Wind-solar off-grid direct current hydrogen production system based on dynamic gradient control

Abstract

The application belongs to the technical field of renewable energy and hydrogen energy preparation, and provides a wind-solar off-grid direct-current hydrogen production system based on dynamic gradient control, comprising: a hardware perception layer for integrating wind-solar off-grid direct-current hydrogen production system hardware; a data communication layer for acquiring key operating parameters of the hardware perception layer; a system control layer for adopting a multi-time scale hierarchical control structure combined with key operating parameters, constructing a dynamic gradient feedback control model, and performing dynamic gradient control on the wind-solar off-grid direct-current hydrogen production system. The three-layer control of the system forms a top-down instruction transmission mechanism, ensuring the operation economy and stability of the wind-solar off-grid direct-current hydrogen production system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of renewable energy and hydrogen energy preparation, and particularly relates to a wind-solar off-grid direct-current hydrogen production system based on dynamic gradient control. BACKGROUND

[0002] With the deepening of global energy transformation, green hydrogen has become an important carrier for global carbon emission reduction due to its characteristics of being clean, efficient and renewable. The large-scale off-grid "green electricity hydrogen production" system containing wind energy, solar energy and hydrogen energy has become an important development direction of hydrogen-based energy industry. However, the output power of wind-solar power has great volatility and intermittency due to the influence of weather, day and night and seasonal changes, which brings challenges to the stable operation of the off-grid system. In the process of electric energy collection, transmission and storage of the existing wind-solar off-grid direct-current system, it is difficult to effectively control the instantaneous fluctuation to realize efficient energy matching between the wind-solar power and the hydrogen production device. In addition, due to the lack of multi-time scale system dynamic control method, the requirements of instantaneous response and global coordination of the system cannot be met, resulting in low energy utilization rate and poor economic benefit of the system. In view of the above problems, a new type of control system for the wind-solar off-grid direct-current hydrogen production system is needed to realize real-time regulation of wind, solar, energy storage and hydrogen production units on multiple time scales, to realize real-time monitoring and coordinated control of system energy flow, and to improve the stability and operating efficiency of the system. SUMMARY

[0003] In view of the problems existing in the prior art, the application provides a wind-solar off-grid direct-current hydrogen production system based on dynamic gradient control, which realizes real-time regulation of the working state of wind, photovoltaic, energy storage and hydrogen production units on multiple time scales by constructing three-layer dynamic regulation strategies of short cycle, medium cycle and long cycle, realizes real-time monitoring and coordinated control of system energy flow, and improves the stability and operating efficiency of the system.

[0004] A wind-solar off-grid direct-current hydrogen production system based on dynamic gradient control, comprising:

[0005] A hardware perception layer for integrating the hardware of the wind-solar off-grid direct-current hydrogen production system;

[0006] A data communication layer for obtaining key operating parameters of the hardware perception layer;

[0007] A system control layer for constructing a dynamic gradient feedback control model by using a multi-time scale hierarchical control structure combined with the key operating parameters, and performing dynamic gradient control on the wind-solar off-grid direct-current hydrogen production system;

[0008] The system control layer comprises:

[0009] The top-level control layer uses a minute-long control cycle, builds an optimization objective function using historical key operating parameters, and calculates the operating trajectory vector of the hardware perception layer in combination with system constraints to generate a long-term energy scheduling plan.

[0010] The middle control layer is used to adopt a second-level medium-time control cycle, based on the operation trajectory vector and the key operation parameters, to correct the long-cycle energy scheduling plan through logical judgment and rolling correction mechanism, and obtain voltage gradient information;

[0011] The bottom control layer is used to adopt a preset short-time control cycle, combine the voltage gradient information and the voltage state error to build the dynamic gradient feedback control model, and correct the bus voltage based on the dynamic gradient feedback control model.

