DC bus voltage fluctuation suppression method of renewable energy hydrogen production system

Abstract

The invention discloses a DC bus voltage fluctuation suppression method of a renewable energy hydrogen production system, and relates to the technical field of renewable energy electrolysis hydrogen production. The method comprises the following steps: collecting system operation parameters in real time; bus voltage disturbance power is estimated based on a disturbance observer, and frequency decomposition is carried out; fusing the ARIMA model and the LSTM model to predict the load power, and calculating a prediction deviation; dynamically setting upper and lower limits of a bus voltage control interval according to the predicted deviation and the operation state; a multi-objective optimization function is constructed, bus voltage disturbance power, prediction deviation and a control limit value are comprehensively considered, and power regulation instructions of battery energy storage, a fuel cell and an electrolytic cell are solved; and executing power regulation control to realize bus voltage stabilization. The method can improve the grading response capability of voltage disturbance, the prediction accuracy of load power change, the adaptability of the control boundary and the overall operation stability of the system, and is suitable for the electric energy quality control of the hydrogen production system in a multi-source fluctuation scene.

Description

Technical Field

[0001] The present invention relates to the technical field of renewable energy electrolysis hydrogen production, and particularly to a method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system. Background Art

[0002] With the acceleration of the energy structure transformation, the penetration rate of renewable energy (such as photovoltaic, wind power, etc.) in the power system is increasing day by day. However, due to its inherent volatility and unpredictability, the power quality problems (such as voltage fluctuation, frequency disturbance, etc.) are becoming increasingly prominent. Especially in the hydrogen production system driven by renewable energy, its impact on the DC bus voltage stability of the system is particularly significant.

[0003] In such a system, the DC bus is connected to multiple core power components, including photovoltaic arrays, wind turbines, fuel cells, electrolyzers, energy storage batteries, and control inverters. Since the output of renewable energy is greatly affected by environmental factors (such as sunlight intensity, temperature, wind speed, etc.), the power fluctuation on the load side of the system is frequent, and the bus voltage is prone to violent fluctuations, which in turn affects the electrolysis hydrogen production efficiency, the safety of the stack, and the service life of the energy storage equipment.

[0004] In the prior art, voltage limit control with a fixed threshold or a charge-discharge scheduling strategy based on simple rules is usually adopted, lacking the ability to perceive the operating state of the system; the traditional disturbance response method does not distinguish the frequency characteristics of disturbances, resulting in the same response mechanism for high-frequency and low-frequency disturbances, and the control resource allocation is unreasonable; in the face of sensor failures, communication anomalies, etc., the control system often lacks the ability of autonomous fault tolerance and voltage stability maintenance.

[0005] Therefore, there is an urgent need for a comprehensive method that can perform multi-objective adaptive optimization control based on the system operating state, disturbance prediction, and load dynamic changes to improve the operating stability and energy scheduling efficiency of the renewable energy hydrogen production system under complex working conditions. Summary of the Invention

[0006] Aiming at the above problems, the present invention proposes a method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system, which reasonably allocates energy storage resources according to factors such as disturbance frequency and load prediction to achieve the rapid stability of the DC bus voltage.

[0007] The present invention realizes the above object through the following technical solutions:

[0008] A method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system, the method comprising:

[0009] Real-time collecting key system operating parameters, including the DC bus voltage V dc , current, renewable energy power, current load power P load(t), State of Charge (SOC) of the battery, hydrogen storage pressure P H2 , ambient temperature S I and light intensity T;

[0010] Estimate the bus voltage disturbance power d(t) through a disturbance observer and decompose it into a high-frequency disturbance component d high (t) and a low-frequency disturbance component d low (t);

[0011] Based on the load power data within the historical time, predict the next moment's load power P load (t + 1) through a hybrid load prediction algorithm and calculate the prediction deviation △P diff ;

[0012] According to the key operating parameters of the system and the prediction deviation △P diff , dynamically set the upper and lower limits V high (t), V low (t);

[0013] Construct a multi-objective optimization function with the bus voltage disturbance power d(t), prediction deviation △P diff , and the upper and lower limits V high (t), V low (t) as inputs to solve the power regulation command values of the battery energy storage, fuel cell, and electrolyzer

[0014] According to the power regulation command values Generate a control command and execute power output regulation.

[0015] As a preferred embodiment of the present invention, the disturbance observer uses the following dynamic gain model to estimate the bus voltage disturbance power d(t):

[0016]

[0017] In the formula, C is the equivalent capacitance of the bus; is the current voltage change rate, t represents time; K d (t) is the observer gain coefficient; sat() is the saturation function; V ref is the target reference voltage of the DC bus;

[0018] Among them, the expression of the observer gain coefficient K d (t) is:

[0019]

[0020] Wherein, K0 is the basic gain constant; K1 is the voltage change rate gain coefficient; K2 is the voltage deviation gain coefficient; V dc (t) is the current bus voltage.

