Frequency response control method for proton exchange membrane electrolyzer considering hydrogen confinement in oxygen

Abstract

The present invention proposes a frequency response control method for a proton exchange membrane electrolyzer considering the constraint of hydrogen in oxygen, which belongs to the technical field of hydrogen energy storage application in power systems; the present invention proposes a calculation formula for the hydrogen content in oxygen in the anode channel of a proton exchange membrane electrolyzer, which constitutes the constraint of hydrogen in oxygen; a holographic target feedback nonlinear frequency dynamic response control strategy considering the constraint of hydrogen in oxygen is proposed, and an expanded state observer is used to avoid the complex calculation process of partial system parameter acquisition and nonlinear control law; the dynamic frequency response of a proton exchange membrane electrolyzer considering the constraint of hydrogen in oxygen is analyzed. The present invention studies the dynamic frequency response control method of a proton exchange membrane electrolyzer, which provides certain guiding significance for the promotion of my country's green and low-carbon energy transformation and the coupling of hydrogen energy with the power system.

Description

Technical Field

[0001] The present invention belongs to the technical field of hydrogen energy storage application in power systems, and in particular relates to a frequency dynamic response control method for a proton exchange membrane electrolyzer taking into account hydrogen constraints in oxygen. Background Art

[0002] With the advancement of my country's green and low-carbon energy transformation, the coupling of hydrogen energy and power systems will show a rapid development trend. In the process of building a new power system, the electrolyzer load has the characteristics of large capacity, high flexibility, and strong regulation ability, which can provide frequency regulation auxiliary services for the power grid. Research on the response of proton exchange membrane electrolyzers to grid frequency will help decarbonize the power system and improve the stability of low-inertia power systems.

[0003] The study of proton exchange membrane electrolyzer's participation in grid frequency response currently only considers the external electrical characteristics of the electrolyzer, namely the voltage and current characteristics, and has never been combined with the internal operating characteristics of the electrolyzer. To fill this gap, the present invention proposes to use the hydrogen content in the anode channel oxygen as an operating constraint.

[0004] Traditional control methods for proton exchange membrane electrolyzers participating in grid frequency response all revolve around droop control designed around the active power-frequency characteristics of the electrolyzer. This control method has a fixed proportional coefficient and cannot fully release the adjustable capacity of the electrolyzer. Summary of the invention

[0005] The object of the present invention is to provide a proton exchange membrane electrolyzer frequency dynamic response control method which can take into account the constraints of the internal operating characteristics of the proton exchange membrane electrolyzer, has a fast adjustment speed, and can achieve system frequency recovery when the adjustable power of the proton exchange membrane electrolyzer is sufficient.

[0006] To achieve the above objectives, the present invention proposes a frequency dynamic response control method for a proton exchange membrane electrolyzer considering the constraint of hydrogen in oxygen, aiming at the potential exploration and safety control problem of the electrolyzer load response to the frequency of the power system. A calculation formula for the hydrogen content in oxygen in the anode channel of the proton exchange membrane electrolyzer is proposed to constitute the constraint of hydrogen in oxygen; a holographic target feedback nonlinear frequency dynamic response control strategy considering the constraint of hydrogen content in oxygen is proposed, and an extended state observer is used to avoid the complex calculation process of partial system parameter acquisition and nonlinear control law; the dynamic response of the proton exchange membrane electrolyzer frequency considering the constraint of hydrogen in oxygen is analyzed. The study of the dynamic response control method of the proton exchange membrane electrolyzer frequency provides certain guiding significance for the promotion of my country's green and low-carbon energy transformation and the coupling of hydrogen energy and the power system.

