A method, apparatus and equipment for load adjustment of a water electrolysis hydrogen production cluster

Abstract

This specification relates to the field of energy and chemical engineering technology, and in particular to a load adjustment method, apparatus, and equipment for a water electrolysis hydrogen production cluster. The method includes: calculating the temperature priority of each electrolyzer based on temperature data from multiple electrolyzers in the water electrolysis hydrogen production cluster; the temperature priority of an electrolyzer is positively correlated with its temperature data; determining multiple operating electrolyzers from among the multiple electrolyzers based on hydrogen production demand data or load consumption data; establishing an electrolyzer energy consumption scheduling optimization model according to a preset optimization algorithm; minimizing the total energy consumption of the multiple operating electrolyzers using the electrolyzer energy consumption scheduling optimization model, with hydrogen production demand data or load consumption data as constraints, to obtain a load allocation scheme; and adjusting the load of the multiple operating electrolyzers according to the load allocation scheme.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of energy and chemical technology, specifically to a load adjustment method, apparatus, and equipment for an electrolytic water hydrogen production cluster. Background Technology

[0002] In the field of hydrogen production through water electrolysis, large-scale production relies on the coordinated operation of multiple electrolyzer clusters, which is key to maximizing industrial-grade hydrogen production capacity and economic benefits. With the continuous decline in renewable energy electricity costs and the rapid development of the hydrogen energy industry, modern hydrogen production plants typically adopt modular designs, constructing efficient hydrogen production systems by connecting dozens or even hundreds of electrolyzers in parallel. This clustered operation mode not only flexibly adapts to fluctuating power inputs and enables precise control of hydrogen production, but also improves the overall reliability of the system through redundant design.

[0003] In existing technologies, electrolytic cell cluster control strategies are mostly based on fixed rotation or timing logic, such as starting electrolytic cells in numerical order or relying on sensors to monitor parameters in real time. However, existing methods mainly rely on running time / cycles, without considering the actual energy consumption and equipment status of the electrolytic cells. Frequent start-ups and shutdowns cause some electrolytic cells to operate in an inefficient range for extended periods, making it impossible to optimize load distribution in real time. This significantly increases system energy consumption and shortens equipment lifespan. Summary of the Invention

[0004] The purpose of the embodiments in this specification is to provide a load adjustment method, apparatus, and equipment for an electrolytic water hydrogen production cluster, so as to optimize the load distribution of the electrolytic water hydrogen production cluster in real time, reduce the increase in energy consumption and the shortening of equipment life caused by frequent start-ups and shutdowns.

[0005] To solve the above-mentioned technical problems, the specific technical solutions of the embodiments in this specification are as follows:

[0006] On the one hand, the embodiments of this specification provide a load adjustment method for an electrolytic water hydrogen production cluster, including:

[0007] Based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster, the temperature priority of each electrolyzer is calculated; the temperature priority of an electrolyzer is positively correlated with its temperature data.

[0008] Based on hydrogen production demand data or load consumption data, multiple operating electrolyzers are determined from the multiple electrolyzers using temperature priority.

[0009] An energy consumption scheduling optimization model for electrolytic cells is established based on a preset optimization algorithm.

[0010] Using hydrogen production demand data or load consumption data as constraints, the total energy consumption of multiple operating electrolyzers is minimized by an electrolyzer energy consumption scheduling optimization model to obtain a load allocation scheme.

[0011] Adjust the load of multiple operating electrolytic cells according to the load distribution plan.

[0012] On another front, embodiments of this specification provide a load adjustment device for an electrolytic water hydrogen production cluster, comprising:

[0013] The calculation module is used to calculate the temperature priority of each electrolyzer in the water electrolysis hydrogen production cluster based on the temperature data of multiple electrolyzers; the temperature priority of the electrolyzer is positively correlated with its temperature data.

[0014] The determination module is used to determine multiple operating electrolyzers from the multiple electrolyzers based on hydrogen production demand data or load consumption data, using temperature priority.

[0015] A module is established to build an energy consumption scheduling optimization model for electrolytic cells based on a preset optimization algorithm;

[0016] The optimization module is used to minimize the total energy consumption of multiple operating electrolyzers by using an electrolyzer energy consumption scheduling optimization model, with hydrogen production demand data or load consumption data as constraints, to obtain a load allocation scheme.

[0017] The adjustment module is used to adjust the load of multiple operating electrolytic cells according to the load distribution scheme.

[0018] In another aspect, a computer device is provided, including a memory for storing computer programs and a processor for executing the computer programs to implement the above-described load adjustment method for the water electrolysis hydrogen production cluster.

[0019] Furthermore, embodiments of this specification also provide a computer program product, which, when run by the processor of a computer device, executes instructions for any of the methods described above.

[0020] As can be seen from the technical solutions provided in the embodiments of this specification above, the embodiments of this specification can calculate the temperature priority of each electrolyzer based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster; the temperature priority of the electrolyzer is positively correlated with its temperature data; based on hydrogen production demand data or load consumption data, multiple operating electrolyzers are determined from the multiple electrolyzers using the temperature priority; an electrolyzer energy consumption scheduling optimization model is established according to a preset optimization algorithm; using the hydrogen production demand data or load consumption data as constraints, the total energy consumption of the multiple operating electrolyzers is minimized using the electrolyzer energy consumption scheduling optimization model to obtain a load allocation scheme; and the load of the multiple operating electrolyzers is adjusted according to the load allocation scheme. Compared with existing methods, the embodiments of this specification can determine the temperature priority of different electrolyzers based on the temperature distribution characteristics of multiple electrolyzers in the water electrolysis hydrogen production cluster, and can use hot start priority logic to reduce the additional energy loss of shutdown and restart under the premise of optimal energy consumption. The embodiments in this specification integrate energy-optimization logic and hot-start priority logic, which can optimize the load distribution of the water electrolysis hydrogen production cluster in real time, reduce the increase in energy consumption and the shortening of equipment life caused by frequent start-stop, and greatly improve the efficiency of water electrolysis hydrogen production. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below.

[0022] Figure 1 This is a flowchart of a load adjustment method for a water electrolysis hydrogen production cluster provided in the embodiments of this specification;

[0023] Figure 2 This is an overall logic flowchart of a load adjustment method for a water electrolysis hydrogen production cluster provided in the embodiments of this specification;

[0024] Figure 3 This is a flowchart of a method for determining the temperature priority of an electrolytic cell, as provided in the embodiments of this specification.

[0025] Figure 4 This is a flowchart of a method for generating candidate load allocation schemes for a water electrolysis hydrogen production cluster, as provided in the embodiments of this specification.

[0026] Figure 5 This is a flowchart of a method for generating candidate load allocation schemes for a water electrolysis hydrogen production cluster, as provided in the embodiments of this specification.

[0027] Figure 6 This is a flowchart of a method for determining the energy consumption priority of an electrolytic cell, as provided in the embodiments of this specification.

[0028] Figure 7This is a flowchart of a method for adjusting the start-up and shutdown state of an electrolytic cell, as provided in the embodiments of this specification.

[0029] Figure 8 This is a flowchart of a method for selecting a load allocation scheme for a water electrolysis hydrogen production cluster, as provided in the embodiments of this specification.

[0030] Figure 9 This is a flowchart of a method for determining the start-up priority of an electrolytic cell group, as provided in an embodiment of this specification.

[0031] Figure 10 This is a flowchart of a load adjustment method for an electrolytic cell group provided in the embodiments of this specification;

[0032] Figure 11 This is a schematic diagram of the structural composition of a load adjustment device for an electrolytic water hydrogen production cluster provided in the embodiments of this specification;

[0033] Figure 12 This is a schematic diagram of the structural composition of the computer device provided in the embodiments of this specification. Detailed Implementation

[0034] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0035] In some embodiments, a water electrolysis hydrogen production cluster may include multiple electrolyzers operating in parallel, and a gas-liquid separation and purification system, an energy management system, and a control system configured separately for each electrolyzer. The energy management system collects real-time data on wind and solar power generation, the state of charge of the energy storage system, and hydrogen demand, and generates a target value for the total hydrogen production or a constraint value for the total electricity consumption of the electrolyzer cluster through dynamic calculation. Based on the target value or constraint value, the control system determines the start-up and shutdown status of each electrolyzer. Subsequently, for electrolyzers in operation, the control system allocates specific operating loads according to their load-energy consumption characteristic curves; for electrolyzers in shutdown status, the control system calculates the maintenance temperature threshold based on their shutdown duration and a preset thermodynamic model to reduce energy loss during restart.

[0036] In some embodiments, the control system can set three types of time parameters: sorting period M (in minutes); measurement period P (in minutes); and load switching period Q (in minutes), where M ≥ P ≥ Q. The sorting period M is used for global energy consumption evaluation and priority reordering. Within each sorting period, the control system updates the energy consumption data of all electrolyzers across the entire load range. The measurement period P is defined as the minimum continuous operating time for a single energy consumption data acquisition. When an electrolyzer operates continuously at a certain load point for more than P minutes, the energy consumption data for that period is considered valid, meaning it can be included in the calculation. The load switching period Q is the interval between control command issuances. Every Q minutes, the start / stop status of the electrolyzers and load allocation are adjusted based on the latest sorting results. For example, when M = 1440 minutes, P = 15 minutes, and Q = 10 minutes are set, the control system updates the global sorting once a day, collects valid energy consumption data every 15 minutes, and performs load adjustments every 10 minutes. The priority calculation in the sorting period M depends on slowly varying parameters such as energy consumption; a long period can avoid excessive response noise. In the measurement cycle P, load, temperature, and other status monitoring are faster than in the sorting cycle, ensuring the timeliness of data for decision-making, but slower than in the switching cycle to filter out instantaneous fluctuations. In the load switching cycle Q, power commands can quickly track grid dispatch. A sorting cycle M ≥ measurement cycle P prevents frequent priority jumps that could cause load oscillations, while a measurement cycle P ≥ load switching cycle Q ensures that switching actions are based on steady-state data. Overall, long cycles optimize global energy efficiency, short cycles ensure rapid frequency regulation, and medium cycles bridge the two, achieving slow optimization followed by fast execution.

[0037] Figure 1 This is a flowchart illustrating a load adjustment method for a water electrolysis hydrogen production cluster provided in the embodiments of this specification. Figure 2 This is an overall logic flowchart of a load adjustment method for a water electrolysis hydrogen production cluster provided in the embodiments of this specification. In specific implementation, it includes the following steps:

[0038] S101: Calculate the temperature priority of each electrolyzer based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster; the temperature priority of an electrolyzer is positively correlated with its temperature data.