[0021] As a preferred embodiment of the present invention, the bus voltage disturbance power d(t) is frequency decomposed by wavelet transform or Fourier transform, and the disturbance with a frequency component higher than the cut-off frequency fc is defined as the high-frequency disturbance component d high (t), which is responded by the battery energy storage system; the disturbance with a frequency component lower than the cut-off frequency fc is defined as the low-frequency disturbance component d low (t), which is jointly responded by the electrolyzer and the fuel cell; the cut-off frequency fc is adaptively set according to the power fluctuation period.

[0022] As a preferred embodiment of the present invention, the hybrid load prediction algorithm is specifically an ARIMA-LSTM dynamic fusion model, and the implementation method is as follows:

[0023] Input the historical load power data {P load (t-N+1),..., P load (t)} at N collected moments, and perform data preprocessing operations, including normalization, filtering out spike outliers and smoothing, and output the cleaned sequence data for modeling;

[0024] Construct the disturbance amplitude factor D s and the dominant period length C p , and the formula is:

[0025]

[0026] Wherein, i is the sliding index, representing the i-th time difference value; P load (t-i+1), P load (t-i) are the load powers at the (t-i+1)-th and (t-i)-th moments respectively; f peak is the maximum non-dc frequency component obtained by applying the fast Fourier transform to the historical load power data;

[0027] Set the period recognition range set Z = {6h, 12h, 24h}, and the tolerance is ±15%;

[0028] If the disturbance amplitude factor D s is less than the preset load disturbance amplitude threshold and the dominant period length C p satisfies: such that ∣C p -z∣≤0.15·z, then preferentially select the ARIMA model as the main model and the LSTM model as the secondary prediction path, otherwise select the LSTM model as the main model and the ARIMA model as the secondary prediction path;

[0029] Calculate the historical residual standard deviation and confidence weight of the ARIMA model and the LSTM model respectively, and finally generate the predicted value P of the load power at the next moment load (t + 1), and the formula is:

[0030]

[0031] In the formula, σ ARIMA , ω ARIMA , are respectively the historical residual standard deviation, confidence weight, and prediction result of the ARIMA model; σ LSTM , ω LSTM , are respectively the historical residual standard deviation, confidence weight, and prediction result of the LSTM model;

[0032] The prediction deviation △P diff is calculated by the formula: △P diff = P load (t + 1)-P load (t).

[0033] As a preferred solution of the present invention, the upper and lower limits V high (t), V low (t) of the dynamic setting bus voltage control interval are determined by the following method:

[0034] Initialization setting: Set the target reference voltage of the DC bus to V ref , and set the basic voltage tolerance to △V base , so as to determine the initial control interval as: V low (0)= V ref -△V base , V high (0)= V ref +△V base , where V high (0), V low (0) are respectively the initial upper limit value and lower limit value of the bus voltage;

[0035] Real-time parameter acquisition: In each control cycle, the state of charge SOC of the battery, the hydrogen storage pressure P H2 and the prediction deviation △P diff are acquired in real time;

[0036] Interval adjustment rule setting: Based on the judgment results of the following three types of working conditions, adjust the upper and lower limits of the bus voltage control interval:

[0037] 1) If the state of charge SOC of the battery is less than the lower threshold SOC low of the state of charge of the battery, then set Vlow V(t)=V ref +△V SOC , where △V SOC is the upward shift of the lower voltage limit in the SOC protection state;

[0038] 2) If the hydrogen storage pressure P H2 is greater than the maximum allowable pressure P of the hydrogen tank H2,max , then set V high (t)=V ref -△V H2 , where △V H2 is the downward adjustment of the upper voltage limit when the hydrogen pressure is too high;

[0039] 3) If the predicted deviation △P diff is greater than the load power disturbance threshold P thresh and lasts for more than three sampling periods, then expand the bus voltage control range to:

[0040] V low (t)=V ref -(△V base +△V ext );

[0041] V high (t)=V ref +(△V base +△V ext );

[0042] where, △V ext is the extended voltage deviation range;

[0043] Hysteresis interval judgment: If the interval change in the current period satisfies:

[0044] |V low (t)-V low (t - 1)| < ε v ;

[0045] |V high (t)-V high (t - 1)| < ε v ;

[0046] then keep the upper and lower limits of the bus voltage in the previous period unchanged;

[0047] In the formula, V high (t - 1), V low (t - 1) are respectively the upper and lower limits of the bus voltage control range in the previous period, and ε v is the minimum adjustment threshold of the voltage lower limit.

