Electric hydrogen production frequency modulation system frequency control method based on model predictive control

Abstract

The invention belongs to the field of power system frequency modulation, and particularly discloses an electric hydrogen production frequency modulation system frequency control method based on model predictive control. According to the method, the prediction model is established based on the dynamic characteristics of the electrolytic cell power, so that the frequency and inertia of the next moment can be predicted according to the current state, and the influence of the power hysteresis characteristic is quantified; based on the prediction result, the additional frequency modulation power at the next moment is optimized and solved by utilizing a multi-target cost function to comprehensively plan frequency deviation, inertia back-compensation, a droop coefficient and the like, and the core is to convert prediction information into an optimized decision so as to actively compensate the lag effect and coordinate multiple control targets; and finally, according to the optimization decision, an accurate power control signal is generated and executed, so that the power regulation of the electrolytic cell can accurately match the real-time frequency support requirement of the system. Compared with the prior art, the problem of control mismatch caused by insufficient consideration of the power lag characteristic is solved, so that the frequency oscillation of the power grid is effectively suppressed.

Description

Technical Field

[0001] This application belongs to the field of power system frequency regulation, and more specifically, relates to a frequency control method for an electric hydrogen production frequency regulation system based on model predictive control. Background Technology

[0002] In isolated or industrial park-level renewable energy microgrids, the synchronous inertia constant is typically less than 2s. Sudden changes in load or wind / solar output within 1 second can cause the Rate of Change of Frequency (RoCoF) to exceed 1Hz / s and the lowest frequency to drop below 50Hz, posing risks to relay protection and critical loads. Fast Frequency Response (FFR), with its millisecond-second response time compared to conventional primary frequency regulation, is considered a key means to improve the frequency safety margin of low-inertia microgrids. However, in isolated or industrial park-level microgrids with a high proportion of renewable energy, the inertia constant is usually less than 2s. Load or wind / solar fluctuations within 1 second can cause RoCoF to exceed 1Hz / s and drop sharply below 50Hz, making the 8-10s ramp-up speed of traditional FFR insufficient to support this in a timely manner.

[0003] Based on this, electrolyzer technology can not only achieve bidirectional conversion of electrical and hydrogen energy, but also effectively solve the problem of wind curtailment caused by the volatility of new energy sources. For example, the alkaline electrolyzer (AEL) consists of anode and cathode electrodes, and ion transport is achieved through the electrolyte. Due to the special physical characteristics of the electrolyzer, the non-uniform distribution of positive and negative charges at the electrode-electrolyte interface forms an electrochemical double layer (EDL) structure. This causes an energy barrier to electron flow when the electrolyzer current changes abruptly, especially when the power changes rapidly to participate in a fast frequency response, resulting in a response delay. Consequently, power fluctuations are dominated by the EDL delay.

[0004] However, due to the relatively lagging power response of AEL (Automatic Electrolyte), existing integrated inertia control methods can only select relatively conservative control parameters to avoid system frequency oscillations. They do not adequately consider the control requirements for supporting frequency recovery, and the supporting role of electro-hydrogenation in frequency recovery is not fully realized. Furthermore, the frequency may drop twice due to the rapid decay of frequency modulation power during the frequency recovery phase caused by the AEL lag. Therefore, it is necessary to develop optimized control strategies for AEL electrolyzers. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this application is to provide a frequency control method for an electric hydrogen production frequency regulation system based on model predictive control. This method aims to solve the technical problem of grid frequency oscillation and poor stability caused by insufficient consideration of frequency regulation control requirements due to the power lag characteristics of traditional electric hydrogen production frequency regulation systems.

[0006] To achieve the above objectives, in a first aspect, this application provides a frequency control method for an electro-hydrogen production frequency modulation system based on model predictive control, comprising: constructing a predictive model of the electro-hydrogen production frequency modulation system based on the power dynamic characteristics of the electrolyzer; during frequency control, obtaining the system frequency and system inertia at the next moment through the predictive model based on the system frequency deviation, additional frequency modulation power, electrolyzer power, and system equivalent inertia at the current moment; obtaining the additional frequency modulation power at the next moment through an additional frequency modulation power control command planning model based on the system frequency and system inertia at the next moment, and generating a power control signal at the next moment based on the additional frequency modulation power at the next moment and the power control signal at the current moment; wherein, the additional frequency modulation power control command planning model is a model that solves for the additional frequency modulation power at the next moment through a cost function based on the system frequency deviation target, the system support inertia resource replenishment target, the electrolyzer power deviation target, and the system droop control parameter change target; and adjusting the power input of the electrolyzer at the next moment through the power control signal at the next moment to achieve frequency modulation.

[0007] In one embodiment, a predictive model for an electro-hydrogen production frequency regulation system is constructed based on the power dynamic characteristics of the electrolyzer, including: constructing a prime mover frequency regulation power model based on the parameters of the generator prime mover in the system; constructing a discretized frequency response model with electro-hydrogen production participation based on the system frequency response model and the prime mover frequency regulation power model; constructing a power frequency response model based on the power dynamic characteristics of the electrolyzer and performing first-order response delay processing to obtain an electrolyzer power change model; constructing a system frequency prediction model based on the electrolyzer power change model and the discretized frequency response model; and constructing a system inertia prediction model based on the electrolyzer power change model and the system power imbalance model.

[0008] In one embodiment, the system frequency prediction model is: .

[0009] In one embodiment, the system inertia prediction model is as follows: .

[0010] In one embodiment, the cost function based on the system frequency deviation target, the system support inertial resource replenishment target, the electrolyzer power deviation target, and the system droop control parameter change target is: .

[0011] In one embodiment, based on the system frequency and system inertia at the next moment, the additional frequency modulation power at the next moment is obtained through an additional frequency modulation power control command planning model, and a power control signal for the next moment is generated based on the additional frequency modulation power at the next moment and the power control signal at the current moment. This includes: inputting the system frequency and system inertia at the next moment into the additional frequency modulation power control command planning model and outputting the planning result; when the planning result is unsolvable, outputting zero additional frequency modulation power and setting the power control signal for the next moment as the power control signal for the current moment; when the planning result is solvable, obtaining the additional frequency modulation power based on the planning result and superimposing it with the power control signal for the current moment to obtain a superimposed value; when the superimposed value is within a preset superimposed range, outputting the superimposed value as the power control signal for the next moment; when the superimposed value is outside the preset superimposed range, outputting the limit value of the preset superimposed range as the power control signal for the next moment.