[0007] The specific technical solution of the method for controlling the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen constraint in oxygen of the present invention is as follows:

[0008] A method for controlling the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen confinement in oxygen comprises the following steps:

[0009] When the proton exchange membrane electrolyzer responds to the grid frequency, the operation is subject to the constraint of hydrogen in oxygen in the anode channel, and the constraint of hydrogen in oxygen is composed of the concentration of hydrogen in oxygen in the anode channel of the proton exchange membrane electrolyzer and a safety threshold;

[0010] When the proton exchange membrane electrolyzer responds to the grid frequency, a holographic target feedback nonlinear control method is used, and an extended state observer is used to observe the system frequency change, and the expanded observation disturbance term is used to obtain the final system control law.

[0011] Further, the hydrogen in the oxygen is constrained to be:

[0012]

[0013] Where: HTO is the hydrogen content of oxygen in the anode channel of the proton exchange membrane electrolyzer; C H,a is the hydrogen concentration in the anode channel; C O,a is the oxygen concentration in the anode channel; α is the safety threshold of hydrogen in oxygen in the electrolyzer.

[0014] Furthermore, the safety threshold, the design of the safety threshold of hydrogen content in oxygen takes into account that in actual engineering production, what is monitored is the hydrogen content in oxygen in the subsequent auxiliary equipment of the electrolyzer, and its value is slightly lower than the hydrogen content in oxygen in the anode channel of the electrolyzer, that is, when the hydrogen content in oxygen in the anode channel reaches the safety threshold, the hydrogen content in oxygen in the auxiliary equipment has not reached the limit, leaving a certain safety margin, which meets the needs of safe production.

[0015] Furthermore, if it is monitored that the hydrogen content in oxygen has reached the safety threshold, emergency measures must be taken to speed up the flow rate of the anode and cathode mixture to accelerate the decrease of the hydrogen content in oxygen and stipulate that during the period when the hydrogen content in oxygen exceeds the safety threshold, the electrolyzer temporarily exits the frequency modulation and its power remains unchanged until the hydrogen content in oxygen is less than or equal to the safety threshold, and then continues to return to the frequency modulation process.

[0016] Furthermore, the emergency measures are designed taking into account that after the use of the measures, the hydrogen content in the anode channel oxygen will not immediately return to below the safety threshold, but a margin has been left for auxiliary equipment, etc. and exceeding it within the short time range of a primary frequency modulation with a response in seconds is acceptable from the perspective of safe operation.

[0017] Furthermore, the holographic target feedback nonlinear control, the specific control method is as follows:

[0018] Step A1: Determine the control target of the system, wherein the control target is that the grid frequency change tends to zero; select the grid frequency change, the grid frequency change and the electrolyzer power change as the state variable, the output variable and the control input respectively;

[0019] Step A2: construct a single-input single-output nonlinear control system equation, and incorporate the control input into the rotor motion equation of the grid equivalent synchronous generator to determine whether the control input is a first-order relative-order output of the nonlinear system;

[0020] Step A3: Construct a Brunovsky standard dynamic system based on the output variables and control objectives, and construct an optimization performance index function;

[0021] Step A4: To achieve system stability and optimal performance, determine the weights of the pre-controlled variables to obtain the feedback control law. Here, the feedback control law obtained is the Brunovsky standard type.

[0022] Step A5: The system control law is derived from the rotor motion equation of the equivalent synchronous generator of the coupled power grid, the Brunovsky standard form and the feedback control law. The system control law obtained here is the system control law without introducing the extended state observer.

[0023] Furthermore, the single-input single-output nonlinear system equation is:

[0024] Where: is the time derivative vector of the state variable; y is the output variable vector; u is the control input; x is the column vector of the state variable set; r(x) is the state vector of the system; s(x) is the coefficient vector of the control input, and h(x) is the output vector;

[0025] The rotor motion equation of the grid equivalent synchronous generator is:

[0026] Where: Δω is the frequency change of the system; t is the time; dΔω / dt is the derivative of the frequency change of the system with respect to time; ΔP M is the change in active output of the synchronous generator; ΔP L is the load power change; ΔP pem is the power change of the electrolyzer; D is the load frequency coefficient, which is equal to the load change percentage divided by the frequency change percentage; M is the system equivalent inertia time constant; Δω is the system frequency change. It should be noted that the rotor speed per unit value of the synchronous generator is equal to the grid frequency per unit value; load specifically refers to other loads that do not include the electrolyzer;

