Source network hydrogen storage system double-temperature-zone solid hydrogen storage device and hydrogen storage method

Abstract

The invention discloses a source network hydrogen storage system double-temperature-zone solid hydrogen storage device which comprises a hydrogen storage tank body, a composite hydrogen storage material layer is arranged in the hydrogen storage tank body, a double-temperature-zone temperature control module is arranged on the outer side wall of the hydrogen storage tank body, a multi-parameter sensing unit is embedded in the composite hydrogen storage material layer, and the source network hydrogen storage system double-temperature-zone solid hydrogen storage device further comprises a regulation and control system. The regulation and control system is electrically connected with the double-temperature-zone temperature control module and the multi-parameter sensing unit. The problems that in the prior art, an existing hydrogen storage device adopts a single heating module, precise temperature control within a wide temperature zone range cannot be achieved, when the load of a source network fluctuates, hydrogen release rate adjustment lags behind, the matching degree with the power grid is insufficient, and the overall efficiency of the system is affected are solved. And the coordination with the source network hydrogen storage system is realized through the regulation and control system. The invention further discloses a double-temperature-zone solid hydrogen storage method of the source network hydrogen storage system.

Description

Technical Field

[0001] This invention belongs to the technical field of hydrogen energy storage and integrated source-grid-load-storage device, and relates to a dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system. This invention also relates to a dual-temperature zone solid-state hydrogen storage method for a source-grid hydrogen storage system. Background Technology

[0002] The integrated source-grid-hydrogen storage system, through the "photovoltaic-electricity-hydrogen" energy conversion link, uses surplus electricity to produce hydrogen through water electrolysis and store it during off-peak hours, and converts hydrogen energy into electricity through fuel cells during peak hours, realizing the spatial and temporal transfer of renewable energy. It is regarded as the core technical path to solve the above problems.

[0003] Hydrogen energy storage is a core component of a hydrogen energy source-grid-storage system, and the mainstream technologies currently include: 1. High-pressure gaseous hydrogen storage: It has a lower cost, but the hydrogen storage density is less than 5 wt%, and there is a risk of high-pressure leakage, requiring complex safety protection measures; 2. Cryogenic liquid hydrogen storage: It has a high hydrogen storage density (about 12 wt%), but it requires more than 30% of the total hydrogen energy to maintain a low temperature of -253℃, resulting in extremely low energy efficiency; 3. Solid-state hydrogen storage: Hydrogen storage through chemical or physical adsorption using materials such as metal alloys and chemical hydrides has advantages such as high hydrogen storage density (5-10wt%) and good safety (working pressure ≤3MPa), and has become a research hotspot.

[0004] Current solid-state hydrogen storage technology still faces three major bottlenecks: 1. Material performance contradictions: Traditional magnesium-based hydrogen storage materials have a theoretical hydrogen storage capacity of up to 7.6 wt%, but in practical applications, they can only achieve effective hydrogen release at temperatures above 250°C, resulting in poor kinetic performance; while titanium-based alloys can release hydrogen at room temperature, but their capacity is only 3-4 wt%, which is difficult to meet the high-density energy storage requirements of power grid systems.

[0005] 2. Insufficient temperature control accuracy: Existing hydrogen storage devices mostly use a single heating module, which cannot achieve precise temperature control over a wide temperature range. When the power grid load fluctuates, the hydrogen release rate adjusts lags, resulting in a matching degree with the power grid of less than 60%, affecting the overall system efficiency.

[0006] 3. Low level of intelligence: Material research and development relies on trial and error, and it takes 5-10 years from laboratory to industrialization; system operation relies on manual experience to adjust parameters, and it is impossible to dynamically optimize the hydrogen storage and release strategy according to the real-time changes in renewable energy output, which restricts the efficiency of source-grid coordination. Summary of the Invention

[0007] The purpose of this invention is to provide a dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system. This invention solves the problems of existing hydrogen storage devices using a single heating module, which cannot achieve precise temperature control over a wide temperature range, and the hydrogen release rate adjustment lagging when the source-grid load fluctuates, resulting in insufficient matching with the grid and affecting the overall system efficiency. Furthermore, this invention achieves coordination with the source-grid hydrogen storage system through a control system.

[0008] Another objective of this invention is to provide a dual-temperature zone solid-state hydrogen storage method for a source-grid hydrogen storage system.

[0009] The technical solution adopted in this invention is a dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system, including a hydrogen storage tank, a composite hydrogen storage material layer disposed inside the hydrogen storage tank, a dual-temperature zone temperature control module disposed on the outer wall of the hydrogen storage tank, a multi-parameter sensing unit embedded in the composite hydrogen storage material layer, and a control system, which is electrically connected to the dual-temperature zone temperature control module and the multi-parameter sensing unit respectively.

[0010] Furthermore, the composite hydrogen storage material layer uses a composite hydrogen storage material, which is specifically prepared according to the following method: Step 1, prepare MXene powder; Step 2: Using the MXene powder prepared in Step 1 as a substrate and transition metal chloride as a precursor, single-atom doping with transition metal is performed to obtain doped MXene. Step 3: Prepare magnesium-based quaternary alloy powder; Step 4: After mixing the doped MXene, magnesium-based alloy powder and carbon nanotubes at a certain mass ratio, ethanol is added and then ultrasonically dispersed, freeze-dried and plasma sintered in sequence to obtain the composite hydrogen storage material.

[0011] Further, step 1 specifically involves mixing Ti3AlC2 powder with 40% hydrofluoric acid at a molar ratio of 1:5 to 1:8, hydrothermally reacting at 180-220℃ for 12-24 hours, centrifuging and washing until pH=6-7, and vacuum drying at 50-70℃ for 10-14 hours to obtain MXene powder. Step 2 specifically involves: using atomic layer deposition technology, with the MXene powder prepared in step 1 as the substrate and transition metal chloride as the precursor, depositing at 150-200℃ for 10-30 cycles, controlling the single atom loading to 0.5-2.0 at%, to obtain the doped MXene. The transition metal TM is at least one of Fe, Co, Ni, and Cu, doped in single-atom form with a loading of 0.5-2.0 at%; Step 3 specifically involves: using Mg, Al, Zn, and TM metal particles, arc melting is performed under argon protection with a current of 200-300A. After casting, the particles are ball-milled to a particle size of 5-10μm to obtain magnesium-based quaternary alloy powder. The general chemical formula of the magnesium-based quaternary alloy phase is Mg. a Al b Zn c TM d Where a+b+c+d=1, 0.6≤a≤0.8, 0.05≤b≤0.15, 0.05≤c≤0.15, 0.05≤d≤0.1; Step 4 is as follows: Mix the doped MXene, magnesium-based alloy powder and carbon nanotubes at a mass ratio of (4-6):(3-5):1, add ethanol and ultrasonically disperse at 400-600W power for 2-4h, freeze dry and then discharge plasma sinter at 350-450℃ and 30-50MPa for 5-10min to obtain the composite hydrogen storage material. Carbon nanotubes have a diameter of 5-20 nm, a length of 1-5 μm, and a mass percentage of 3-5%.

[0012] Furthermore, the inner wall of the hydrogen storage tank is coated with a 0.1-0.2mm thick polytetrafluoroethylene anti-corrosion layer. The inside of the hydrogen storage tank is equipped with a honeycomb flow guiding structure with flow guiding holes of 8-12mm in diameter and hole spacing of 20-30mm. Hydrogen inlet and outlet are respectively provided at both ends of the hydrogen storage tank. Hydrogen inlet solenoid valve and hydrogen outlet solenoid valve are respectively provided at the hydrogen inlet and outlet. A flow regulating valve is also connected in series at the hydrogen outlet. The flow regulating valve has an adjustment range of 0.1-2L / min and an adjustment accuracy of ±0.01L / min.

[0013] Furthermore, the dual-temperature zone temperature control module includes a low-temperature control unit and a high-temperature control unit, which are arranged alternately from top to bottom on the outer wall of the hydrogen storage tank. The low-temperature control unit and the high-temperature control unit are attached to the outer wall of the hydrogen storage tank by insulating and thermally conductive silicone, and the thermal conductivity of the bonding surface is ≥2.0W / (m²). K), the low temperature control unit is a semiconductor cooling chip with a power of 300-800W, a temperature control range of -20℃ to 50℃, and adopts PID control with an accuracy of ±1℃; the high temperature control unit is a graphene electric heating chip with a power of 500-1500W, a temperature control range of 50℃ to 200℃, and a heating rate of 5-10℃ / min. The low temperature control unit and the high temperature control unit are electrically connected to the control system respectively.

