Electrolysis adaptive adjustment method for MW-level off-grid photovoltaic hydrogen production system

Abstract

The invention discloses an electrolysis self-adaptive adjusting method for an MW-level off-grid photovoltaic hydrogen production system. The electrolysis self-adaptive control developed by using a dynamic programming algorithm can reasonably match the preheating, operation and hot standby conditions of the hydrogen production system according to the current photovoltaic power generation amount, and the photovoltaic utilization efficiency and the hydrogen production yield are improved. Compared with unified control of the electrolysis device, the grouping control mode achieves following of the hydrogen production working condition and photovoltaic power generation, and the fluctuation adaptability of the electrolysis device is improved.

Description

Technical Field

[0001] This invention relates to the technical field of hydrogen production, and in particular to an adaptive regulation method for electrolysis in a MW-level off-grid photovoltaic hydrogen production system. Background Technology

[0002] With the transformation of the global energy structure, developing renewable energy has become an important measure for major countries around the world to ensure energy security, strengthen environmental protection, and address climate change. Due to the intermittent and unstable nature of renewable energy, energy storage is generally required to mitigate fluctuations and reduce renewable energy waste. Hydrogen energy, as a clean and efficient energy carrier, can balance the fluctuations in renewable energy generation through electricity-hydrogen conversion, achieving integrated supply across "source-grid-load-storage." This provides a technological approach to solving the problem of large-scale renewable energy consumption and promoting the formation and development of a green hydrogen energy industry chain.

[0003] Currently, hydrogen production technology based on renewable energy-based water electrolysis has received widespread attention. It converts and stores fluctuating clean electricity into high-quality hydrogen, providing an effective way to improve the efficiency and reliability of renewable energy utilization. Alkaline water electrolysis (AWE) technology is the most mature, with a long lifespan, large scale, and relatively low equipment cost, making it the mainstream hydrogen production method on the market. However, AWE is difficult to start / stop quickly and adjust the hydrogen production rate, making it challenging to coordinate with renewable energy sources that have rapidly fluctuating characteristics, thus reducing the efficiency of renewable energy utilization. This problem is particularly prominent for MW-level systems.

[0004] Therefore, in order to improve the utilization rate of renewable energy and enhance the overall performance and economic benefits of hydrogen production systems, a control method that adaptively adjusts the electrolysis state is needed to adapt to the volatility of renewable energy and improve the operational stability of the system. Summary of the Invention

[0005] This invention provides an adaptive adjustment method for electrolysis in a MW-level off-grid photovoltaic hydrogen production system. The purpose is to adaptively adjust the electrolysis state to adapt to the volatility of renewable energy and improve the operational stability of the system.

[0006] Firstly, an adaptive regulation method for electrolysis in a MW-level off-grid photovoltaic hydrogen production system is provided, including:

[0007] If the current time t is to the left of the peak illumination time T, then determine the maximum number n electrolytic cells that can be turned on at the current time t. max (t);

[0008] If n maxIf (t)≤0, then the number of electrolyzers i(t) to be in hot standby is determined based on the current illumination conditions. Multiple values ​​for i(t) are then used to calculate the hydrogen production power P at the current moment for each value of i(t). 制氢 (t)=P pv (t)-P 热备 *i(t); if n max If n(t) > 0, then determine multiple values ​​for the number of additional electrolyzers n(t) to be activated at the current time t, and calculate the hydrogen production power P at the current time for each value of n(t). 制氢 (t)=min[(P 运行 *m(t)),(P pv (t)-P 预热 *n(t))];where P pv (t) represents the photovoltaic power generation at time t, P 热备 P is the hot standby power for a single electrolytic cell. 运行 P represents the hydrogen production power of a single electrolyzer. 预热 Let m(t) be the power consumption for AWE preheating, and m(t) be the number of electrolytic cells in operation at the current time t.

[0009] If the current time t is to the right of the peak illumination time T, then determine the number of electrolysis stations that need to be turned off at the current time t.

[0010] After calculating all time points, by applying the objective function opt = max∫P 制氢 (t) seeks the maximum value to obtain the electrolysis operation path with the maximum hydrogen production power, and obtains the number of electrolyzers in operation / hot standby / preheating at each moment.