[0012] Secondly, this application provides an electro-hydrogen frequency modulation system, comprising: a modeling module for building a predictive model of the electro-hydrogen frequency modulation system based on the power dynamic characteristics of the electrolyzer; a calculation module for obtaining the system frequency and system inertia at the next moment through the predictive model based on the system frequency deviation, additional frequency modulation power, electrolyzer power, and system equivalent inertia at the current moment during frequency control; and for obtaining the additional frequency modulation power at the next moment through an additional frequency modulation power control command planning model based on the system frequency and system inertia at the next moment, and generating a power control signal at the next moment based on the additional frequency modulation power at the next moment and the power control signal at the current moment; wherein, the additional frequency modulation power control command planning model is a model that solves for the additional frequency modulation power at the next moment through a cost function based on the system frequency deviation target, the system support inertia resource replenishment target, the electrolyzer power deviation target, and the system droop control parameter change target; and a frequency modulation module for adjusting the power input of the electrolyzer at the next moment through the power control signal at the next moment to achieve frequency modulation.

[0013] Thirdly, this application provides an electrolytic cell control device, comprising: at least one memory for storing a program; and at least one processor for executing the program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to execute the method described in the first aspect or any possible implementation thereof.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0015] Fifthly, this application provides a computer program product that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0016] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0017] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art: This application effectively solves the control mismatch problem caused by insufficient consideration of power lag characteristics in traditional electro-hydrogen frequency regulation systems, leading to grid frequency oscillations. Specifically, firstly, a predictive model is built based on the dynamic power characteristics of the electrolyzer, which can proactively predict the system state, thus overcoming the inherent defects of lag. Based on this, during control, the frequency and inertia of the next moment are obtained using the predictive model from the current system state, enabling prediction of future system behavior and providing a basis for formulating control strategies in advance. Secondly, an instruction model using a multi-objective cost function programming is used to solve for the optimal additional frequency regulation power. This design simultaneously balances multiple objectives such as rapidly smoothing frequency deviations, maintaining system inertia levels, ensuring stable electrolyzer operation, and optimizing droop control parameters, thereby avoiding secondary oscillations that may be caused by single-objective control and ensuring a fast and smooth frequency regulation process. Finally, the calculated additional power is combined with the current control signal to convert it into the control signal for the next moment to adjust the electrolyzer power input, thus forming a closed-loop optimized control loop.

[0018] Compared with existing technologies, this method combines prediction and multi-objective optimization to proactively incorporate and compensate for the power lag characteristics of the electrolyzer and its impact into the control considerations. Ultimately, it achieves faster, more accurate, and more stable support for the power grid frequency and significantly suppresses frequency oscillations. Attached Figure Description

[0019] Figure 1 This is a diagram of the fast frequency response control framework for AEL-type electro-hydrogen production load based on MPC provided in the embodiments of this application; Figure 2 This is a diagram of the dynamic response model of electric hydrogen production with converter control provided in the embodiments of this application; Figure 3 This is a typical electro-hydrogen production frequency control structure diagram provided in the embodiments of this application; Figure 4 This is a microgrid frequency response model diagram with the participation of electro-hydrogen production provided in the embodiments of this application; Figure 5This is one of the flowcharts illustrating the frequency control method for an electric hydrogen production frequency modulation system based on model predictive control provided in this application embodiment; Figure 6 This is the second flowchart of the frequency control method for an electric hydrogen production frequency modulation system based on model predictive control provided in the embodiments of this application; Figure 7 This is a comparison chart of the effects of the proposed solution and the integrated inertia control under a given frequency fluctuation, as provided in the embodiments of this application. Figure 8 This is a wind speed fluctuation diagram set under a given working condition, as provided in the embodiments of this application; Figure 9 This is a comparison chart showing the effects of the proposed scheme of electric hydrogen production participating in frequency response under wind speed and frequency fluctuations, integrated inertia control, and adaptive integrated inertia control provided in the embodiments of this application. Figure 10 This is a wind speed fluctuation diagram set under a given complex working condition, as provided in the embodiments of this application; Figure 11 This is a comparison chart showing the effects of the proposed solution, integrated inertia control, and adaptive integrated inertia control under a given complex working condition, as provided in the embodiments of this application. Figure 12 This is a control schematic diagram of the electrolytic cell control device provided in the embodiments of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] In this article, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The symbol " / " in this article indicates that the related objects are in an "or" relationship; for example, A / B means A or B.

[0022] The terms "first" and "second," etc., used in the specification and claims herein are used to distinguish different objects, not to describe a specific order of objects. For example, "first response message" and "second response message," etc., are used to distinguish different response messages, not to describe a specific order of response messages.

[0023] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0024] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more, for example, multiple processing units means two or more processing units, multiple elements means two or more elements, etc.

[0025] Because the power response of AEL (Alternating Elasticity Variable) cells is relatively lagging, existing integrated inertia control methods can only select relatively conservative control parameters to avoid system frequency oscillations. Furthermore, they do not adequately consider the control requirements for frequency recovery, and the supporting role of electro-hydrogen production in frequency recovery is not fully realized. Moreover, the frequency may drop again due to the rapid decay of frequency regulation power during the frequency recovery phase caused by AEL lag. Existing islanded microgrids need to control the frequency regulation load to provide rapid power support to avoid exceeding the tripping threshold; then, they should recover to above the frequency anomaly threshold as quickly as possible to reduce the impact of the cumulative effect of frequency deviation. However, the recovery process of inertial frequency regulation resources will additionally increase the primary frequency regulation burden of the system. On the other hand, when the frequency is higher than the anomaly threshold, an excessively large droop coefficient may also cause over-frequency regulation, affecting normal user operation and the recovery of electro-hydrogen production frequency regulation capacity, potentially causing frequency oscillations. The electro-hydrogen production loads widely deployed in microgrids are of the AEL type. To fully explore the frequency regulation response potential of existing electro-hydrogen production, and considering that the power lag characteristics of AEL electro-hydrogen production limit its frequency regulation potential, this patent conducts research on optimized control strategies for AEL electrolyzers.

[0026] Based on this, the embodiments of this application will be described below with reference to the accompanying drawings. This application first analyzes the participation of AEL-type electro-hydrogen production loads in fast frequency response; please refer to... Figure 1 , Figure 1 This is a framework diagram of AEL-type electrohydrogen production load participating in fast frequency response control based on MPC provided in the embodiments of this application.

[0027] Understandably, in Figure 1 The system diagram including the hydrogen production load also includes: a microgrid dominated by new energy sources, and a BKB converter connecting the electrolyzer to the microgrid. This BKB converter consists of a grid-side rectifier, a DC voltage regulator, and a generator-side Buck converter. Refers to the fan speed; This refers to the inertial frequency modulation capacity that has been put into frequency response. This refers to the power control commands attached to the MPC.

[0028] Understandably, inertial response energy includes the kinetic energy of synchronous machine rotors, wind turbine rotors, and other inertial response resources, such as energy storage under VSG or virtual inertial control, and loads. Since the dynamic response characteristics of these types of equipment are linearly related to the rate of change of power or frequency, these inertial response resources play a crucial role in low-inertia microgrids by mitigating RoCoF, preventing the frequency from rapidly exceeding the tripping threshold, and buying time for primary frequency regulation. However, these inertial response resources will also absorb power from the grid during the frequency recovery process, becoming a major obstacle to frequency recovery.