[0014] Furthermore, the multi-parameter sensing unit includes a distributed optical fiber temperature sensor. Five monitoring points are uniformly embedded in the composite hydrogen storage material layer along the axis of the hydrogen storage tank. The measurement range of the distributed optical fiber temperature sensor is -50℃ to 300℃, the resolution is 0.1℃, and the response time is ≤1s. The miniature pressure sensor is installed at the top center of the hydrogen storage tank, with a measurement range of 0-5MPa, an accuracy of ±0.02MPa, and a sampling frequency of 10Hz. A hydrogen concentration sensor, installed at the hydrogen outlet, has a measurement range of 0-100%vol and a resolution of 0.1%vol. The electrochemical workstation integrated module includes a three-electrode system: the working electrode is a composite hydrogen storage material, the reference electrode is a saturated calomel electrode, and the counter electrode is a platinum sheet. The testing frequency range is 10. -2 -10 5 Hz; Electrical connections for the integrated control system of distributed fiber optic temperature sensors, miniature pressure sensors, hydrogen concentration sensors, and electrochemical workstation modules.

[0015] Furthermore, the control system includes an STM32H743 microprocessor, a machine learning inference engine, a data storage module, a power drive unit, a PID control module, and a source-network coordination interface. The STM32H743 microprocessor is electrically connected to the machine learning inference engine, the data storage module, the PID control module, the power drive unit, and the source-network coordination interface, respectively. The PID control module is electrically connected to the power drive unit. The STM32H743 microprocessor is also electrically connected to a multi-parameter sensing unit. The PID control module is electrically connected to the dual-temperature zone temperature control module and is used to output a PWM signal to regulate its power. The STM32H743 microprocessor has a main frequency of ≥400MHz, and the machine learning inference engine has a built-in trained LSTM prediction model and reinforcement learning optimization model; the data storage module has a capacity of ≥32GB, and the power drive unit has an output voltage of 0-24V and a current of 0-10A. The source-network coordination interface includes a DC / DC converter and a communication protocol conversion unit. The DC / DC converter is electrically connected to the communication protocol conversion unit, which is in turn electrically connected to the STM32H743 microprocessor. The DC / DC converter has an input voltage range of 24-72V and an output power of 0-1000W. The communication protocol conversion unit supports Modbus-RTU and OPC UA protocols and connects to the source-network hydrogen storage system scheduling platform via 4G / 5G wireless communication or Ethernet. The data transmission rate is ≥1Mbps and the response time is ≤1s. The source-network coordination interface also includes a power quality monitoring unit, which monitors the input voltage fluctuation ≤±10%, harmonic distortion rate ≤5%, and frequency deviation ≤±0.5Hz in real time. When the thresholds are exceeded, a protection mechanism is activated to cut off the external power supply and switch to the backup power supply.

[0016] Furthermore, the outer side of the hydrogen storage tank is also wrapped with a composite insulation layer, which includes an aluminum foil reflective layer, an aluminum silicate fiber layer, and a stainless steel protective layer arranged sequentially from the inside to the outside.

[0017] The second technical solution adopted in this invention is a dual-temperature zone solid-state hydrogen storage method for a source-grid hydrogen storage system, which uses the aforementioned dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system, including standby mode, hydrogen storage mode, hydrogen release mode, maintenance mode, and dynamic optimization process.

[0018] Furthermore, the standby mode is as follows: when the system is under low load and no hydrogen storage or release is required, the control system controls the dual-temperature zone temperature control module to maintain the internal temperature of the hydrogen storage tank at 25±5℃, with a power consumption of ≤50W; both the hydrogen input solenoid valve and the hydrogen output solenoid valve are in the closed state; the flow regulating valve is in the minimum opening or closed state; the distributed fiber optic temperature sensor, miniature pressure sensor, hydrogen concentration sensor, and electrochemical workstation integrated module in the multi-parameter sensing unit continuously collect data and transmit it to the STM32H743 microprocessor; the STM32H743 microprocessor triggers a self-test program every 30 minutes, and performs diagnostic analysis on key parameters through the machine learning inference engine; The hydrogen storage mode is as follows: When the power grid system is in a low-power period, the control system receives a scheduling signal from the power grid hydrogen storage system scheduling platform through the power grid coordination interface and starts the hydrogen storage program. The STM32H743 microprocessor controls the power drive unit to start the low-temperature control unit of the dual-temperature zone temperature control module to maintain the temperature at 20-30℃; at the same time, it controls the opening of the hydrogen input solenoid valve, and hydrogen enters the hydrogen storage tank through the hydrogen input port; the multi-parameter sensing unit monitors the pressure in real time. When the micro pressure sensor detects that the pressure reaches 2-2.5MPa, the STM32H743 microprocessor controls the closing of the hydrogen input solenoid valve; the hydrogen storage rate is controlled at 0.5-1L / min by the flow regulating valve. The hydrogen release mode is as follows: When the power grid system is in peak electricity consumption, the control system receives hydrogen release commands and load demands from the power grid hydrogen storage system scheduling platform through the power grid collaborative interface; the STM32H743 microprocessor controls the power drive unit to switch temperature zones according to the load size: if the load is ≤2kW, the low temperature control unit is activated to maintain the temperature at 50-100℃, and the hydrogen output solenoid valve and flow regulating valve are controlled to make the hydrogen release rate at 0.1-0.5L / min; if the load is >2kW, the high temperature control unit is activated to maintain the temperature at 100-200℃, and the hydrogen output solenoid valve and flow regulating valve are controlled to make the hydrogen release rate at 1-2L / min. The dynamic optimization process is as follows: The machine learning inference engine is periodically invoked; the STM32H743 microprocessor acquires temperature (T), pressure (P), and flow rate (hydrogen release rate (V)) data collected by the multi-parameter sensing unit every 5 minutes, and inputs it into the LSTM model built into the machine learning inference engine to predict the hydrogen release rate for the next 10-30 minutes; based on the prediction results, the STM32H743 microprocessor dynamically adjusts the PWM signal output from the power drive unit to the dual-temperature zone temperature control module through the PID adjustment module to optimize the temperature control power, so that the hydrogen release rate matches the load change in advance; for example, if it is predicted that the load will increase to 3kW in 10 minutes, the temperature will be raised from 80℃ to 120℃ in advance to ensure that the hydrogen release rate increases synchronously. The prediction results are sent to the dispatching platform through the source-grid coordination interface to assist the grid in load distribution, thereby improving the overall system efficiency by 15-20%.

[0019] The maintenance mode is as follows: After the system has accumulated 100 hours of operation, the STM32H743 microprocessor controls the electrochemical workstation integrated module to perform electrochemical impedance spectroscopy detection and transmits the data to the machine learning inference engine for analysis. If the impedance increase is greater than 20%, the STM32H743 microprocessor controls the power drive unit to drive the dual-temperature zone temperature control module to execute the in-situ activation program and perform temperature cycling. The in-situ activation program is as follows: 1. Deep cooling stage: The low-temperature control unit 31 is activated to lower the overall temperature of the composite hydrogen storage material layer to -20℃ at a rate of not less than 5℃ / min, and hold it at this temperature for 10 minutes. This stage aims to utilize the "thermal shock" effect to generate stress at the microscopic interface through the difference in the volume shrinkage coefficients of the material and the hydride, thereby peeling off part of the dense hydride passivation layer. 2. Temperature activation stage: Switch to the high-temperature control unit 32 to raise the temperature to 100℃ at a rapid heating rate of 8-10℃ / min, and hold it at this temperature for 15 minutes. Within this temperature and time frame, the material's lattice vibration intensifies, providing desorption kinetic energy for physically adsorbed hydrogen and clearing hydrogen diffusion channels. Simultaneously, it promotes the partial decomposition of chemically adsorbed hydrides, re-exposing fresh metal active sites. 3. Steady-state recovery phase: Stop heating and, if necessary, activate the low-temperature control unit 31 for auxiliary temperature control, allowing the temperature to steadily drop back to 20℃ (near room temperature), hold for 5 minutes, and complete one cycle.