[0048] As a preferred embodiment of the present invention, the expression of the multi-objective optimization function is:

[0049]

[0050] Wherein, J(t) is the value of the optimization objective function, representing the total control cost of the control system at the current moment; is the target charge-discharge power of the battery energy storage at the current moment, is the target value of the fuel cell output power at the current moment, is the power that the electrolyzer should absorb at the current moment; V dc (t) is the DC bus voltage at the current moment, V ref is the target reference voltage of the DC bus; α, β, γ, and δ are the battery power response weight coefficient, the hydrogen energy system response weight coefficient, the voltage stability weight coefficient, and the battery power change rate penalty weight respectively.

[0051] As a preferred solution of the present invention, the generating control instructions and performing power output regulation specifically include:

[0052] Smoothing sliding mode control of the battery energy storage system:

[0053] Using the hyperbolic tangent function to replace the traditional sign function to generate charge-discharge instructions and suppress the chattering of control instructions;

[0054] Dynamically limiting the charge-discharge power according to the state of charge SOC of the battery. When the SOC is higher than 90%, only discharging is allowed. When it is lower than 20%, only charging is allowed;

[0055] Hydrogen flow closed-loop control of the fuel cell:

[0056] By real-time monitoring of the fuel cell output current and hydrogen flow rate, calculating the theoretical hydrogen demand, and superimposing the low-frequency disturbance component d low (t) to generate a feed-forward compensation flow instruction;

[0057] Using a PID controller to close-loop regulate the hydrogen flow rate to ensure that the hydrogen utilization rate is not less than 95% and avoid fuel starvation or waste;

[0058] Dynamically adjusting the output power instruction according to the stack voltage and conversion efficiency, and matching the bus voltage demand through a boost DC-DC converter;

[0059] MPPT-feed-forward composite control of the electrolyzer:

[0060] Based on the maximum power point tracking MPPT algorithm, periodically disturbing the input voltage of the electrolyzer, and combining with the low-frequency disturbance component d low (t) to dynamically correct the hydrogen production power;

[0061] Dynamically limiting the hydrogen production power according to the pressure of the hydrogen storage tank. When the pressure exceeds the limit, forced shutdown is performed. When the pressure is insufficient, it is increased to the rated power.

[0062] As a preferred embodiment of the present invention, the method further includes an embedded PI regulation link, which in real time takes the bus voltage deviation △V = V ref - V dc as the input to generate a regulated power △P PI , and superimposes it on the control command;

[0063] When any of the following abnormal states is detected:

[0064] Sensor data interruption or out-of-limit;

[0065] The controller communication interruption exceeds the set duration;

[0066] The energy storage unit has no power response for more than two sampling periods;

[0067] Automatically start the fault tolerance logic, and control the battery or fuel cell to maintain the bus voltage within the tolerance range at a preset power until the main control system resumes.

[0068] As a preferred embodiment of the present invention, the regulated power △P PI satisfies:

[0069]

[0070] wherein, △P PI (t) is the current regulated power of the PI feedback; K p is the proportional gain coefficient; V ref is the target reference voltage of the DC bus; K j is the integral gain coefficient; τ is the integral variable.

[0071] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system as described above.

[0072] The beneficial effects of the present invention are as follows: By comprehensively introducing disturbance observation, frequency decomposition response, load prediction, dynamic control interval setting and multi-objective optimization control strategies, it is possible to achieve precise identification and hierarchical response to bus voltage disturbances, improve the foresight and accuracy of load prediction, dynamically adapt to the voltage control limits under different working conditions, and through the collaborative optimization of the power regulation commands of the battery energy storage, fuel cell and electrolyzer by the multi-objective function, effectively improve the operation stability, control intelligence and response robustness of the system. At the same time, the method has an abnormal detection and fault tolerance mechanism, which can ensure the stability of the bus voltage in case of key equipment failure or signal interruption, and has good engineering adaptability and application value as a whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0074] Wherein:

[0075] Figure 1 is the method flow chart of the present invention. Detailed implementation manners

[0076] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions of the embodiments of the present invention with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention fall within the scope of protection of the present invention.