[0012] Preferably, the key operating parameters include wind power generation data, photovoltaic power generation data, charge and discharge data, state of charge information, electrolysis hydrogen production power and bus voltage data;

[0013] The hardware perception layer includes:

[0014] A wind power generation unit, configured to generate wind power and generate wind power generation data;

[0015] A photovoltaic power generation unit, used for performing photovoltaic power generation and generating photovoltaic power generation data;

[0016] Energy storage unit, used to store wind and photovoltaic power generation and track system load and wind and solar fluctuations, generating charge and discharge data and state of charge information;

[0017] an electrolytic hydrogen production unit, configured to produce hydrogen by electrolysis based on the electrical energy to generate electrolytic hydrogen production power;

[0018] The DC power distribution module is used to realize DC energy transmission and voltage and current regulation between the wind power generation unit, photovoltaic power generation unit, energy storage unit and electrolytic hydrogen production unit, and generate bus voltage data;

[0019] A sampling module, used for collecting the key operating parameters in real time;

[0020] A communication unit, configured to transmit the key operating parameters in real time;

[0021] The monitoring and control module is used to perform dynamic gradient control on the wind-solar off-grid DC hydrogen production system.

[0022] Preferably, the optimization objective function is constructed based on the unit operation and maintenance cost of energy storage, energy storage output, unit hydrogen production income, electrolyzer working efficiency, hydrogen production efficiency, unit wind and solar power abandonment penalty cost, wind and solar power abandonment power and a preset long-term control period.

[0023] Preferably, the system constraints comprise a system energy balance constraint, a storage output constraint, a storage boundary constraint, an electrolyzer working interval constraint, and a hydrogen production flexibility gradient constraint; wherein,

[0024] The energy balance constraint is calculated based on predicted wind power and predicted photovoltaic power.

[0025] The storage output constraint is calculated based on a minimum storage output value and a maximum storage output value.

[0026] The storage boundary constraint is calculated based on an initial state of charge of the storage, a storage capacity, a minimum allowable state of charge of the storage, and a maximum allowable state of charge of the storage.

[0027] The electrolyzer working interval constraint is calculated based on a minimum allowable electrolyzer power value and a maximum allowable electrolyzer power value.

[0028] The hydrogen production flexibility gradient constraint is calculated based on a maximum allowable electrolyzer adjustment value.

[0029] Preferably, the optimization objective function is solved to obtain an electrolyzer operation reference value and a storage operation reference value, and in combination with a bus voltage set value, an operation trajectory vector is obtained.

[0030] Preferably, in the middle control layer, the process of modifying the long-period energy scheduling scheme through logical judgment and a rolling correction mechanism comprises:

[0031] A system state vector is calculated based on the key operation parameters.

[0032] A gradient information is obtained by difference calculation of the system state vector; wherein, the gradient information comprises wind and light power generation gradient information and voltage gradient information.

[0033] Based on the wind and light power generation gradient information, a wind and light power generation short-time power estimation is performed using first-order Taylor expansion to obtain a wind and light power generation short-time power estimation result.

[0034] A power calculation result is obtained by calculating a power gap or surplus value based on the wind and light power generation short-time power estimation result and electrolyzer hydrogen production power.

[0035] When the power calculation result is greater than zero, it represents power surplus, and it is determined whether the electrolyzer hydrogen production power can be increased under the premise of satisfying the electrolyzer working interval constraint and the hydrogen production flexibility gradient constraint. If yes, the electrolyzer hydrogen production power at this time is calculated. Otherwise, it is determined whether the storage unit can absorb the surplus power under the premise of satisfying the storage output constraint and the storage boundary constraint. If yes, the storage absorbs the surplus power, and the storage absorption power is calculated. Otherwise, the surplus power is discarded through a wind and light discarding method.

[0036] When the power calculation result is less than zero, it represents that there is a power gap, and it is judged whether the energy storage unit can release power under the premise of meeting the energy storage output constraint and the energy storage boundary constraint, if yes, the power is released for the electrolytic tank to maintain the hydrogen production plan, and the release power of the energy storage unit is calculated, otherwise, the hydrogen production power of the electrolytic tank is reduced, and the hydrogen production power of the electrolytic tank at this time is calculated.

