricity to balance the inter-slot operation, resulting in unnecessary operating costs.

[0038] Example 2

[0039] Based on Example 1, this example provides a real-time scheduling and compensation method for renewable energy hydrogen production systems, the flowchart of which can be found here. Figure 5 The method may include the following steps: S1. Obtain the electrolyzer cluster scheduling plan, real-time grid electricity price signal, real-time operating status of the electrolyzer cluster, and real-time output of renewable energy generated by the optimized scheduling model used by the scheduling platform; S2. Compare the real-time output of the renewable energy with the minimum operating power threshold of the electrolyzer, and combine the real-time electricity price signal of the power grid with the real-time operating status of the electrolyzer cluster to generate a compensation decision instruction based on the trust scheduling principle. S3. The compensation decision instruction is sent to the power control system of the electrolytic cell cluster, and the scheduling control system performs real-time correction on the periodically generated electrolytic cell cluster scheduling plan according to the compensation decision instruction.

[0040] Preferably, the optimized scheduling model generates scheduling plans over a relatively long period, typically 5 to 30 minutes. However, the real-time scheduling compensation method for a renewable energy hydrogen production system in this embodiment issues instructions in a period of seconds or less, thus perfectly solving the problem that the computational delay of the optimized scheduling model cannot meet real-time control requirements. Furthermore, the real-time scheduling compensation method for a renewable energy hydrogen production system in this embodiment can be triggered at the end of each rolling optimization cycle, ensuring that corrections are always made based on the latest version of the scheduling plan.

[0041] Example 3

[0042] Based on Examples 1 and 2, this example verifies the performance of the system in Example 1 and the method in Example 2. To comprehensively verify the effectiveness and adaptability of the system under different operating conditions, firstly, based on the actual annual operating data of the wind farm, the K-means clustering algorithm is used to perform pattern recognition and classification of the daily wind power curves. Secondly, by calculating the Euclidean distance between samples, the most representative daily load curves are identified. Finally, the annual data is divided into eight typical daily categories with significant characteristics, and their classification and characteristics are shown in the appendix. Figure 6 As shown, this classification covers all typical operating conditions of wind farms, including extremely high output, stable output, fluctuating output, and extremely low output, providing a scientific basis for comprehensively verifying the adaptability and stability of scheduling strategies in different scenarios.

[0043] To further verify the system's robustness to wind power prediction errors, this embodiment employs a simulation method based on random error injection. The specific implementation process is as follows: 1) Data preparation: Select the wind power output data (time resolution 5 minutes) of the above eight typical days as the real values, use them to drive system simulation and serve as the benchmark for performance evaluation.

[0044] 2) Simulated Prediction Data Construction: An iterative random noise injection method is used, with typical daily wind power output data as input, to construct simulated wind power prediction data. By dynamically adjusting the error generation coefficients with mean absolute error (MAE) and root mean square error (RMSE) as target parameters, a simulated prediction sequence integrating trend error components and composite noise is generated. In this embodiment, the prediction data accuracy range is MAE 0.16-0.25 and RMSE 0.21-0.35 to simulate the output effect of typical power prediction models in engineering practice. The results after error construction for 8 typical days are shown in Table 2.

[0045] Table 2. Error construction results for 8 typical days. , After the above performance verification is performed, simulation testing can be carried out. In this embodiment, the specific simulation testing steps are as follows: 1) Input the generated simulation prediction data into the upper-level optimization scheduling model (MILP model). This model is based on the prediction data with errors and is optimized in a rolling cycle of 15 minutes to generate a scheduling plan for the electrolytic cell cluster.

[0046] 2) Input the real data of a typical day as the real-time output of renewable energy into the real-time dispatch and compensation system described in this invention.

[0047] 3) The rule-based compensation decision module corrects the scheduling plan generated by the optimization model in real time based on real wind power output data and issues the final control command.

[0048] The above process effectively simulates the scenario in actual engineering where predictions may deviate but the scheduling system needs to operate reliably, providing a reliable test environment for verifying the robustness of a real-time scheduling compensation system for a renewable energy hydrogen production system in this embodiment.

[0049] Finally, the adaptability and stability of the real-time scheduling compensation system for renewable energy hydrogen production systems described in this embodiment under different wind resource characteristics can be fully verified. This embodiment's test covers eight typical days obtained through cluster analysis to simulate various scenarios that a wind farm may encounter during its year-round operation. During the test, the optimized scheduling model is executed on a rolling basis with a 15-minute cycle, and the rule-based compensation decision module described in this invention is used to correct the scheduling plan in real time.