[0077] As Figure 1 shown, it is an embodiment of the present invention. This embodiment provides a method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system, including the following steps:

[0078] S1: Real-time collect key system operation parameters, including the DC bus voltage V dc , current, renewable energy power, current load power P load (t), state of charge of the battery SOC, hydrogen storage pressure P H2 , ambient temperature S I , light intensity T, etc.;

[0079] S2: Estimate the bus voltage disturbance power d(t) through a disturbance observer, and decompose it into a high-frequency disturbance component d high (t) and a low-frequency disturbance component d low (t);

[0080] Specifically, the disturbance observer uses the following dynamic gain model to estimate the bus voltage disturbance power d(t):

[0081]

[0082] In the formula, C is the equivalent capacitance of the bus; is the current voltage change rate, t represents time; K d (t) is the observer gain coefficient; sat() is the saturation function; V ref is the target reference voltage of the DC bus;

[0083] Among them, the observer gain coefficient K d (t) has the following expression:

[0084]

[0085] In the formula, K0 is the basic gain constant; K1 is the voltage change rate gain coefficient; K2 is the voltage deviation gain coefficient; V dc (t) is the current bus voltage.

[0086] Perform frequency decomposition on the bus voltage disturbance power d(t) through wavelet transform or Fourier transform, and define the disturbance with a frequency component higher than the cut-off frequency fc as the high-frequency disturbance component d high (t), which is responded by the battery energy storage system; define the disturbance with a frequency component lower than the cut-off frequency fc as the low-frequency disturbance component d low (t), which is jointly responded by the electrolyzer and the fuel cell; the cut-off frequency fc is adaptively set according to the power fluctuation period.

[0087] S3: Based on the load power data within the historical time, predict the next moment's load power P load (t + 1) through the hybrid load prediction algorithm, and calculate the prediction deviation △P diff .

[0088] The hybrid load prediction algorithm of this embodiment introduces a time-series adaptive discrimination mechanism to analyze the disturbance amplitude and periodicity of the input data, and dynamically selects ARIMA or LSTM as the main predictor. ARIMA is used to reduce latency in the prediction stable section, LSTM is called in the mutation section to enhance the non-linear modeling ability, and a confidence-weighted mechanism is introduced to improve the prediction accuracy at the mutation edge moment. The final prediction result is used for feed-forward regulation in the bus voltage regulation to enhance the system's leading response ability.

[0089] In one of the embodiments, the hybrid load prediction algorithm is specifically an ARIMA-LSTM dynamic fusion model, and the implementation method is as follows:

[0090] S31: Input the historical load power data {P load (t - N + 1),..., P load (t)} at N collected moments, and perform data preprocessing operations, including normalization, filtering out spike outliers, and smoothing processing (such as double exponential weighted moving average), and output the cleaned sequence data for modeling;

[0091] S32: Construct the disturbance amplitude factor D s and the dominant period length C p , and the formula is:

[0092]

[0093] In the formula, i is the sliding index, representing the i-th time difference value; P load (t - i + 1) and P load (t - i) are the load powers at the (t - i + 1)-th moment and the (t - i)-th moment respectively; f peak is the maximum non - DC frequency component obtained by applying the fast Fourier transform to the historical load power data;

[0094] Set the period recognition range set Z = {6h, 12h, 24h}, and the tolerance is ±15%;

[0095] S33: Judgment rule (adjustable): If the disturbance amplitude factor D s is less than the preset load disturbance amplitude threshold and the dominant period length C p satisfies: such that ∣C p - z∣≤0.15·z (that is, the dominant period length C p falls within ±15% of any typical period in the period recognition range set), then preferentially select the ARIMA model as the main model and the LSTM model as the secondary prediction path; otherwise, select the LSTM model as the main model and the ARIMA model as the secondary prediction path;

[0096] If the sampling period is 1 minute, the preset load disturbance amplitude threshold can be set to 5kW (that is, a load change amplitude within one minute not exceeding 5kW is considered stable). The load disturbance amplitude threshold can be set between 3kW and 10kW in combination with the actual load fluctuation amplitude statistics. ARIMA is proficient in period modeling but not good at dealing with sudden fluctuations, and LSTM is good at non - linear modeling but may be over - fitted or insensitive to short - term fluctuations. Combining the two can achieve better prediction performance under multiple working conditions, especially in: renewable energy scenarios; hybrid load (constant power + random load) scenarios; systems where complex weather changes affect load fluctuations.

[0097] S34: Calculate the historical residual standard deviation and confidence weight of the ARIMA model and the LSTM model respectively, and finally generate the predicted value P load (t + 1) of the load power at the next moment. The formula is:

[0098]

[0099] In the formula, σ ARIMA and ω ARIMA and are the historical residual standard deviation, confidence weight, and prediction result of the ARIMA model respectively; σ LSTM and ω LSTM and are the historical residual standard deviation, confidence weight, and prediction result of the LSTM model respectively;

[0100] S35: Prediction deviation △P diff The calculation formula for it is: △P diff = P load (t + 1) - P load (t).