[0037] Preferably, in the bottom control layer, the process of constructing the dynamic gradient feedback control model comprises:

[0038] Based on the real-time obtained bus voltage data, the voltage state error is calculated;

[0039] Based on the voltage state error and the adaptive adjustment factor, the proportional, differential and integral dynamic gain matrix based on gradient information is calculated respectively;

[0040] Based on the dynamic gain matrix, the voltage state error and the voltage gradient information, the dynamic gradient feedback control model is constructed.

[0041] Compared with the prior art, the beneficial effects of the present application are: the system three-layer control forms a top-down instruction transmission mechanism, ensuring the economic efficiency and stability of the wind-solar off-grid direct current hydrogen production system. The top control layer uses wind-solar prediction for forward-looking power scheduling, generates the operation reference trajectory of the large-capacity energy storage unit and the electrolytic tank to guide the long-term operation of the system, realizes the economic optimization scheduling scheme of the system, the middle control generates gradient information and performs short-term prediction on the wind-solar power data, realizes the tracking and correction of the operation reference trajectory, ensures the dynamic energy balance of the system under wind-solar fluctuation, the bottom control uses the dynamic gradient feedback control model to control the real-time response of the fast-response energy storage unit to the frequency oscillation caused by wind-solar power fluctuation, maintains transient voltage stability, and ensures the stable operation of the system. The hierarchical optimization across different time scales, the online adaptive setting mechanism of parameters and the dynamic control logic of the bottom control in the present application together constitute the significant innovation of the present application which is different from the prior art, effectively improving the comprehensive performance of the system under complex working conditions. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the present application, the following briefly introduces the drawings needed in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0043] Figure 1 The hierarchical diagram of the wind-solar off-grid direct current hydrogen production system based on dynamic gradient control in the embodiments of the present application;

[0044] Figure 2This is a multi-scale hierarchical control strategy architecture for a wind-solar off-grid DC hydrogen production system based on dynamic gradient control according to an embodiment of the present invention;

[0045] Figure 3 Schematic diagrams comparing the performance of the multi-scale control strategy and traditional control under different operating scenarios of the embodiment of the present invention; (a) is a schematic diagram for comparing hydrogen production; (b) is a schematic diagram for comparing economic benefits; (c) is a schematic diagram for comparing system utilization; and (d) is a schematic diagram for comparing performance improvement percentages.

[0046] Figure 4 Schematic diagram of the system operation under a typical day scenario of an embodiment of the present invention; wherein, (a) is a schematic diagram of the power curve; (b) is a schematic diagram of the energy storage and power abandonment curve; (c) is a schematic diagram of the energy storage SOC curve; and (d) is a schematic diagram of the cumulative hydrogen production curve. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] Example 1:

[0050] like Figure 1 As shown, a wind-solar off-grid DC hydrogen production system based on dynamic gradient control includes: a hardware perception layer, a data communication layer and a system control layer.

[0051] The hardware perception layer is used to integrate the hardware of the wind-solar off-grid DC hydrogen production system.

[0052] The data communication layer is used to obtain key operating parameters of the hardware perception layer.

[0053] The system control layer is used to adopt a multi-time scale hierarchical control structure combined with key operating parameters to build a dynamic gradient feedback control model to perform dynamic gradient control on the wind-solar off-grid DC hydrogen production system.

[0054] A further implementation method is that the key operating parameters include wind power generation data, photovoltaic power generation data, charging and discharging data, state of charge information, electrolysis hydrogen production power and bus voltage data.

[0055] The hardware perception layer comprises a wind power generation unit, a photovoltaic power generation unit, an energy storage unit, an electrolytic hydrogen production unit, a direct current power distribution module, a sampling module, a communication unit, and a monitoring and control module.