[0011] In conjunction with the first aspect, in some implementations of the first aspect, in n max When (t)≤0, the transfer function of m(t) is: m(t+1)=v(t)+i(t), where m(t+1) is the number of electrolytic cells operating at the next time t+1, v(t) is the minimum number of electrolytic cells currently operating normally, and v(t)=floor(P pv (t) / P 运行 ).

[0012] In conjunction with the first aspect, in some implementations of the first aspect, 0≤i(t)≤m(t)-v(t), v(t)=floor(P) pv (t) / P 运行 v(t) is the minimum number of electrolytic cells currently in normal operation, and floor() is the floor function.

[0013] In conjunction with the first aspect, in some implementations of the first aspect, in n maxWhen (t) > 0, the transfer function of m(t) is: m(t+1) = m(t) + n(t), where m(t+1) is the number of electrolytic cells to be started at the next time t+1, n(t) is the number of electrolytic cells to be started at the current time t, and 0 ≤ n(t) ≤ n max (t).

[0014] In conjunction with the first aspect, in some implementations of the first aspect, when t > T, the transfer function of m(t) is: m(t+1) represents the number of electrolytic cells operating at the next time t+1, and P pv (t+1) represents the photovoltaic power generation at time t+1.

[0015] In conjunction with the first aspect, in some implementations of the first aspect, when t > T, P 制氢 (t)=min[(P 运行 *m(t)),P pv (t)].

[0016] In conjunction with the first aspect, in some implementations of the first aspect, P 制氢 The optimization constraints for (t) include: 0.5*P 运行 *m(t)≤P 制氢 (t)≤P 运行 *m(t).

[0017] In conjunction with the first aspect, in some implementations of the first aspect, n max (t)=floor((P pv (t)-P 运行 *m(t)) / P 预热 m(t) represents the number of electrolytic cells in operation at the current time t, and floor() is the floor function.

[0018] In conjunction with the first aspect, in some implementations of the first aspect, J is formed for time t-1. t-1 The traversal results, J t-1 The number of electrolytic cells already started in each of the traversal results is different; in J t-1 The j-th traversal result t-1 In each traversal result, the number of electrolytic cells that have been started is m(t,j). t-1 ), j t-1 The range of values ​​is [0, J]. t-1 According to m(t,j) t-1 Determine the maximum number of electrolytic cells n that can be activated at the current time t. max (t,j t-1 );

[0019] In n max (t,jt-1 When )≤0, based on the j-th t-1 The traversal results were executed in total. The loop iterates through the data. v(t) = floor(P) pv (t) / P 运行 ), where based on the j-th t-1 The first traversal result The traversal results For the number of hot standby units in the 1st The value obtained in the second traversal The hydrogen production power at time t is at the t-th The value obtained in the next iteration Value Therefore, based on the j-th t-1 The transfer function for the number of electrolytic cells already started, based on the traversal results, is:

[0020] After completing all J at time t-1 t-1 After traversing the results, a total of The number of electrolytic cells m(t+1) at time t+1 is calculated; duplicate results m(t+1) are combined and removed to obtain the final J of the number of electrolytic cells m(t+1) at time t+1. t The traversal results, where J t The j-th traversal result t In each iteration, the number of electrolytic cells that have been started is m(t+1,j). t ), j t The range of values ​​is [0, J]. t ],

[0021] In conjunction with the first aspect, in some implementations of the first aspect, J is formed for time t-1. t-1 The traversal results, J t-1 The number of electrolytic cells already started in each of the traversal results is different; in J t-1 The j-th traversal result t-1 In each traversal result, the number of electrolytic cells that have been started is m(t,j). t-1 ), j t-1 The range of values ​​is [0, J]. t-1 According to m(t,j) t-1 Determine the maximum number of electrolytic cells n that can be activated at the current time t. max (t,j t-1 );

[0022] In n max (t,j t-1 When ) > 0, based on the j-tht-1 The traversal results were executed in total. The loop iterates through the data. n max (t,j t-1 ) = floor((P pv (t)-P 运行 *m(t,j t-1 )) / P 预热 ; where based on the j-th t-1 The first traversal result The transfer function for the number of electrolytic cells already started, based on the traversal results, is: The number of electrolytic cells to be started at time t is the number of cells to be started at time t. The value obtained in the next iteration The hydrogen production power at time t is at the t-th The value obtained in the next iteration Value