[0050] This embodiment selects two representative traditional scheduling strategies in the current field as comparison benchmarks: S1 (Traditional Daisy Chain Strategy): Based on traditional rule-based control strategies. This strategy uses fixed power thresholds and priority sequences to control the start-up and shutdown of electrolyzers. Although it has good real-time performance, it cannot adapt to dynamic electricity prices and complex operating conditions, resulting in poor economic efficiency.

[0051] S2 (Equal Power Distribution Strategy): A simple equal power distribution strategy. When the total power exceeds the minimum start-up power of the electrolyzer cluster, this strategy distributes the wind power equally to all operable electrolyzers. It is simple to implement but ignores the nonlinear efficiency characteristics of the electrolyzers, resulting in low overall efficiency.

[0052] The operational results for eight typical days are shown in Table 3. As can be seen from the data in Table 3, under the eight typical day operational scenarios representing different wind resource characteristics, the real-time scheduling compensation system (S3) for renewable energy hydrogen production systems in this embodiment exhibits excellent and stable comprehensive performance: Hydrogen production: On the vast majority of typical days (except for days 1 and 8, which are on par with or slightly better than S2), the hydrogen production of the system of this invention is significantly higher than that of the two traditional strategies. Especially on typical days 2, 4, 6, and 7 when wind energy fluctuates drastically, its advantage is extremely obvious. For example, on typical day 2, the hydrogen production is 20.4% higher than that of the optimal traditional strategy (S1).

[0053] Wind energy utilization rate: The system of this invention achieved the highest wind energy utilization rate on all eight typical days, indicating that its regular compensator can most effectively schedule the electrolyzer cluster to absorb fluctuating renewable energy and minimize the "wind curtailment" phenomenon. In particular, on typical days 2 and 4, when the S2 strategy performed poorly, the wind energy utilization rate of the strategy of this invention was significantly improved compared to S2.

[0054] System hydrogen production efficiency: The system hydrogen production efficiency of this invention was the highest in all eight typical days. This indicator comprehensively reflects the economics of the energy conversion process, and its leading advantage proves that this invention, while pursuing high hydrogen production, ensures high efficiency in the energy conversion process and avoids efficiency losses caused by frequent start-ups and shutdowns or operation in inefficient ranges.

[0055] The simulation results above clearly demonstrate that even with errors in the predicted data, the real-time scheduling compensation system for renewable energy hydrogen production systems provided in this embodiment can effectively adapt to various operating scenarios, from high-output stable days to low-output fluctuating days. Through second-level real-time correction of the optimized scheduling plan, the system successfully bridges the gap between the computational delay of the optimization model and the real-time control requirements of engineering. It can reliably improve the system's hydrogen production, wind energy utilization rate, and overall operating efficiency in different scenarios, proving its excellent robustness, adaptability, and engineering application value.