[0101] S4: Based on the system's key operating parameters and the prediction deviation △P diff , dynamically set the upper and lower limits V high (t), V low (t).

[0102] Furthermore, step S4 includes:

[0103] S41: Initial setting

[0104] Set the target reference voltage of the DC bus to V ref , and set the basic voltage tolerance to △V base , thereby determining the initial control interval as: V low (0) = V ref - △V base , V high (0) = V ref + △V base , where V high (0), V low (0) are the initial upper and lower limit values of the bus voltage respectively, that is, the upper and lower limit values of the bus voltage control set when the system starts or first enters the voltage control logic;

[0105] S42: Real-time parameter acquisition

[0106] In each control period, the state of charge SOC of the battery, the hydrogen storage pressure P H2 and the prediction deviation △P diff are acquired in real time;

[0107] S43: Interval adjustment rule setting

[0108] Based on the judgment results of the following three types of operating conditions, adjust the upper and lower limits of the bus voltage control interval:

[0109] 1) If the state of charge SOC of the battery is less than the lower threshold SOC of the state of charge low , then set V low (t) = V ref + △V SOC to inhibit further discharge of the battery; where △V SOC is the upward shift amount of the voltage lower limit in the SOC protection state. When the actual SOC is lower than SOC lowWhen the battery is discharging, disable or limit the battery discharge to prevent over-discharge damage to the battery; artificially raise the minimum bus control value to reduce the battery burden;

[0110] 2) If the hydrogen storage pressure P H2 is greater than the maximum allowable pressure P H2,max of the hydrogen tank, then set V high (t) = V ref - △V H2 to limit the electrolyzer from continuing to produce hydrogen; where △V H2 is the downward adjustment amount of the voltage upper limit when the hydrogen pressure is too high; P H2,max is used to prevent the electrolyzer from continuing to operate when the hydrogen storage system is saturated and ensure the safety of the system; when it is detected that the hydrogen pressure exceeds the upper limit, the voltage upper limit voltage drop amount △V H2 of the bus for preventing continuous hydrogen production is used to inhibit the electrolyzer from continuing to receive power and produce hydrogen, indirectly reducing the bus voltage;

[0111] 3) If the predicted deviation △P diff is greater than the load power disturbance threshold P thresh (i.e., the critical value for the system to judge that the load changes violently) and lasts for more than three sampling periods, then expand the bus voltage control range to:

[0112] V low (t) = V ref - (△V base + △V ext );

[0113] V high (t) = V ref + (△V base + △V ext );

[0114] Among them, △V ext is the extended voltage deviation range, which is used to enhance the dynamic response ability of the system to large power disturbances;

[0115] S44: Hysteresis interval judgment

[0116] If the interval change in the current period satisfies:

[0117] |V low (t) - V low (t - 1)| < ε v ;

[0118] |V high (t) - V high (t - 1)| < ε v ;

[0119] Then keep the upper and lower limits of the bus voltage in the previous period unchanged to avoid frequent switching due to small fluctuations;

[0120] Wherein, V high (t - 1) and V low (t - 1) are respectively the upper and lower limits of the bus voltage control interval in the previous cycle, and ε v is the minimum adjustment threshold of the voltage lower limit.

[0121] The updated voltage upper and lower limits V high (t) and V low (t) are used as the upper and lower limit boundaries for bus disturbance mode recognition judgment, the adjustment trigger conditions of each energy storage device, and the reference signal of the feedback controller.

[0122] S5: Construct a multi-objective optimization function with the bus voltage disturbance power d(t), the prediction deviation △P diff , and the upper and lower limits V high (t), V low (t) as inputs to solve the power adjustment command values of the battery energy storage, fuel cell, and electrolyzer

[0123] The expression of the multi-objective optimization function is:

[0124]

[0125] Wherein, J(t) is the value of the optimization objective function, representing the total control cost of the control system at the current moment; is the target charge and discharge power of the battery energy storage at the current moment, is the target value of the fuel cell output power at the current moment, is the power that the electrolyzer should absorb at the current moment; V dc (t) is the DC bus voltage at the current moment, V ref is the target reference voltage of the DC bus; α, β, γ, and δ are respectively the battery power response weight coefficient, the hydrogen energy system response weight coefficient, the voltage stability weight coefficient, and the battery power change rate penalty weight.

[0126] This multi-objective optimization function can be solved by methods such as the weighted linear method, the particle swarm optimization algorithm, and NSGA-II (Non-dominated Sorting Genetic Algorithm).