[0023] After completing all J at time t-1 t-1 After traversing the results, a total of The number of electrolytic cells m(t+1) at time t+1 is calculated; duplicate results m(t+1) are combined and removed to obtain the final J of the number of electrolytic cells m(t+1) at time t+1. t The traversal results, where J t The j-th traversal result t In each iteration, the number of electrolytic cells that have been started is m(t+1,j). t ), j t The range of values ​​is [0, J]. t ],

[0024] Compared with the prior art, the solution provided by the present invention has at least the following beneficial technical effects:

[0025] 1. The electrolysis adaptive control developed using dynamic programming algorithm can reasonably match the preheating, operation, and hot standby status of the hydrogen production system according to the current photovoltaic power generation, thereby improving photovoltaic utilization efficiency and hydrogen production output; 2. Compared with the unified control of the electrolysis unit, the group control method of this invention realizes the following of hydrogen production conditions and photovoltaic power generation, improving the fluctuation adaptability of the electrolysis unit. Attached Figure Description

[0026] Figure 1 The flowchart illustrates an adaptive adjustment method for electrolysis in a MW-level off-grid photovoltaic hydrogen production system provided by this invention.

[0027] Figure 2 The path planning graph for dynamic programming control when t≤12 is shown.

[0028] Figure 3 This is a schematic diagram showing the operating status and hydrogen production capacity of an electrolysis unit under average annual sunlight intensity.

[0029] Figure 4 This diagram illustrates the operating status and hydrogen production capacity of the electrolysis unit during sunny summer weather.

[0030] Figure 5 This is a schematic diagram showing the operating status and hydrogen production capacity of the electrolysis unit on a cloudy or rainy day. Detailed Implementation

[0031] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0032] MW-level photovoltaic hydrogen production systems typically include multiple parallel electrolysis hydrogen production systems. To achieve photovoltaic power absorption and stable control of the electrolysis system while optimizing hydrogen production efficiency, this invention provides an adaptive adjustment method for electrolysis, satisfying the requirements of load following the source and the adaptability of the electrolysis hydrogen production system to changes in photovoltaic power generation. Photovoltaic forecasting combined with dynamic optimization of the number of electrolyzers efficiently absorbs photovoltaic energy for hydrogen production, ensuring the adaptability of the electrolysis hydrogen production system to changes in photovoltaic power generation.

[0033] like Figure 1 As shown, the present invention provides an adaptive regulation method for electrolysis of a MW-level off-grid photovoltaic hydrogen production system. The specific steps of the regulation method are as follows.

[0034] Step 1: If the current time t is to the left of the peak illumination time T, then determine the maximum number n of electrolytic cells that can be turned on at the current time t. max (t).

[0035] This adjustment method assumes that the variation of the total light intensity throughout the day over time is known. In step 1, since the current time t has not yet reached the peak light intensity T, the maximum hydrogen production power P that the electrolyzer can achieve at the current time t can be used as a reference. 运行 *m(t) calculates the maximum number of electrolytic cells n that can be activated at the current time t. max (t). Once the electrolytic cell is turned on, it means that the electrolytic cell can be in any of the following states: preheating (start-up requires power), running, or hot standby.

[0036] n max The way to determine (t) is, for example, n max (t)=floor((P pv (t)-P 运行 *m(t)) / P 预热 floor() is the floor function, P pvP(t) represents the photovoltaic power generation at time t, m(t) represents the number of electrolytic cells that have been started (i.e., are running) at time t, and P 运行 P represents the hydrogen production power of a single electrolyzer (this application defaults to taking the peak hydrogen production power), i.e., the peak power of each electrolyzer. 预热 Power consumption for AWE preheating.