[0039] In some embodiments, the temperature data includes first temperature data for each electrolyzer in a historical sorting cycle, second temperature data in a historical measurement cycle, and third temperature data in the current load switching cycle.

[0040] The historical sorting period can be one or more prior to the current load switching period. Within the historical sorting period, the effective temperature data of each electrolyzer measured in multiple measurement periods can be statistically analyzed. By fusing these temperature data, the first temperature data can be obtained, which can reflect the long-term thermal state trend of the electrolyzer.

[0041] The historical measurement period can be one or more measurement periods preceding the current load switching period. Within the historical measurement period, the effective temperature data of each electrolyzer measured during that period can be acquired as secondary temperature data, which can reflect the short-term thermal dynamics of the electrolyzer.

[0042] The third temperature data can be the real-time temperature data of the electrolytic cell measured during the current load switching cycle. It can provide real-time temperature feedback to ensure the instantaneous safety of control commands.

[0043] Figure 3 This is a flowchart illustrating a method for determining the temperature priority of an electrolytic cell, as provided in the embodiments of this specification. In practice, it includes the following steps:

[0044] S1011: Based on the fusion result of the first temperature data, the second temperature data and the third temperature data, determine the fourth temperature data for each electrolytic cell.

[0045] In some embodiments, a fourth temperature data for each electrolytic cell is determined based on the fusion result of the first temperature data, the second temperature data, and the third temperature data.

[0046] By extrapolating the trends of temperature data from historical sorting and measurement periods, temperature changes in future periods can be predicted, helping to mitigate overheating risks in advance. Furthermore, the fusion results inherit historical trends and are overlaid with real-time corrections, preventing priority shifts caused by measurement noise.

[0047] A weighted moving average can be used to generate the fourth temperature data. When generating the fourth temperature data, the weights of different temperature data can be allocated based on the differences between real-time temperature data and historical trend temperature data. For example, when the difference between real-time temperature data and historical trend temperature data is large, the stability of the electrolyzer operation should be prioritized; that is, real-time temperature data should be given priority, and the weight of the first temperature data should be increased. Conversely, when the difference between real-time temperature data and historical trend temperature data is small, the energy efficiency of the electrolyzer operation should be prioritized; that is, the weight of historical trend temperature data can be appropriately increased to avoid short-term fluctuations.

[0048] S1012: Based on the fourth temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster, the multiple electrolyzers are divided using a preset temperature threshold to obtain multiple hot cells and multiple cold cells.

[0049] In some embodiments, based on fourth temperature data from multiple electrolyzers in the water electrolysis hydrogen production cluster, the multiple electrolyzers are divided using a preset temperature threshold to obtain multiple hot-state cells and multiple cold-state cells. Dynamically dividing the cold / hot-state cells based on temperature thresholds can reflect the thermal inertia state of the equipment in real time. Furthermore, the cells can be sorted according to the temperature data of the hot-state cells to maximize the utilization of preheating energy, shorten start-up time, and reduce additional energy consumption.

[0050] In some embodiments, a temperature threshold is determined based on the operating data of multiple electrolyzers in the water electrolysis hydrogen production cluster during the current load switching cycle; the multiple electrolyzers are then divided using the temperature threshold to obtain multiple hot cells and multiple cold cells.

[0051] During each load switching cycle, operational data can be collected or calculated simultaneously. This data may include one or more of the following: real-time current density, operating time, ambient temperature, recent average load rate, and cell age. Furthermore, statistical characteristics of the water electrolysis hydrogen production cluster can be calculated based on the operational data of multiple electrolyzers within the cluster. These characteristics may include one or more of the following: average temperature of all cells, temperature standard deviation, and maximum / minimum temperature. Theoretical or semi-empirical relationships between temperature and strongly correlated parameters such as current density and ambient temperature can be established based on physical models or empirical formulas to calculate the theoretical or expected operating temperature range for the current load switching cycle, using its boundaries as temperature thresholds.

[0052] In some embodiments, a Gaussian process is used to fit the cluster temperature distribution based on fourth temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster; and a probability boundary for distinguishing between hot and cold cells is determined based on the statistical characteristics of the cluster temperature distribution.

[0053] Fixed preset thresholds cannot adapt to the dynamic changes in the operating state of the electrolytic cell. This may lead to misjudgment of hot and cold cells during the transition phase, or fail to accurately reflect the optimal thermal state boundary for different cells and operating conditions.

[0054] A Gaussian process can be used to fit the fourth temperature data of multiple electrolyzers in a water electrolysis hydrogen production cluster to obtain a cluster temperature distribution curve or a cluster temperature distribution histogram. The statistical characteristics of the cluster temperature distribution can include at least the mean and variance of the distribution curve / hitogram. Therefore, the mean and variance of the distribution curve / hitogram can be used to determine the probability boundaries for distinguishing between hot and cold cells. For example, the sum of the mean and 1.5 times the standard deviation can be used as the probability boundary for hot cells, and the difference between the mean and 1.5 times the standard deviation can be used as the probability boundary for cold cells. That is, cells above the mean + 1.5 times the standard deviation are considered hot cells, and cells below the mean - 1.5 times the standard deviation are considered cold cells. For each load switching cycle, the above steps can be used to calculate the cluster temperature distribution in real time, and then the temperature threshold for distinguishing between hot and cold cells can be dynamically adjusted according to the probability boundaries of the cluster temperature distribution. By incorporating the changes in the overall thermal state distribution of the cluster into the temperature threshold construction process, the classification accuracy can be significantly improved, adapting to different operating conditions and environmental changes, and reducing misjudgments, especially in dynamic processes such as startup and load changes.

[0055] S1013: Determine the temperature priority of the multiple hot-state tanks based on the comparison results of the temperature data of the multiple hot-state tanks.

[0056] The temperature priority of an electrolyzer can be positively correlated with the magnitude of its temperature data. Specifically, a temperature priority number HN(a) can be assigned to each electrolyzer, with the following rules: for all cold cells, HN(a) is forcibly set to 0. For hot cells, they can be sorted in descending order based on their fourth temperature data, with higher temperatures resulting in smaller HN(a) numbers.

[0057] S102: Based on hydrogen production demand data or load consumption data, multiple operating electrolyzers are determined among the multiple electrolyzers using temperature priority.

[0058] Figure 4 This is a flowchart illustrating a method for generating candidate load allocation schemes for a water electrolysis hydrogen production cluster, as provided in the embodiments of this specification. In practice, it includes the following steps:

[0059] S1021: Calculate the first operating scale data of the electrolyzer during the current load switching cycle based on hydrogen production demand data or load absorption data.

[0060] In some embodiments, the first operating scale data of the electrolyzer during the current load switching cycle is calculated based on hydrogen production demand data or load absorption data.

[0061] Hydrogen production demand data can include the expected hydrogen production during the current load switching cycle. The initial operational scale data can include the theoretical number of electrolyzers operating during the current load switching cycle. The spatiotemporal dimensions of the hydrogen production data can be clearly defined and converted into volumetric flow rate under standard operating conditions. Combined with the rated capacity and performance degradation coefficient of the electrolyzers, the actual capacity of a single electrolyzer can be determined. Then, the number of electrolyzers can be calculated using the capacity balance equation, i.e., the initial operational scale data of the electrolyzers expected during the current load switching cycle.

[0062] S1022: Obtain the second operating scale data of the electrolyzer during the historical load switching cycle.

[0063] In some embodiments, second operating scale data of the electrolyzer during historical load switching cycles are obtained.

[0064] Historical load switching cycles can be one or more load switching cycles preceding the current load switching cycle. It can also retrieve the second operating scale data of the electrolyzers operating within the previous one or more load switching cycles.

[0065] S1023: Obtain the energy consumption priority of each electrolytic cell within the historical sorting period.

[0066] In some embodiments, the energy consumption priority of each electrolyzer within a historical sorting period is obtained.

[0067] The historical sorting period can be one or more sorting periods before the current load switching period. During the historical sorting period, the effective energy consumption data of each electrolytic cell measured in multiple measurement periods can be statistically analyzed, these energy consumption data can be fused, and sorted from small to large according to the energy consumption data of different electrolytic cells to obtain the energy consumption priority levels of each electrolytic cell. That is, the smaller the energy consumption data, the higher the energy consumption priority level.

[0068] S1024: Determine multiple operating electrolytic cells among the multiple electrolytic cells according to the comparison result of the first operating scale data and the second operating scale data, and the temperature priority levels and energy consumption priority levels of the multiple electrolytic cells.

[0069] In some embodiments, multiple operating electrolytic cells are determined among the multiple electrolytic cells according to the comparison result of the first operating scale data and the second operating scale data, and the temperature priority levels and energy consumption priority levels of the multiple electrolytic cells.

[0070] By comparing the first operating scale data and the second operating scale data in real time, the number of cells that need to be scaled up or down in the system can be quickly identified, thereby ensuring that the hydrogen production amount accurately tracks the preset target. By combining the composite criterion of temperature priority level and energy consumption priority level, the cells to be started and stopped can be intelligently selected, taking into account both the response speed and the operating economy.

[0071] In the stage of adjusting the start-stop state, first compare the first operating scale data R(q + 1) with the second operating scale data R(q): If R(q + 1) > R(q), additional electrolytic cells need to be started, and they can be sorted according to the temperature priority level and energy consumption priority level, and the hot and low-energy-consuming cells are preferentially started; if R(q + 1) < R(q), the redundant equipment is shut down according to the energy consumption priority level and temperature priority level.

[0072] S103: Establish an electrolytic cell energy consumption scheduling optimization model according to a preset optimization algorithm.

[0073] In some embodiments, an electrolytic cell energy consumption scheduling optimization model is established according to a preset optimization algorithm.

[0074] After determining multiple operating electrolytic cells in the current load switching period, an electrolytic cell energy consumption scheduling optimization model for the multiple operating electrolytic cells can be established using an optimization algorithm. The real-time power consumption data of each operating electrolytic cell, unit: kWh / Nm³; the equipment efficiency curve; and the hot / cold state characteristic differences can be collected. For example, with the minimum total energy consumption of the multiple operating electrolytic cells as the objective function, an electrolytic cell energy consumption scheduling optimization model is constructed, and the constraint conditions can include one or more of the total hydrogen production constraint, the total power consumption constraint for the electrolytic water hydrogen production cluster, the load-hydrogen production constraint relationship, the single-cell load adjustment range, the dynamic response time difference between the hot cells and the cold cells, etc.

[0075] In some embodiments, the following formula can be used to construct an energy consumption scheduling optimization model for electrolyzers:

[0076] ;

[0077] In the formula, from top to bottom, are the optimization objective function, the constraint relationship between the hydrogen production of a single electrolyzer and its load, the total hydrogen production constraint, the upper and lower limits of the load constraint, and the load adjustment step size constraint.