[0027] The Brunovsky canonical form is:

[0028] Where: v is the pre-controlled variable; is the derivative of the system frequency change with respect to time;

[0029] The optimization performance index function is:

[0030] Where: J is the optimization performance index function; a is the pre-controlled variable weight; v is the pre-controlled variable; I is the first component of the Lie derivative matrix of the output quantity, which specifically refers to the system frequency change in this patent; Q is the state weight matrix; is the integral of time from zero to infinity;

[0031] The feedback control law is: v = -Δω

[0032] In the formula: Δω is the frequency change of the system; v is the pre-controlled variable;

[0033] The system control law: ΔP pem =ΔP M -ΔP L +(MD)Δω

[0034] Where: ΔP M is the change in active output of the synchronous generator; ΔP L is the load power change; ΔP pem is the power change of the electrolytic cell; D is the load frequency coefficient, which is equal to the load change percentage divided by the frequency change percentage; M is the system equivalent inertia time constant; Δω is the system frequency change.

[0035] Further, the method of using the extended state observer is as follows:

[0036] Step B1: Incorporate the control input into the rotor motion equation of the grid equivalent synchronous generator, and regard the term that does not contain the control input as a disturbance term;

[0037] Step B2: Use an extended state observer to observe the frequency change of the power grid, the extended term is a disturbance term, and a second-order extended state observer equation is constructed;

[0038] Step B3: Determine the bandwidth size;

[0039] Step B4: Combine the Brunovsky standard form, feedback control law and extended state observer equation and substitute the observation value of the extended state observer to obtain the final system control law. Here, the final system control law is obtained after the introduction of the extended state observer and simplification.

[0040] Here is a special note: there are three control laws in this paper: 1. Brunofsky standard feedback control law; 2. System control law without the introduction of extended state observer (corresponding to step A5); 3. The final system control law obtained after simplification by introducing the extended state observer (corresponding to step B4).

[0041] Further, it is characterized in that:

[0042] The rotor motion equation of the grid equivalent synchronous generator is modified as follows:

[0043] Where: Δω is the frequency change of the system; t is time; dΔω / dt is the derivative of the frequency change of the system with respect to time; f(x) is the disturbance term; ΔP pem is the power change of the electrolytic cell; M is the equivalent inertia time constant of the system;

[0044] The extended state observer equation is:

[0045] Where: e is the observation error; m 1 is the observed value of frequency change; m 2 is the observed value of the disturbance term; is the time derivative of the observed value of the system frequency variation; is the time derivative of the observed value of the disturbance term; Δω is the frequency change of the system; u is the input vector; M is the equivalent inertia time constant of the system;

[0046] β 1 , β 2 is the observation error coefficient:

[0047]

[0048] Where: β 1 , β 2 is the observation error coefficient; ω 0 For bandwidth, an appropriate value is usually selected to ensure convergence speed;

[0049] The final system control law is: ΔP pem =M(m 1 +m 2 )

[0050] Where: f(x) is the disturbance term; ΔP pem is the change in electrolytic cell power; M is the system equivalent inertia time constant; Δω is the change in system frequency.

[0051] The method for dynamic frequency response control of a proton exchange membrane electrolyzer taking into account hydrogen constraints in oxygen of the present invention has the following advantages: the method of the present invention can solve the potential mining and safety control problems of the electrolyzer load response to the power system frequency, improve the stability of the power system, and is a nonlinear control method with practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a simulation system diagram of four machines and two zones with a wind power penetration rate of 25%.

[0053] Figure 2 This is the system frequency diagram without considering the sudden increase in hydrogen confinement load in oxygen.

[0054] Figure 3 This is the power diagram of the electrolyzer without considering the sudden increase in hydrogen confinement load in oxygen.

[0055] Figure 4 This is a diagram of the system frequency change rate without considering the sudden increase in hydrogen confinement load in oxygen.