[0020] The beneficial effects of this invention are: (1) Excellent hydrogen storage performance Through the synergistic effect of TM-modified MXene and magnesium-based quaternary alloy, the composite hydrogen storage material layer achieves a room temperature hydrogen storage capacity of 5.5-6.8 wt%, far exceeding that of traditional high-pressure gaseous hydrogen storage (<5 wt%); the hydrogen release at 100-200℃ is ≥5.0 wt%, solving the problem of "difficulty in balancing high capacity and low hydrogen release temperature" for magnesium-based materials.

[0021] (2) Dual-zone precise control The device employs a combination of semiconductor cooling chips and graphene heating elements to achieve a wide temperature range of 20℃ to 200℃ with a temperature control accuracy of ±2℃. The hydrogen release response time is ≤30s, which is 50% better than traditional devices (>60s), and it can match the load fluctuations of the power grid in real time.

[0022] (3) The LSTM equalization prediction model is used to realize the active prediction of hydrogen release trend. The PID adjustment module dynamically optimizes the temperature control strategy, so that the matching degree between hydrogen release rate and grid load is increased from 60% to more than 90%. The digital twin module supports offline simulation, reduces the number of physical experiments by more than 50%, and accelerates device iteration.

[0023] (4) System synergy By deeply integrating with the source-grid hydrogen storage system through standardized communication interfaces, it can participate in multiple application scenarios such as grid peak shaving and backup power supply; the overall energy efficiency of the device reaches 70-75%, which is 10-15 percentage points higher than that of traditional solid-state hydrogen storage systems. Attached Figure Description

[0024] Figure 1 This is a three-dimensional structural diagram of the dual-temperature zone solid hydrogen storage device of the source-grid hydrogen storage system of the present invention; Figure 2 This is a layout diagram of the dual-temperature zone temperature control module in the dual-temperature zone solid hydrogen storage device of the source-grid hydrogen storage system of the present invention; Figure 3 This is a circuit connection diagram of the dual-temperature zone solid hydrogen storage device of the source-grid hydrogen storage system of the present invention; Figure 4 This is a schematic diagram of the microstructure of the composite hydrogen storage material layer prepared in Example 3 of the present invention; Figure 5 This is a timing diagram of the hydrogen storage and release mode switching in Embodiment 5 of the present invention; Figure 6 This is a comparison chart of the prediction performance of the LSTM model in Embodiment 7 of the present invention.

[0025] In the diagram: 1. Hydrogen storage tank, 2. Composite hydrogen storage material layer, 3. Dual-temperature zone temperature control module, 4. Parameter sensing unit, 5. Control system, 6. Hydrogen input solenoid valve, 7. Hydrogen output solenoid valve, 8. Flow regulating valve; 31. Low temperature control unit; 32. High temperature control unit; 41. Distributed fiber optic temperature sensor; 42. Miniature pressure sensor; 43. Hydrogen concentration sensor; 44. Electrochemical workstation integrated module; 51. STM32H743 microprocessor; 52. Machine learning inference engine; 53. Data storage module; 54. Power drive unit; 55. Source-network interface; 56. PID control module. 551. DC / DC converter; 552. Communication protocol conversion unit. Detailed Implementation

[0026] The following detailed description is provided in conjunction with specific implementation methods.

[0027] Example 1 The present invention relates to a dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system, the structure of which is as follows: Figure 1-3 As shown, it includes a hydrogen storage tank 1, a composite hydrogen storage material layer 2 inside the hydrogen storage tank 1, a dual-temperature zone temperature control module 3 on the outer wall of the hydrogen storage tank 1, a multi-parameter sensing unit 4 embedded in the composite hydrogen storage material layer 2, and a control system 5 located outside the hydrogen storage tank 1. The control system 5 is electrically connected to the dual-temperature zone temperature control module 3 and the multi-parameter sensing unit 4, respectively.

[0028] The composite hydrogen storage material layer uses a composite hydrogen storage material, which is prepared according to the following method: Step 1, prepare MXene powder; Step 2: Using the MXene powder prepared in Step 1 as a substrate and transition metal chloride as a precursor, single-atom doping with transition metal is performed to obtain doped MXene. Step 3: Prepare magnesium-based quaternary alloy powder; Step 4: After mixing the doped MXene, magnesium-based alloy powder, and carbon nanotubes at a certain mass ratio, ethanol is added, followed by ultrasonic dispersion, freeze-drying, and plasma sintering to obtain a composite hydrogen storage material. Its microstructure is as follows: Figure 4 As shown.

[0029] Step 1 specifically involves mixing Ti3AlC2 powder with 40% hydrofluoric acid at a molar ratio of 1:5 to 1:8, hydrothermally reacting at 180-220℃ for 12-24 hours, centrifuging and washing until pH=6-7, and vacuum drying at 50-70℃ for 10-14 hours to obtain MXene powder. MXene has a high specific surface area (≥200m² / g) and abundant surface functional groups, which can provide a large number of active sites for hydrogen adsorption. Step 2 specifically involves: using atomic layer deposition technology, with the MXene powder prepared in step 1 as the substrate and transition metal chloride as the precursor, depositing at 150-200℃ for 10-30 cycles, controlling the single atom loading to 0.5-2.0 at%, to obtain the doped MXene. Transition metal TM is at least one of Fe, Co, Ni, and Cu, doped in single-atom form with a loading of 0.5-2.0 at; the metal atoms are anchored by forming coordination bonds with the functional groups on the MXene surface, thereby regulating the electronic state of the material, reducing the hydrogen adsorption energy (optimized to -0.2~0 eV), and improving the hydrogen storage kinetic performance; Step 3 specifically involves: using Mg, Al, Zn, and TM metal particles, arc melting is performed under argon protection with a current of 200-300A. After casting, the particles are ball-milled to a particle size of 5-10μm to obtain magnesium-based quaternary alloy powder. The general chemical formula of the magnesium-based quaternary alloy phase is Mg. a Al b Zn c TM d Where a+b+c+d=1, 0.6≤a≤0.8, 0.05≤b≤0.15, 0.05≤c≤0.15, 0.05≤d≤0.1; Magnesium provides high hydrogen storage capacity, Al and Zn improve the mechanical properties of the alloy, and TM further catalyzes the hydrogen release reaction, reducing the hydrogen release temperature to 100-200℃. Step 4 is as follows: Mix the doped MXene, magnesium-based alloy powder and carbon nanotubes at a mass ratio of (4-6):(3-5):1, add ethanol and ultrasonically disperse at 400-600W power for 2-4h, freeze dry and then discharge plasma sinter at 350-450℃ and 30-50MPa for 5-10min to obtain the composite hydrogen storage material. Carbon nanotubes have a diameter of 5-20 nm, a length of 1-5 μm, and a mass percentage of 3-5%.

[0030] The inner wall of the hydrogen storage tank 1 is coated with a 0.1-0.2mm thick polytetrafluoroethylene anti-corrosion layer to prevent hydrogen from reacting with the tank to form hydrides. The hydrogen storage tank 1 has a honeycomb flow guiding structure inside, with flow guiding holes of 8-12mm in diameter and hole spacing of 20-30mm. Hydrogen inlet and outlet are respectively provided at both ends of the hydrogen storage tank 1. Hydrogen inlet solenoid valve 6 and hydrogen outlet solenoid valve 7 are respectively provided at the hydrogen inlet and outlet. A flow regulating valve 8 is also connected in series at the hydrogen outlet. The flow regulating valve 8 has an adjustment range of 0.1-2L / min and an adjustment accuracy of ±0.01L / min. Hydrogen inlet solenoid valve 6, hydrogen outlet solenoid valve 7, and flow regulating valve 8 are all electrically connected to the control system 5.

[0031] The hydrogen storage tank 1 is made of 316L stainless steel, which has excellent resistance to hydrogen embrittlement. It has a volume of 50-200L and a design working pressure of 2-3MPa, which meets the medium-pressure hydrogen storage requirements of the power grid system.

[0032] The honeycomb-shaped flow guiding structure specifically involves setting 3-5 flow guiding plates inside the hydrogen storage tank 1. The flow guiding plates have evenly distributed flow channels with a diameter of 5-8 mm. The flow channels are at an angle of 30° to the axis, which promotes the uniform diffusion of hydrogen to the material layer and avoids excessively high local concentrations.