[0078] S104: Using hydrogen production demand data or load consumption data as constraints, the total energy consumption of multiple operating electrolyzers is minimized using the electrolyzer energy consumption scheduling optimization model to obtain the load allocation scheme.

[0079] Figure 5 This is a flowchart illustrating a method for generating candidate load allocation schemes for a water electrolysis hydrogen production cluster, as provided in the embodiments of this specification. In practice, it includes the following steps:

[0080] S1041: Using hydrogen production demand data or load consumption data as constraints, the total energy consumption of multiple operating electrolyzers is minimized using an electrolyzer energy consumption scheduling optimization model to obtain the minimum energy consumption load allocation scheme for the multiple operating electrolyzers.

[0081] In some embodiments, using hydrogen production demand data or load consumption data as constraints, an electrolyzer energy consumption scheduling optimization model is used to minimize the total energy consumption of multiple operating electrolyzers, thereby obtaining the minimum energy consumption load allocation scheme for the multiple operating electrolyzers.

[0082] Based on the electrolyzer energy consumption scheduling optimization model, a preset optimization algorithm can be used to iteratively solve the optimization objective of minimizing total energy consumption, thereby obtaining the minimum energy load allocation scheme for multiple operating electrolyzers. The preset optimization algorithm can be a heuristic search algorithm such as particle swarm optimization or genetic algorithm, which will not be elaborated here. Hydrogen production demand data can be transformed into a specific hydrogen production target, i.e., setting the total hydrogen production S (in Nm³) that the operating electrolyzers must meet per unit time. This target can be decomposed to each operating electrolyzer, ensuring that the sum of the hydrogen production of each cell equals S. Furthermore, in some embodiments, the load absorption data of the water electrolysis hydrogen production cluster can be used as a constraint condition. The electrolyzer energy consumption scheduling optimization model can be used to minimize the total energy consumption of multiple operating electrolyzers, thereby obtaining the minimum energy load allocation scheme for multiple operating electrolyzers. The load absorption data can include the total power absorbed by the water electrolysis hydrogen production cluster, which can limit the maximum electrical power O (in kW) that the water electrolysis hydrogen production cluster can consume within that time period. The sum of the real-time power of each electrolyzer must not exceed O. Alternatively, by combining hydrogen production demand data and load consumption data as constraints, an electrolyzer energy consumption scheduling optimization model can be used to minimize the total energy consumption of multiple operating electrolyzers, thus obtaining the minimum energy consumption load allocation scheme for multiple operating electrolyzers, which will not be elaborated here.

[0083] In the iterative solution process using the electrolyzer energy consumption scheduling optimization model, the base load can be prioritized for hot cells, as their energy cost for maintaining a hot state after startup is lower than that of cold cells with frequent start-stop cycles. Cold cells can then bear peak loads based on fluctuations in hydrogen production demand. For cells of the same type with similar energy consumption characteristics, the crossover and mutation operator in a genetic algorithm can be used to optimize the load allocation ratio. For example, hot cells can be sorted by energy consumption from low to high and allocated increasing loads accordingly, while cold cells can be sorted by energy consumption from high to low and allocated decreasing loads accordingly. Simultaneously, a lifespan balancing factor is introduced to prevent premature aging caused by long-term full-load operation of a single cell. A penalty function is used to control the cumulative operating time difference between cells within a preset difference threshold. The final minimum energy consumption load allocation scheme includes parameters such as the load allocation ratio of each cell, the expected total power consumption, and the hydrogen production fluctuation rate.

[0084] S1042: Generate multiple candidate load allocation schemes corresponding to the minimum energy consumption load allocation scheme based on the preset energy consumption deviation.

[0085] In some embodiments, multiple candidate load allocation schemes corresponding to the minimum energy consumption load allocation scheme are generated based on a preset energy consumption deviation.

[0086] After obtaining the minimum energy consumption load allocation scheme, multiple candidate load allocation schemes can be generated through dynamic adjustment strategies based on a preset energy consumption deviation range. During the generation of candidate load allocation schemes, a multi-dimensional adjustment strategy can be adopted. On the one hand, the load allocation ratio of individual electrolyzers can be fine-tuned, prioritizing load reduction for hot or cold cells with higher unit hydrogen production power consumption, transferring the reduced load to equipment with better energy efficiency. On the other hand, the operational stability and lifespan of the equipment can be considered to avoid excessive adjustments leading to excessive load fluctuations on individual units. Furthermore, different load adjustment schemes can be formulated based on the differences in hot / cold characteristics of the electrolyzers. For hot cells, since the energy cost of maintaining a hot state after startup is lower than that of cold cells with frequent start-stop cycles, the load proportion of hot cells can be appropriately increased within the allowable energy consumption range. For cold cells, peak-shaving loads are assigned according to fluctuations in hydrogen production demand. Through iterative calculations using optimization algorithms, it is ensured that each candidate load allocation scheme meets the preset energy consumption deviation requirements and also complies with constraints such as the single-cell load adjustment range. The final output of multiple candidate load allocation schemes includes detailed energy consumption data, load allocation ratios for each electrolyzer, and expected hydrogen production for each scheme.

[0087] S1043: Based on the load data of the multiple operating electrolyzers during the historical load switching cycle, select the candidate load allocation scheme with the smallest relative load change from the multiple candidate load allocation schemes as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

[0088] In some embodiments, based on the load data of the plurality of operating electrolyzers during historical load switching cycles, the candidate load allocation scheme with the smallest relative load change is selected from the plurality of candidate load allocation schemes as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

[0089] By selecting the candidate load allocation scheme with the smallest relative load change from multiple candidate load allocation schemes as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster, the wear and tear on core components such as electrolyzer electrodes and electrolytes caused by drastic load fluctuations can be minimized, the frequency of equipment start-up and shutdown and energy consumption impact can be reduced, and the stable output of hydrogen production can be guaranteed, providing support for the long-term efficient operation of the hydrogen production cluster.

[0090] Real-time load records of all operating electrolyzers during historical load switching cycles can be retrieved from a preset dynamic energy consumption database, and the average load value of each electrolyzer during that cycle can be calculated. For each candidate load allocation scheme, the target load value allocated to each electrolyzer is extracted, and the relative change rate of the load per cell is calculated. The weighted average of the relative change rates of all electrolyzers is taken to obtain the overall relative change index of the scheme. The smaller the index, the smoother the connection between the scheme and the historical load. Finally, the optimal scheme is selected based on the relative change index. By comparing the relative change indices of all candidate schemes, the scheme with the smallest index can be selected as the load allocation scheme for the current switching cycle. If multiple schemes have the same index, the maximum relative change rate of a single cell is further compared, and the scheme with the smaller maximum change rate is selected first; if they are still the same, the scheme with a higher fit to the historical load curve can be selected by referring to the historical load fluctuation trend.

[0091] S105: Adjust the load of multiple operating electrolytic cells according to the load distribution plan.

[0092] In some embodiments, the loads of multiple operating electrolyzers are adjusted according to the load allocation scheme.

[0093] Based on the load allocation scheme of the water electrolysis hydrogen production cluster during the current load switching cycle, the current operating status of multiple electrolyzers can be comprehensively monitored and evaluated, including key parameters such as real-time voltage, current intensity, cell temperature, and electrolyte concentration of each electrolyzer, thus clarifying the current load level of each electrolyzer. Subsequently, the load of multiple electrolyzers is finely adjusted according to the load level set in the load allocation scheme. For electrolyzers with excessive load, their load can be reduced by decreasing the input current or optimizing the power supply lines; for electrolyzers with insufficient load, the current input is increased accordingly or their priority in the power supply network is adjusted to ensure that the load of each electrolyzer meets the expected scheme, thereby achieving efficient, stable, and safe operation of the entire electrolysis system.

[0094] Figure 6 This is a flowchart illustrating a method for determining the energy consumption priority of an electrolytic cell, as provided in the embodiments of this specification. In practice, it includes the following steps:

[0095] S10231: Obtain energy consumption data for each load point of each electrolytic cell in multiple historical measurement cycles within the historical sorting period.

[0096] In some embodiments, each electrolyzer includes multiple load points. For electrolyzers numbered 1 to T in a water electrolysis hydrogen production cluster, the control system can establish a dynamic energy consumption map of the entire load range for each electrolyzer. The lower limit of the operating load of electrolyzer a (1<=a<=T) can be defined as MIN(a), in units of %; and the upper limit as MAX(a), in units of %; where MIN(a) and MAX(a) are both integers between 0 and 100, and satisfy MIN(a)≤MAX(a). Within the load range [MIN(a),MAX(a)], the control system can discretize it into N(a)=MAX(a)-MIN(a)+1 load points, with corresponding load values ​​of {b1=MIN(a),b2=MIN(a)+1,…,bn=MAX(a)}. During the initial run, the initial unit hydrogen production energy consumption value of each load point can be obtained through test data, denoted as EC0(a,b), and stored in a preset electrolyzer energy consumption map database.

[0097] In some embodiments, energy consumption data for each load point of each electrolyzer during multiple historical measurement periods are obtained within a historical sorting period.

[0098] Based on the measured energy consumption data of each electrolyzer at multiple load points, the actual energy consumption characteristics under different loads can be accurately reflected, avoiding the one-sidedness of single-point evaluation.

[0099] During the operation of the water electrolysis hydrogen production cluster, the control system can collect energy consumption data for each electrolyzer within each historical sequencing period, using the measurement period P as a reference. Specifically, when electrolyzer a operates continuously at load point b for P minutes, this period can be recorded as a valid data segment Period(x,a,b,1), and the total power consumption Power(x,a,b,1) within Period(x,a,b,1) (in kWh) and the corresponding hydrogen production H2Prod(x,a,b,1) (in Nm³) can be obtained. The energy consumption data of electrolyzer a at load point b within the historical sequencing period x can be calculated using the following formula:

[0100] ;

[0101] Within the x-th historical sorting period, for electrolytic cell a, all data segments {Period(x,a,b,1),Period(x,a,b,2),…,Period(x,a,b,n)} at load point b, as well as the corresponding energy consumption data {EC(x,a,b,1),EC(x,a,b,2),…,EC(x,a,b,n)}, can be recorded, where n is a positive integer greater than or equal to 0.

[0102] In some embodiments, the energy consumption data of each electrolyzer at each load point is calculated based on the energy consumption data of each electrolyzer at multiple time periods at each load point. For electrolyzer a, if energy consumption data for n valid time periods appears at load point b within the x-th sorting period, the energy consumption data of electrolyzer a in the x-th sorting period can be determined based on the fusion result of the energy consumption data for these n valid time periods. The fusion result can be a weighted average result, a maximum result, etc., of the energy consumption data for these n valid time periods.