[0029] Further, please refer to Figure 2 , Figure 2 This is a diagram illustrating the dynamic response model of electro-hydrogen production with converter control provided in an embodiment of this application. Figure 2 The dynamic response model for electro-hydrogen production shown in the figure consists of several parts. ez Refers to the voltage at the electrolytic cell port; i ez Refers to the current at the electrolytic cell port; u eabc The input voltage of the grid-side rectifier in the electric hydrogen production equipment; Refers to the reference input current at the electrolytic cell port; P ez These refer to the reference value and actual input power of the electrolytic cell, respectively; C ezb The capacitance value inside the Buck converter; L ezb This refers to the capacitance value inside the Buck converter; u edc C edc These refer to the DC bus voltage and capacitance value within the electric hydrogen power electronic converter, respectively; d refers to the duty cycle of the Buck converter. Refers to the internal resistance of the system; , These represent the double-layer capacitance effect on the electrode surfaces of the positive and negative electrodes, respectively. , These represent the charge transfer resistance between ions and electrons at the positive and negative electrodes, respectively. This is the reversible voltage required for a chemical reaction.

[0030] Understandably, under steady-state conditions, the model can be simplified to a static model that ignores capacitance effects and combines all resistances. However, under transient conditions, each EDL branch will reflect the dynamic characteristics of the electrolyzer. Figure 2 As can be seen from the circuit shown, when the current i ez When sudden changes are required, the EDL phenomenon acts as an energy barrier to electron flow, resulting in a response delay. If the rapid change in the electrolytic hydrogen production power is forcibly controlled, it may cause the DC bus capacitor voltage to exceed the limit or damage the equipment. Key parameters in the model, such as reverse voltage and internal resistance, are directly affected by the electrolyzer stack pressure and stack temperature, as shown in formulas (1) to (3): (1); (2); (3).

[0031] In the formula, The terminal voltage of the electrolytic cell stack varies over time; The internal resistance of a system that varies with pressure; Refers to the EDL branch voltage; Refers to the reference reversible voltage; This refers to the current internal pressure of the electrolytic cell; This refers to the rated pressure inside the electrolytic cell stack; The internal resistance of the reference system. The proportionality coefficient of the system's internal resistance as pressure changes.

[0032] For further analysis of typical frequency control structures for electro-hydrogen production, please refer to [reference needed]. Figure 3 , Figure 3 This is a typical frequency control structure diagram for electro-hydrogen production provided in an embodiment of this application. Wherein, D... AEL H AEL These refer to the droop and virtual inertia coefficients of AEL electro-hydrogen production, respectively. This refers to the additional frequency modulation power of AEL electro-hydrogen production. This refers to the rated power of the AEL electro-hydrogen production system. The integrated inertia control in the diagram is based on the filtered system frequency deviation value. The frequency response value of the electrogenerated hydrogen load is calculated and sent to the Buck controller after receiving the electrogenerated hydrogen power command. The Buck controller adjusts the duty cycle to change the input current of the electrogenerated hydrogen load, thereby changing the power consumed by the electrolyzer. The change in generator-side power causes DC bus voltage fluctuations. The grid-side rectifier adjusts its power control to maintain a stable capacitor voltage, thus reducing the input power along with the electrolyzer and providing equivalent frequency regulation power to the microgrid. The expression for the existing frequency response controller can be obtained from the figure: (4).

[0033] Specifically, the power electronic components of electro-hydrogen production equipment, including Buck converters and AC / DC rectifiers, all possess good dynamic response performance and the ability to quickly and accurately adjust power on a millisecond-level timescale. However, due to the unique EDL phenomenon of the electrolyzer, the power of the electro-hydrogen production equipment cannot change rapidly. Therefore, the overall dynamic characteristics are determined by the power dynamic characteristics of the electrolyzer. Based on... Figure 2 The equivalent circuit diagram of the electrolytic cell is given by equations (1) to (3). The power dynamic characteristics of the electrolytic cell can be approximated as a transfer function with first-order dynamic characteristics, as shown in equation (5): (5).

[0034] The time delay characteristic of the electrolytic cell can be approximated as a transfer function with first-order dynamic characteristics, as shown in equation (6): (6).

[0035] In the formula, Let be the time constant of the electrolyzer, then the power response model expression of the electrolyzer is shown in equation (7): (7).

[0036] Ultimately, based on the above content, we can... Figure 3 The electro-hydrogen production model can be transformed into a frequency response model; please refer to [reference needed]. Figure 4 , Figure 4 This is a frequency response model diagram of a microgrid with the participation of electro-hydrogen production, provided in an embodiment of this application.

[0037] against Figure 3 To address the shortcomings and improvement needs of existing typical models, this application proposes a frequency control method for an electric hydrogen production frequency modulation system based on model predictive control. Please refer to [reference needed]. Figure 5 , Figure 5 This is one of the flowcharts illustrating the frequency control method for an electric hydrogen production frequency modulation system based on model predictive control provided in this application. In this embodiment, the frequency control method for an electric hydrogen production frequency modulation system based on model predictive control includes steps S10 to S40.

[0038] Step S10: Build a predictive model for the frequency modulation system of electro-hydrogen production based on the power dynamic characteristics of the electrolyzer.

[0039] It should be noted that, based on the above discussion and background technology, the electro-hydrogen frequency regulation system is essentially a composite system, participating in grid frequency regulation through electro-hydrogen production. The predictive model for this composite system is designed to predict the system's state at the next moment.

[0040] Understandably, the state of a system is based on specific quantitative criteria, typically clearly characterized by several core parameters. Among these parameters, frequency is a crucial indicator, reflecting the balance between power supply and demand within the system; frequency changes directly reflect power fluctuations. Inertia is equally critical, representing the system's ability to maintain stable operation in the face of power changes; the magnitude of inertia affects the ease with which the system responds to sudden power adjustments. Therefore, the predictive model for an electric hydrogen production frequency regulation system should at least include a frequency model for the next time step and an inertia model for the next time step.

[0041] It should be noted that the predictive model of a system can usually be obtained from the frequency response model built for the system. Specifically, the frequency response model is a quantitative description of how a system reacts to frequency changes; it details the dynamic behavior and interrelationships of the system's components under different frequency fluctuations. Through this established frequency response model, key information and data closely related to the system's future state can be extracted. After a series of analyses and processing, the predictive model of the system can be obtained, providing strong support for accurately predicting the system's operating state in subsequent moments.