[0033] The dual-temperature zone temperature control module 3 includes a low-temperature control unit 31 and a high-temperature control unit 32. The low-temperature control unit 31 and the high-temperature control unit 32 are arranged alternately from top to bottom on the outer wall of the hydrogen storage tank 1. The low-temperature control unit 31 and the high-temperature control unit 32 are bonded to the outer wall of the hydrogen storage tank 1 using insulating and thermally conductive silicone. The thermal conductivity of the bonding surface is ≥2.0 W / (m²). The low-temperature control unit (K) is a semiconductor refrigeration chip with a power of 300-800W, a temperature control range of -20℃ to 50℃, and PID control with an accuracy of ±1℃. It is used for hydrogen storage and low-load hydrogen release. The high-temperature control unit is a graphene electric heating element with a power of 500-1500W, a temperature control range of 50℃ to 200℃, and a heating rate of 5-10℃ / min. It is used for rapid hydrogen release under high load. The low-temperature and high-temperature control units are electrically connected to the control system 5. The two units are alternately distributed on the outside of the tank and bonded together with insulating and thermally conductive silicone, with a thermal conductivity ≥2.0W / (m²). K) ensures efficient heat transfer, enables precise control over a wide temperature range, and adapts to different load requirements.

[0034] The multi-parameter sensing unit 4 includes a distributed optical fiber temperature sensor 41. Five monitoring points of the distributed optical fiber temperature sensor 41 are uniformly embedded in the composite hydrogen storage material layer 2 along the axis of the hydrogen storage tank 1. The measurement range of the distributed optical fiber temperature sensor 41 is -50℃ to 300℃, the resolution is 0.1℃, the response time is ≤1s, and it captures the temperature distribution. Miniature pressure sensor 42 is installed at the top center inside the hydrogen storage tank 1. It has a measurement range of 0-5MPa, an accuracy of ±0.02MPa, a sampling frequency of 10Hz, and reflects changes in the amount of hydrogen stored. The hydrogen concentration sensor 43 is installed at the hydrogen outlet, with a measurement range of 0-100%vol and a resolution of 0.1%vol. The catalytic combustion sensor is located in the output pipeline to quickly detect the risk of leakage. Electrochemical workstation integrated module 44 includes a three-electrode system: the working electrode is a composite hydrogen storage material, the reference electrode is a saturated calomel electrode, and the counter electrode is a platinum sheet. The test frequency range is 10. -2 -10 5 Hz, to acquire parameters such as hydrogen adsorption energy and exchange current density in real time, and to evaluate material performance; 41. Distributed fiber optic temperature sensor; 42. Miniature pressure sensor; 43. Hydrogen concentration sensor; 44. Electrochemical workstation integrated module; 5. Control system; 6. Electrical connection.

[0035] The control system 5 includes an STM32H743 microprocessor 51, a machine learning inference engine 52, a data storage module 53, a power drive unit 54, a PID control module 56, and a source-network interface 55. The STM32H743 microprocessor 51 is electrically connected to the machine learning inference engine 52, the data storage module 53, the PID control module 56, the power drive unit 54, and the source-network interface 55, respectively. The PID control module 56 is electrically connected to the power drive unit 54. The hydrogen input solenoid valve 6, the hydrogen output solenoid valve 7, and the flow control valve 8 are all electrically connected to the STM32H743 microprocessor 51.

[0036] The STM32H743 microprocessor 51 is also electrically connected to the multi-parameter sensing unit 4; the STM32H743 microprocessor 51 is electrically connected to the distributed fiber optic temperature sensor 41, the miniature pressure sensor 42, the hydrogen concentration sensor 43, and the electrochemical workstation integrated module 44, respectively. The PID adjustment module 56 is electrically connected to the dual-temperature zone temperature control module 3 and is used to output a PWM signal to adjust its power. The power drive unit 54 is electrically connected to the semiconductor cooling chip and the graphene electric heating chip respectively. The STM32H743 microprocessor has a main frequency of ≥400MHz, a machine learning inference engine with a built-in trained LSTM prediction model and reinforcement learning optimization model, a data storage module with a capacity of ≥32GB, and a power drive unit with an output voltage of 0-24V and a current of 0-10A. The source-network coordination interface 55 includes a DC / DC converter 551 and a communication protocol conversion unit 552. The DC / DC converter 551 is electrically connected to the communication protocol conversion unit 552, and the communication protocol conversion unit 552 is electrically connected to the STM32H743 microprocessor 51. The DC / DC converter 551 has an input voltage range of 24-72V and an output power of 0-1000W. The communication protocol conversion unit 552 supports Modbus-RTU and OPC UA protocols and connects to the source-network hydrogen storage system scheduling platform via 4G / 5G wireless communication or Ethernet. The data transmission rate is ≥1Mbps and the response time is ≤1s. The source-network coordination interface 55 also includes a power quality monitoring unit, which monitors the input voltage fluctuation ≤±10%, harmonic distortion rate ≤5%, and frequency deviation ≤±0.5Hz in real time. When the threshold is exceeded, a protection mechanism is activated to cut off the external power supply and switch to the backup power supply.

[0037] The hydrogen storage tank (1) is also wrapped with a composite insulation layer, which includes an aluminum foil reflective layer, an aluminum silicate fiber layer, and a stainless steel protective layer arranged sequentially from the inside to the outside.

[0038] Example 2 The present invention relates to a dual-temperature zone solid-state hydrogen storage method for a source-grid hydrogen storage system, which uses the dual-temperature zone solid-state hydrogen storage device of the source-grid hydrogen storage system in Example 1, including standby mode, hydrogen storage mode, hydrogen release mode, maintenance mode, and dynamic optimization process.

[0039] In standby mode, the system is under low load and does not require hydrogen storage or release. At this time, the control system 5 controls the dual-temperature zone temperature control module 3 to maintain the internal temperature of the hydrogen storage tank 1 at 25±5℃, with a power consumption of ≤50W. The hydrogen input solenoid valve 6 and the hydrogen output solenoid valve 7 are both in the closed state. The flow regulating valve 8 is in the minimum opening or closed state. The distributed fiber optic temperature sensor 41, the miniature pressure sensor 42, the hydrogen concentration sensor 43, and the electrochemical workstation integrated module 44 in the multi-parameter sensing unit 4 continuously collect data and transmit it to the STM32H743 microprocessor 51. The STM32H743 microprocessor 51 triggers a self-test program every 30 minutes and performs diagnostic analysis on key parameters through the machine learning inference engine 52. The hydrogen storage mode is as follows: When the power grid system is in a low-power period, the control system 5 receives the scheduling signal from the power grid hydrogen storage system scheduling platform through the power grid coordination interface 54 and starts the hydrogen storage program. The STM32H743 microprocessor 51 controls the power drive unit 54 to start the low-temperature control unit 31 of the dual-temperature zone temperature control module 3 to maintain the temperature at 20-30℃. At the same time, it controls the opening of the hydrogen input solenoid valve 6, and hydrogen enters the hydrogen storage tank 1 through the hydrogen input port. The multi-parameter sensing unit 4 monitors the pressure in real time. When the micro pressure sensor 42 detects that the pressure reaches 2-2.5MPa, the STM32H743 microprocessor 51 controls the closing of the hydrogen input solenoid valve 6. The hydrogen storage rate is controlled at 0.5-1L / min by the flow regulating valve 8. Hydrogen storage: The hydrogen produced by the electrolyzer is dried (dew point ≤ -40℃) and purified (purity ≥ 99.99%) before entering the hydrogen storage tank through the inlet valve; in the composite material layer, the hydrogen first adheres to the MXene surface through physical adsorption, and then reacts with the magnesium-based alloy through chemical adsorption to generate hydrides. The hydrogen release mode is as follows: When the power grid system is in peak electricity demand, the control system 5 receives hydrogen release commands and load demands from the power grid hydrogen storage system scheduling platform through the power grid coordination interface 54; the STM32H743 microprocessor 51 controls the power drive unit 54 to switch the temperature zone according to the load size: if the load is ≤2kW, the low temperature control unit 31 is activated to maintain the temperature at 50-100℃, and the hydrogen output solenoid valve 7 and flow regulating valve 8 are controlled to make the hydrogen release rate at 0.1-0.5L / min; if the load is >2kW, the high temperature control unit 32 is activated to maintain the temperature at 100-200℃, and the hydrogen output solenoid valve 7 and flow regulating valve 8 are controlled to make the hydrogen release rate at 1-2L / min to ensure stable output. The fuel cell converts hydrogen energy into DC power, which is then converted into AC power by the inverter and fed into the grid, with a power generation efficiency ≥45%. The dynamic optimization process is as follows: the machine learning inference engine 52 is periodically invoked; the STM32H743 microprocessor 51 acquires temperature, pressure, and flow data collected by the multi-parameter sensing unit 4 every 5 minutes, and inputs it into the LSTM model built into the machine learning inference engine 52 to predict the hydrogen release rate in the next 10-30 minutes; based on the prediction results, the STM32H743 microprocessor 51 dynamically adjusts the PWM signal output by the power drive unit 54 to the dual-temperature zone temperature control module 3 through the PID adjustment module 56 to optimize the temperature control power and make the hydrogen release rate match the load change in advance; The maintenance mode is as follows: After the system has accumulated 100 hours of operation, the STM32H743 microprocessor 51 controls the electrochemical workstation integrated module 44 to perform electrochemical impedance spectroscopy detection and transmits the data to the machine learning inference engine 52 for analysis. If the impedance increase is greater than 20%, the STM32H743 microprocessor 51 controls the power drive unit 54 to drive the dual-temperature zone temperature control module 3 to execute the in-situ activation program and perform temperature cycling. The in-situ activation program is as follows: 1. Deep cooling stage: The low temperature control unit 31 is activated to lower the overall temperature of the composite hydrogen storage material layer 2 to -20℃ at a rate of not less than 5℃ / min, and maintain this temperature for 10 minutes. This stage aims to utilize the "thermal shock" effect. 1. Stress is generated at the microscopic interface by the difference in volume shrinkage coefficient between the material and the hydride, which peels off part of the dense hydride passivation layer; 2. Temperature activation stage: switch to high temperature control unit 32, raise the temperature to 100℃ at a rapid heating rate of 8-10℃ / min, and hold for 15 minutes. At this temperature and time, the lattice vibration of the material intensifies, which on the one hand provides desorption kinetic energy for physically adsorbed hydrogen and opens up hydrogen diffusion channels; on the other hand, it promotes the partial decomposition of chemically adsorbed hydrides and re-exposes fresh metal active sites; 3. Steady-state recovery stage: stop heating and start low temperature control unit 31 for auxiliary temperature control as needed, so that the temperature drops steadily back to room temperature (i.e., around 20℃), hold for 5 minutes, and complete one cycle.