[0103] In some embodiments, based on a preset deviation threshold, it can be determined whether the energy consumption data of each electrolytic cell at each load point is abnormal. For the energy consumption data EC(x,a,b) of electrolytic cell a in the xth sorting cycle, it can be compared with the energy consumption data EC(x-1,a,b) of electrolytic cell a in the (x-1)th sorting cycle. If the comparison result exceeds the preset deviation threshold, an equipment abnormality alarm is triggered; otherwise, the energy consumption data of electrolytic cell a at load point b can be updated to EC(x,a,b).

[0104] In some embodiments, for discrete load points not covered by electrolytic cell a, such as load point b... k and load point b k+1 If there is no valid time period for energy consumption data, the linear interpolation algorithm can be used to complete the energy consumption data, as shown in the following formula:

[0105] ;

[0106] In some embodiments, for load points that violate the energy consumption data coverage of any valid time period, the energy consumption data of the previous period can be maintained unchanged, that is:

[0107] ;

[0108] By calculating the energy consumption data of each electrolyzer at multiple load points using the above logic, the energy consumption map database of the electrolyzer can be dynamically updated, ensuring the real-time nature of the energy consumption data and the consistency of the electrolyzer status.

[0109] S10232: Determine the energy consumption data of each electrolyzer based on the fusion results of energy consumption data of each load point of each electrolyzer over multiple historical measurement periods.

[0110] In some embodiments, the energy consumption data of each electrolyzer is determined based on the fusion results of energy consumption data of each load point of each electrolyzer over multiple historical measurement periods.

[0111] By integrating multi-load data to generate comprehensive energy efficiency indicators for each tank, an objective basis for prioritization can be provided, significantly improving the reliability of decision-making.

[0112] At the end of each historical sorting period x, the energy consumption data of electrolyzer a at all load points can be merged to obtain the energy consumption data for each electrolyzer. The merging method can be averaging, weighted averaging, or load forecast weighting of the energy consumption data at all load points. For example, the average energy consumption data of electrolyzer a at all available load points can be calculated using the full-range averaging method to obtain the energy consumption data for each electrolyzer, as shown in the following formula:

[0113] ;

[0114] In the formula, N(a) = MAX(a) - MIN(a) + 1 is the total number of load points.

[0115] Alternatively, a pre-defined electrolyzer load prediction model F=duty(x-1,x-2,…) can be used to predict the theoretical load of electrolyzer a within each historical ranking period x, assigning higher weights to load points close to the theoretical load. The pre-defined electrolyzer load prediction model can use the average load of multiple consecutive electrolyzers prior to historical ranking period x and the expected hydrogen production for each historical ranking period as inputs to predict the theoretical load of each electrolyzer within historical ranking period x.

[0116] In some embodiments, a Gaussian distribution centered on a preset electrolyzer load prediction model can be defined, as shown in the following formula:

[0117] ;

[0118] in This is a load fluctuation tolerance parameter, for example, 5%-10%.

[0119] In some embodiments, the weighting coefficients of each load point of the electrolytic cell within the historical sorting period x can be calculated based on the Gaussian distribution. The weighted average energy consumption data of electrolytic cell a across all available load points can be calculated using the weighted average method to obtain the energy consumption data for each electrolytic cell, as shown in the following formula:

[0120] ;

[0121] Based on the Gaussian distribution to characterize the confidence interval of load forecasting, load fluctuations are transformed into probabilistic weights, which better reflects the stochastic characteristics of actual operating scenarios. Differential weighting is achieved through load point weighting coefficients, strengthening the data contribution of high-frequency operating ranges, weakening interference from edge conditions, and improving the practicality of comprehensive energy efficiency indicators. The weighting coefficients are updated periodically with the forecasting model, automatically tracking equipment performance evolution, and the fusion of weighted average energy consumption data with predicted probabilities and actual measurements avoids misjudgments caused by extreme operating conditions, supporting more stable priority ranking. Overall, through probabilistic modeling of the load-energy consumption relationship, a statistically significant optimization basis is provided for the energy efficiency management of water electrolysis hydrogen production clusters under fluctuating environments.

[0122] S10233: Based on the comparison results of energy consumption data of multiple electrolyzers in the water electrolysis hydrogen production cluster, determine the energy consumption priority of the multiple electrolyzers.

[0123] In some embodiments, the energy consumption priority of the multiple electrolyzers is determined based on the comparison results of energy consumption data of multiple electrolyzers in the water electrolysis hydrogen production cluster.

[0124] Based on the global energy efficiency comparison of the water electrolysis hydrogen production cluster, the energy consumption priority of each electrolyzer is determined, laying a data foundation for subsequent dynamic adjustment of the operation strategy, ensuring that high-efficiency cells take the lead in bearing the core load, and low-efficiency cells serve as flexible adjustment units to maximize the system's energy efficiency.

[0125] The energy consumption data of multiple electrolyzers in the water electrolysis hydrogen production cluster can be sorted from smallest to largest within a historical sorting period x, generating an energy consumption sorting number EN(a). EN(a)=1 indicates the electrolyzer with the lowest energy consumption in historical sorting period x, and the numbers are arranged in ascending order. For example, when the energy consumption data is sorted from smallest to largest as electrolyzers e, d, and c, the corresponding numbers are EN(e)=1, EN(d)=2, and EN(c)=3.

[0126] In some embodiments, the first energy consumption data of each electrolyzer in the historical ranking period, the second energy consumption data in the historical measurement period, and the third energy consumption data in the current load switching period can be obtained; the energy consumption ranking data of each electrolyzer is determined based on the fusion result of the first energy consumption data, the second energy consumption data, and the third energy consumption data; and the energy consumption priority of the plurality of electrolyzers is determined based on the energy consumption ranking data of each electrolyzer in the water electrolysis hydrogen production cluster.

[0127] The first energy consumption data of the historical ranking period can reflect the long-term energy consumption characteristics under different loads. The second energy consumption data of the historical measurement period reflects the recent stability of the electrolyzer's energy consumption. The third energy consumption data of the current period shows the real-time energy consumption status of the electrolyzer. Combining the three can make the assessment of the electrolyzer's energy consumption more comprehensive and objective, and avoid misjudgment due to one-sided data.

[0128] Three types of energy consumption data can be acquired in different time periods: The first type of energy consumption data can come from historical sorting periods, including the total power consumption of each electrolyzer in that period, the power consumption per unit of hydrogen production, and the energy consumption change corresponding to load fluctuations. The data acquisition interval is, for example, 5 minutes / time to ensure coverage of energy consumption characteristics under different load conditions; The second type of energy consumption data can come from historical measurement periods, including not only basic power consumption data, but also energy consumption recovery curves after equipment maintenance and energy consumption deviation values ​​under different ambient temperatures; The third type of energy consumption data can be a real-time energy consumption snapshot of the current load switching period, including preheating energy consumption, instantaneous power consumption at the current load rate, and energy consumption deviation rate compared with the same period in history, reflecting the real-time energy efficiency status of the equipment.

[0129] A weighted fusion algorithm can be used to assign appropriate weights to the three types of data for fusion, which will not be elaborated here. When generating energy consumption ranking data based on the fusion results, a multi-dimensional scoring method can be used: for example, energy consumption per unit of hydrogen production can be used as the core indicator, and scores can be assigned in order of increasing value; supplemented by energy consumption volatility and energy efficiency degradation rate. When finally determining energy consumption priorities, the ranked data can be mapped to energy consumption priority levels.

[0130] Figure 7 This is a flowchart of a method for adjusting the start-up and shutdown state of an electrolytic cell, as provided in the embodiments of this specification. In specific implementation, it includes the following steps:

[0131] S10241: If the first operating scale data is greater than the second operating scale data, start multiple non-operating hot tanks in sequence according to temperature priority until the number of start-ups is equal to the difference between the first operating scale data and the second operating scale data; if the total number of non-operating hot tanks is less than the difference between the first operating scale data and the second operating scale data, start multiple non-operating cold tanks in sequence according to energy consumption priority until the number of start-ups is equal to the difference between the first operating scale data and the second operating scale data.

[0132] In some embodiments, if the first operating scale data is greater than the second operating scale data, multiple non-operating hot tanks are started sequentially according to temperature priority until the number of started tanks equals the difference between the first operating scale data and the second operating scale data; if the total number of non-operating hot tanks is less than the difference between the first operating scale data and the second operating scale data, multiple non-operating cold tanks are started sequentially according to energy consumption priority until the number of started tanks equals the difference between the first operating scale data and the second operating scale data.

[0133] When the target operating scale of the current load switching cycle q+1, i.e., the first operating scale data, is greater than the actual operating scale of the historical load switching cycle q, i.e., the second operating scale data, the electrolytic cell startup operation can be performed in the following order: Prioritize selecting non-operating electrolytic cells in a hot standby state, i.e., non-operating hot cells. Activate hot cells sequentially according to a preset temperature priority; the number of cells started is equal to the difference between the first and second operating scale data, denoted as Δ. If the total number of non-operating hot cells is less than the difference, i.e., the number of available hot cells < Δ, then switch to the cold cell startup mode. That is, activate cold cells sequentially according to energy consumption priority. The number of cold cells started is the difference between the difference and the number of already activated hot cells, ensuring that the total number of newly added cells equals Δ. For example, if Δ = 5 and only 3 hot cells are available, then after starting all 3 hot cells, start 2 optimal cold cells according to energy consumption priority.

[0134] Specifically, hot cells, due to their ability to maintain operating temperature, have short start-up times and low energy consumption, thus receiving absolute priority. Cold cells, on the other hand, require a preheating process, resulting in high start-up costs, and are only activated when hot cell resources are exhausted. Therefore, when the first operating scale data exceeds the second operating scale data, it means that the number of currently operating electrolyzers cannot meet the preset hydrogen production demand, requiring the activation of additional equipment. In this case, a temperature priority strategy can be prioritized to sort the unoperated hot cells. Hot cells can be defined as electrolyzers with short downtime and whose cell temperature remains at a high level. These devices, due to their high residual heat, can quickly recover to operating temperature, reducing start-up time and energy consumption. According to the preset temperature priority rules, the unoperated hot cells are activated sequentially until the number of activated equipment reaches the difference between the first and second operating scale data. If the total number of unoperated hot cells is less than this difference, it means that even after activating all hot cells, the required production capacity cannot be supplemented, and cold cells are activated as a supplement. At this point, an energy consumption priority strategy can be switched to evaluate and prioritize the unoperated cold cells. Prioritize starting up cold-state tanks with lower energy consumption per unit of hydrogen production and better energy efficiency. During startup, continuously monitor the total number of started equipment until the number of newly started equipment completely matches the difference between the first and second operating scale data, thereby ensuring that the actual hydrogen production of the water electrolysis hydrogen production cluster reaches the preset target.