[0056] Figure 5 This is a graph of the hydrogen content in oxygen.

[0057] Figure 6 It is the system frequency diagram considering the sudden increase of hydrogen confinement load in oxygen.

[0058] Figure 7 It is the power diagram of the electrolyzer under the sudden increase of hydrogen confinement load in oxygen.

[0059] Figure 8 It is a diagram of the system frequency change rate under the sudden increase of hydrogen confinement load in oxygen. DETAILED DESCRIPTION

[0060] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of a frequency dynamic response control method for a proton exchange membrane electrolyzer considering hydrogen constraint in oxygen in conjunction with the accompanying drawings.

[0061] The frequency response control method proposed in the present invention is a nonlinear control method, which is faster than the traditional control method. When the adjustable capacity of the electrolyzer is sufficient, an appropriate control target can be set to restore the system frequency. The nonlinear characteristics of the system can be retained, and it is suitable for power systems with high nonlinearity, avoiding the truncation error caused by using a linear control method after approximate linearization.

[0062] The present invention provides a method for controlling the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen constraints in oxygen, and its specific implementation includes the following steps:

[0063] (1) Hydrogen confinement in oxygen in proton exchange membrane electrolyzer

[0064] When a proton exchange membrane electrolyser responds to grid frequency, its operation is constrained by the hydrogen content of oxygen in the anode channel, i.e.:

[0065]

[0066] Where: HTO is the hydrogen content of oxygen in the anode channel of the proton exchange membrane electrolyzer; C H,a is the hydrogen concentration in the anode channel; C O,a is the oxygen concentration in the anode channel; α is the safety threshold of hydrogen in oxygen in the electrolyzer.

[0067] It is necessary to explain here that in actual engineering production, the HTO in the subsequent auxiliary equipment of the electrolyzer is monitored, and its value is slightly lower than the HTO in the anode channel of the electrolyzer, that is, when the HTO in the anode channel reaches α, the HTO in the auxiliary equipment has not reached the limit, leaving a certain safety margin, which meets the safety production requirements. If it is monitored that HTO reaches α, emergency measures need to be taken to quickly reduce HTO. The measures taken are to increase the flow rate of the cathode mixture. The increase in the cathode mixture flow rate reduces the amount of cathode hydrogen, the cathode and anode partial pressures are reduced, and the amount of hydrogen permeating from the cathode to the anode is greatly reduced, which accelerates the decline of HTO. At this time, HTO will not immediately return to below α. On the one hand, there is a margin for auxiliary equipment, etc.; on the other hand, although HTO exceeds the α limit after the flow rate is increased, it is within the short time range of a frequency modulation with a second-level response, so it is acceptable from the perspective of safe operation. Formula (1) is a safety constraint, and it is stipulated that during the period when HTO exceeds α, the electrolyzer temporarily exits the frequency modulation, and its power remains unchanged until HTO ≤ α, and then it continues to return to the frequency modulation process.

[0068] (2) Nonlinear frequency dynamic response control strategy based on holographic target feedback

[0069] First, determine the system control target. To meet the requirements of frequency response performance, the control target is that the change in grid frequency tends to zero. Select the grid frequency change as the output variable and state variable of the nonlinear system, and the change in electrolyzer power as the control input. A single-input single-output nonlinear control system of the following form can be established:

[0070]

[0071] Where: is the time derivative vector of the state variable; y is the output variable vector; u is the control input; x is the column vector of the state variable set; r(x) is the state vector of the system; s(x) is the coefficient vector of the control input, and h(x) is the output vector.