[0042] The core issue in this application lies in the power hysteresis characteristic of the electrolyzer, a key component of the electro-hydrogen frequency regulation system. Power hysteresis refers to the fact that when responding to a power change command, the electrolyzer's actual power output cannot immediately reach the required level; instead, it requires a certain period to gradually adjust. This time delay significantly impacts the operation of the entire electro-hydrogen frequency regulation system. Therefore, the power dynamic characteristics of the electrolyzer must be carefully considered when building the system's predictive model to ensure that the model accurately reflects the actual operation of the electro-hydrogen frequency regulation system. This will achieve high accuracy in the predictive model and provide a reliable basis for stable system operation and precise control.

[0043] It's important to note that when dealing with an electric hydrogen production frequency regulation system, the predictive models will inevitably differ depending on the components or even if the components are the same but the specifications vary. This is because different components and specifications of equipment will exhibit unique parameters and operating patterns in terms of power characteristics, response speed, and operating efficiency. These differences will directly affect the structure and parameter settings of the predictive model. Therefore, when first encountering an electric hydrogen production frequency regulation system, since there is no readily available suitable predictive model, it is necessary to build a suitable predictive model from scratch based on the specific components and equipment specifications of the system. This ensures that the model can accurately simulate and predict the system's operating state.

[0044] In subsequent processes, if the system undergoes no structural changes, such as the addition or removal of key components or the replacement of equipment, its overall characteristics and operational patterns remain relatively stable. In this case, the previously established predictive model can still accurately reflect the system's operation, eliminating the need to build a new predictive model and allowing the existing model to be used directly for system prediction and analysis.

[0045] In one feasible implementation, this application provides a method for implementing step S10. Based on Figure 1 and Figure 4 The proposed electro-hydrogen production frequency modulation system is analyzed and explained.

[0046] First, a frequency regulation power model of the prime mover needs to be built based on the parameters of the generator prime mover in the system.

[0047] It should be noted that the predictive model for the frequency regulation system of the electric hydrogen production system first requires obtaining the frequency response model of the entire system, while correct analysis requires first building the frequency regulation power model of the generator prime mover in the system. The building process is as follows.

[0048] Understandably, measuring power changes within a microgrid system requires additional communication links, making real-time optimization difficult for models incorporating this element. Therefore, using models that include equivalent speed governors and turbine time constants is preferable. The first-order model approximates the frequency modulation power of the prime mover. The approximate result is shown in equation (8): (8).

[0049] It is understandable that the prime mover frequency modulation power model is discretized to obtain the discretized prime mover model as shown in equation (9): (9).

[0050] in, T represents the system frequency deviation at the current moment. S R is the sampling period; R is the frequency-effective droop coefficient of the speed controller; T g The time constant is the equivalent governor and turbine.

[0051] Specifically, on a smaller time scale, such as conditions, Compared to the previous moment The changes are relatively small, and the changes are mainly affected by the measurable frequency variations. Impact. Therefore, equation (9) can be simplified to: (10).

[0052] Understandably, due to the change in power value The rate of change of the disturbance power varies with the speed regulation process of the synchronous machine, making it difficult to accurately model the system disturbance power. Therefore, this paper treats the change in disturbance power as an error.

[0053] Secondly, based on the system frequency response model and the prime mover frequency modulation power model, a discretized frequency response model with the participation of electric hydrogen production is constructed.

[0054] Understandably, please refer to Figure 4The control structure for the frequency characteristics of the isolated microgrid uses a classic frequency response aggregation simplification method to simulate the frequency response characteristics of the synchronous generator governor, turbine, and load, and synchronous generator ramp rate constraints are set simultaneously. The system frequency response model is shown in equation (11) below: (11); Where D is the frequency response coefficient of the load; H is the equivalent moment of inertia of the prime mover; For the motor input power before participating in FFR; The additional power generated by the prime mover after a disturbance event; This refers to the system disturbance power. (See...) That is, the equivalent frequency-modulated power provided by the motor, H and D, can be obtained through system parameter identification technology. In order to realize online control of the motor input power, the frequency response model of equation (11) is discretized using the Euler method: (12).

[0055] It is understandable that the changes in frequency, prime mover frequency modulation power, and motor output power at time k+1 are obtained as shown in equation (13): (13).

[0056] It is understandable that the change in prime mover power in equation (10) and the discretized frequency response model of the system in equation (12) are combined as shown in equation (14): (14); In the formula, This represents the change in frequency modulation power of the prime mover; This represents the change in disturbance power. For the overall error, see equation (15): (15). This represents the change in disturbance power; This refers to the frequency modulation power error of the prime mover.

[0057] It is understandable that by combining equations (14) and (15), the frequency response model of the system with flexible load participation is shown in equation (16): (16).

[0058] Furthermore, a power frequency response model is built based on the power dynamic characteristics of the electrolyzer and a first-order response delay is performed to obtain a model of the power change of the electrolyzer.

[0059] Understandably, given the slow power response of AEL-type electro-hydrogen production equipment, the control strategy employs additional power compensation. The frequency response is optimized by using the response model Equation 3 under the integrated inertia control of electro-hydrogen production, and after processing the first-order response delay, the electrolyzer power change model is obtained. It can be approximately equivalent to the following equation (17): (17).

[0060] Then, based on the electrolytic cell power change model and the discretized frequency response model, a system frequency prediction model is built.

[0061] It is understandable that by combining equations (16) and (17), the frequency prediction model of the microgrid with AEL-based hydrogen production is shown in equation (18): (18); in, (19); (20); (twenty one); (twenty two); (twenty three); (twenty four); (25); (26); (27).

[0062] in, The frequency prediction result for the system at the next time step; This represents the system frequency deviation at the current moment. Add frequency modulation power to the current moment; This represents the system frequency deviation at the previous moment. t is the current power of the electrolyzer; D is the frequency response coefficient of the load; H is the equivalent moment of inertia of the prime mover; TS is the sampling period; τHE is the delay coefficient of the electrolyzer; DAEL is the droop coefficient of the electrolyzer for hydrogen production; HAEL is the virtual inertia coefficient of the electrolyzer for hydrogen production; R is the equivalent active droop coefficient of the governor; Tg is the time constant of the equivalent governor and turbine. This represents the change in disturbance power; This refers to the frequency modulation power error of the prime mover.

[0063] Finally, based on the electrolytic cell power change model and the system power imbalance model, a system inertia prediction model is built.

[0064] It should be noted that since the inertial energy of a microgrid is actually the sum of the energy that opposes the change in frequency of the system, the change in this energy at a certain moment can be considered as the product of the power imbalance at that moment and time. The change in the imbalance energy in the system at this moment is shown in equation (28): (28).

[0065] in, This refers to the power imbalance in the system.

[0066] It is understandable that the existing power deviation of the system can be regarded as the product of the rate of change of frequency and the system inertia, while the power imbalance of the system also needs to take into account the change of the power of electro-hydrogen production under MPC control. Therefore, equation (29) can be rewritten as: (29).