[0135] S10242: If the multiple hot tanks that are not in operation have the same temperature priority, start the multiple hot tanks that are not in operation in sequence according to the energy consumption priority of the multiple hot tanks.

[0136] In some embodiments, if multiple non-operating hot baths have the same temperature priority, the multiple non-operating hot baths are started sequentially according to their energy consumption priority.

[0137] When multiple inactive hot cells have the same temperature priority (i.e., their downtime and residual temperature are similar, making it impossible to determine the startup order based on hot-start efficiency), an energy consumption priority strategy can be used to determine the startup order. This means prioritizing the hot cell with the lowest unit hydrogen production power consumption and optimal energy efficiency to ensure that while meeting capacity expansion needs, the overall system energy consumption is minimized. If equipment with the same energy consumption priority exists, maintenance records and remaining lifespan can be further considered to prioritize the electrolyzer in better condition, achieving efficient resource utilization and balanced equipment lifespan management.

[0138] S10243: If the first operating scale data is less than the second operating scale data, the multiple hot-state tanks in operation are shut down in reverse order according to temperature priority until the number of shut-down tanks is equal to the difference between the second operating scale data and the first operating scale data; if the total number of hot-state tanks in operation is less than the difference between the second operating scale data and the first operating scale data, the multiple cold-state tanks in operation are shut down in reverse order according to energy consumption priority until the number of shut-down tanks is equal to the difference between the second operating scale data and the first operating scale data.

[0139] In some embodiments, if the first operating scale data is less than the second operating scale data, multiple hot-state tanks are shut down in reverse order according to temperature priority until the number of shut-down tanks equals the difference between the second operating scale data and the first operating scale data; if the total number of operating hot-state tanks is less than the difference between the second operating scale data and the first operating scale data, multiple cold-state tanks are shut down in reverse order according to energy consumption priority until the number of shut-down tanks equals the difference between the second operating scale data and the first operating scale data.

[0140] While meeting the requirements for scale reduction, a tiered shutdown strategy can be used to achieve optimal system energy efficiency and equipment lifespan protection, avoiding frequent start-ups and shutdowns of cold-state cells. When the target operating scale of the current load switching cycle q+1 is less than the actual operating scale of the historical load switching cycle q, i.e., the first operating scale data R(q+1) is less than the second operating scale data R(q), the electrolytic cell startup operation can be performed in the following order: Prioritize the currently operating cold-state cells, i.e., shut down cold-state cells sequentially according to the reverse order of energy consumption priority. The number of cells shut down is equal to the difference between the second and first operating scale data, denoted as Δ. If the total number of operating cold-state cells is less than the difference, i.e., the number of operating cold-state cells < Δ, the system can automatically switch to the hot-state cell shutdown mode. That is, the hot-state cells can be shut down sequentially according to the reverse order of preset temperature priority. The number of cells shut down is the difference between the difference and the number of already shut-down cold-state cells, ensuring that the total number of shut-down cells equals Δ. For example, if Δ=5 and only 3 cold tanks are running, then all 3 cold tanks should be shut down in reverse order, and then the 2 hot tanks with the highest cost should be shut down in reverse order according to temperature priority.

[0141] Specifically, shutting down the most energy-intensive operating cold cells maximizes the retention of high-efficiency production units. Hot cells with low temperature priority, which can be considered electrolyzers with late start-up times or short cumulative operating times, can be prioritized for shutdown to reduce temperature maintenance losses and operating costs. Through a dual optimization mechanism of reverse-order shutdown of cold cells based on energy consumption and reverse-order shutdown of hot cells based on priority, the need for scale reduction is met while maximizing the retention of high-value hot production units and reducing operating costs. Therefore, when the first operating scale data is less than the second operating scale data, it indicates that the number of currently operating electrolyzers exceeds the actual hydrogen production demand, requiring the reasonable shutdown of some equipment to reduce energy consumption and operating costs. In this case, an energy consumption priority strategy can be prioritized to evaluate and screen operating cold cells. Through real-time monitoring and historical data analysis, cold cells with high energy consumption and low efficiency can be prioritized for shutdown. The system can sequentially mark the devices to be shut down according to the order of unit hydrogen production power consumption (highest to lowest) and energy efficiency ratio (lowest to highest) of each cold-state tank, continuously executing shutdown operations until the number of shut-down cold-state tanks equals the difference between the second and first operating scale data. If, after all operating cold-state tanks are shut down, the total number of shut-down devices still does not reach the target difference (i.e., the total number of operating hot-state tanks is less than the required shutdown number), a temperature priority reverse order strategy can be switched to process the remaining hot-state tanks. The temperature priority reverse order logic focuses on equipment restart costs and efficiency, prioritizing the shutdown of hot-state tanks that require long preheating times and high energy consumption to reach operating status. Based on temperature priority, hot-state tanks with long preheating times and high energy consumption can be arranged sequentially and shut down gradually until the total number of shut-down devices completely matches the difference between the second and first operating scale data, thereby achieving efficient and energy-saving operation of the water electrolysis hydrogen production cluster while meeting hydrogen production needs.

[0142] S10244: If multiple hot tanks in operation have the same temperature priority, the multiple hot tanks in operation shall be shut down in reverse order according to their energy consumption priority.

[0143] In some embodiments, if the multiple hot tanks in operation have the same temperature priority, the multiple hot tanks in operation are shut down in reverse order according to their energy consumption priority.

[0144] When multiple hot tanks in operation have the same temperature priority, it means that the time and energy consumption for each device to recover to working status after shutdown are basically the same, making it difficult to determine the shutdown order based on hot-start characteristics. In this case, energy consumption priority can be used as the core decision-making basis to reverse-select the equipment that needs to be shut down. That is, the hot tank with the highest unit hydrogen production power consumption and the lowest energy efficiency ratio can be prioritized for shutdown. These devices often have potential problems such as electrode aging and electrolyte concentration imbalance, and continued operation will lead to a significant increase in energy costs. If the energy consumption data of some hot tanks are similar, the maintenance frequency and fault warning information of the equipment can be further compared. Priority can be given to shutting down the equipment with high maintenance requirements and high potential fault risks, so as to ensure that while reducing production capacity, the dual goals of cluster energy consumption optimization and equipment life cycle management are achieved, and the efficient and stable operation of the water electrolysis hydrogen production cluster is maintained.

[0145] Figure 8 This is a flowchart illustrating a method for selecting a load allocation scheme for a water electrolysis hydrogen production cluster, as provided in the embodiments of this specification. In practice, it includes the following steps:

[0146] S1031: Based on the load data of the multiple operating electrolyzers during the historical load switching cycle and the similarity of the load data of the multiple electrolyzers in each candidate load allocation scheme, calculate the load fluctuation data of each candidate load allocation scheme.

[0147] In some embodiments, Kronecker can be used based on the load data of the plurality of operating electrolyzers during historical load switching cycles and the load data of the plurality of electrolyzers in each candidate load allocation scheme. The function calculates the mismatch number of multiple electrolyzers in each candidate load allocation scheme; based on the mismatch number of multiple electrolyzers in each candidate load allocation scheme, it calculates the load fluctuation corresponding to each candidate load allocation scheme.

[0148] Using Kronek The binary nature of the function allows for the rapid identification of electrolytic cells requiring load adjustment, transforming a continuous optimization problem into a discrete state matching problem and significantly reducing computational complexity. Furthermore, the mismatch number directly corresponds to the actual number of electrolytic cells requiring operation, facilitating practical verification and debugging.

[0149] Kronek The function is a binary discrete function with two input parameters. Its output is 1 when the two input parameters are identical, and 0 otherwise. Therefore, Kronecker can be used. The function calculates the matching number between multiple electrolyzers in a water electrolysis hydrogen production cluster and multiple electrolyzers in each candidate load allocation scheme within a historical load switching cycle. Specifically, it calculates the number of electrolyzers whose load in each candidate load allocation scheme remains unchanged relative to the historical load switching cycle. The specific calculation can be performed using the following formula:

[0150] ;

[0151] In the formula, For the first historical load switching cycle Load data for each electrolytic cell For the candidate load allocation scheme, the first The load data for each electrolyzer is obtained. Furthermore, based on the total number N of electrolyzers in the water electrolysis hydrogen production cluster, the mismatch number of multiple electrolyzers in each candidate load allocation scheme can be calculated, thus yielding the load fluctuation corresponding to each candidate load allocation scheme.

[0152] .

[0153] In some embodiments, the load fluctuation corresponding to each candidate load allocation scheme is calculated based on the load data of the plurality of operating electrolyzers during the historical load switching cycle and the similarity of the load data of the plurality of electrolyzers in each candidate load allocation scheme, as well as the energy consumption priority and temperature priority of the plurality of electrolyzers.

[0154] For multiple pre-generated candidate load allocation schemes, the multidimensional similarity between each scheme and the actual load distribution during historical load switching cycles can be calculated. Methods for calculating multidimensional similarity include Euclidean distance sum of squares, Manhattan distance sum, and cosine distance, which will not be elaborated here. Furthermore, higher weights can be assigned to high-efficiency cells based on energy consumption priorities to ensure priority load matching for critical electrolyzers, and higher weights can be assigned to hot cells based on the adjustment costs of hot and cold cells to ensure priority load matching for hot cells.

[0155] In some embodiments, the load fluctuation corresponding to each candidate load allocation scheme can be calculated using the following formula:

[0156] ;

[0157] In the formula, Indicates the number of historical load switching cycles. Energy consumption priority of each electrolytic cell Indicates the number of historical load switching cycles. Temperature priority of each electrolytic cell These represent the historical load switching cycles. Each electrolytic cell is either a hot cell or a cold cell. This represents the preset weighting coefficients. For the first historical load switching cycle Load data for each electrolytic cell For the candidate load allocation scheme, the first Load data for each electrolytic cell.

[0158] S1032: Select the candidate load allocation scheme with the smallest load fluctuation data as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

[0159] In some embodiments, the candidate load allocation scheme with the smallest load fluctuation data is selected as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

[0160] By quantifying the similarity between candidate solutions and the current load distribution, and prioritizing the solution with the smallest adjustment range, the impact of power surges on the power grid can be effectively suppressed. Minimizing load fluctuations reduces the frequency of electrolyzer operating state switching, implicitly preserving the steady-state operating range of high-efficiency cells, avoiding the additional losses caused by forced switching from inefficient conditions, and thus preventing accelerated material aging due to frequent thermal cycling. Furthermore, the similarity index calculated based on real-time data can adapt to wind and solar power fluctuations, maintaining the stability of the hydrogen production process. Overall, gradual adjustments based on the load redistribution process can balance system safety and operational economy.