[0056] The wind power generation unit is configured to generate wind power generation data by generating wind power. In this embodiment, the wind power generation unit is equipped with a frequency converter and a wind speed and direction sensor to realize real-time acquisition and preliminary adjustment of wind energy signals, and the rated power of the wind power generation unit is 100 kW.

[0057] The photovoltaic power generation unit is configured to generate photovoltaic power generation data by generating photovoltaic power. In this embodiment, the photovoltaic power generation unit uses the maximum power point tracking (MPPT) technology to realize real-time energy capture through an illumination intensity sensor, and the rated power of the photovoltaic power generation unit is 80 kW.

[0058] The energy storage unit is configured to store the wind power and photovoltaic power, track system load and wind and light fluctuations, and generate charge and discharge data and state of charge information. The energy storage unit includes a large-capacity energy storage unit and a fast-response energy storage unit, both of which generate charge and discharge data and state of charge information.

[0059] The large-capacity energy storage unit is a battery configured to store wind power and photovoltaic power on a medium and long-term scale, responsible for medium and long-term energy storage, maintaining SOC balance and smoothing power trends, and the maximum capacity of the battery is 240 kWh.

[0060] The fast-response energy storage unit is a super capacitor configured to quickly track system load or wind and light fluctuations to ensure stable bus voltage, and the maximum capacity of the super capacitor is 10 kWh.

[0061] The electrolytic hydrogen production unit is configured to produce electrolytic hydrogen power based on electrical energy. In this embodiment, the electrolytic hydrogen production unit uses an alkaline electrolytic cell with a rated power of 150 kW and a hydrogen production rate of 0.018 kg / kWh.

[0062] The direct current power distribution module is configured to realize direct current energy transmission and voltage and current adjustment among the wind power generation unit, the photovoltaic power generation unit, the energy storage unit, and the electrolytic hydrogen production unit, and to generate bus voltage data.

[0063] The sampling module is configured to acquire key operating parameters in real time.

[0064] The communication unit is used for real-time transmission of the key operation parameters. In the embodiment, the sampling module and the communication unit can sample the key operation parameters in real time and transmit the key operation parameters in real time through a high-speed bus or an industrial communication protocol. The Modbus TCP / IP protocol is selected for the top control layer and the middle control layer to adapt to the asynchronous transmission requirement of the decision instruction, and the Profinet IRT protocol is selected for the bottom control layer to establish a hard real-time communication channel.

[0065] The monitoring and control module is used for performing dynamic gradient control on the wind-solar off-grid DC hydrogen production system. The monitoring and control module integrates data acquisition, communication and an embedded processor, and is responsible for performing a multi-time scale control algorithm based on dynamic gradient control.

[0066] The data communication layer acquires the key operation parameters from the hardware perception layer by using the sampling module and the communication unit, and transmits the key operation parameters to the system control layer in real time through a high-speed bus or an industrial communication protocol, so as to realize collection, summarization, processing and transmission of various monitoring data, and accurately issue the instruction generated by the system control layer to the execution mechanism.

[0067] The data communication layer acquires wind power generation data, photovoltaic power generation data, energy storage unit charging and discharging data, state of charge information, electrolytic hydrogen production power, bus voltage and other key operation parameters from the hardware perception layer by using the sampling module and the communication unit, transmits the key operation parameters to the system control layer in real time through a high-speed bus or an industrial communication protocol by fusing historical wind and light generation data, and accurately issues the instruction generated by the system control layer to the execution mechanism.

[0068] Further, the system control layer comprises a top control layer, a middle control layer and a bottom control layer. Figure 2