[0037] Step 2, based on step 1, if n max If (t)≤0, it means that no additional electrolytic cells need to be started. Based on this, the need for hot standby is determined according to the current illumination level. If the current moment is at the left end of the peak illumination period and the current illumination is lower than the previous moment, hot standby is required. The hot standby will terminate and the system will return to normal operation once the illumination recovers. Hot standby means temporarily stopping operation to maintain the electrolytic cell temperature with lower power consumption, keeping the electrolytic cell in standby mode so that it can resume normal operation at any time after the illumination intensity increases. The number of cells that can be hot-standby i(t) ≤ the number of currently operating electrolytic cells m(t) - the minimum value v(t) of the number of electrolytic cells that should be operating normally. The minimum value v(t) of the number of cells operating when hot standby is required = floor(P) pv (t) / P 运行 For different hot standby units i(t), calculate the hydrogen production power at the current moment = photovoltaic power generation power - electrolyzer hot standby power. The number of electrolyzers working at the next moment m(t+1) is the minimum number of electrolyzers that should be operating normally v(t) + the number of hot standby units i(t), and then enter the next moment cycle.

[0038] Based on the analysis of AWE's operating characteristics, the power balance is as follows:

[0039] P pv ≥P 预热 *n 预热 +P 运行 *n 运行 +P 热备 *n 热备

[0040] Among them, P pv For photovoltaic power generation, P 预热 n represents the power consumption for preheating a single electrolytic cell. 预热 P represents the number of electrolytic cells in the preheated state. 运行 n represents the hydrogen production power of a single electrolyzer under operating conditions. 运行 P represents the number of electrolytic cells in operation. 热备 n represents the power consumption for hot standby of a single electrolytic cell. 热备 This refers to the number of electrolytic cells in hot standby mode. In other words, there can be three types of electrolytic cells at any given time: one type is in a preheating state to complete startup preparations, and its power consumption corresponds to P. 预热The second type of electrolytic cell is in operation, and its power consumption corresponds to P. 运行 The three types of electrolytic cells are in hot standby mode, and their power consumption corresponds to P. 热备 .

[0041] When hot standby is required, the transfer function is m(t+1) = v(t) + i(t), t ≤ T. Here, m is the state variable, i.e., the number of electrolytic cells already started (starting an electrolytic cell means it is in operation or hot standby mode); m(t+1) is the number of electrolytic cells started at the next time step t+1; and T is the moment when the solar radiation curve has a significant peak. In other words, the number of electrolytic cells started at the next time step remains unchanged, and the cycle continues into the next time step.

[0042] If hot standby is required, since there is no electrolytic cell ready to be started (i.e., in a preheated state), therefore P 预热 *n 预热 The value is zero. For ease of calculation, let the hydrogen production power P at the current time t be set. 制氢 (t) satisfies:

[0043] P 制氢 (t)=P pv (t)-P 热备 *i(t)

[0044] i(t) represents the number of hot standby electrolytic cells at the current time t. The number of hot standby cells must satisfy the following condition: the number of hot standby cells ≤ the number of cells currently in operation. That is, i(t) is an integer and its value range is [0, m(t) - v(t)].

[0045] The number of hot standby units can be obtained through optimization calculation. During optimization, the hot standby unit number i(t) can be iterated within the selectable range [0, m(t)] to calculate the hydrogen production power P at the current time t for different hot standby unit numbers. 制氢 (t).

[0046] Step 3, based on step 1, if n max If n(t) > 0, it means that at the current time t, an additional number of electrolytic cells need to be turned on, and the number of additional electrolytic cells turned on is n(t) ≤ n. max (t).

[0047] When it is determined that the electrolytic cells need to be started, the transfer function is: m(t+1) = m(t) + n(t), t ≤ T. Here, m is the state variable, i.e., the number of electrolytic cells already started; m(t+1) is the number of electrolytic cells started at the next time t+1; and m(t) is the number of electrolytic cells started at the current time t. n is the decision variable, i.e., the number of electrolytic cells to be started; n(t) is the number of electrolytic cells to be started at the current time t. max(t) is used to constrain the maximum value of n(t). That is, the number of electrolytic cells that have been started in the next moment is the number of electrolytic cells that have been started at present plus the number of additional electrolytic cells that have been started.