[0067] Further transforming equation (29) into discrete form, we get: (30).

[0068] Equation (30) consists of two parts: one is the product of the system frequency (synchronous machine speed) and the equivalent inertia, where the equivalent inertia includes the rotational inertia within the system and the frequency-power inertia simulated by virtual inertial control or VSG control; the other is the work done by the power change of the electric hydrogen production load at this moment. It is evident that equation (30) conforms to the actual situation. Therefore, the predictive relationship for the system inertia can be obtained as follows: (31).

[0069] in, The predicted value of the system frequency change under the participation of electro-hydrogen production is obtained by further processing equation (31) to obtain the system inertia prediction model: (32).

[0070] in, (33); (34); (35); (36).

[0071] in, The inertia prediction result for the system at the next moment; This represents the system frequency deviation at the current moment. Add frequency modulation power to the current moment; This represents the system frequency deviation at the previous moment. The system's equivalent inertia at the current moment; D is the frequency response coefficient of the load; H is the equivalent rotational inertia of the prime mover; T S τ is the sampling period;HE D is the delay factor of the electrolytic cell; AEL H is the sag coefficient for hydrogen production by electrolysis. AEL R is the virtual inertia coefficient for hydrogen production by electrolysis; R is the active power equivalent droop coefficient of the speed governor; T g For the equivalent governor and turbine time constants; A1, B2 and C1, please refer to equations (24), (25) and (26).

[0072] Step S20: When performing frequency control, based on the current system frequency deviation, additional frequency modulation power, electrolytic cell power, and system equivalent inertia, the system frequency and system inertia at the next moment are obtained through a prediction model.

[0073] Understandably, when performing frequency control, the system frequency and system inertia at the next moment can be obtained by substituting the current system frequency deviation, additional frequency modulation power, electrolytic cell power, and system equivalent inertia into the two models respectively. The system frequency deviation, electrolytic cell power, and system equivalent inertia can be obtained through measurement or calculation of the current state; the additional frequency modulation power is obtained through step S30, and can be defaulted to 0 at the start-up moment.

[0074] Step S30: Based on the system frequency and system inertia at the next moment, obtain the additional frequency modulation power at the next moment through the additional frequency modulation power control command planning model, and generate the power control signal at the next moment based on the additional frequency modulation power at the next moment and the power control signal at the current moment.

[0075] It should be noted that successfully obtaining the specific value of the system frequency and the relevant data of the system inertia at the next moment means that the possible operating conditions of the system in the next state have been effectively predicted. Based on this prediction result, and with the help of the additional frequency modulation power control command planning model, the additional frequency modulation power required at the next moment can be accurately calculated. Subsequently, by implementing corresponding control operations, any deviations or anomalies that may occur in the system can be corrected in a timely and accurate manner, thereby ensuring the stable and reliable operation of the system.

[0076] It is understood that in this application, the additional frequency modulation power control command planning model is a model that solves for the additional frequency modulation power at the next moment by using a cost function based on the system frequency deviation target, the system support inertia resource replenishment target, the electrolyzer power deviation target, and the system droop control parameter change target.

[0077] Specifically, regarding the system frequency deviation target, excessive frequency deviation can lead to serious problems such as equipment damage and power outages. By appropriately adjusting the additional power, the power imbalance is precisely compensated to minimize the system frequency deviation, ensuring stable system operation near the rated frequency, rather than aimlessly depleting equipment resources.

[0078] Specifically, regarding the goal of replenishing system-supporting inertial resources, with the large-scale integration of new energy sources, the system's equivalent inertia decreases, making inertial resources crucial for suppressing rapid frequency changes. The additional power aims to achieve optimal replenishment of system-supporting inertial resources with minimal additional power input, thereby increasing the system's equivalent inertia.

[0079] Specifically, regarding the target for electrolyzer power deviation, the power lag of the electrolyzer affects the frequency modulation effect and the electro-hydrogen production process. The additional power should minimize the power deviation of the electrolyzer to ensure that the electrolyzer operates under relatively stable conditions and improve the efficiency and quality of electro-hydrogen production.

[0080] Specifically, regarding the target change in system droop control parameters, droop control is a commonly used frequency modulation control strategy. The change in system droop control parameters affects the frequency modulation resource response characteristics and power allocation effect. The additional power should minimize the change in system droop control parameters to achieve flexible and stable frequency modulation control.

[0081] Based on this, the planning problem for solving the MPC of the proposed AEL electric hydrogen production load is shown in equation (37): (37).

[0082] It is understandable that the four terms in this planning problem refer to the reduction of frequency deviation, the replenishment of support inertia resources, the change in electrolytic cell power deviation, and the droop control coefficient. The inequalities in the conditions imply that the changes in frequency, frequency deviation, electrolytic cell power deviation, and droop control coefficient should be within the maximum and minimum allowable ranges.

[0083] Specifically, the cost function based on the system frequency deviation target, the system support inertial resource replenishment target, the electrolyzer power deviation target, and the system droop control parameter change target is: (38).

[0084] in: (39); (40); (41); (42).

[0085] in, , These are the predicted values ​​for the system frequency and the system's consumed inertial capacity, respectively, with the participation of electro-hydrogen production. The rated frequency; This represents the deviation between the current electrolytic cell power and the rated power. This represents the change in power between the current moment and the previous moment. , , These are the weight matrices for the system frequency deviation, the system inertial capacity deviation, and the electro-hydrogen power deviation, respectively. , , These are the normalization coefficients for RoCoF, frequency, and rotational speed, respectively, used to unify the various physical dimensions to the same order of magnitude.

[0086] It should be noted that in the sum of the four terms of the cost function, the first term represents the system frequency deviation target through the frequency deviation values ​​at the next three time points; the second term represents the inertia at the next three time points to support the replenishment of inertial resources, which means reducing the consumed inertial capacity deviation value, which is equivalent to encouraging AEL to continuously provide frequency modulation power during the frequency recovery phase to help the fan / synchronous machine, etc., replenish inertial energy; the third term is the limitation on the power deviation of the electrolyzer to maintain the stability of the electrolyzer equipment.

[0087] It should be noted that the fourth item takes into account minimizing the change in the droop control coefficient. The principle of droop control in this model is to multiply the difference between the current frequency and the rated frequency by the droop coefficient as the power reduction value control signal and send it to the electrolytic cell. Therefore, the control includes both the current moment and the future moment. This controls the effect of droop control on the overall control, thus achieving the goal of correcting the change in the droop coefficient.

[0088] Specifically, the weights can be adjusted based on the key or most important objectives that need to be considered during the process, or they can be used to achieve optimal matching of the scenario through model algorithms.

[0089] Furthermore, equation (38) is transformed into the standard form of MPC controller operation, i.e. The quadratic programming equation to be solved is used as the planning model in this application: (43).