[0069] ​The top-level control layer uses a minute-long control cycle and historical key operating parameters to construct an optimization objective function. Combined with system constraints, it calculates the operating trajectory vector of the hardware perception layer and generates a long-term energy scheduling plan to maximize the system's hydrogen production efficiency. Specifically, the data communication layer uses sampling modules and communication units to obtain key operating parameters such as wind power generation data, photovoltaic power generation data, energy storage unit charge and discharge data, and charge status information, electrolysis hydrogen production power, and bus voltage from the hardware perception layer. It then integrates historical wind and photovoltaic power generation data and transmits it to the system control layer in real time via a high-speed bus or industrial communication protocol. The instructions generated by the system control layer are then accurately issued to the actuators. In this embodiment, the control cycle of the top-level control layer is 15 minutes. The top-level control layer adopts the LSTM-TCN hybrid prediction model. The LSTM module of the prediction model processes the time series wind speed and irradiance data with a time step of 15 minutes. The TCN module captures the long-term periodic characteristics of meteorological data through an expanded convolution kernel with a 2n expansion coefficient. At the same time, the Attention mechanism is designed based on the SENet architecture to dynamically weighted fuse the numerical weather forecast data to ensure that the weight variance is controlled within 0.15.

[0070] A further implementation method is to construct an optimization objective function based on the unit operation and maintenance cost of energy storage, energy storage output, unit hydrogen production income, electrolyzer operating efficiency, hydrogen production efficiency, unit wind and solar power curtailment penalty cost, wind and solar power curtailment power, and a preset long-term control period. The optimization objective function is as follows:

[0071] ,

[0072] Where, is the unit operation and maintenance cost of the battery, To provide power to the battery, is the unit hydrogen production income, For the working efficiency of the electrolyzer, is the hydrogen production power, The penalty cost for unit wind and solar curtailment, is the abandoned wind and solar power, For long-term control cycle.

[0073] A further implementation method is that the system constraints include system energy balance constraints, energy storage output constraints, energy storage boundary constraints, electrolyzer working range constraints and hydrogen production flexible gradient constraints.

[0074] Calculate the energy balance constraints based on the predicted wind power and photovoltaic power:

[0075] ,

[0076] Where, For wind power prediction, To predict the power of photovoltaic.

[0077] Based on the minimum value of the energy storage output and the maximum value of the energy storage output, the energy storage output constraint is calculated:

[0078] ,

[0079] In the formula, is the minimum value of the battery output, is the maximum value of the battery output.

[0080] Based on the initial state of charge of the energy storage, the energy storage capacity, the minimum value of the energy storage state of charge allowance and the maximum value of the energy storage state of charge allowance, the energy storage boundary constraint is calculated:

[0081] ,

[0082] In the formula, is the initial state of charge of the battery, is the battery capacity, is the minimum value of the battery state of charge allowance, is the maximum value of the battery state of charge allowance.

[0083] Based on the minimum value of the electrolytic tank allowable power and the maximum value of the electrolytic tank allowable power, the electrolytic tank working interval constraint is calculated:

[0084] ,

[0085] In the formula, is the minimum value of the electrolytic tank allowable power, is the maximum value of the electrolytic tank allowable power.

[0086] Based on the maximum value of the electrolytic tank allowable adjustment, the hydrogen production flexible gradient constraint is calculated:

[0087] ,

[0088] In the formula, is the maximum value of the electrolytic tank allowable adjustment.

[0089] Further embodiments are that the optimization objective function is solved to obtain the electrolytic tank operation reference value , the energy storage operation reference value , combined with the bus voltage set value , to obtain the operation trajectory vector :

[0090] .

[0091] The top control layer will to It is periodically sent to the medium-cycle or short-cycle controller to guide the system to evolve towards the predetermined goal in the future. is a long-term control cycle, t represents the current time variable, and t0 represents the control start time. In this embodiment, For 15 minutes.

[0092] The middle control layer is used to control the cycle in seconds (10s) based on the trajectory vector As well as key operating parameters, the long-term energy scheduling plan is corrected through logical judgment and rolling correction mechanism, and voltage gradient information is obtained.

[0093] A further implementation method is that, in the middle control layer, the process of correcting the long-term energy scheduling plan through logical judgment and rolling correction mechanism includes:

[0094] Calculate the system state vector based on key operating parameters :

[0095] ,

[0096] Where, is the real-time wind power generation power, is the real-time photovoltaic power generation power, It is the real-time bus voltage data.