[0048] If it is determined that the electrolytic cell needs to be turned on, since there is no electrolytic cell in hot standby mode, therefore P 热备 The value is zero. For ease of calculation, let the hydrogen production power P at the current time t be set. 制氢 (t) is calculated according to the following formula:

[0049] P 制氢 (t)=min[(P 运行 *m(t)),(P pv (t)-P 预热 *n(t))]

[0050] The number of electrolytic cells to be started can be obtained through optimization calculation. During the optimization calculation, the number of electrolytic cells to be started, n(t), can be within the selectable range [0, n]. max The process iterates through the range [t] and calculates the hydrogen production power P at the current time t for different numbers of electrolyzers to be started, n(t). 制氢 (t).

[0051] Step 4: If the current time t is to the right of the peak illumination time T, then determine the number of electrolysis stations that need to be turned off at the current time t.

[0052] If the current time t has exceeded the peak time T of photovoltaic power generation, then the number of electrolyzers needs to be shut down one by one. The current hydrogen production power = min(current maximum hydrogen production, photovoltaic power generation) and enter the next time cycle.

[0053] If it is determined that the electrolytic cell needs to be shut down, the transfer function is: t>T. Where m is the state variable, i.e., the number of electrolytic cells already started, m(t+1) is the number of electrolytic cells started at the next time t+1, round() is the floor function, and P... pv (t+1) represents the photovoltaic power generation at the next time point t+1, P 运行 This represents the hydrogen production power of a single electrolyzer.

[0054] Hydrogen production power P at current time t 制氢 (t) is calculated according to the following formula:

[0055] P 制氢 (t)=min[(P 运行 *m(t)),P pv (t)]

[0056] Step 5: After calculating all time points, use the objective function opt = max∫P 制氢(t) seek the maximum value to obtain the electrolysis operation path with the maximum hydrogen production power, and then obtain the number of electrolyzers in operation / hot standby / preheating at each moment.

[0057] This invention, considering the overall resource utilization efficiency of the system, uses the electrolyzer grouping scheme as the optimization decision variable and the effective daily hydrogen production power as the objective function opt. The constraints in this problem are:

[0058]

[0059] Where N is the total number of electrolytic cells configured, P 运行 This refers to the hydrogen production capacity of each electrolyzer.

[0060] Combination Figure 2 This invention achieves the objective function P through dynamic programming. 制氢 The optimization of (t) is achieved through dynamic programming. Starting from the initial stage, different nodes can be selected for each stage. A node represents a state variable, specifically the number of operating electrolyzers. A definite hydrogen production power can be calculated from the current node to the next stage node. By summing and comparing the hydrogen production power across all stages and paths, the electrolyzer operating path with the maximum total hydrogen production power is finally determined. Figure 2 The optimal path, represented by the thick solid line, determines the number of electrolytic cells in operation at each time point.

[0061] Objective function P 制氢 The optimization process of (t) is referenced. Figure 2 For each time step t, the specific values ​​of the variables i(t) and n(t) during the traversal depend on the calculation results of the previous time step.

[0062] For example, suppose J is formed for time t-1. t-1 The traversal results, J t-1 The number of electrolytic cells already started in each of the traversal results is different. Assume that in J... t-1 The j-th traversal result t-1 In each traversal result, the number of electrolytic cells that have been started is m(t,j). t-1 ), j t-1 The range of values ​​is [0, J]. t-1 ].

[0063] Then, with the jth t-1 Taking the traversal result as an example, execute step 1, based on m(t,j) t-1 Determine the maximum number of electrolytic cells n that can be turned on at the current time t. max (t,j t-1 If n max (t,j t-1 If n ≤ 0, execute the loop traversal in step 2; if nmax (t,j t-1 If )>0, execute the loop traversal in step 3.

[0064] If we execute the loop iteration in step 2, assuming step 2 is based on the j-th... t-1 The traversal results were executed in total. The loop iterates through the data. v(t) = floor(P) pv (t) / P 运行 ), where based on the j-th t-1 The first traversal result The traversal results For the number of hot standby units in the 1st The value obtained in the next iteration The hydrogen production power at time t is at the t-th The value obtained in the next iteration Value Therefore, based on the j-th t-1 The transfer function for the number of electrolytic cells already started, based on the traversal results, is:

[0065] After completing all J at time t-1 t-1 After traversing the results, a total of The number of electrolytic cells m(t+1) at time t+1 is calculated. Duplicate counts m(t+1) are combined and removed to obtain the final J value representing the number of electrolytic cells m(t+1) at time t+1. t The traversal results, where J t The j-th traversal result t In each iteration, the number of electrolytic cells that have been started is m(t+1,j). t ), j t The range of values ​​is [0, J]. t ],

[0066] If we execute the loop iteration in step 3, assuming step 3 is based on the j-th... t-1 The traversal results were executed in total. The loop iterates through the data. n max (t,j t-1 ) = floor((P pv (t)-P 运行 *m(t,j t-1 )) / P 预热 Among them, based on the j-th t-1 The first traversal result The transfer function for the number of electrolytic cells already started, based on the traversal results, is: The number of electrolytic cells to be started at time t is the number of cells to be started at time t. The value obtained in the next iteration The hydrogen production power at time t is at the t-th The value obtained in the next iteration Value

[0067] After completing all J at time t-1 t-1 After traversing the results, a total of The number of electrolytic cells m(t+1) at time t+1 is calculated. Duplicate counts m(t+1) are combined and removed to obtain the final J value representing the number of electrolytic cells m(t+1) at time t+1. t The traversal results, where J t The j-th traversal result t In each iteration, the number of electrolytic cells that have been started is m(t+1,j). t ), j t The range of values ​​is [0, J]. t ],

[0068] For the j-th t-1 After each traversal result completes all iterations of step 2 or step 3 above, a total of J are formed at the current time t. t The traversal results, where J t The number of electrolytic cells that have been started in each of the traversal results is different. In J t Among the traversal results, there may be a certain number of electrolytic cells that have been started, m(t,j). t There are multiple sources for ), in other words, the traversal results with the same state variables in the transfer function can be combined to avoid redundancy, while retaining P during the combination. 制氢 Find the largest traversal result (t, k) and delete the others. Then set t = t + 1 to perform the optimization calculation for the next time step.

[0069] Determine if t is still to the left of the peak illumination time T. If so, iterate through the loop as described above. If not, use J... t-1 Based on the traversal results, proceed to step 4.

[0070] Dynamic programming is used to perform a depth-first search during the first half of the photovoltaic (PV) power generation period, searching within the range of available PV units per hour to find the solution with the minimum amount of wasted solar power. Since PV power generation is zero between 0:00 and 5:00, data from this period is removed and not included in the calculations, significantly improving the computational efficiency of the dynamic programming algorithm. The control strategy calculated using dynamic programming can determine the number of AWE (Activated Photovoltaic Welfare) units and hydrogen production capacity throughout the entire process, ensuring optimal daily hydrogen production. The control method provided by this invention rationally allocates the start-up, power regulation, hot standby, and shutdown conditions for each electrolysis unit based on the operating characteristics of the AWE unit and the local sunlight conditions, enabling the overall hydrogen production capacity to better couple with PV fluctuations, optimizing PV power generation utilization, and increasing hydrogen production output.

[0071] Example 1

[0072] According to the project configuration, the hydrogen production system consists of 16 electrolysis units, each with a capacity of 5MW. The electrolysis units are divided into 16 groups, meaning that each electrolysis unit is separately controlled for startup, operation, and hot standby. Figure 3 This is a schematic diagram showing the operating status and hydrogen production capacity of an electrolysis unit under average annual sunlight intensity. Figure 3 This diagram illustrates the operating status and hydrogen production capacity of the electrolysis unit during sunny summer weather. Figure 4 This is a schematic diagram showing the operating status and hydrogen production capacity of the electrolysis unit on a cloudy or rainy day.

[0073] An adaptive regulation method for electrolysis in a MW-level off-grid photovoltaic hydrogen production system, the specific control steps of which are as follows:

[0074] 1) During the early morning weak photovoltaic power generation phase, the sunlight in this area begins as early as 5 a.m. and gradually increases. However, due to the weak initial light intensity, it does not reach the starting power of the AWE system.