[0161] Each candidate load allocation scheme corresponds to a load fluctuation value. The smaller the value, the higher the stability of the corresponding candidate load allocation scheme in the electrolyzer load allocation. Therefore, all candidate load allocation schemes can be sorted in ascending order of load fluctuation value, and the candidate load allocation scheme with the smallest load fluctuation can be selected as the load allocation scheme for the water electrolysis hydrogen production cluster.

[0162] In addition, if multiple schemes have the same fluctuation value, you can choose the scheme with higher participation in the hot bath to reduce the adjustment loss of the cold bath; or the scheme with more uniform historical running time to extend the equipment life.

[0163] In some embodiments, if the load fluctuations of multiple candidate load allocation schemes are the same, the load fluctuation of each electrolyzer in each candidate load allocation scheme is calculated; based on the load fluctuation of each electrolyzer in each candidate load allocation scheme, the maximum load fluctuation of the electrolyzer and the energy consumption priority of the electrolyzer to which the maximum load fluctuation belongs are determined; the candidate load allocation scheme with the lowest energy consumption priority is used as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

[0164] When multiple candidate load allocation schemes exhibit similar load fluctuations, a detailed analysis at the electrolyzer level can be used to determine the optimal scheme. This involves precisely calculating the load fluctuation of each electrolyzer in each candidate load allocation scheme, specifically the difference between the load of each electrolyzer during historical load switching cycles and the load of each electrolyzer in the candidate load allocation scheme. Subsequently, the electrolyzer with the largest load fluctuation is selected from each candidate scheme, and its energy consumption priority is associated with it. In the final decision, the energy consumption priorities of the electrolyzers with the largest load fluctuations in each candidate load allocation scheme are compared, and the candidate scheme with the lowest corresponding energy consumption priority is selected as the final load allocation scheme. If multiple schemes have the same energy consumption priority for the electrolyzers with the largest fluctuations, the energy consumption performance of the second-largest fluctuating electrolyzers or the load fluctuation balance of all electrolyzers can be further compared. This ensures that, under the premise of consistent load fluctuations, the scheme with lower fluctuations in high-energy-consuming equipment and better system energy efficiency is prioritized, achieving dual optimization of equipment lifespan and energy costs. For example, when increasing the total load, the load of electrolyzers with lower energy consumption priorities is adjusted first, rather than low-energy-consuming electrolyzers with higher energy consumption priorities.

[0165] In some embodiments, the determination of multiple operating electrolyzers and load allocation can be performed synchronously. In this synchronous mode, the start-up and shutdown decisions of the electrolyzers and load allocation parameters can be incorporated into the same optimization model and solved simultaneously using a multi-objective algorithm. For example, when constructing a model with the objective of minimizing total energy consumption, the model outputs both the number and specific numbers of the electrolyzers to be put into operation, and simultaneously provides the load allocation ratio for each operating cell. During the model solution process, the energy efficiency characteristics and start-up and shutdown costs of the equipment can be correlated in real time, ensuring that the selection of operating cells and load allocation are mutually compatible—for example, prioritizing the allocation of a high proportion of load to the already determined high-efficiency hot cells, while simultaneously using load demand to deduce the required number of operating cells, avoiding the problem of an excessive number of operating cells leading to low-load, inefficient operation or an insufficient number to meet hydrogen production demand. This synchronous mode helps improve decision-making efficiency and quickly respond to grid power fluctuations or sudden changes in hydrogen production demand, and is particularly suitable for highly dynamic scenarios such as renewable energy hydrogen production.

[0166] In other embodiments, the determination and load allocation of multiple operating electrolyzers can also be performed asynchronously. The asynchronous mode can be divided into a two-step process: first determining the operating cells, then allocating the load. In the first step, based on constraints such as preset hydrogen production and grid absorption capacity, combined with the temperature priority and basic energy efficiency level of the electrolyzers, a list of electrolyzers to be operated in this round is determined. At this stage, specific load allocation is not involved; only the start-up and shutdown status of the equipment is determined. In the second step, within the determined operating cell range, a specific load ratio is allocated to each operating cell based on a refined energy consumption optimization model to ensure that total energy consumption is minimized. This asynchronous mode can more clearly define the decision-making logic and facilitate the introduction of different optimization objectives in stages. For example, when determining the operating cells, the focus can be on equipment start-up speed and stability, while when allocating the load, the focus can be on refined energy consumption control, making it suitable for scenarios with high requirements for equipment lifespan management. Furthermore, the asynchronous mode can flexibly handle system failures: if a determined operating cell suddenly fails, the load of other operating cells can be readjusted only during the load allocation stage, without recalculating the entire list of operating cells, thus improving the system's fault tolerance.

[0167] In some embodiments, a water electrolysis hydrogen production cluster may include multiple electrolyzer groups. The load of the water electrolysis hydrogen production cluster can be allocated on a group basis.

[0168] Figure 9 This is a flowchart of a method for determining the start-up priority of an electrolytic cell group, as provided in the embodiments of this specification. In specific implementation, it includes the following steps:

[0169] S901: Calculate the energy consumption priority of each electrolytic cell group based on the energy consumption data of multiple electrolytic cells in each electrolytic cell group.

[0170] In some implementation examples, the energy consumption priority of each electrolytic cell group is calculated based on the energy consumption data of multiple electrolytic cells in each electrolytic cell group.

[0171] In a water electrolysis hydrogen production cluster, each electrolyzer group can contain multiple electrolyzers. The startup priority of each electrolyzer group can be calculated by collecting energy consumption and temperature data from all electrolyzers within each group, combined with the overall operating status of the group. A weighted comprehensive scoring method can be used, assigning different weights to energy consumption and temperature data to construct a scoring model. The startup priority score of each electrolyzer group can be expressed as a weighted sum of energy consumption and temperature scores. The energy consumption score can be assigned based on the overall unit hydrogen production power consumption of the electrolyzer group, ranked from low to high. The temperature score can be assessed considering the proportion of hot cells in the electrolyzer group and the overall temperature stability of the group; a higher score indicates a higher startup priority. Alternatively, the energy consumption priority can be calculated based on the inverse sum of the total energy consumption data of all electrolyzers in each group; that is, the larger the sum of the inverses of the total energy consumption data, the higher the energy consumption priority.

[0172] S902: Calculate the temperature priority and thermal stability of each electrolytic cell group based on the temperature data of multiple electrolytic cells in each group.

[0173] In some embodiments, the temperature priority and thermal stability of each electrolytic cell group are calculated based on the temperature data of multiple electrolytic cells in each electrolytic cell group.

[0174] The degree of temperature fluctuation among multiple electrolyzers in each electrolyzer group can be used as the basis for calculating thermal stability; that is, the greater the fluctuation, the lower the thermal stability. Furthermore, based on the temperature data of multiple electrolyzers in each electrolyzer group, the temperature priority of each electrolyzer group can be calculated using the following formula:

[0175] ;

[0176] In the formula, Δt is the average preheating time for each electrolytic cell group, and τ is the thermal time constant. Temperature priority can describe the effect of the preheating degree of each electrolytic cell group on energy efficiency. When Δt is much greater than τ, Approximately equal to 1, the negative impact of the temperature field on energy efficiency is negligible; when Δt is much smaller than τ, Approaching 0, it suppresses the priority start-up of cold-state equipment, reducing preheating energy consumption. The thermal time constant τ reflects the thermal inertia of the equipment and can be determined through experiments or simulations, which will not be elaborated here.

[0177] S903: Calculate the start-up priority of each electrolytic cell group based on its energy consumption priority, temperature priority, and thermal stability.

[0178] Prioritizing electrolytic cell groups reduces the optimization complexity of large-scale clusters and improves computational efficiency. By coupling energy consumption priority, temperature priority, and thermal stability in a three-dimensional manner, the limitations of traditional single-index ranking are overcome, achieving synergistic optimization of economy, agility, and safety. The contribution weights of the three dimensions can be automatically adjusted according to the operating scenario, forming an adaptive inter-group scheduling strategy. Furthermore, independent calculation of priorities for each dimension avoids cross-interference, ensuring that high-temperature, high-efficiency cell groups obtain the optimal comprehensive score, while also identifying potential risk groups.

[0179] In some embodiments, the startup priority of each electrolytic cell group can be calculated using the following formula based on the energy consumption priority, temperature priority, and thermal stability of each electrolytic cell group:

[0180] ;

[0181] In the formula, This indicates the energy consumption priority of each electrolytic cell group. This indicates the thermal stability of each electrolytic cell group. This indicates the temperature priority of each electrolytic cell group. By sorting the start-up priorities of multiple electrolytic cell groups, the start-up order can be determined, prioritizing hot start-up, high efficiency, and stability.

[0182] Figure 10 This is a flowchart of a load adjustment method for an electrolytic cell group provided in the embodiments of this specification. In specific implementation, it includes the following steps:

[0183] S1001: Determine multiple candidate load allocation schemes for the water electrolysis hydrogen production cluster based on the startup priority of multiple electrolyzer groups.

[0184] In some embodiments, multiple candidate load allocation schemes for the water electrolysis hydrogen production cluster are determined based on the startup priority of multiple electrolyzer groups.

[0185] Based on the startup priority of each electrolytic cell group, multiple candidate load allocation schemes can be generated. When formulating a scheme, load can be preferentially allocated to cells with higher startup priority, while also considering the capacity differences and equipment characteristics of different cells. For example, high-priority cells can handle 60%-80% of the base load. Low-priority cells are responsible for peak shaving or standby. The generated candidate schemes include different combinations of load allocation ratios. For example, Scheme 1 assigns 70% load to high-priority cells, with medium and low-priority cells each handling 15%. Scheme 2 adjusts this to 80% load for high-priority cells and 20% for low-priority cells; further details are omitted here.

[0186] S1002: Based on the load data of multiple electrolyzers in each electrolyzer group, select the candidate load allocation scheme with the smallest load fluctuation as the load allocation scheme for the water electrolysis hydrogen production cluster.

[0187] In some embodiments, based on the load data of multiple electrolyzers in each electrolyzer group, the candidate load allocation scheme with the smallest load fluctuation is selected as the load allocation scheme for the water electrolysis hydrogen production cluster.

[0188] By minimizing load fluctuations between groups, the overall power jump of the cluster can be reduced, grid compatibility can be enhanced, and an effective balance between computational efficiency and control accuracy can be achieved.

[0189] The system can select the optimal solution from multiple candidate schemes, monitor the load data of each electrolytic cell group in real time, and calculate the load fluctuation value of each electrolytic cell group after the implementation of each candidate scheme. By comparing the load fluctuation data of each scheme, the candidate scheme with the smallest load fluctuation is selected as the final load allocation scheme.