[0097] Perform differential calculation on the system state vector to obtain gradient information , real-time analysis of the changing trends of wind and solar power output, battery status and hydrogen production load. Among them, the gradient information includes wind and solar power generation gradient information and voltage gradient information:

[0098] .

[0099] Assuming that the change of wind and solar power output is approximately linear, based on the wind and solar power generation gradient information, the first-order Taylor expansion is used to estimate the short-term power of wind and solar power generation, and the short-term power estimation result of wind and solar power generation is obtained:

[0100] ,

[0101] Where, is the real-time wind power generation power, is the real-time photovoltaic power generation power, It is the mid-time control cycle.

[0102] Based on the short-term power estimation results of wind and solar power generation and the hydrogen production power of the electrolyzer, the power gap or surplus value is calculated to obtain the power calculation results:

[0103] .

[0104] When the power calculation result is greater than zero ( ), representing power surplus (indicating that wind and light power generation is surplus, and the battery needs to absorb or increase hydrogen production power), it is judged whether the electrolyzer hydrogen production power can be increased under the premise of meeting the electrolyzer working interval constraint and the hydrogen production flexible gradient constraint. If it is met, the electrolyzer hydrogen production power at this time is calculated:

[0105] .

[0106] Otherwise, it is judged whether the excess power absorbed by the battery unit meets the energy storage output constraint and the energy storage boundary constraint. If it is met, the excess power is absorbed, and the absorption power of the energy storage unit is calculated:

[0107] ,

[0108] Otherwise, the excess power is discarded by the wind and light abandonment method.

[0109] When the power calculation result is less than zero ( ), representing a power gap (indicating that wind and light power generation is insufficient, and the battery needs to release or reduce hydrogen production power), it is judged whether the battery unit can release power under the premise of meeting the energy storage output constraint and the energy storage boundary constraint. If it is met, the power is released for the electrolyzer to maintain the hydrogen production plan, and the release power of the energy storage unit is calculated:

[0110] .

[0111] Otherwise, the electrolyzer hydrogen production power is reduced, and the electrolyzer hydrogen production power at this time is calculated:

[0112] .

[0113] The bottom control layer is used to adopt a preset short-time control period, construct a dynamic gradient feedback control model in combination with voltage gradient information and voltage state error for instantaneous changes of environmental parameters such as wind speed and light, and correct the bus voltage based on the dynamic gradient feedback control model to ensure the stability of the bus voltage and prevent the system from being instantaneously unbalanced.

[0114] Further embodiments are that in the bottom control layer, the process of constructing the dynamic gradient feedback control model includes:

[0115] Based on the real-time acquired bus voltage data , the voltage state error is calculated:

[0116] .

[0117] Based on the voltage state error and an adaptive adjustment factor, respectively, calculate the proportional, differential and integral dynamic gain matrix based on gradient information.

[0118] The dynamic gain matrix based on gradient information is calculated by the following formula:

[0119] ,

[0120] In the formula, , , is the initial gain, , , is an adaptive adjustment factor. In the present embodiment, the adaptive adjustment factor is calibrated based on expert experience and multi-scenario experiments.

[0121] Based on the dynamic gain matrix, the voltage state error and the voltage gradient information, a dynamic gradient feedback control model is constructed:

[0122] ,

[0123] In the formula, , , are the proportional, differential and integral dynamic gain matrix based on gradient information, respectively.

[0124] Further, the output power of the super capacitor is:

[0125] .

[0126] The dynamic gain matrix based on gradient information can adaptively adjust the control gain according to the fluctuation of the system state change rate, and accelerate the system response speed.

[0127] Further, the dynamic gradient feedback control model can make the actual execution unit respond quickly, quickly suppress power fluctuations and maintain voltage stability.