[0075] 2) During the gradual enhancement phase of photovoltaic power generation, the photovoltaic power generation in this region generally shows a gradual upward trend before 12:00. When the minimum start-up power of the AWE (30% of rated power) is reached, the photovoltaic power generation is prioritized for AWE preheating. After 1 hour of AWE preheating, it can switch to normal hydrogen production mode. Based on the current photovoltaic power generation and the number of AWE units preheated in the previous hour, the maximum number of AWE units that can be turned on at the current moment can be determined. The AWE hydrogen production mode adjustment range is between 50% and 100%. The possible number of units that can be turned on is traversed to calculate the hydrogen production power under different numbers of units turned on. Finally, through a global control strategy, the number of electrolyzers that need to be turned on at that moment under the maximum global hydrogen production power is calculated.

[0076] 3) During the phase of gradually weakening photovoltaic power generation, generally, the photovoltaic power generation in this region shows a gradual decreasing trend after 12:00. During this phase, there is no need to consider the preheating of AWE (Aqueous Power Generation Engine); simply shut down AWE sequentially based on the current photovoltaic power generation. After 18:00, if the photovoltaic power generation is lower than 50% of the rated power of AWE, all AWEs will be shut down until the photovoltaic power generation reaches 0, at which point the entire hydrogen production system will be completely shut down.

[0077] 4) During periods of fluctuation in photovoltaic power generation, such as cloudy or rainy weather, photovoltaic power generation may experience [various issues]. Figure 5 The diagram illustrates a scenario where photovoltaic (PV) power generation changes are not monotonic. Specifically, before 12:00, if the current PV power generation is lower than the previous moment's, especially if the current power generation is less than 2.5 * m (the number of AWE units in operation), some AWE electrolysis units need to be shut down. They can then be restarted as PV power generation increases. Shutting down AWEs during this phase differs from shutting them down after 12:00; in this case, AWEs will not be restarted for the rest of the day. Shutting down AWEs during periods of PV power fluctuation due to insufficient current power generation is only temporary. To ensure that AWEs can be restarted as quickly as PV power generation increases, additional power is needed to maintain the necessary startup temperature. The globally optimized control strategy calculates the electrolysis operation path at maximum hydrogen production power by considering different startup and hot standby conditions at each moment, thus determining the number of units requiring hot standby at the current moment.

[0078] The above preferred solution is for a small number of electrolyzers. If the number of electrolyzers in the hydrogen production system increases, the present invention can also perform secondary grouping, that is, keep the operating status of several electrolyzers consistent. This can simplify control, but will affect the control accuracy.

[0079] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope defined in the claims of the present invention.

Claims

1. An adaptive adjustment method for electrolysis in a MW-level off-grid photovoltaic hydrogen production system, characterized in that, include: If the current time t is to the left of the peak illumination time T, then determine the maximum number n electrolytic cells that can be turned on at the current time t. max (t); If n max If (t)≤0, then the number of electrolyzers i(t) to be in hot standby is determined based on the current illumination conditions. Multiple values ​​for i(t) are then used to calculate the hydrogen production power P at the current moment for each value of i(t). 制氢 (t)=P pv (t)-P 热备 *i(t); if n max If n(t) > 0, then determine multiple values ​​for the number of additional electrolyzers n(t) to be activated at the current time t, and calculate the hydrogen production power P at the current time for each value of n(t). 制氢 (t)=min[(P 运行 *m(t)),(P pv (t)-P 预热 *n(t))];where P pv (t) represents the photovoltaic power generation at time t, P 热备 P is the hot standby power for a single electrolytic cell. 运行 P represents the hydrogen production power of a single electrolyzer. 预热 Let m(t) be the power consumption for AWE preheating, and m(t) be the number of electrolytic cells in operation at the current time t. If the current time t is to the right of the peak illumination time T, then determine the number of electrolysis stations that need to be turned off at the current time t. After calculating all time points, by applying the objective function opt = max∫P 制氢 (t) seeks the maximum value to obtain the electrolysis operation path with the maximum hydrogen production power, and obtains the number of electrolyzers in operation / hot standby / preheating at each moment.

2. The method according to claim 1, characterized in that, In n max When (t)≤0, the transfer function of m(t) is: m(t+1)=v(t)+i(t), where m(t+1) is the number of electrolytic cells operating at the next time t+1, v(t) is the minimum number of electrolytic cells currently operating normally, and v(t)=floor(P pv (t) / P 运行 ).