[0190] Specifically, the load fluctuation value of each electrolyzer in each candidate scheme can be calculated using standard deviation or range. By comparing the overall load fluctuation amplitude of the electrolyzer group, the scheme with the smallest fluctuation can be selected. If the fluctuations of multiple schemes are the same, the fluctuation distribution of individual cells can be further analyzed: the energy consumption priority of the electrolyzer with the largest load fluctuation in each scheme can be calculated, and the scheme with the smallest energy consumption priority of the cell with the largest fluctuation can be selected; if they are still the same, the load fluctuation balance of all cells can be compared to ensure that the load distribution is both stable and energy efficiency optimized.

[0191] In some embodiments, the load adjustment of any electrolyzer group in the water electrolysis hydrogen production cluster can be performed according to the above method.

[0192] S1003: Adjust the load of multiple electrolytic cells in each electrolytic cell group according to the load distribution scheme.

[0193] In some embodiments, the load of multiple electrolytic cells in each electrolytic cell group is adjusted according to the load allocation scheme.

[0194] After determining the load allocation scheme for each electrolytic cell group, the load can be further redistributed according to the energy consumption priority and temperature priority of different electrolytic cells in each electrolytic cell group, which will not be elaborated here.

[0195] As can be seen from the above, the load adjustment method for the water electrolysis hydrogen production cluster provided in the embodiments of this specification can calculate the temperature priority of each electrolyzer based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster; the temperature priority of the electrolyzer is positively correlated with its temperature data; based on hydrogen production demand data or load consumption data, multiple operating electrolyzers are determined among the multiple electrolyzers using the temperature priority; an electrolyzer energy consumption scheduling optimization model is established according to a preset optimization algorithm; using the hydrogen production demand data or load consumption data as constraints, the total energy consumption of the multiple operating electrolyzers is minimized using the electrolyzer energy consumption scheduling optimization model to obtain a load allocation scheme; and the load of the multiple operating electrolyzers is adjusted according to the load allocation scheme. Compared with existing methods, the embodiments of this specification can determine the temperature priority of different electrolyzers based on the temperature distribution characteristics of multiple electrolyzers in the water electrolysis hydrogen production cluster, and can use hot start priority logic to reduce the additional energy loss of shutdown and restart under the premise of optimal energy consumption. The embodiments in this specification integrate energy-optimization logic and hot-start priority logic, which can optimize the load distribution of the water electrolysis hydrogen production cluster in real time, reduce the increase in energy consumption and the shortening of equipment life caused by frequent start-stop, and greatly improve the efficiency of water electrolysis hydrogen production.

[0196] The following are two specific embodiments of this specification:

[0197] Taking the A-type water electrolysis hydrogen production cluster consisting of 32 alkaline electrolyzers as an example, the system parameters are as follows:

[0198] Time period: Sorting period M = 1440 minutes, measurement period P = 15 minutes, load switching period Q = 10 minutes. Load range: MIN = 30%, MAX = 90% for all electrolytic cells, with discrete load points at 30%, 31%, ..., 90%. Energy consumption threshold: S% = 5%, temperature threshold: T_cold = 50℃, adjustment step: Δbmax = 5%, energy consumption deviation: sc = 1%. Initial energy consumption data: Initial EC0(a,b) for each cell was obtained through a 72-hour benchmark test. Some example data are shown in Table 1.

[0199] Table 1

[0200]

[0201] 1. When the load switching cycle arrived, a load switch was performed for a period of time m to m+1.

[0202] Initial state, i.e., time period m. Operating electrolyzers: E1-E20, a total of 20 units, with a total hydrogen production H2Prod(m) = 12,000 Nm³ / h. Shutdown electrolyzers: E21-E32, where E21-E25 are in hot standby mode (HOTBACK=1, temperature T≥60℃), and E26-E32 are in cold state (T<50℃). Load allocation: Each operating cell operates at 70% load, with a unit energy consumption of approximately 3.95-4.10 kWh / Nm³. Demand change, i.e., time period m+1: A sudden increase in wind and solar power generation increases the total hydrogen production demand to H2Prod(m+1) = 15,000 Nm³ / h, requiring the addition of 5 electrolyzers. Startup priority determination: Hot standby slots HN(a) sorting: E21 (T=68℃) → E22 (T=65℃) → E23 (T=63℃) → E24 (T=61℃) → E25 (T=59℃). Cold standby slots EN(a) sorting: E26 (EC_avg=4.05) → E27 (EC_avg=4.08) → E28 (EC_avg=4.12) → E32 (EC_avg=4.35). Startup logic: 5 units need to be started. HN(a)=1-3 are started first, i.e., E21-E23. The remaining 2 units are started with EN(a)=1-2, i.e., E26-E27. Final operating slots: E1-E20 + E21-E23 + E26-E27, a total of 25 units. Load allocation optimization: Minimize total energy consumption, constrained H2Prod = 15,000 Nm³ / h. The optimal load allocation is obtained by solving for:

[0203] E1-E20: The load is increased from 70% to 72%, Δb=2%≤Δb_max.

[0204] E21 - E23: The starting back - load is set to 68%. E26 - E27: The starting back - load is set to 65%. Total energy consumption: PowerSum = 15,000×3.98 kWh / Nm³ = 59,700 kWh / h.

[0205] Result comparison: Traditional rotation strategy: PowerSum = 15,000×4.12 kWh / Nm³ = 61,800 kWh / h. The present invention saves 2,100 kWh / h of energy, a reduction of 3.4%.

[0206] 2. Re - sorting at the end of the sorting cycle.

[0207] During the 24th sorting cycle, electrolyzer E1 runs for 8 hours at a load of 70%. Calculate its average energy consumption. Comparing with the previous cycle EC(23, E1, 70%) = 3.92 kWh / Nm³, it is determined to be normal aging, and update EC(E1, 70%) = 3.98.

[0208] Predict the next cycle load duty(25) = 75% using the predicted load weighting method. The sorting results are shown in Table 2:

[0209] Table 2

[0210]

[0211] Temperature priority update: After shutdown, the temperatures of E21 - E25 are 62℃, 58℃, 55℃, 51℃, 49℃ respectively. Among them, for E25, T = 49℃ < Tcold, it exits the hot standby state. HN(a) sorting update: E21(62℃) → E22(58℃) → E23(55℃) → E24(51℃).

[0212] Based on the above - mentioned load adjustment method for the electrolytic water hydrogen production cluster, this specification also presents an embodiment of the load adjustment device for the electrolytic water hydrogen production cluster. As Figure 11 shown, the load adjustment device 1100 for the electrolytic water hydrogen production cluster may specifically include the following modules:

[0213] Calculation module 1101, configured to calculate the temperature priority of each electrolyzer according to the temperature data of multiple electrolyzers in the electrolytic water hydrogen production cluster; the temperature priority of the electrolyzer is positively correlated with its temperature data.

[0214] Determination module 1102, configured to determine multiple operating electrolyzers among the multiple electrolyzers based on the hydrogen production demand data or load consumption data, using the temperature priority.

[0215] Establishment module 1103, configured to establish an electrolyzer energy consumption scheduling optimization model according to a preset optimization algorithm.

[0216] The optimization module 1104 is used to minimize the total energy consumption of multiple operating electrolyzers by using the electrolyzer energy consumption scheduling optimization model with hydrogen production demand data or load consumption data as constraints, and obtain a load allocation scheme.

[0217] Adjustment module 1105 is used to adjust the load of multiple operating electrolytic cells according to the load distribution scheme.

[0218] In some embodiments, the temperature data includes first temperature data of each electrolyzer in a historical sorting cycle, second temperature data in a historical measurement cycle, and third temperature data in the current load switching cycle; the calculation module 1101 described above can specifically be used for:

[0219] Based on the fusion results of the first temperature data, the second temperature data, and the third temperature data, the fourth temperature data for each electrolytic cell is determined.

[0220] Based on the fourth temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster, the multiple electrolyzers are divided using a preset temperature threshold to obtain multiple hot cells and multiple cold cells.

[0221] Based on the comparison results of the temperature data of the multiple hot baths, the temperature priority of the multiple hot baths is determined.

[0222] In some embodiments, the determining module 1102 described above can be specifically used for:

[0223] Based on hydrogen production demand data or load absorption data, calculate the first operating scale data of the electrolyzer during the current load switching cycle;

[0224] Obtain the second operating scale data of the electrolyzer during historical load switching cycles;

[0225] Obtain the energy consumption priority of each electrolyzer within the historical sorting period;

[0226] Based on the comparison results of the first and second operating scale data and the temperature and energy consumption priorities of the multiple electrolytic cells, multiple operating electrolytic cells are determined among the multiple electrolytic cells.

[0227] In some embodiments, the determining module 1102 described above can also be used for:

[0228] The process of obtaining the energy consumption priority of each electrolyzer within the historical sorting period includes:

[0229] Acquire energy consumption data for each load point of each electrolyzer within multiple historical measurement periods during the historical sorting period;

[0230] Based on the fusion results of energy consumption data of each load point of each electrolyzer over multiple historical measurement periods, the energy consumption data of each electrolyzer is determined.

[0231] Based on the comparison results of energy consumption data of multiple electrolyzers in the water electrolysis hydrogen production cluster, the energy consumption priority of the multiple electrolyzers is determined.

[0232] In some embodiments, the determining module 1102 described above can also be used for:

[0233] If the first operating scale data is greater than the second operating scale data, the multiple non-operating hot tanks are started sequentially according to temperature priority until the number of start-up tanks is equal to the difference between the first operating scale data and the second operating scale data.

[0234] If the total number of non-operational hot tanks is less than the difference between the first and second operating scale data, multiple non-operational cold tanks will be started sequentially according to energy consumption priority until the number of started tanks equals the difference between the first and second operating scale data.

[0235] In some embodiments, the determining module 1102 described above can also be used for:

[0236] If the first operating scale data is less than the second operating scale data, the multiple cold tanks in operation are shut down in reverse order according to energy consumption priority until the number of shut-down tanks is equal to the difference between the second operating scale data and the first operating scale data.

[0237] If the total number of operating hot tanks is less than the difference between the second operating scale data and the first operating scale data, the multiple operating hot tanks are shut down in reverse order according to temperature priority until the number of shut-down tanks equals the difference between the second operating scale data and the first operating scale data.

[0238] In some embodiments, the optimization module 1104 described above can be specifically used for:

[0239] Using hydrogen production demand data or load consumption data as constraints, the total energy consumption of multiple operating electrolyzers is minimized using an electrolyzer energy consumption scheduling optimization model, thereby obtaining the minimum energy consumption load allocation scheme for the multiple operating electrolyzers.