[0128] Figure 3 is a schematic diagram for comparing the performance of the multi-scale control strategy and the traditional control under different operating scenarios. (a) is a schematic diagram for comparing the hydrogen production amount; (b) is a schematic diagram for comparing the economic benefit; (c) is a schematic diagram for comparing the system utilization rate; (d) is a schematic diagram for comparing the performance improvement percentage. In order to comprehensively verify the performance of the hierarchical control strategy proposed in the present application, four different operating scenarios covering "typical day", "wind abundant day", "light abundant day" and "extreme weather condition" are constructed for testing. The simulation results show that, compared with the traditional PI control method, the control strategy proposed in the present application improves the average hydrogen production amount by 8.42% and the economic benefit by 7.24%.

[0129] Figure 4 The system operation schematic diagram under the typical day scenario of the embodiment of the application; wherein, (a) is a power curve schematic diagram; (b) is a storage and abandoned electricity curve schematic diagram; (c) is a storage SOC curve schematic diagram; (d) is a cumulative hydrogen production amount curve schematic diagram. Through the dynamic process display of each key variable (wind power generation unit output power, photovoltaic power generation unit output power, electrolyzer power, load power, storage power, storage SOC, cumulative hydrogen production amount) of the system under the typical day scenario, it is further confirmed that the control strategy proposed in the application has high efficiency and adaptability under the actual operation conditions.

[0130] In summary, the innovation of the wind-solar off-grid DC hydrogen production system based on dynamic gradient control proposed in the application is:

[0131] 1. A multi-time scale dynamic hierarchical control is introduced in the wind-solar off-grid DC hydrogen production system, which takes into account both long-period energy optimization and short-period instantaneous dynamic stability control.

[0132] 2. An adaptive control mechanism based on dynamic gradient information feedback is proposed for the wind-solar off-grid DC hydrogen production system, which establishes a nonlinear correlation between the hydrogen production equipment climbing rate and the real-time SOC, and realizes the flexible linkage adjustment of the storage unit and the hydrogen production load.

[0133] 3. A dynamic gradient feedback control model is designed for the wind-solar off-grid scenario, which drives the storage system to quickly respond to power fluctuations, suppresses frequency oscillation and maintains transient voltage stability.

[0134] The above-described embodiments only describe the preferred modes of the application, and do not limit the scope of the application. Without departing from the design spirit of the application, various modifications and improvements to the technical solutions of the application made by those skilled in the art shall fall within the protection scope determined by the claims of the application.

Claims

1. A wind-solar off-grid DC hydrogen production system based on dynamic gradient control, characterized in that: include: Hardware perception layer, used to integrate the hardware of wind-solar off-grid DC hydrogen production system; The data communication layer is used to obtain the key operating parameters of the hardware perception layer; The system control layer is used to adopt a multi-time scale hierarchical control structure in combination with the key operating parameters to construct a dynamic gradient feedback control model to perform dynamic gradient control on the wind-solar off-grid DC hydrogen production system; The system control layer includes: The top-level control layer uses a minute-long control cycle, builds an optimization objective function using historical key operating parameters, and calculates the operating trajectory vector of the hardware perception layer in combination with system constraints to generate a long-term energy scheduling plan. The middle control layer is used to adopt a second-level medium-time control cycle, based on the operation trajectory vector and the key operation parameters, to correct the long-cycle energy scheduling plan through logical judgment and rolling correction mechanism, and obtain voltage gradient information; A bottom control layer is configured to adopt a preset short-time control period, combine the voltage gradient information and the voltage state error to construct the dynamic gradient feedback control model, and correct the bus voltage based on the dynamic gradient feedback control model; In the middle control layer, the process of correcting the long-period energy scheduling plan through logical judgment and rolling correction mechanism includes: Calculating a system state vector based on the key operating parameters; Performing differential calculation on the system state vector to obtain gradient information; wherein the gradient information includes wind and solar power generation gradient information and voltage gradient information; Based on the wind-solar power generation gradient information, using a first-order Taylor expansion, perform wind-solar power generation short-term power estimation to obtain a wind-solar power generation short-term power estimation result; Based on the short-term power estimation result of wind and solar power generation and the hydrogen production power of the electrolyzer, a power gap or surplus value is calculated to obtain a power calculation result; When the power calculation result is greater than zero, it indicates excess power. It is determined whether the electrolyzer hydrogen production power can be increased while satisfying the electrolyzer operating range constraint and the hydrogen production flexible gradient constraint. If so, the electrolyzer hydrogen production power at this time is calculated. Otherwise, it is determined whether the energy storage unit can absorb the excess power to satisfy the energy storage output constraint and the energy storage boundary constraint. If so, the excess power is absorbed and the energy storage absorption power is calculated. Otherwise, the excess power is discarded by abandoning wind and solar power. When the power calculation result is less than zero, it means that there is a power gap. It is necessary to determine whether the energy storage unit can release power while meeting the energy storage output constraint and the energy storage boundary constraint. If so, the power is released to the electrolyzer to maintain the hydrogen production plan, and the released power of the energy storage unit is calculated; otherwise, the hydrogen production power of the electrolyzer is reduced, and the hydrogen production power of the electrolyzer at this time is calculated.