3. The method according to claim 1, characterized in that, 0≤i(t)≤m(t)-v(t), v(t)=floor(P pv (t) / P 运行 v(t) is the minimum number of electrolytic cells currently in normal operation, and floor() is the floor function.

4. The method according to claim 1, characterized in that, In n max When (t) > 0, the transfer function of m(t) is: m(t+1) = m(t) + n(t), where m(t+1) is the number of electrolytic cells to be started at the next time t+1, n(t) is the number of electrolytic cells to be started at the current time t, and 0 ≤ n(t) ≤ n max (t).

5. The method according to claim 1, characterized in that, When t > T, the transfer function of m(t) is: m(t+1) represents the number of electrolytic cells operating at the next time t+1, and P pv (t+1) represents the photovoltaic power generation at time t+1.

6. The method according to claim 1, characterized in that, When t > T, P 制氢 (t)=min[(P 运行 *m(t)),P pv (t)].

7. The method according to claim 1, characterized in that, P 制氢 The optimization constraints for (t) include: 0.5*P 运行 *m(t)≤P 制氢 (t)≤P 运行 *m(t).

8. The method according to claim 1, characterized in that, n max (t)=floor((P pv (t)-P 运行 *m(t)) / P 预热 m(t) represents the number of electrolytic cells in operation at the current time t, and floor() is the floor function.

9. The method according to claim 1, characterized in that, For the formation of J at time t-1 t-1 The traversal results, J t-1 The number of electrolytic cells already started in each of the traversal results is different; in J t-1 The j-th traversal result t-1 In each traversal result, the number of electrolytic cells that have been started is m(t,j). t-1 ), j t-1 The range of values ​​is [0, J]. t-1 According to m(t,j) t-1 Determine the maximum number of electrolytic cells n that can be activated at the current time t. max (t,j t-1 ); In n max (t,j t-1 When )≤0, based on the j-th t-1 The traversal results were executed in total. The loop iterates through the data. v(t) = floor(P) pv (t) / P 运行 ), where based on the j-th t-1 The first traversal result The traversal results For the number of hot standby units in the 1st The value obtained in the next iteration The hydrogen production power at time t is at the t-th The value obtained in the next iteration Value Therefore, based on the j-th t-1 The transfer function for the number of electrolytic cells already started, based on the traversal results, is: After completing all J at time t-1 t-1 After traversing the results, a total of The number of electrolytic cells m(t+1) at time t+1 is calculated; duplicate results m(t+1) are combined and removed to obtain the final J of the number of electrolytic cells m(t+1) at time t+1. t The traversal results, where J t The j-th traversal result t In each iteration, the number of electrolytic cells that have been started is m(t+1,j). t ), j t The range of values ​​is [0, J]. t ], 10. The method according to claim 1, characterized in that, For the formation of J at time t-1 t-1 The traversal results, J t-1 The number of electrolytic cells already started in each of the traversal results is different; in J t-1 The j-th traversal result t-1 In each traversal result, the number of electrolytic cells that have been started is m(t,j). t-1 ), j t-1 The range of values ​​is [0, J]. t-1 According to m(t,j) t-1 Determine the maximum number of electrolytic cells n that can be activated at the current time t. max (t,j t-1 ); In n max (t,j t-1 When ) > 0, based on the j-th t-1 The traversal results were executed in total. The loop iterates through the data. n max (t,j t-1 ) = floor((P pv (t)-P 运行 *m(t,j t-1 )) / P 预热 ; where based on the j-th t-1 The first traversal result The transfer function for the number of electrolytic cells already started, based on the traversal results, is: The number of electrolytic cells to be started at time t is the number of cells to be started at time t. The value obtained in the next iteration The hydrogen production power at time t is at the t-th The value obtained in the next iteration Value After completing all J at time t-1 t-1 After traversing the results, a total of The number of electrolytic cells m(t+1) at time t+1 is calculated; duplicate results m(t+1) are combined and removed to obtain the final J of the number of electrolytic cells m(t+1) at time t+1. t The traversal results, where J t The j-th traversal result t In each iteration, the number of electrolytic cells that have been started is m(t+1,j). t ), j t The range of values ​​is [0, J]. t ],

Images