[0127] S6: Generate control commands according to the power adjustment command values and execute power output adjustment.

[0128] Step S6 specifically includes:

[0129] Smoothing sliding mode control of the battery energy storage system:

[0130] Use the hyperbolic tangent function to replace the traditional sign function to generate charge and discharge commands, suppressing the chattering of the control commands;

[0131] Dynamically limit the charge and discharge power according to the state of charge (SOC) of the battery. When the SOC is higher than 90%, only discharging is allowed; when it is lower than 20%, only charging is allowed.

[0132] Closed-loop control of the hydrogen flow of the fuel cell:

[0133] By real-time monitoring of the output current and hydrogen flow rate of the fuel cell, calculate the theoretical hydrogen demand, and superimpose a low-frequency disturbance component d low (t) to generate a feedforward compensation flow command;

[0134] Adopt a PID controller to close-loop regulate the hydrogen flow rate to ensure that the hydrogen utilization rate is not less than 95%, and avoid fuel starvation or waste;

[0135] Dynamically adjust the output power command according to the stack voltage and conversion efficiency, and match the bus voltage demand through a boost DC-DC converter;

[0136] MPPT-feedforward composite control of the electrolyzer:

[0137] Based on the maximum power point tracking (MPPT) algorithm, periodically perturb the input voltage of the electrolyzer, and combine the low-frequency disturbance component d low (t) to dynamically correct the hydrogen production power;

[0138] Limit the hydrogen production power in real time according to the pressure of the hydrogen storage tank. When the pressure exceeds the limit, force a shutdown; when the pressure is insufficient, increase it to the rated power.

[0139] In another specific embodiment, the method further includes an embedded PI regulation link, which takes the bus voltage deviation △V = V ref -V dc as the input, generates a regulation power △P PI , and superimposes it on the control command;

[0140] The regulation power △P PI satisfies:

[0141]

[0142] where △P PI (t) is the current regulation power of the PI feedback; K p is the proportional gain coefficient; V ref is the target reference voltage of the DC bus; K j is the integral gain coefficient; τ is the integral variable;

[0143] When any of the following abnormal states is detected:

[0144] Sensor data interruption or out-of-limit;

[0145] The communication interruption of the controller exceeds the set duration;

[0146] The energy storage unit has no power response for more than two sampling periods;

[0147] Automatically start the fault-tolerant logic, and control the battery or fuel cell to maintain the bus voltage within the tolerance range at a preset power until the main control system recovers.

[0148] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, it can be implemented in whole or in part in the form of a computer program product, which includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another.

[0149] In addition, in each embodiment of the present application, the functional units can be integrated into one processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. If the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a magnetic disk, an optical disk, etc. Therefore, the embodiments of the present invention also provide a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, it implements a method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system as described above.

[0150] In summary, the present invention realizes the accurate estimation and classification response of the DC bus voltage disturbance power. By introducing a disturbance observer and performing frequency decomposition on the disturbance power, high-frequency disturbances and low-frequency disturbances can be separated and processed, and the battery energy storage system and the fuel cell / electrolyzer system are respectively responsible for the response tasks, realizing the reasonable allocation of disturbance energy and effectively suppressing the high-frequency pulsation and low-frequency drift of the bus voltage.

[0151] The present invention improves the load dynamic prediction ability and enhances the control foresight. By adopting a load prediction algorithm that fuses the ARIMA and LSTM deep learning models, combined with the dominant period determination and disturbance amplitude factor extraction mechanisms, while ensuring the prediction accuracy, the system's perception ability of future load change trends is significantly improved, and the predictability and robustness of the control strategy are enhanced.

[0152] The present invention realizes the adaptive dynamic adjustment of the bus voltage control range. By comprehensively considering various operating parameters such as the state of charge (SOC) of the battery, the hydrogen storage pressure, and the prediction deviation of the load power, it can dynamically set the upper and lower limits of the bus voltage control under different operating conditions, avoiding the frequent mis-triggering or control failure caused by improper setting of the traditional static limit control, and improving the flexibility and reliability of the bus voltage control.

[0153] The present invention constructs a multi-objective optimization function to achieve coordinated power regulation control. The proposed multi-objective optimization function simultaneously considers multiple factors such as the battery power response cost, the hydrogen energy system response cost, the voltage deviation stability, and the power change rate penalty. Based on the comprehensive balance of the system operation cost and the control effect, it can dynamically solve the optimal power regulation command to achieve the coordinated and efficient operation among the electrolyzer, the fuel cell, and the battery energy storage unit.