[0090] in: (44); (45); (46).

[0091] In the formula, , These refer to the upper and lower limits of the variation in AEL electro-hydrogen production power, respectively. , These refer to the upper and lower limits of power change at each moment.

[0092] In one feasible implementation, this application provides one way to implement step S30. Please refer to... Figure 6 , Figure 6 This is the second schematic flowchart of the frequency control method for an electro-hydrogen production frequency modulation system based on model predictive control provided in this application embodiment. Step S30 includes steps S31 to S34.

[0093] Step S31: Input the system frequency and system inertia of the next moment into the additional frequency modulation power control command planning model, and output the planning result.

[0094] Understandably, the MPC controller updates the parameters of equation (44) in each sampling period, solves equation (43) under constraints through quadratic programming, obtains the optimal control sequence in the future finite time domain, and takes the first term of the calculation result as the MPC output result, i.e., the additional frequency modulation power. .

[0095] Step S32: When the planning result is unsolvable, the output additional frequency modulation power is zero, and the power control signal at the next moment is set to the power control signal at the current moment.

[0096] Understandably, when the quadratic programming problem has no solution and the constraints are not met, the electric hydrogen production load maintains its current power operation and is solved again. The reference power value of the grid-side rectifier is shown in equation (47): (47).

[0097] Step S33: When the planning result is a solution, obtain the additional frequency modulation power based on the planning result and then superimpose it with the power control signal at the current moment to obtain the superimposed value. When the superimposed value is within the preset superimposed range, output the superimposed value as the power control signal at the next moment.

[0098] Step S34: When the superimposed value is outside the preset superimposed range, output the limit value of the preset superimposed range as the power control signal for the next moment.

[0099] It is understandable that when a quadratic programming problem has a solution and satisfies the constraints: MPC output results Add this to the combined inertia control value from the previous step to obtain the optimized solution for the input power. As shown in equation (48): (48).

[0100] Understandably, flexible loads using electric hydrogen production have a large margin for both power up- and down-adjustment; therefore, a minimum input power value is set. It is 0.4 pu, the maximum value The power control signal sent by the MPC controller is 1.5 pu. The expression is shown in (49): (49).

[0101] Step S40: Adjust the power input of the electrolytic cell at the next moment by using the power control signal at the next moment to achieve frequency modulation.

[0102] It is understandable that the solution obtained Send to the Buck converter, via Figure 2 By converting the classic buck control into a duty cycle, the power input of the electrolytic cell can be adjusted.

[0103] In one specific embodiment, this application provides an analysis under a specific operating condition, please refer to... Figure 7 See Table 1. Figure 7 The frequency response capabilities of the electrogenerated hydrogen load using the integrated inertia control method and the proposed MPC method were compared under the condition that a 20kW load was put into the microgrid and the instantaneous power increase was 16% of the microgrid's rated capacity.

[0104] Table 1:

[0105] Understandably, as can be seen from the comparison, the electro-hydrogen production system begins to respond at the beginning of the frequency disturbance, reducing the power deficit of the microgrid and suppressing the frequency drop by decreasing the electro-hydrogen production power. Due to the dynamic characteristics of the AEL electrolyzer, neither the existing integrated inertia control nor the MPC control method proposed in this paper can further increase the power change rate to suppress RoCoF. In the first 2 seconds before the frequency drop, the system RoCoFs are 0.195Hz / s and 0.185Hz / s, respectively, with little difference. The lowest system frequencies are 49.61Hz and 49.68Hz, respectively, differing by only 0.07Hz. As shown at point ① in the figure, the electro-hydrogen production power decreases slowly, gradually decreasing to 20kW about 3 seconds after the disturbance, significantly slower than the inertial response wind turbine providing power support with a rapid change of 40kW within 0.2s. However, subsequently at point ②, the support effect of the MPC-controlled electro-hydrogen production equipment for frequency recovery gradually becomes apparent, with the system frequency recovery speed significantly faster than the integrated inertia control method. This is because the MPC provides additional frequency modulation power commands to continuously expand the equivalent frequency modulation power, thus addressing the problem of the system's primary frequency modulation burden gradually increasing due to the gradual reduction of the fan's frequency modulation power under virtual inertia control.

[0106] Considering that the existing standard design frequency regulation threshold is within ±0.1Hz to 0.2Hz, the duration of frequencies below this threshold is 16.18s using the original method. However, when using the frequency response with the present invention, this duration is shortened to 10.19s, a reduction of 63%, significantly supporting rapid frequency recovery. The proposed method also suppresses the frequency drop depth and reduces the frequency recovery time. The integral of the time below the threshold and the frequency exceeding the limit is the frequency anomaly area. When using the proposed MPC method, the frequency anomaly area (I) is only 37.14% of the frequency anomaly area (I+II) when using the integrated inertia response. In addition, it saves the frequency regulation capacity of the wind turbine, increasing the minimum wind turbine speed from 7.3 rad / s under the original integrated inertia control to 7.7 rad / s under MPC control, reserving more inertial response resources. It is evident that the present invention can effectively reduce the frequency fluctuation of the system under single power step conditions and improve the frequency security of the microgrid.

[0107] And based on Figure 8 When a wind speed is set, rapid fluctuations in wind speed cause the microgrid's power to change by approximately 20% from the 2nd second to the 10th second, resulting in a gradual drop in the microgrid frequency. A comparison is made using integrated inertia control, an adaptive inertia control method originally used for the frequency response of electrolytic aluminum, and the MPC method proposed in this application. The participation of the electro-hydrogen load in the frequency response and its support capability for primary frequency regulation are observed. This simulates the power variation of wind turbines under unstable wind speeds. System frequency-power comparison is shown below. Figure 9 As shown in Table 2, the comparison of various indicators is presented.

[0108] Table 2:

[0109] It is understandable that, based on the proposed optimized control strategy, this invention adds frequency modulation power commands to continuously support system frequency recovery, causing the frequency to stop declining and rebound as quickly as possible. It can be seen that... Figure 9 The lowest frequency at point ③ is 49.62Hz, higher than 49.51Hz with the integrated inertia control method and 49.59Hz with adaptive integrated inertia control. Furthermore, the control system of this invention with the added MPC controller remains below 49.75Hz for 9.14 seconds, while the original integrated inertia control system takes 15.38 seconds to recover to 49.8Hz, approximately 1.8 times longer than this invention. Although the adaptive integrated inertia control method dynamically adjusts the control coefficient based on the frequency response state, increasing the lowest frequency by approximately 37% to 49.59Hz, it fails to adequately support the primary frequency modulation of the system because it does not consider continuously expanding the frequency modulation output to accelerate frequency recovery, thus failing to effectively reduce the cumulative effect of system frequency deviation.