[0072] Incorporating the control input into the rotor motion equation of the grid equivalent synchronous generator, the grid frequency dynamic equation including the control quantity is obtained:

[0073]

[0074] Where: Δω is the frequency change of the system; t is the time; dΔω / dt is the derivative of the frequency change of the system with respect to time; ΔP M is the change in active output of the synchronous generator; ΔP L is the load power change; ΔP pem is the power change of the electrolyzer; D is the load frequency coefficient, which is equal to the load change percentage divided by the frequency change percentage; M is the system equivalent inertia time constant; Δω is the system frequency change. It should be noted that the rotor speed per unit value of the synchronous generator is equal to the grid frequency per unit value; load specifically refers to other loads that do not include the electrolyzer.

[0075] By comparing formula (2) and formula (3), we can know that:

[0076]

[0077] Where: Δω is the frequency change of the system; ΔP M is the change in active output of the synchronous generator; ΔP L is the load power change; D is the load frequency coefficient, which is equal to the load change percentage divided by the frequency change percentage; M is the system equivalent inertia time constant; r(x) is the state quantity of the system; s(x) is the coefficient of the control input quantity, and h(x) is the output quantity.

[0078] Formula (3) can be transformed into:

[0079]

[0080] Where: r(x), s(x), h(x) are nonlinear function quantities; h(x) is the output quantity; u is the input quantity vector; L is the Lie derivative operator; L r h(x) means the Lie derivative of the output quantity with respect to the system state quantity; It means the Lie derivative of the output quantity with respect to the system state quantity, but does not take the derivative.

[0081]

[0082] Where: M is the equivalent inertia time constant of the system. It shows that Δω is the first-order relative output of the nonlinear system, and the system is controllable.

[0083] According to the output variables and control objectives, the Brunovsky standard dynamic system can be constructed:

[0084]

[0085] Where: v is the pre-controlled variable; is the time derivative of the system frequency change.

[0086] The equilibrium point of the system state variables is zero. If the system can be asymptotically stable, the frequency change of the power grid can approach zero.

[0087] Since the Brunovsky standard dynamic system is a linear steady-state system, the control law of the system can be obtained by linear quadratic optimal control, and its optimization performance index function is

[0088]

[0089] Where: J is the optimization performance index function; a is the pre-controlled variable weight; v is the pre-controlled variable; I is the first component of the Lie derivative matrix of the output quantity, which specifically refers to the system frequency change in this patent; Q is the state weight matrix; is the integral of time from zero to infinity. In order to achieve system stability and optimal performance, its feedback control law can be expressed as:

[0090] v=-Δω (9)

[0091] Where: Δω is the frequency change of the system; v is the pre-controlled variable.

[0092] Combining equations (3), (7) and (9), the control law of the system is derived as follows:

[0093]

[0094] Where: ΔP M is the change in active output of the synchronous generator; ΔP L is the load power change; ΔP pem is the power change of the electrolytic cell; D is the load frequency coefficient, which is equal to the load change percentage divided by the frequency change percentage; M is the system equivalent inertia time constant; Δω is the system frequency change.

[0095] In engineering applications, ΔP M , ΔP L It is difficult to obtain and D. In order to avoid the control law being completely dependent on the system parameters and to simplify the nonlinear control calculation, an extended state observer is used to observe the system state variables and expand the disturbance term to assist the control.

[0096] First, the terms in equation (3) that do not contain the control input u are considered as disturbance terms, and the equation is transformed into

[0097]

[0098] Where: Δω is the frequency change of the system; t is time; dΔω / dt is the derivative of the frequency change of the system with respect to time; f(x) is the disturbance term; ΔP pemis the power change of the electrolytic cell; M is the equivalent inertia time constant of the system.

[0099] The extended state observer is used to observe the frequency change of the power grid. The extended term is the disturbance term. The second-order extended state observer is constructed as:

[0100]

[0101] Where: e is the observation error; m 1 is the observed value of frequency change; m 2 is the observed value of the disturbance term; is the time derivative of the observed value of the system frequency variation; is the time derivative of the observed value of the disturbance term; Δω is the frequency change of the system; u is the input vector; M is the equivalent inertia time constant of the system; β 1 , β 2 is the observation error coefficient:

[0102]

[0103] Where: β 1 , β 2 is the observation error coefficient; ω 0 For the bandwidth, an appropriate value is usually selected to ensure the convergence speed.