[0154] The present invention enhances the abnormal handling and fault tolerance capabilities of the control system. By setting an embedded PI feedback adjustment link and various types of abnormal recognition mechanisms (such as sensor data interruption, control communication failure, energy storage response failure, etc.), when a failure occurs in the key equipment of the system, the fault tolerance logic can be automatically triggered, and the backup control strategy can be called to maintain the stability of the bus voltage, significantly improving the safety and continuous operation ability of the system.

[0155] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various changes or substitutions, and these should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system, characterized in that, The method includes: Real-time acquisition of key system operating parameters, including DC bus voltage V dc , current, renewable energy power, current load power P load (t), battery state of charge SOC, hydrogen storage pressure P H2 、Ambient temperature S I and light intensity T; Estimate the bus voltage disturbance power d(t) through a disturbance observer and decompose it into a high-frequency disturbance component d high (t) and a low-frequency disturbance component d low (t); Based on the load power data within the historical time, predict the load power P at the next moment through a hybrid load prediction algorithm load (t + 1), and calculate the prediction deviation △P diff ; Based on the key operating parameters of the system and the prediction deviation △P diff , dynamically set the upper and lower limits V high (t), V low (t); Construct a multi-objective optimization function, using the bus voltage disturbance power d(t), prediction deviation △P diff , and the upper and lower limits V high (t), V low (t) as inputs to solve the power regulation command values of the battery energy storage, fuel cell, and electrolyzer According to the power adjustment command value Generate a control command and perform power output adjustment.

2. The method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 1, wherein, The disturbance observer estimates the bus voltage disturbance power d(t) using the following dynamic gain model: Where C is the equivalent capacitance of the busbar; is the current voltage change rate, t represents time; K d (t) is the observer gain coefficient; sat() is the saturation function; V ref is the target reference voltage of the DC busbar; Among them, the observer gain coefficient K d (t) is expressed as: where K0 is the basic gain constant; K1 is the voltage change rate gain coefficient; K2 is the voltage deviation gain coefficient; V dc (t) is the current bus voltage.

3. The method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 2, characterized in that, The bus voltage disturbance power d(t) is frequency-decomposed by wavelet transform or Fourier transform, and the disturbance with a frequency component higher than the cut-off frequency fc is defined as the high-frequency disturbance component d high (t), which is responded to by the battery energy storage system; Define the disturbance with a frequency component lower than the cut-off frequency fc as the low-frequency disturbance component d low (t), which is jointly responded by the electrolyzer and the fuel cell; the cut-off frequency fc is adaptively set according to the power fluctuation period.

4. A method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 1, characterized in that The specific hybrid load prediction algorithm is an ARIMA-LSTM dynamic fusion model, and the implementation method is as follows: Input the historical load power data {P load (t - N + 1),..., P load (t)} at N collected moments, and perform data preprocessing operations, including normalization, filtering out spike outliers, and smoothing, and output the cleaned sequence data for modeling; Construct the perturbation amplitude factor D s and the dominant period length C p , and the formula is as follows: where i is a sliding index representing the i-th time difference value; P load (t - i + 1) and P load (t - i) are the load powers at the (t - i + 1)-th and (t - i)-th moments respectively; f peak is the maximum non-DC frequency component obtained by applying the fast Fourier transform to the historical load power data; Set the period recognition range set Z = {6h, 12h, 24h}, and the tolerance is ±15%; If the disturbance amplitude factor D s is less than a preset load disturbance amplitude threshold and the dominant period length C p satisfies: such that |C p - z| ≤ 0.15·z, then preferably select the ARIMA model as the main model and the LSTM model as the secondary prediction path; otherwise, select the LSTM model as the main model and the ARIMA model as the secondary prediction path; Calculate the historical residual standard deviation and confidence weight of the ARIMA model and the LSTM model respectively, and finally generate the predicted value P of the load power at the next moment load (t + 1), and the formula is: Where, σ ARIMA , ω ARIMA , are respectively the historical residual standard deviation, confidence weight, and prediction result of the ARIMA model; σ LSTM , ω LSTM , are respectively the historical residual standard deviation, confidence weight, and prediction result of the LSTM model; Prediction deviation △P diff The calculation formula is: △P diff = P load (t + 1) - P load (t).