[0040] Example 3 Based on Example 1, the composite hydrogen storage material of the present invention was prepared and its performance was tested, specifically as follows: Dataset Construction: 800 sets of published data on MXene-based hydrogen storage materials (sources: Journal of Energy Chemistry, Renewable Energy, etc.) were collected, covering four transition metals: Fe, Co, Ni, and Cu, with doping amounts of 0.5-2.0 at%, and magnesium-based alloy ratios of a=0.6-0.8, b,c,d=0.05-0.15; 200 sets of data from self-made samples were added, and hydrogen storage capacity and hydrogen release temperature were tested using a Sievert-type hydrogen storage instrument. DFT calculations were performed using VASP software to obtain quantitative characteristics such as interlayer electrostatic energy (-5~5eV) and hydrogen adsorption energy (-1~0eV), which were then fused with component parameters to form an 88-dimensional input matrix. Model training: An ensemble model was built using GBDT (XGBoost library) and NN (3-layer fully connected network). The 5-fold cross-validation results showed that the hydrogen storage capacity prediction error was 2.3% and the hydrogen release temperature prediction error was 3.1%. Multi-objective optimization: With the objectives of "maximizing hydrogen storage capacity and maximizing hydrogen release at 100℃", the optimal solution was obtained through optimization using the NSGA-II algorithm: Ni doping amount 1.2at%, magnesium-based alloy composition Mg0.7Al0.1Zn0.1Ni0.1, predicted room temperature hydrogen storage capacity 6.3wt%, hydrogen release at 100℃ 3.2wt%, and hydrogen release at 200℃ 5.9wt%.

[0041] The preparation method is as follows: ①Preparation of Ti3C2(OH)2MXene: 10g of Ti3AlC2 powder (99% purity, 200 mesh) was slowly added to 200mL of 40% hydrofluoric acid solution and magnetically stirred for 30min; then transferred to a polytetrafluoroethylene reactor and hydrothermally reacted at 180℃ for 24h. The reaction product was washed with deionized water by centrifugation (8000 rpm, 5 min) until pH=6, and then vacuum dried at 60℃ for 12 h to obtain MXene powder with an interlayer spacing of 1.0 nm. X-ray diffraction (XRD) showed that the (002) peak was located at 2θ=7.8°.

[0042] ②Ni single-atom doping: An atomic layer deposition system (Beneq TFS200) was used with Ni(COD)₂ as the precursor and Ar as the carrier gas (flow rate 50 sccm). The deposition temperature was 180 °C, the precursor pulse time was 3 s, the purging time was 10 s, and the cycle was 20 times. The Ni loading was determined to be 1.2 at by inductively coupled plasma mass spectrometry (ICP-MS). X-ray photoelectron spectroscopy (XPS) showed that the Ni 2p³ / 2 peak was located at 855.2 eV, indicating that Ni formed a coordination bond with the -OH group on the MXene surface.

[0043] ③ Preparation of magnesium-based quaternary alloys: Weigh the metal raw materials according to the formula: Mg0.7Al0.1Zn0.1Ni0.1 (Mg granules 99.95%, Al flakes 99.99%, Zn granules 99.99%, Ni powder 99.99%); place them in a copper crucible in an electric arc furnace, and evacuate to 5 × 10⁻⁶. -3Pa, high-purity Ar (99.999%) was introduced to 0.05 MPa, and the mixture was arc-melted 4 times (3 min each time) to ensure uniform composition; the ingot was vacuum annealed at 350℃ for 4 h, and then ball-milled (agate jar, ZrO2 balls) for 4 h at a speed of 400 r / min and a ball-to-material ratio of 10:1 to obtain alloy powder with a particle size of 8 μm.

[0044] ④ Composite molding: Ni-MXene and magnesium-based alloy powder were mixed at a mass ratio of 1:4, and anhydrous ethanol (solid-liquid ratio 1:5) was added. The mixture was ultrasonically dispersed for 2 hours (power 500W, frequency 30kHz). After freeze-drying (-50℃, 12h), the mixture was shaped in a spark plasma sintering furnace (SPS-20T): heating rate: 100℃ / min. Sintering temperature: 400℃; pressure: 40MPa; holding time: 8min; atmosphere: high-purity Ar (flow rate 200sccm); a composite block with a diameter of 50mm×5mm and a density of 4.2g / cm³ was obtained.

[0045] The performance tests of the composite block prepared in this embodiment are as follows: ① Hydrogen storage capacity: In the Sievert hydrogen storage instrument, after adsorption at 25℃ and 3MPa for 24h, the hydrogen storage capacity is 6.2wt%, which is consistent with the model prediction. ② Hydrogen release kinetics: Temperature programmed desorption (TPD) experiments showed that the hydrogen release was 3.2 wt% at 50-100℃ (peak temperature 85℃) and 5.8 wt% at 100-200℃ (peak temperature 150℃). ③ Cyclic stability: After 50 cycles of hydrogen absorption at 25℃ and hydrogen release at 150℃, the capacity retention rate is 92%, which is better than that of traditional magnesium-based materials (70-80%). ④ Microstructure: High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) showed that Ni atoms were uniformly dispersed on the MXene surface without agglomeration; X-ray diffraction (XRD) confirmed that the magnesium-based alloy and MXene formed a heterojunction with no obvious reactive phase formation.

[0046] Example 4 Based on Example 3, this embodiment specifically specifies the assembly and commissioning of the solid hydrogen storage device, as follows: 316L stainless steel sheet (5mm thick) is selected, and after laser cutting and rolling, it is welded into a cylindrical tank with a diameter of 300mm × 1000mm and a volume of 70L. The welding is carried out by argon arc welding (protective gas Ar 99.99%), and the weld is tested by penetrant testing (PT) to ensure that there are no pores or cracks. After sandblasting the inner wall (roughness Ra=3.2μm), a 0.15mm thick polytetrafluoroethylene coating was applied by electrostatic spraying and cured at 200℃ for 2 hours. The coating adhesion (cross-cut test) reached level 1. Three guide plates (316L material, 3mm thick) were welded inside, with a plate spacing of 300mm and a flow channel diameter of 6mm. The hydrogen distribution uniformity reached 90% through computational fluid dynamics (CFD) simulation.