[0104] Through equations (7), (9) and (12), we can deduce:

[0105] ΔP pem =Mf(x)+MΔω (14)

[0106] Where: f(x) is the disturbance term; ΔP pem is the change in electrolytic cell power; M is the system equivalent inertia time constant; Δω is the change in system frequency.

[0107] Then bring in the observation value of the extended state observer to get the final control law:

[0108] ΔP pem =M(m 1 +m 2 ) (15)

[0109] Where: P pem is the power change of the electrolytic cell; M is the equivalent inertia time constant of the system; m 1 is the observed value of the system frequency change; m 2 is the observed value of the disturbance term.

[0110] It should be noted that the parameter system equivalent inertia time constant in the formula does not need to obtain a specific value, but only needs to be roughly estimated, because the extended state observer tracks the error and automatically compensates for the error.

[0111] Specific examples:

[0112] Based on the Matlab / simulink platform, a four-machine two-area system with a wind power penetration rate of 25% was used for simulation verification. The system wiring is as follows Figure 1 As shown. There are two thermal power units in area 1, one thermal power unit in area 2, and an equivalent 900MVA wind farm. The two areas are connected by a 220km double-circuit π-type tie line. The electrolyzer load is connected at node 4, which consists of a transformer, AC / DC grid-connected inverter, DC / DC converter and proton exchange membrane electrolyzer. Before the load surge, the operating power of thermal power units 1-3 was 700MW, 706MW, and 700MW, the operating power of wind turbines was 719MW, and the operating power of proton exchange membrane electrolyzers was 400MW. When the hydrogen constraint in oxygen is not considered, the load of system node 4 surged by 300MW in the 10th second.

[0113] Depend on Figure 2 It can be seen that the lowest point of the system frequency is 0.079Hz higher than that of the no-control strategy, and the use of holographic target feedback nonlinear frequency dynamic response control can restore the frequency to 50Hz.

[0114] Depend on Figure 3 It can be seen that the holographic target feedback nonlinear frequency dynamic response control strategy can fully mobilize the adjustable capacity of the electrolyzer, adjust the output to the maximum, the minimum output of the electrolyzer is about 120MW, and the adjustment amount reaches 280MW.

[0115] Depend on Figure 4 It can be seen that in the frequency response stage after the inertial response, the maximum frequency change rate of the holographic target feedback nonlinear frequency dynamic response control can reach 0.091Hz / s, indicating that the frequency recovery speed of the holographic target feedback nonlinear frequency dynamic response control is fast. At 40s, the load of node 4 suddenly increased by 300MW, and at this time, the holographic target feedback nonlinear frequency dynamic response control began to act.

[0116] Depend on Figure 5 It can be seen that when HTO ≥ 2%, the flow rates of the cathode and anode mixtures increase rapidly, the flow rate of the anode mixture increases from 0.43 m / s to 0.48 m / s, and the flow rate of the cathode mixture increases from 0.01 m / s to 0.02 m / s, and HTO decreases rapidly.

[0117] Depend on Figure 6 , 7 It can be seen that the load suddenly increased at 40s, the electrolytic cell power decreased rapidly and the HTO increased rapidly. This is because the electrolytic cell current i decreased rapidly, resulting in the H generated by the anode and cathode.2 and O 2 The rate decreases at the same time, so the oxygen content in the anode channel decreases, and the oxygen concentration C O,a decreases; although the amount of hydrogen permeating from the cathode also decreases, the hydrogen concentration C in the anode channel H,a The permeation is also decreasing, but it is a slow accumulation process, far less than the anode itself producing O 2 The flow rate of the cathode and anode mixture is increased at the same time and the electrolyzer exits the frequency modulation. The power remains unchanged at this moment and the system frequency no longer recovers quickly. After the flow rate increases, HTO begins to decrease. It is worth noting that the decrease rate of HTO is faster than Figure 5 The display is fast because the electrolyzer output is constrained and operates at constant power. Figure 5 The electrolyzer has been participating in the frequency modulation with a continuously decreasing power, and there is no trend of increasing HTO. At 43s, that is, when HTO reaches 2%, the electrolyzer quickly reduces its own power and participates in the frequency modulation again, repeating the above analysis process of HTO increase. The electrolyzer power decreases, and the upward trend of HTO given by the reduction of electrolyzer power is much greater than the downward trend of HTO given by the increase of the flow rate of the anode and cathode mixture. HTO returns to more than 2% again, and the electrolyzer can no longer participate in the frequency modulation. At the same time, the power drop caused by the electrolyzer power tracking the reference value is equivalent to a sudden load reduction for the system, so the system frequency rises rapidly.