[0240] Based on the preset energy consumption deviation, multiple candidate load allocation schemes corresponding to the minimum energy consumption load allocation scheme are generated;

[0241] Based on the load data of the multiple operating electrolyzers during the historical load switching cycle, the candidate load allocation scheme with the smallest relative load change is selected from the multiple candidate load allocation schemes as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

[0242] In some embodiments, the optimization module 1104 described above can also be used for:

[0243] Based on the load data of the multiple operating electrolyzers during the historical load switching cycle and the similarity of the load data of the multiple electrolyzers in each candidate load allocation scheme, the load fluctuation data of each candidate load allocation scheme is calculated.

[0244] The candidate load allocation scheme with the smallest load fluctuation data is selected as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

[0245] This specification also provides a computer device for a load adjustment method of a water electrolysis hydrogen production cluster, including a processor and a memory for storing processor-executable instructions. Specifically, the processor can perform the following tasks according to the instructions: calculating the temperature priority of each electrolyzer based on temperature data from multiple electrolyzers in the water electrolysis hydrogen production cluster; the temperature priority of an electrolyzer is positively correlated with its temperature data; determining multiple operating electrolyzers based on hydrogen production demand data or load consumption data using the temperature priority; establishing an electrolyzer energy consumption scheduling optimization model according to a preset optimization algorithm; minimizing the total energy consumption of the multiple operating electrolyzers using the electrolyzer energy consumption scheduling optimization model with hydrogen production demand data or load consumption data as constraints, thereby obtaining a load allocation scheme; and adjusting the load of the multiple operating electrolyzers according to the load allocation scheme.

[0246] To execute the above instructions more accurately, please refer to... Figure 12 As shown in the embodiments of this specification, another specific computer device 1200 is also provided, wherein the computer device 1200 includes a network communication port 1201, a processor 1202 and a memory 1203, and the above structures are connected by internal cables so that the various structures can perform specific data interaction.

[0247] The processor 1202 can be specifically used to calculate the temperature priority of each electrolyzer based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster; the temperature priority of an electrolyzer is positively correlated with its temperature data; based on hydrogen production demand data or load consumption data, multiple operating electrolyzers are determined from the multiple electrolyzers using the temperature priority; an electrolyzer energy consumption scheduling optimization model is established according to a preset optimization algorithm; using the hydrogen production demand data or load consumption data as constraints, the electrolyzer energy consumption scheduling optimization model is used to minimize the total energy consumption of the multiple operating electrolyzers to obtain a load allocation scheme; and the load of the multiple operating electrolyzers is adjusted according to the load allocation scheme.

[0248] The memory 1203 can be used to store the corresponding instruction program.

[0249] In this embodiment, the network communication port 1201 can be a virtual port bound to different communication protocols, thereby enabling the sending or receiving of different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0250] In this embodiment, the processor 1202 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. This specification is not limiting.

[0251] In this embodiment, the memory 1203 includes volatile memory and non-volatile memory. The memory 1203 can include multiple layers. In digital systems, anything that can store binary data can be a memory; in integrated circuits, a circuit with storage function but no physical form is also called a memory, such as RAM, FIFO, etc.; in a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0252] This specification also provides a computer program product, including at least one instruction or at least one program segment, wherein the at least one instruction or the at least one program segment is loaded and executed by a processor to achieve the following: Figure 1 The method shown.

[0253] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0254] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0255] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0256] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational tasks to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The task is a function specified in one or more boxes.

[0257] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. 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 load adjustment method for a water electrolysis hydrogen production cluster, characterized in that, include: Based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster, calculate the temperature priority of each electrolyzer. The temperature priority of the electrolyzer is positively correlated with its temperature data; the temperature data includes the first temperature data of each electrolyzer in the historical sorting period, the second temperature data in the historical measurement period, and the third temperature data in the current load switching period; The step of calculating the temperature priority of each electrolyzer based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster includes: determining the fourth temperature data of each electrolyzer based on the fusion result of the first temperature data, the second temperature data, and the third temperature data; dividing the multiple electrolyzers into multiple hot-state cells and multiple cold-state cells using a preset temperature threshold based on the fourth temperature data of the multiple electrolyzers in the water electrolysis hydrogen production cluster; and determining the temperature priority of the multiple hot-state cells based on the comparison result of the temperature data of the multiple hot-state cells. Based on hydrogen production demand data or load consumption data, multiple operating electrolyzers are determined from the multiple electrolyzers using temperature priority. This determination includes: calculating first operating scale data for the electrolyzers within the current load switching cycle based on the hydrogen production demand data or load consumption data; obtaining second operating scale data for the electrolyzers within historical load switching cycles; obtaining the energy consumption priority of each electrolyzer within historical sorting cycles; and determining multiple operating electrolyzers based on the comparison results of the first and second operating scale data, as well as the temperature priority and energy consumption priority of the multiple electrolyzers. An energy consumption scheduling optimization model for electrolytic cells is established based on a preset optimization algorithm. Using hydrogen production demand data or load consumption data as constraints, the total energy consumption of multiple operating electrolyzers is minimized by an electrolyzer energy consumption scheduling optimization model to obtain a load allocation scheme. Adjust the load of multiple operating electrolytic cells according to the load distribution plan.

2. The method according to claim 1, characterized in that, Each of the electrolytic cells includes multiple load points; The process of obtaining the energy consumption priority of each electrolyzer within the historical sorting period includes: Acquire energy consumption data for each load point of each electrolyzer within multiple historical measurement periods during the historical sorting period; Based on the fusion results of energy consumption data of each load point of each electrolyzer over multiple historical measurement periods, the energy consumption data of each electrolyzer is determined. Based on the comparison results of energy consumption data of multiple electrolyzers in the water electrolysis hydrogen production cluster, the energy consumption priority of the multiple electrolyzers is determined.

3. The method according to claim 1, characterized in that, The step of determining multiple operating electrolytic cells from among the multiple electrolytic cells based on the comparison results of the first and second operating scale data and the temperature and energy consumption priorities of the multiple electrolytic cells includes: If the first operating scale data is greater than the second operating scale data, the multiple non-operating hot tanks are started sequentially according to temperature priority until the number of start-up tanks is equal to the difference between the first operating scale data and the second operating scale data. If the total number of non-operational hot tanks is less than the difference between the first and second operating scale data, multiple non-operational cold tanks will be started sequentially according to energy consumption priority until the number of started tanks equals the difference between the first and second operating scale data.

4. The method according to claim 1, characterized in that, The step of determining multiple operating electrolytic cells from among the multiple electrolytic cells based on the comparison results of the first operating scale data and the second operating scale data, as well as the temperature priority and energy consumption priority of the multiple electrolytic cells, further includes: If the first operating scale data is less than the second operating scale data, the multiple cold tanks in operation are shut down in reverse order according to energy consumption priority until the number of shut-down tanks is equal to the difference between the second operating scale data and the first operating scale data. If the total number of operating hot tanks is less than the difference between the second operating scale data and the first operating scale data, the multiple operating hot tanks are shut down in reverse order according to temperature priority until the number of shut-down tanks equals the difference between the second operating scale data and the first operating scale data.

5. The method according to claim 1, characterized in that, The process involves using hydrogen production demand data or load absorption data as constraints, and employing an electrolyzer energy consumption scheduling optimization model to minimize the total energy consumption of multiple operating electrolyzers, resulting in a load allocation scheme, including: Using hydrogen production demand data or load consumption data as constraints, the total energy consumption of multiple operating electrolyzers is minimized using an electrolyzer energy consumption scheduling optimization model, thereby obtaining the minimum energy consumption load allocation scheme for the multiple operating electrolyzers. Based on the preset energy consumption deviation, multiple candidate load allocation schemes corresponding to the minimum energy consumption load allocation scheme are generated; Based on the load data of the multiple operating electrolyzers during the historical load switching cycle, the candidate load allocation scheme with the smallest relative load change is selected from the multiple candidate load allocation schemes as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

6. The method according to claim 5, characterized in that, The step of selecting the candidate load allocation scheme with the smallest relative load change from among the multiple candidate load allocation schemes based on the load data of the multiple operating electrolyzers during historical load switching cycles as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster includes: Based on the load data of the multiple operating electrolyzers during the historical load switching cycle and the similarity of the load data of the multiple electrolyzers in each candidate load allocation scheme, the load fluctuation data of each candidate load allocation scheme is calculated. The candidate load allocation scheme with the smallest load fluctuation data is selected as the load allocation scheme for the current load switching cycle of the water electrolysis hydrogen production cluster.

7. A load adjustment device for a water electrolysis hydrogen production cluster, characterized in that, include: The calculation module is used to calculate the temperature priority of each electrolyzer based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster. The temperature priority of the electrolyzer is positively correlated with its temperature data; the temperature data includes the first temperature data of each electrolyzer in the historical sorting period, the second temperature data in the historical measurement period, and the third temperature data in the current load switching period; The step of calculating the temperature priority of each electrolyzer based on the temperature data of multiple electrolyzers in the water electrolysis hydrogen production cluster includes: determining the fourth temperature data of each electrolyzer based on the fusion result of the first temperature data, the second temperature data, and the third temperature data; dividing the multiple electrolyzers into multiple hot-state cells and multiple cold-state cells using a preset temperature threshold based on the fourth temperature data of the multiple electrolyzers in the water electrolysis hydrogen production cluster; and determining the temperature priority of the multiple hot-state cells based on the comparison result of the temperature data of the multiple hot-state cells. A determination module is used to determine multiple operating electrolyzers from a plurality of electrolyzers based on hydrogen production demand data or load consumption data, using temperature priority. The determination of multiple operating electrolyzers based on hydrogen production demand data or load consumption data, using temperature priority, includes: calculating first operating scale data of the electrolyzers within the current load switching cycle based on the hydrogen production demand data or load consumption data; obtaining second operating scale data of the electrolyzers within historical load switching cycles; obtaining the energy consumption priority of each electrolyzer within a historical sorting cycle; and determining multiple operating electrolyzers based on the comparison result of the first and second operating scale data, as well as the temperature priority and energy consumption priority of the plurality of electrolyzers. A module is established to build an energy consumption scheduling optimization model for electrolytic cells based on a preset optimization algorithm; The optimization module is used to minimize the total energy consumption of multiple operating electrolyzers by using an electrolyzer energy consumption scheduling optimization model, with hydrogen production demand data or load consumption data as constraints, to obtain a load allocation scheme. The adjustment module is used to adjust the load of multiple operating electrolytic cells according to the load distribution scheme.

8. A computer device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the method of any one of claims 1-6.

Images