[0047] The low-temperature control unit of the dual-temperature zone temperature control module uses four semiconductor cooling chips (model TEC1-12706, size 40×40mm), each with a power of 500W; thermally conductive silicone (thermal conductivity 3.0W / (m²)) is used. K)) is attached to the lower part of the tank body with a spacing of 6cm, and connected to the PID adjustment module 56 (accuracy ±0.1℃). The high temperature control unit uses 4 graphene heating elements (size 100×50mm, resistance 10Ω), with a power of 1000W / element; it is attached to the upper part of the tank body and is alternately distributed with the low temperature unit, and is connected by high temperature resistant silicone wire (temperature resistance 200℃).

[0048] The hydrogen storage tank 1 is externally wrapped with an 80mm thick aluminum silicate fiber blanket (thermal conductivity 0.028W / (m²)). K), encased in a 0.5mm thick 304 stainless steel plate, with an IP65 protection rating.

[0049] A distributed fiber optic sensor (model FBG-DTS-100) is inserted into a Φ3mm quartz tube along the tank axis, with the tube end fixed to a guide plate. Four monitoring points are set up (top, upper middle, lower middle, and bottom). The fiber optic cable is connected to a demodulator (accuracy ±0.5℃, sampling frequency 1Hz) via an ST interface. The miniature pressure sensor (model PX409-3.5MPaG) is installed on the top of the tank via an M10 threaded interface and outputs a 4-20mA signal. The hydrogen concentration sensor (model EHS-100) is installed in the pipeline after the output valve, with a measurement range of 0-100%vol and a response time of ≤1s; The electrochemical workstation integrated module uses a micro electrochemical workstation (model CHI760E-Mini) connected to the composite hydrogen storage material layer via probes to monitor parameters such as open-circuit voltage and AC impedance in real time.

[0050] Hardware setup of control system 5: Main control board: STM32H743IIT6, external 16GBeMMC storage, 4.3-inch touch screen; Power module: PWM control (frequency 10kHz), output voltage 0-24V, current 0-10A; Communication module: 4G module (Huawei ME909s-821), supports TCP / IP protocol; all circuits are integrated in a waterproof box (IP65), power supply voltage 24VDC, power consumption ≤50W; Software development: The machine learning inference engine incorporates a trained LSTM model and a DQN (Directed QN) optimization model. The LSTM model takes temperature, pressure, and flow rate data from the past 10 minutes as input and outputs a predicted hydrogen release rate for the next 30 minutes with an error of ≤5%. The data storage module has a capacity of ≥32GB and supports local data caching and cloud synchronization. The power drive unit outputs voltage 0-24V and current 0-10A, adjusting the power of the temperature control module via a PWM signal. Operating System: FreeRTOS real-time system, task scheduling cycle 1ms; Data Acquisition: 8 analog signals are synchronously acquired through AD7606 chip, sampling frequency 10Hz; LSTM Model: Based on TensorFlow Lite, with 128 neurons in the input layer, 64 neurons in the hidden layer, and 1 neuron in the output layer (hydrogen release rate); PID Parameters: Tuned using the Ziegler-Nichols method, Kp=1.2, Ki=0.08, Kd=0.15; System debugging: Temperature control accuracy: With a set temperature of 100℃ and continuous monitoring for 24 hours, the actual temperature fluctuation range is 99.2-100.8℃, with an accuracy of ±0.8℃; Response time: It takes 28 seconds to heat from 25℃ to 100℃, which meets the design requirements (≤30s). Model prediction: Comparing the measured hydrogen release rate with the LSTM prediction, the prediction error is 4.8% at 10 minutes and 5.2% at 30 minutes. Communication test: 500 communications with the source network scheduling platform (simulated) with a success rate of 100% and an average latency of 800ms.

[0051] Example 5 The application test of the dual-temperature zone solid-state hydrogen storage device in the source-grid hydrogen storage system of Example 4 in the source-grid hydrogen storage system is as follows: (1) Test platform setup A 10kW-level hydrogen storage microsystem is constructed, comprising: renewable energy input: 10kW photovoltaic panel (Voc=380V) + 5kW wind turbine; electrolyzer: proton exchange membrane (PEM) electrolyzer, with a hydrogen production rate of 1.5Nm³ / h and a power of 5kW; hydrogen storage device: the device of this invention (70L) + backup high-pressure hydrogen storage cylinder (30MPa, 50L); power generation unit: 5kW fuel cell (PEMFC), with an efficiency of 45%. Load simulation: resistive-inductive load cell (0-5kW continuously adjustable); monitoring system: data acquisition unit (Agilent 34970A) + host computer software; (2) Hydrogen storage mode test (simulating nighttime operating conditions) ① Experimental conditions: Photovoltaic simulated power output 3kW, wind power simulated output 2kW, electrolytic cell operating at full load; ② Process record: At 18:00, the hydrogen storage mode is activated, and the cryogenic unit stabilizes the temperature at 25℃. From 18:00 to 24:00, the pressure increased from 0.1 MPa to 2.0 MPa, with a hydrogen storage capacity of approximately 5.6 m³. The system automatically shuts down at 24:00, with a hydrogen storage efficiency (hydrogen storage capacity × high calorific value / electrical energy consumed) of 72%.

[0052] (3) Hydrogen release mode test (simulating daytime operating conditions) ①Low load test (8:00-10:00): With the load set at 1.5kW, the system activates the cryogenic unit, maintaining the temperature at 80℃. The hydrogen release rate is 0.3 L / min, the fuel cell output is stable, and the voltage fluctuation is ±2%. The cumulative power generation in 2 hours is 3kWh, with an average efficiency of 45%.

[0053] ②High-load test (10:00-12:00): When the load increases to 4kW, the system switches to the high-temperature unit, and the temperature rises to 150℃. Hydrogen release rate 1.8 L / min, fuel cell output response time <5 s; The cumulative power generation in 2 hours is 8 kWh, with an average efficiency of 44%.

[0054] (4) Dynamic response test Simulated load change (1kW→4kW→2kW): At 12:00, the load suddenly increased from 1kW to 4kW, and the system temperature rose from 80℃ to 140℃ within 10 seconds, with the hydrogen release rate increasing to 1.7L / min simultaneously. At 12:10, the load was reduced to 2kW, the temperature was adjusted to 100℃ within 30 seconds, and the hydrogen release rate stabilized at 0.8L / min. Throughout the process, the fuel cell output voltage fluctuated by ≤5%, and no overvoltage or undervoltage phenomena occurred.

[0055] The timeline diagram for the switching between hydrogen storage and release modes is as follows: Figure 5 As shown.

[0056] (5) Long-term operation test Continuous 7-day hydrogen storage and release cycle (8 hours of hydrogen storage and 10 hours of hydrogen release per day): The hydrogen storage capacity retention rate is 95%, with no significant decay; Temperature control accuracy is consistently maintained within ±1℃; The average matching degree with the grid load is 89%, which is 43% higher than that of traditional devices (62%).

[0057] Example 6 Based on Example 4, and comparing the performance with traditional hydrogen storage technologies, three typical hydrogen storage technologies were selected for comparison with the device of the present invention. The results are shown in the table below:

[0058] As can be seen from the table, the device of the present invention has significant comprehensive advantages in key indicators such as hydrogen storage capacity, response time, and system energy efficiency, and is especially suitable for the dynamic energy storage needs of source-grid-hydrogen storage systems.

[0059] Example 7 Based on Example 2, the training process of the machine learning inference engine includes: (1) Dataset construction: Collect more than 1,000 sets of composition parameters (MXene interlayer spacing, alloy ratio, doping amount), operating parameters (temperature, pressure) and performance data (hydrogen storage capacity, hydrogen release rate) of hydrogen storage materials, of which 80% is used as training set and 20% is used as test set; (2) Feature engineering: Quantitative features such as interlayer electrostatic energy and hydrogen adsorption energy are obtained by density functional theory (DFT) calculation and fused with experimental parameters to form a 128-dimensional input feature matrix; (3) Model training: The LSTM model (3 hidden layers, 64 nodes per layer) was trained using the Adam optimizer (learning rate 0.001), and the hyperparameters were optimized by 5-fold cross-validation until the validation set error was ≤5%; (4) Reinforcement learning optimization: The reward function is “maximizing the matching degree between hydrogen release rate and load + minimizing energy consumption”. The training is completed after 10,000 iterations and the decision delay is ≤100ms.