[0118] like Figure 8 As shown in the figure, the frequency change rate also increases rapidly at this time. At 47s, when HTO drops to 2% again, because the electrolyzer power does not drop much, the rising trend of HTO given by the reduction of electrolyzer power is smaller than the falling trend of HTO given by the increase of anode and cathode mixture flow rate, so the frequency modulation can be re-entered, and the frequency is basically restored in the end.

[0119] In summary, the present invention proposes a nonlinear frequency response control method for proton exchange membranes that takes into account the constraint of hydrogen in oxygen. By calculating the concentration of hydrogen in oxygen in the anode channel to form the constraint of hydrogen in oxygen, this method fills the gap that the internal operating characteristics of the current proton exchange membrane electrolyzer response to the grid frequency have never been considered. The proposed nonlinear control method can effectively increase the lowest point of the system frequency and reduce the frequency change rate in the inertial response stage; in the frequency response stage, nonlinear control can fully release the adjustable capacity of the electrolyzer and accelerate the frequency recovery speed; when the adjustable capacity of the electrolyzer is large enough, the frequency can be restored. When considering the constraint of hydrogen in oxygen, the capacity of the electrolyzer to participate in the frequency response is limited, which reduces its frequency modulation effect.

[0120] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A control method for the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen confinement in oxygen, characterized in that: The following steps are involved: When the proton exchange membrane electrolyzer responds to the grid frequency, the operation is subject to the constraint of hydrogen in oxygen in the anode channel, and the constraint of hydrogen in oxygen is composed of the concentration of hydrogen in oxygen in the anode channel of the proton exchange membrane electrolyzer and a safety threshold; The hydrogen constraints in the oxygen are: Where: HTO is the hydrogen content of oxygen in the anode channel of the proton exchange membrane electrolyzer; C H,a is the hydrogen concentration in the anode channel; C O,a is the oxygen concentration in the anode channel; α is the safety threshold of hydrogen in oxygen in the electrolyzer; The safety threshold, if it is monitored that the hydrogen content in oxygen reaches the safety threshold, emergency measures need to be taken to speed up the flow rate of the anode and cathode mixture to accelerate the decrease of the hydrogen content in oxygen and stipulate that during the period when the hydrogen content in oxygen exceeds the safety threshold, the electrolyzer temporarily exits the frequency modulation, and its power remains unchanged until the hydrogen content in oxygen is less than or equal to the safety threshold, and then continues to return to the frequency modulation process; When the proton exchange membrane electrolyzer responds to the grid frequency, a holographic target feedback nonlinear control method is used, and an extended state observer is used to observe the system frequency change, and the expanded observation disturbance term is used to obtain the final system control law; The holographic target feedback nonlinear control method is as follows: Step A1: Determine the control target of the system, wherein the control target is that the grid frequency change tends to zero; select the grid frequency change, the grid frequency change and the electrolyzer power change as the state variable, the output variable and the control input respectively; Step A2: construct a single-input single-output nonlinear control system equation, and incorporate the control input into the rotor motion equation of the grid equivalent synchronous generator to determine whether the control input is a first-order relative-order output of the nonlinear system; Step A3: Construct a Brunovsky standard dynamic system based on the output variables and control objectives, and construct an optimization performance index function; Step A4: To achieve system stability and optimal performance, determine the weights of the pre-controlled variables to obtain the feedback control law; The feedback control law is: v = -Δω In the formula: Δω is the frequency change of the system; v is the pre-controlled variable; Step A5: The system control law is derived from the rotor motion equation of the equivalent synchronous generator of the grid, the Brunovsky standard form and the feedback control law; The method of using the extended state observer is as follows: Step B1: Incorporate the control input into the rotor motion equation of the grid equivalent synchronous generator, and regard the term that does not contain the control input as a disturbance term; Step B2: Use an extended state observer to observe the frequency change of the power grid, the extended term is a disturbance term, and a second-order extended state observer equation is constructed; Step B3: Determine the bandwidth size; Step B4: Combine the Brunovsky standard form, feedback control law and extended state observer equations and substitute the observation values ​​of the extended state observer to obtain the final system control law.