[0056] Table 3 Simulation Results ,

[0057] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A real-time dispatch compensation system for a renewable energy hydrogen production system, characterized in that, include: The data acquisition and input module is used to acquire the electrolyzer cluster scheduling plan, real-time grid electricity price signal, real-time operating status of the electrolyzer cluster, and real-time output of renewable energy generated by the optimized scheduling model used by the scheduling platform, and output them to the rule compensation decision module. The rule-based compensation decision module stores the trust scheduling principle, which is used to compare the real-time output of the renewable energy with the minimum operating power threshold of the electrolyzer, and combine the real-time electricity price signal of the power grid with the real-time operating status of the electrolyzer cluster to generate a compensation decision instruction based on the trust scheduling principle and output it to the control instruction output module. The control command output module is used to send compensation decision commands to the power control system of the electrolytic cell cluster. The power control system corrects the periodically generated electrolytic cell cluster scheduling plan in real time according to the compensation decision commands. When the rule-based compensation decision module compares the real-time output of the renewable energy source with the minimum operating power threshold of the electrolyzer, it includes: When the real-time output of the renewable energy is lower than the minimum operating power threshold of the electrolyzer, the system determines whether the current situation is economically viable based on the electrolyzer cluster scheduling plan of the scheduling platform. If the scenario is one with economic benefits, a compensation decision instruction is output to maintain or execute the electrolytic cell cluster scheduling plan of the scheduling platform. If the scenario is not economically viable, a compensation decision instruction is output to switch the affected electrolyzer to hot standby or shutdown state based on the real-time output of the renewable energy source. The trust scheduling principle executes the following steps in a loop with a second-level cycle: Obtain the latest scheduling plan issued by the optimized scheduling platform and verify the current position in the scheduling plan; For each electrolytic cell, if the electrolytic cell is in a stopped state, and the scheduling plan indicates that the electrolytic cell should be switched to a cold start process, then the electrolytic cell cold start process will be started; if the scheduling plan indicates that the electrolytic cell should be kept in a stopped state, then the electrolytic cell should be kept in a stopped state. If the electrolytic cell is in the cold start process, the scheduling plan will not be considered, and the electrolytic cell will be maintained to complete the cold start process. If the electrolyzer is in hot standby mode, consider whether the real-time output of renewable energy allocated to the electrolyzer exceeds the minimum power threshold. If it does, the electrolyzer is switched to hot start-up mode according to the dispatch plan. If the power allocated to the electrolyzer does not meet the minimum power threshold, but the dispatch plan still instructs the electrolyzer to switch to hot start-up mode, then it is switched to hot start-up mode according to the dispatch plan. If the dispatch plan instructs the electrolyzer to be switched to shutdown mode, determine whether the real-time renewable energy output can meet the electrolyzer's hot standby power consumption. If it can meet the electrolyzer's hot standby power consumption, then the electrolyzer is kept in hot standby mode instead of being shut down according to the dispatch plan. If it does not meet the requirements, then it is switched to shutdown mode according to the dispatch plan. If the dispatch plan instructs the electrolyzer to be kept in hot standby mode, then it is kept in hot standby mode according to the dispatch plan. If the electrolytic cell is in the hot start process, then maintain the hot start of the electrolytic cell; If the electrolyzer is in operation, consider whether the real-time output of renewable energy allocated to the electrolyzer exceeds the minimum power threshold. If it does, maintain the electrolyzer's operation according to the dispatch plan. If the dispatch plan does not issue a dispatch instruction to maintain operation when the real-time output of renewable energy meets the electrolyzer's operating requirements (i.e., if the power allocated to the electrolyzer does not meet the minimum power threshold and the dispatch plan indicates that it should be switched to shutdown), consider whether the renewable energy output meets the electrolyzer's hot standby power consumption. If it does, prioritize switching to hot standby; otherwise, switch to shutdown according to the dispatch plan. If the dispatch plan indicates that it should be switched to hot standby, then switch the electrolyzer to hot standby according to the dispatch plan. If the dispatch plan indicates that the electrolyzer should be maintained in operation when renewable energy power is insufficient, determine whether the current grid electricity price is during off-peak hours. If not, switch the electrolyzer to hot standby; if it is during off-peak hours, maintain the electrolyzer's operation according to the dispatch plan.

2. A real-time dispatch compensation system for a renewable energy hydrogen production system as claimed in claim 1, wherein, The economically beneficial scenario refers to the operating condition that is expected to generate economic benefits, which is calculated by the optimized scheduling model based on the real-time electricity price of the power grid, the predicted value of renewable energy power, and the operating status information of the electrolyzer cluster.

3. A real-time dispatch compensation system for a renewable energy hydrogen production system as claimed in claim 1, wherein, Maintaining or executing the electrolyzer cluster scheduling plan of the scheduling platform includes: controlling the system to purchase electricity from the grid to maintain the operation or start-up of the load when the real-time output of renewable energy is lower than the required power.

4. A real-time dispatch compensation system for a renewable energy hydrogen production system as claimed in claim 1, wherein, The optimized scheduling model is a mixed integer linear programming, dynamic programming, or stochastic programming model. The optimized scheduling model adopts a rolling solution method.

5. A real-time dispatch compensation system for a renewable energy hydrogen production system as claimed in claim 1, wherein, The electrolytic cell cluster scheduling plan includes the start / stop status of each electrolytic unit, the target power command, and the planned switching sequence.

6. A real-time scheduling compensation method for a renewable energy hydrogen production system, applied to a real-time scheduling compensation system for a renewable energy hydrogen production system according to any one of claims 1-5, characterized in that, Includes the following steps: The system acquires the electrolyzer cluster scheduling plan generated by the optimized scheduling model used by the scheduling platform, the real-time electricity price signal of the power grid, the real-time operating status of the electrolyzer cluster, and the real-time output of renewable energy. The real-time output of the renewable energy source is compared with the minimum operating power threshold of the electrolyzer, and compensation decision instructions are generated based on the trust scheduling principle, taking into account the real-time electricity price signal of the power grid and the real-time operating status of the electrolyzer cluster. The compensation decision command is sent to the power control system of the electrolytic cell cluster, and the power control system corrects the periodically generated electrolytic cell cluster scheduling plan in real time according to the compensation decision command.

7. A real-time dispatch compensation method for a renewable energy hydrogen production system according to claim 6, wherein, The optimized scheduling model generates scheduling plans over a period of 5 to 30 minutes, and the compensation decision instruction issuance period is in the second range.

8. The method for real-time dispatch compensation for a renewable energy hydrogen production system of claim 6, wherein, The method is triggered at the end of each rolling optimization cycle.

Images