[0060] The LSTM prediction model is compared with traditional time series prediction models (ARIMA, SVM), and the results are as follows:

[0061] (2) The load matching degree of the LSTM+PID control strategy of the present invention, manual control, and traditional PID control strategies is compared as follows:

[0062] The present invention adopts an LSTM+PID control strategy, which significantly improves the load matching degree, especially in dynamic load scenarios, by 43 percentage points.

[0063] The prediction performance of the LSTM model of this invention is compared with that of the traditional sequence prediction algorithm ARIMA. The LSTM prediction model can achieve proactive prediction of hydrogen release trends, such as... Figure 6 As shown.

[0064] Example 8 Based on Example 6, the long-term stability test of the invented composite hydrogen storage material was conducted as follows: (1) Cyclic performance: After 500 hydrogen storage and release cycles (hydrogen absorption at 25℃ / hydrogen release at 150℃), the hydrogen storage capacity decreased from 6.2wt% to 5.8wt%, with a retention rate of 93.5%. XRD analysis showed that the diffraction peak intensity of the magnesium-based alloy decreased slightly after cycling, but no new phase was formed, indicating good structural stability.

[0065] (2) Anti-poisoning properties: Trace impurities (CO 50ppm, H2S 10ppm) in hydrogen produced by a simulated electrolyzer were continuously stored and released 20 times; the hydrogen storage capacity decreased to 5.9wt%, a reduction of only 4.8%, indicating that the material has a certain tolerance to impurities.

[0066] (3) Mechanical properties: Three-point bending tests showed that the composite material had a bending strength of 120 MPa, which was superior to that of pure magnesium-based alloy (80 MPa). After 100 thermal cycles (-20℃ to 200℃), no cracks were generated in the material, meeting the long-term operation requirements of the device.

[0067] The application scenarios of the device of the present invention include: (1) Wind power / photovoltaic consumption: During peak wind power / photovoltaic output periods (such as noon and night), hydrogen converted from surplus electricity is stored and released to generate electricity to supplement the power grid during off-peak periods; (2) Grid peak shaving: Participate in grid peak-valley arbitrage, store hydrogen during periods of low electricity prices, and release hydrogen to generate electricity during peak periods to reduce electricity costs; (3) Emergency power supply: As an emergency power supply for the microgrid, it can quickly start the hydrogen release mode to supply power to critical loads (such as hospitals and communication base stations) when the grid fails, with a response time of ≤5min; (4) Hydrogen transportation: In conjunction with hydrogen refueling stations, high-pressure hydrogen is provided to fuel cell vehicles (pressurized to 35MPa by a compressor), and the hydrogen storage device serves as a hydrogen energy buffer hub.

[0068] Example 8 Based on Example 6, the system has multiple security protection mechanisms: Leak detection: The hydrogen concentration sensor monitors in real time. When the concentration is ≥1% vol, all valves are immediately closed and the fan is started for ventilation. Overpressure protection: If the pressure sensor detects a pressure ≥3.5MPa, it will automatically open the safety valve to release pressure and simultaneously cut off the input to the electrolytic cell; Fault alarm: When any sensor data is abnormal, the system will notify the operation and maintenance personnel through audible and visual alarms and remote communication. The fault information is stored in the local database for easy post-event analysis.

Claims

1. A dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system, characterized in that, The device includes a hydrogen storage tank (1), a composite hydrogen storage material layer (2) is provided inside the hydrogen storage tank (1), a dual-temperature zone temperature control module (3) is provided on the outer wall of the hydrogen storage tank (1), a multi-parameter sensing unit (4) is embedded in the composite hydrogen storage material layer (2), and a control system (5) is also included. The control system (5) is electrically connected to the dual-temperature zone temperature control module (3) and the multi-parameter sensing unit (4) respectively.

2. The dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system according to claim 1, characterized in that, The composite hydrogen storage material layer is made of a composite hydrogen storage material, which is prepared according to the following method: Step 1, prepare MXene powder; Step 2: Using the MXene powder prepared in Step 1 as a substrate and transition metal chloride as a precursor, single-atom doping with transition metal is performed to obtain doped MXene. Step 3: Prepare magnesium-based quaternary alloy powder; Step 4: After mixing the doped MXene, magnesium-based alloy powder and carbon nanotubes at a certain mass ratio, ethanol is added and then ultrasonically dispersed, freeze-dried and plasma sintered in sequence to obtain the composite hydrogen storage material.

3. The dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system according to claim 2, characterized in that, Step 1 specifically involves mixing Ti3AlC2 powder with 40% hydrofluoric acid at a molar ratio of 1:5 to 1:8, hydrothermally reacting at 180-220℃ for 12-24 hours, centrifuging and washing until pH=6-7, and vacuum drying at 50-70℃ for 10-14 hours to obtain MXene powder. Step 2 specifically involves: using atomic layer deposition technology, with the MXene powder prepared in step 1 as the substrate and transition metal chloride as the precursor, depositing at 150-200℃ for 10-30 cycles, controlling the single atom loading to 0.5-2.0 at%, to obtain the doped MXene. The transition metal TM is at least one of Fe, Co, Ni, and Cu, doped in single-atom form with a loading of 0.5-2.0 at%; Step 3 specifically involves: using Mg, Al, Zn, and TM metal particles, arc melting under argon protection with a current of 200-300A, followed by ball milling to a particle size of 5-10μm after casting, to obtain magnesium-based quaternary alloy powder; the general chemical formula of the magnesium-based quaternary alloy phase is Mg. a Al b Zn c TM d Where a+b+c+d=1, 0.6≤a≤0.8, 0.05≤b≤0.15, 0.05≤c≤0.15, 0.05≤d≤0.1; Step 4 specifically involves: mixing doped MXene, magnesium-based alloy powder and carbon nanotubes at a mass ratio of (4-6):(3-5):1, adding ethanol and ultrasonically dispersing at 400-600W power for 2-4 hours, freeze-drying and then sintering by discharge plasma at 350-450℃ and 30-50MPa for 5-10 minutes to obtain a composite hydrogen storage material. The carbon nanotubes have a diameter of 5-20 nm, a length of 1-5 μm, and a mass percentage of 3-5%.

4. The dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system according to any one of claims 1-3, characterized in that, The inner wall of the hydrogen storage tank (1) is coated with a 0.1-0.2mm thick polytetrafluoroethylene anti-corrosion layer. The hydrogen storage tank (1) is provided with a honeycomb flow guiding structure inside, with a flow guiding hole diameter of 8-12mm and a hole spacing of 20-30mm. The hydrogen storage tank (1) is provided with a hydrogen inlet and an outlet at both ends. A hydrogen inlet solenoid valve (6) and a hydrogen outlet solenoid valve (7) are provided at the hydrogen inlet and outlet respectively. A flow regulating valve (8) is also connected in series at the hydrogen outlet. The flow regulating valve (8) has an adjustment range of 0.1-2L / min and an adjustment accuracy of ±0.01L / min. The hydrogen inlet solenoid valve (6), the hydrogen outlet solenoid valve (7), and the flow regulating valve (8) are all electrically connected to the control system (5).

5. The dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system according to claim 4, characterized in that, The dual-temperature zone temperature control module (3) includes a low-temperature control unit (31) and a high-temperature control unit (32). The low-temperature control unit (31) and the high-temperature control unit (32) are arranged alternately from top to bottom on the outer wall of the hydrogen storage tank (1). The low-temperature control unit (31) and the high-temperature control unit (32) are attached to the outer wall of the hydrogen storage tank (1) by insulating and thermally conductive silicone, and the thermal conductivity of the bonding surface is ≥2.0W / (m). K), the low temperature control unit is a semiconductor refrigeration chip with a power of 300-800W, a temperature control range of -20℃ to 50℃, and PID control with an accuracy of ±1℃; the high temperature control unit is a graphene electric heating chip with a power of 500-1500W, a temperature control range of 50℃ to 200℃, and a heating rate of 5-10℃ / min. The low temperature control unit and the high temperature control unit are electrically connected to the control system (5) respectively.