2. The system according to claim 1, wherein: The key operating parameters include wind power generation data, photovoltaic power generation data, charge and discharge data, state of charge information, electrolysis hydrogen production power and bus voltage data; The hardware perception layer includes: A wind power generation unit, configured to generate wind power and generate wind power generation data; A photovoltaic power generation unit, used for performing photovoltaic power generation and generating photovoltaic power generation data; Energy storage unit, used to store wind and photovoltaic power generation and track system load and wind and solar fluctuations, generating charge and discharge data and state of charge information; an electrolytic hydrogen production unit, configured to produce hydrogen by electrolysis based on the electrical energy to generate electrolytic hydrogen production power; The DC power distribution module is used to realize DC energy transmission and voltage and current regulation between the wind power generation unit, photovoltaic power generation unit, energy storage unit and electrolytic hydrogen production unit, and generate bus voltage data; A sampling module, used for collecting the key operating parameters in real time; A communication unit, configured to transmit the key operating parameters in real time; The monitoring and control module is used to perform dynamic gradient control on the wind-solar off-grid DC hydrogen production system.

3. The system according to claim 1, wherein: The optimization objective function is constructed based on the unit operation and maintenance cost of energy storage, energy storage output, unit hydrogen production income, electrolyzer working efficiency, hydrogen production efficiency, unit wind and solar power abandonment penalty cost, wind and solar power abandonment power and a preset long-term control period.

4. The system according to claim 3, characterized in that The system constraints include system energy balance constraints, energy storage output constraints, energy storage boundary constraints, electrolyzer working range constraints and hydrogen production flexible gradient constraints; wherein, Calculating the energy balance constraint based on the wind power forecast and the photovoltaic power forecast; Calculating the energy storage output constraint based on the minimum energy storage output and the maximum energy storage output; Calculating the energy storage boundary constraint based on the energy storage initial state of charge, the energy storage capacity, the minimum allowable energy storage state of charge, and the maximum allowable energy storage state of charge; Calculating the working range constraint of the electrolytic cell based on the minimum allowable power value of the electrolytic cell and the maximum allowable power value of the electrolytic cell; The hydrogen production flexibility gradient constraint is calculated based on the maximum allowable adjustment value of the electrolyzer.

5. The system according to claim 1, wherein: The optimization objective function is solved to obtain the electrolytic cell operation reference value and the energy storage operation reference value, and the operation trajectory vector is obtained in combination with the bus voltage setting value.

6. The system according to claim 1, wherein: In the bottom control layer, the process of constructing the dynamic gradient feedback control model includes: Calculate voltage state error based on real-time bus voltage data; Based on the voltage state error and the adaptive adjustment factor, respectively calculating proportional, differential and integral dynamic gain matrices based on gradient information; The dynamic gradient feedback control model is constructed based on the dynamic gain matrix, the voltage state error, and the voltage gradient information.

Images