[0110] The reduced wind speed caused a decrease in turbine output of approximately 40kW, resulting in system frequency recovery relying on a secondary frequency modulation process, which was slow. However, compared to... Figure 9 The proportion of frequency anomalies below 49.9Hz, achieved using this invention, is only 42.84% of the frequency anomaly area under the original integrated inertia control. Furthermore, this invention avoids excessive consumption of the system's inertia response resources. The minimum wind turbine generator speed controlled by this invention during hydrogen production is 7.24 rad / s, higher than the 6.56 rad / s under integrated inertia control and the 6.67 rad / s under the adaptive method, thus avoiding the risk of a secondary frequency drop due to forced exit from frequency support caused by excessively low turbine speed. It is evident that this invention achieves better frequency response control. As shown in Table 2, based on the proposed method, the three key indicators—minimum frequency value, frequency recovery time, and frequency anomaly area proportion—are all superior to both the integrated inertia control method and the adaptive method.

[0111] To further verify the response capability of the electrogenerated hydrogen load under the complex operating conditions of an isolated microgrid, a 30kW load was applied to the isolated microgrid at 15s, followed by a 20kW load at 30s, with continuous wind speed fluctuations. The frequency response of the system was observed under the condition of repeated power changes caused by the two load applications and wind turbine power fluctuations, and various frequency response control methods were compared. A comprehensive comparison was made between inertia control, the adaptive inertia control method originally used for the frequency response of electrolytic aluminum, and the present invention. The random wind speed fluctuations were as follows: Figure 10 As shown, the simulation results are as follows: Figure 11 As shown in Table 3, the comparison of various indicators is presented.

[0112] Table 3:

[0113] It can be seen that under wind speed fluctuations, the wind turbine power is disturbed, and the grid frequency oscillates slightly. Figure 11 The introduction of a 30kW load at point ① triggered a frequency drop in the system, which occurred rapidly. The electro-hydrogen load employing various control strategies reduced its output power in the initial stage of the frequency drop. However, due to the limitations of the dynamic response speed of electro-hydrogen, the RoCoF in the initial response phase could not be effectively suppressed, leading to a rapid frequency drop. Subsequently, the load under the control of this invention continuously increased its frequency modulation power according to the multi-objective control strategy, supporting rapid frequency recovery. However, the electro-hydrogen load under integrated inertia control and adaptive inertia control reduced its frequency modulation power as the frequency drop bottomed out (i.e., the RoCoF negative value decreased), thus lowering its frequency and increasing the primary frequency modulation burden. After the load at point ① was introduced, the lowest system frequency under the proposed MPC control dropped rapidly to below 49.83Hz, higher than the 49.81Hz of adaptive control and the 49.77Hz of integrated inertia control.

[0114] In this embodiment, by employing a model predictive control framework, the control mismatch problem caused by insufficient consideration of power lag characteristics in traditional electro-hydrogen frequency regulation systems, leading to grid frequency oscillations, is effectively solved. Specifically, firstly, a predictive model is built based on the dynamic power characteristics of the electrolyzer. This method can proactively predict the system state, thus overcoming the inherent defects of lag. Based on this, during control, the frequency and inertia of the next moment are obtained using the predictive model from the current system state. This enables the prediction of future system behavior and provides a basis for formulating control strategies in advance. Secondly, an instruction model using a multi-objective cost function programming is used to solve for the optimal additional frequency regulation power. This design simultaneously balances multiple objectives such as rapidly smoothing frequency deviations, maintaining system inertia levels, ensuring the stable operation of the electrolyzer, and optimizing droop control parameters. This avoids secondary oscillations that may be caused by single-objective control, ensuring a fast and smooth frequency regulation process. Finally, the calculated additional power is combined with the current control signal to convert it into the control signal for the next moment to adjust the electrolyzer power input, thus forming a closed-loop optimized control loop.

[0115] Compared with existing technologies, this method combines prediction and multi-objective optimization to proactively incorporate and compensate for the power lag characteristics of the electrolyzer and its impact into the control considerations. Ultimately, it achieves faster, more accurate, and more stable support for the power grid frequency and significantly suppresses frequency oscillations.

[0116] In addition, this application provides an electro-hydrogen frequency modulation system, comprising: a modeling module for building a predictive model of the electro-hydrogen frequency modulation system based on the power dynamic characteristics of the electrolyzer; a calculation module for obtaining the system frequency and system inertia at the next moment through the predictive model based on the system frequency deviation, additional frequency modulation power, electrolyzer power, and system equivalent inertia at the current moment during frequency control; and for obtaining the additional frequency modulation power at the next moment through an additional frequency modulation power control command planning model based on the system frequency and system inertia at the next moment, and generating a power control signal at the next moment based on the additional frequency modulation power at the next moment and the power control signal at the current moment; wherein, the additional frequency modulation power control command planning model is a model that solves for the additional frequency modulation power at the next moment through a cost function based on the system frequency deviation target, the system support inertia resource replenishment target, the electrolyzer power deviation target, and the system droop control parameter change target; and a frequency modulation module for adjusting the power input of the electrolyzer at the next moment to achieve frequency modulation through the power control signal at the next moment.

[0117] It should be understood that the above-described device is used to execute the methods in the above embodiments. The implementation principle and technical effect of the corresponding program modules in the device are similar to those described in the above methods. The working process of the device can be referred to the corresponding process in the above methods, and will not be repeated here.

[0118] Based on the methods in the above embodiments, this application provides an electrolytic cell control device, including: a central processing unit (CPU), a communication interface, an analog-to-digital converter (ADC), a PWM controller, a memory, and a communication bus. The CPU, communication interface, ADC, and memory communicate with each other via the communication bus. The processor can call logical instructions from the memory to execute the methods in the above embodiments.

[0119] Specifically, refer to Figure 12 During operation, the electrolytic cell control equipment can receive current and voltage signals from the electrolytic cell measured on-site via an analog-to-digital converter interface, as well as the AC voltage signal from the system. It can also receive fan speed signals transmitted remotely via a communication interface. By selecting appropriate interfaces for different data types, latency can be minimized. The system frequency is obtained from the AC voltage, and the electrolytic cell power is obtained from the voltage and current signals.

[0120] Next, the control device processes the signal by calling the logic instructions of the above-described embodiment method in the memory through the central processing unit, and outputs the chopper drive signal to the PWM controller. The PWM controller then outputs a high-frequency 0 / 1 signal to the Buck circuit to adjust the power input of the electrolytic cell.

[0121] In addition, the control equipment can also be equipped with human-machine interaction functions. For example, a status monitoring display screen can be added to acquire and intuitively display the received data through the communication bus; or a start / stop switch can be added to indicate whether the central processing unit needs to call logic instructions through the communication bus.