6. The dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system according to claim 5, characterized in that, The multi-parameter sensing unit (4) includes a distributed optical fiber temperature sensor (41). The distributed optical fiber temperature sensor (41) has 5 monitoring points uniformly embedded in the composite hydrogen storage material layer (2) along the axis of the hydrogen storage tank (1). The measurement range of the distributed optical fiber temperature sensor (41) is -50℃ to 300℃, the resolution is 0.1℃, and the response time is ≤1s. Miniature pressure sensor (42), the miniature pressure sensor (42) is installed at the top center inside the hydrogen storage tank (1), with a measurement range of 0-5MPa, an accuracy of ±0.02MPa, and a sampling frequency of 10Hz; A hydrogen concentration sensor (43) is installed at the hydrogen outlet, with a measurement range of 0-100%vol and a resolution of 0.1%vol. The electrochemical workstation integrated module (44) includes a three-electrode system, with the working electrode being a composite hydrogen storage material, the reference electrode being a saturated calomel electrode, and the counter electrode being a platinum sheet. The test frequency range is 10. -2 -10 5 Hz; The distributed optical fiber temperature sensor (41), the miniature pressure sensor (42), the hydrogen concentration sensor (43), the electrochemical workstation integrated module (44), and the control system (5) are electrically connected.

7. The dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system according to claim 6, characterized in that, The control system (5) includes an STM32H743 microprocessor (51), a machine learning inference engine (52), a data storage module (53), a power drive unit (54), a PID control module (56), and a source-network coordination interface (55). The STM32H743 microprocessor (51) is electrically connected to the machine learning inference engine (52), the data storage module (53), the PID control module (56), the power drive unit (54), and the source-network coordination interface (55), respectively. The PID control module (56) is electrically connected to the power drive unit (54). The STM32H743 microprocessor (51) is also electrically connected to the multi-parameter sensing unit (4); The PID adjustment module (56) is electrically connected to the dual-temperature zone temperature control module (3) and is used to output a PWM signal to adjust its power; The STM32H743 microprocessor (51) has a main frequency of ≥400MHz, the machine learning inference engine (52) has a built-in trained LSTM prediction model and reinforcement learning optimization model; the data storage module (53) has a capacity of ≥32GB, and the power drive unit (54) has an output voltage of 0-24V and a current of 0-10A. The source-network coordination interface (55) includes a DC / DC converter (551) and a communication protocol conversion unit (552). The DC / DC converter (551) is electrically connected to the communication protocol conversion unit (552), and the communication protocol conversion unit (552) is electrically connected to the STM32H743 microprocessor (51). The DC / DC converter (551) has an input voltage range of 24-72V and an output power of 0-1000W. The communication protocol conversion unit (552) supports Modbus-RTU and OPC UA protocols and connects to the source-network hydrogen storage system scheduling platform via 4G / 5G wireless communication or Ethernet. The data transmission rate is ≥1Mbps and the response time is ≤1s. The source-network coordination interface (55) also includes a power quality monitoring unit, which monitors the input voltage fluctuation ≤±10%, harmonic distortion rate ≤5%, and frequency deviation ≤±0.5Hz in real time. When the threshold is exceeded, a protection mechanism is activated to cut off the external power supply and switch to the backup power supply.

8. The dual-temperature zone solid-state hydrogen storage device for a source-grid hydrogen storage system according to claim 7, characterized in that, The hydrogen storage tank (1) is also wrapped with a composite insulation layer, which includes an aluminum foil reflective layer, an aluminum silicate fiber layer, and a stainless steel protective layer arranged sequentially from the inside to the outside.

9. A dual-temperature zone solid-state hydrogen storage method for a source-grid hydrogen storage system, characterized in that, The dual-temperature zone solid-state hydrogen storage device of the source-grid hydrogen storage system described in claim 8 includes a standby mode, a hydrogen storage mode, a hydrogen release mode, a maintenance mode, and a dynamic optimization process.

10. The dual-temperature zone solid-state hydrogen storage method for a source-grid hydrogen storage system according to claim 9, characterized in that, The standby mode is as follows: the system is under low load and does not need to store or release hydrogen. At this time, the control system (5) controls the dual-temperature zone temperature control module (3) to maintain the internal temperature of the hydrogen storage tank (1) at 25±5℃, and the power consumption is ≤50W; the hydrogen input solenoid valve (6) and the hydrogen output solenoid valve (7) are both in the closed state; the flow regulating valve (8) is in the minimum opening or closed state; the distributed fiber optic temperature sensor (41), the micro pressure sensor (42), the hydrogen concentration sensor (43) and the electrochemical workstation integrated module (44) in the multi-parameter sensing unit (4) continuously collect data and transmit it to the STM32H743 microprocessor (51); the STM32H743 microprocessor (51) triggers a self-test program every 30 minutes and performs diagnostic analysis on key parameters through the machine learning inference engine (52); The hydrogen storage mode is as follows: When the power grid system is in a low-power period, the control system (5) receives the scheduling signal sent by the power grid hydrogen storage system scheduling platform through the power grid collaborative interface (54) and starts the hydrogen storage program. The STM32H743 microprocessor (51) controls the power drive unit (54) to start the low-temperature control unit (31) of the dual-temperature zone temperature control module (3) to maintain the temperature at 20-30℃. At the same time, it controls the opening of the hydrogen input solenoid valve (6) and the hydrogen enters the hydrogen storage tank (1) through the hydrogen input port. The multi-parameter sensing unit (4) monitors the pressure in real time. When the micro pressure sensor (42) detects that the pressure reaches 2-2.5MPa, the STM32H743 microprocessor (51) controls the closing of the hydrogen input solenoid valve (6). The hydrogen storage rate is controlled at 0.5-1L / min by the flow regulating valve (8). The hydrogen release mode is as follows: when the power grid system is in peak electricity consumption, the control system (5) receives the hydrogen release command and load demand issued by the power grid hydrogen storage system scheduling platform through the power grid collaborative interface (54); the STM32H743 microprocessor (51) controls the power drive unit (54) to switch the temperature zone according to the load size: if the load is ≤2kW, the low temperature control unit (31) is activated to maintain the temperature at 50-100℃, and the hydrogen output solenoid valve (7) and flow regulating valve (8) are controlled to make the hydrogen release rate at 0.1-0.5L / min; if the load is >2kW, the high temperature control unit (32) is activated to maintain the temperature at 100-200℃, and the hydrogen output solenoid valve (7) and flow regulating valve (8) are controlled to make the hydrogen release rate at 1-2L / min; The dynamic optimization process is as follows: the machine learning inference engine (52) is periodically called; the STM32H743 microprocessor (51) acquires the temperature, pressure, and flow data collected by the multi-parameter sensing unit (4) every 5 minutes and inputs it into the LSTM model built into the machine learning inference engine (52) to predict the hydrogen release rate in the next 10-30 minutes; the STM32H743 microprocessor (51) dynamically adjusts the PWM signal output by the power drive unit (54) to the dual-temperature zone temperature control module (3) through the PID adjustment module (56) according to the prediction results, so as to optimize the temperature control power and make the hydrogen release rate match the load change in advance; The maintenance mode is as follows: when the system has accumulated 100 hours of operation, the STM32H743 microprocessor (51) controls the start of the electrochemical workstation integrated module (44) to perform electrochemical impedance spectroscopy detection and transmits the data to the machine learning inference engine (52) for analysis; if the impedance increase is greater than 20%, the STM32H743 microprocessor (51) controls the power drive unit (54) to drive the dual-temperature zone temperature control module (3) to execute the in-situ activation program and perform temperature cycling; The in-situ activation procedure specifically includes:

1. Deep cooling stage: start the low temperature control unit 31 to lower the overall temperature of the composite hydrogen storage material layer (2) to -20℃ at a rate of not less than 5℃ / min, and keep it at this temperature for 10 minutes; 2. Warm activation stage: switch to the high temperature control unit (32) to raise the temperature to 100℃ at a rapid heating rate of 8-10℃ / min, and keep it at this temperature for 15 minutes; 3. Steady-state recovery stage: stop heating, start the low temperature control unit (31) to assist in temperature control, so that the temperature drops back to room temperature steadily, keep it at this temperature for 5 minutes, and complete one temperature cycle.

Images