2. The method for controlling the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen constraint in oxygen according to claim 1, characterized in that: In step A2, the single-input single-output nonlinear system equation is: Where: is the time derivative vector of the state variable; y is the output variable vector; u is the control input; x is the column vector of the state variable set; r(x) is the state vector of the system; s(x) is the coefficient vector of the control input, and h(x) is the output vector.

3. The control method for the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen constraint in oxygen according to claim 1, characterized in that: In step A3, the Brunovsky standard form is: Where: v is the pre-controlled variable; is the derivative of the system frequency change with respect to time; The optimization performance index function is: Where: J is the optimization performance index function; a is the pre-controlled variable weight; v is the pre-controlled variable; I is the first component of the Lie derivative matrix of the output, specifically the frequency change of the system; Q is the state weight matrix; is the integral of time from zero to infinity.

4. The method for controlling the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen constraint in oxygen according to claim 1, characterized in that: In step A5, the rotor motion equation of the grid equivalent synchronous generator is: Where: Δω is the frequency change of the system; t is the time; dΔω / dt is the derivative of the frequency change of the system with respect to time; ΔP M is the change in active output of the synchronous generator; ΔP L is the load power change; ΔP pem is the power change of the electrolyzer; D is the load frequency coefficient, which is equal to the load change percentage divided by the frequency change percentage; M is the system equivalent inertia time constant; the rotor speed per unit value of the synchronous generator is equal to the grid frequency per unit value; the load specifically refers to other loads excluding the electrolyzer; System control law: ΔP pem =ΔP M -ΔP L +(MD)Δω Where: ΔPM is the change in active output of the synchronous generator; ΔPL is the change in load power; ΔPpem is the change in electrolytic cell power; D is the load frequency coefficient, which is equal to the load change percentage divided by the frequency change percentage; M is the system equivalent inertia time constant; Δω is the system frequency change.

5. The method for controlling the frequency dynamic response of a proton exchange membrane electrolyzer considering hydrogen constraint in oxygen according to claim 1, characterized in that: The rotor motion equation of the grid equivalent synchronous generator is modified as follows: Where: Δω is the frequency change of the system; t is time; dΔω / dt is the derivative of the frequency change of the system with respect to time; f(x) is the disturbance term; ΔP pem is the power change of the electrolytic cell; M is the equivalent inertia time constant of the system; The extended state observer equation is: Where: e is the observation error; m1 is the observed value of the frequency change; m2 is the observed value of the disturbance term; is the time derivative of the observed value of the system frequency variation; is the time derivative of the observed value of the disturbance term; Δω is the frequency change of the system; u is the input vector; M is the equivalent inertia time constant of the system; β1, β2 are observation error coefficients: Where: β1, β2 are observation error coefficients; ω0 is the bandwidth, and an appropriate value is usually selected to ensure the convergence speed; The final system control law is: ΔP pem =M(m1+m2) Where: f(x) is the disturbance term; ΔP pem is the change in electrolytic cell power; M is the system equivalent inertia time constant; Δω is the change in system frequency.

Images