5. A method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 1, characterized in that, The upper and lower limits V high (t), V low (t) of the bus voltage control range set dynamically, the method comprising: Initial setting: Set the target reference voltage of the DC bus to V ref , and set the basic voltage tolerance to △V base , so as to determine the initial control range as: V low (0) = V ref - △V base , V high (0) = V ref + △V base , where V high (0), V low (0) are the initial upper limit value and lower limit value of the bus voltage respectively; Real-time parameter acquisition: In each control cycle, the state of charge (SOC) of the battery and the hydrogen storage pressure P are acquired in real time H2 and the prediction deviation △P diff ; Interval adjustment rule setting: Based on the judgment results of the following three types of working conditions, adjust the upper and lower limits of the bus voltage control interval: 1) If the state of charge (SOC) of the battery is less than the lower threshold SOC of the state of charge of the battery low , then set V low (t) = V ref + ΔV SOC , where ΔV SOC is the upward shift amount of the lower voltage limit in the SOC protection state; 2) If the hydrogen storage pressure P H2 is greater than the maximum allowable pressure P H2,max of the hydrogen tank, then set V high (t) = V ref - △V H2 , where △V H2 is the downward adjustment amount of the voltage upper limit when the hydrogen pressure is too high; 3) If the prediction deviation △P diff is greater than the load power disturbance threshold P thresh and lasts for more than three sampling periods, then the bus voltage control range is expanded to: V low V(t)=V ref -(△V base +△V ext ); V high V(t) = V ref + (ΔV base + ΔV ext ); Among them, △V ext is the extended voltage deviation range; Hysteresis interval judgment: If the interval change in the current period satisfies: |V low (t)-V low (t - 1)| < ε v ; |V high (t)-V high (t - 1)| < ε v ; Then keep the upper and lower limits of the bus voltage in the previous period unchanged; Where, V high (t - 1), V low (t - 1) are respectively the upper and lower limits of the bus voltage control interval in the previous cycle, and ε v is the minimum adjustment threshold of the voltage lower limit.

6. A method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 1, characterized in that The expression of the multi-objective optimization function is: In the formula, J(t) is the value of the optimization objective function, representing the total control cost of the control system at the current moment; is the target charge and discharge power of the battery energy storage at the current moment, is the target value of the fuel cell output power at the current moment, is the power that the electrolyzer should absorb at the current moment; V dc (t) is the DC bus voltage at the current moment, V ref is the target reference voltage of the DC bus; α, β, γ, and δ are the battery power response weight coefficient, the hydrogen energy system response weight coefficient, the voltage stability weight coefficient, and the battery power change rate penalty weight, respectively.

7. A method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 6, characterized in that Generating the control command and performing power output adjustment specifically includes: Smoothing sliding mode control of the battery energy storage system: Using the hyperbolic tangent function to replace the traditional sign function to generate charge and discharge commands to suppress the chattering of the control command; Dynamically limit the charge and discharge power according to the state of charge (SOC) of the battery. When the SOC is higher than 90%, only discharging is allowed, and when it is lower than 20%, only charging is allowed; Closed-loop control of the hydrogen flow of the fuel cell: By monitoring the output current and hydrogen flow rate of the fuel cell in real time, calculating the theoretical hydrogen demand, and superimposing the low-frequency disturbance component d low (t) to generate a feedforward compensation flow rate command; Using a PID controller to close-loop regulate the hydrogen flow rate to ensure that the hydrogen utilization rate is not less than 95% to avoid fuel starvation or waste; Dynamically adjust the output power command according to the stack voltage and conversion efficiency, and match the bus voltage demand through a boost DC-DC converter; MPPT-feedforward composite control of the electrolyzer: Periodically perturb the input voltage of the electrolyzer based on the maximum power point tracking (MPPT) algorithm, and combine the low-frequency perturbation component d low (t) to dynamically correct the hydrogen production power; Limit the hydrogen production power in real time according to the pressure of the hydrogen storage tank. When the pressure exceeds the limit, force the unit to stop, and when the pressure is insufficient, increase it to the rated power.

8. The method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 1, characterized in that, The method further includes an embedded PI regulation link, which in real time uses the bus voltage deviation △V = V ref - V dc as the input to generate a regulated power △P PI , which is superimposed on the control command; When any of the following abnormal states is detected: Sensor data interruption or out-of-limit; The controller communication interruption exceeds the set duration; The energy storage unit has no power response for more than two sampling periods; Automatically start the fault-tolerant logic, and control the battery or fuel cell to maintain the bus voltage within the tolerance interval at a preset power until the main control system recovers.

9. The method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system according to claim 8, characterized in that, The adjusted power ΔP PI Satisfies: Where, △P PI (t) is the current regulating power of the PI feedback; K p is the proportional gain coefficient; V ref is the target reference voltage of the DC bus; K j is the integral gain coefficient; τ is the integral variable.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements a method for suppressing the DC bus voltage fluctuation of a renewable energy hydrogen production system as described in any one of claims 1-9.

Images