[0122] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0123] Based on the methods in the above embodiments, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0124] Based on the methods in the above embodiments, this application provides a computer program product that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0125] It is understood that the various numerical designations used in the embodiments of this application are for descriptive convenience only and are not intended to limit the scope of the embodiments of this application.

[0126] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A frequency control method for a model predictive control-based frequency regulation system for electro-hydrogen generation, characterized in that, The method comprises the following steps: building a prediction model of a frequency regulation system of the hydrogen production by electrolysis based on power dynamic characteristics of an electrolyzer; when frequency control is performed, based on a system frequency deviation at a current time, an additional frequency regulation power, an electrolyzer power and an equivalent inertia of the system, a system frequency and a system inertia at a next time are obtained through the prediction model; based on the system frequency and the system inertia at the next time, an additional frequency regulation power at the next time is obtained through an additional frequency regulation power control instruction planning model, and a power control signal at the next time is generated based on the additional frequency regulation power at the next time and a power control signal at the current time; wherein the additional frequency regulation power control instruction planning model is a model for solving the additional frequency regulation power at the next time based on a system frequency deviation target, a system supported inertia resource backfill target, an electrolyzer power deviation target and a system droop control parameter variation target; the power input of the electrolyzer at the next time is adjusted through the power control signal at the next time to realize frequency regulation.

2. The model predictive control based frequency control method for an electrical hydrogen production frequency regulation system as claimed in claim 1, wherein, The method comprises the following steps: building a prime mover frequency regulation power model according to parameters of a prime mover of a generator in the system; building a discretized frequency response model with participation of the hydrogen production by electrolysis according to a system frequency response model and the prime mover frequency regulation power model; building a power frequency response model based on power dynamic characteristics of the electrolyzer and performing first-order response delay processing to obtain an electrolyzer power variation model; building a system frequency prediction model according to the electrolyzer power variation model and the discretized frequency response model; building a system inertia prediction model according to the electrolyzer power variation model and a system power imbalance model.

3. The model predictive control based frequency control method for a frequency regulation system of an electrical hydrogen production plant as claimed in claim 2, wherein, The system frequency prediction model is: ； ； ； ； ； ； ； ； ； ； Wherein, is the system next time frequency prediction result; is the current time system frequency deviation; is the current time additional frequency modulation power; is the last time system frequency deviation; is the current time electrolytic cell power; D is the frequency response coefficient of the load; H is the equivalent rotational inertia of the prime mover; T S is the sampling period; τ HE is the time delay coefficient of the electrolytic cell; D AEL is the droop coefficient of the electrolytic cell hydrogen production; H AEL is the virtual inertia coefficient of the electrolytic cell hydrogen production; R is the frequency active equivalent droop coefficient of the governor; T g is the equivalent governor and turbine time constant; is the disturbance power change amount; is the prime mover frequency modulation power error.

4. The model predictive control based frequency control method for a frequency regulation system of an electrical hydrogen production plant as claimed in claim 2, wherein, The system inertia prediction model is: ； ； ； ； ； ； ； ； wherein, is the system inertia prediction result for the next time instant; is the system frequency deviation at the current time instant; is the additional frequency modulation power at the current time instant; is the system frequency deviation at the previous time instant; is the system equivalent inertia at the current time instant; D is the frequency response coefficient of the load; H is the equivalent rotational inertia of the prime mover; T S is the sampling period; τ HE is the time delay coefficient of the electrolyzer; D AEL is the droop coefficient of the electrolyzer for hydrogen production; H AEL is the virtual inertia coefficient of the electrolyzer for hydrogen production; R is the frequency active equivalent droop coefficient of the governor; T g is the equivalent time constant of the governor and the turbine.

5. The model predictive control based frequency control method for a frequency regulation system of an electrical hydrogen production plant as claimed in claim 1, wherein, a cost function based on a system frequency deviation target, a system supported inertia resource backfill target, an electrolyzer power deviation target and a system droop control parameter variation target is: ； ； ； ； ； Wherein, , are respectively the predicted value of the system frequency under the participation of the electrolytic hydrogen, and the predicted value of the system consumed inertia capacity; is the rated frequency; is the deviation amount of the electrolytic tank power at the current moment from the rated power; is the power change amount between the current moment and the previous moment; , , are respectively the weight matrix of the system frequency deviation value, the weight matrix of the system inertia capacity deviation value, and the weight matrix of the electrolytic hydrogen power deviation value; , , are respectively the normalization coefficients of RoCoF, frequency, and rotating speed, which are used to unify the dimensions of each physical quantity to the same order of magnitude.

6. The model predictive control based frequency control method for a frequency regulation system of an electrical hydrogen production plant as claimed in claim 1, wherein, based on the system frequency and the system inertia at the next time, an additional frequency regulation power at the next time is obtained through an additional frequency regulation power control instruction planning model, and a power control signal at the next time is generated based on the additional frequency regulation power at the next time and a power control signal at the current time, comprising: inputting the system frequency and the system inertia at the next time into the additional frequency regulation power control instruction planning model to output a planning result; when the planning result is no solution, outputting the additional frequency regulation power as zero and setting the power control signal at the next time as the power control signal at the current time; when the planning result is a solution, obtaining the additional frequency regulation power based on the planning result and then superimposing the additional frequency regulation power with the power control signal at the current time to obtain a superimposed value, and when the superimposed value is within a preset superimposed range, outputting the superimposed value as the power control signal at the next time; when the superimposed value is outside the preset superimposed range, outputting a limit value of the preset superimposed range as the power control signal at the next time.

7. An electric hydrogen production frequency modulation system characterized by, The method comprises the following steps: building a prediction model of a frequency regulation system of the hydrogen production by electrolysis based on power dynamic characteristics of an electrolyzer through a modeling module; The solution module is configured to, when frequency control is performed, acquire, based on a system frequency deviation at a current time, an additional frequency modulation power, an electrolytic tank power, and a system equivalent inertia, a system frequency and a system inertia at a next time through the prediction model; and further configured to acquire, based on the system frequency and the system inertia at the next time, an additional frequency modulation power at the next time through an additional frequency modulation power control instruction planning model, and generate a power control signal at the next time based on the additional frequency modulation power at the next time and a power control signal at the current time; wherein the additional frequency modulation power control instruction planning model is a model for solving the additional frequency modulation power at the next time based on a system frequency deviation target, a system support inertia resource backfill target, an electrolytic tank power deviation target, and a system droop control parameter change target The frequency modulation module is configured to adjust a power input of the electrolytic tank at the next time through the power control signal at the next time to achieve frequency modulation.

8. An electrolysis cell control device, characterized by, The computer program product comprises: at least one memory for storing a computer program; at least one processor for executing the program stored in the memory, and when the program stored in the memory is executed, the processor is configured to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. When the computer program runs on the processor, the processor is caused to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterised in that, When the computer program product runs on the processor, the processor is caused to execute the method according to any one of claims 1 to 6.

Images