Thermodynamic inversion-based thermal management method for solid-state hydrogen storage system and system thereof

Abstract

The application discloses a thermodynamic inversion-based solid-state hydrogen storage system thermal management method and system, and relates to the technical field of solid-state hydrogen storage systems. The method comprises the following steps: using all partitioned temperature sensors to collect temperature distribution data of the outer wall of a hydrogen storage tank in real time, reconstructing the temperature distribution data into a continuous virtual temperature field cloud map based on a spatial interpolation algorithm, receiving a target hydrogen flow instruction, simultaneously obtaining current internal state parameters of the hydrogen storage tank, calling a preset thermodynamic-dynamic coupling model, inversely solving an internal target temperature field, and optimizing an optimal target temperature field; performing spatial matching comparison between the optimal target temperature field and the virtual temperature field cloud map, generating temperature deviation signals of each heating module, based on a multi-loop decoupling PID controller, dynamically and independently adjusting input power of the heating module of each partition according to each temperature deviation signal and a thermal coupling interference matrix of adjacent partitions, and synchronously and jointly controlling the opening degree of an air inlet valve and an air outlet valve.

Description

Technical Field

[0001] This application relates to the field of solid-state hydrogen storage system technology, and more specifically, to a thermal management method and system for solid-state hydrogen storage systems based on thermodynamic inversion. Background Technology

[0002] Solid-state hydrogen storage technology is a technique that stores hydrogen in solid materials through physical or chemical processes. It represents a core breakthrough in hydrogen energy storage systems, overcoming the bottlenecks of large-scale, long-term hydrogen storage and transportation with high safety and high density. Compared to high-pressure gaseous and cryogenic liquid hydrogen storage, solid-state hydrogen storage achieves safe, high-density hydrogen storage at room temperature and pressure through material adsorption, fundamentally avoiding high-pressure risks. This provides the physical basis for building long-term, stationary energy storage power stations ranging from hundreds of kilowatt-hours to megawatt-hours across seasons. Furthermore, its high safety and modularity allow for seamless application in various scenarios, including vehicle power, distributed energy, and industrial hydrogen supply, making it a hub connecting green electricity production and end-use applications. This technology not only solves the hydrogen storage problem but also serves as a core engine driving the development of the entire industry chain, including high-performance materials and system integration, possessing irreplaceable value in building a new energy system.

[0003] However, existing solid-state hydrogen storage systems suffer from significant drawbacks in thermal management technology. Their temperature control accuracy is insufficient (typically exceeding ±10℃), leading to uneven temperature distribution in the hydrogen storage material. This crude thermal management approach, on the one hand, easily causes unstable hydrogen absorption and desorption reaction kinetics, resulting in large fluctuations in hydrogen release rate and pressure, making it difficult to meet the downstream equipment's demand for a stable hydrogen source. On the other hand, repeated thermal shocks and localized overheating accelerate the pulverization, phase separation, and activity decay of the hydrogen storage material, severely impairing its cycle life and capacity retention, significantly increasing the system's total lifecycle cost, and becoming a key bottleneck restricting the large-scale application of this technology. To address the aforementioned crude management issues of solid-state hydrogen storage systems, most solutions employ a "whole-heated solid-state hydrogen storage tank with heating wires wound around the outer wall + single-point temperature control" approach. This involves uniformly winding resistance heating wires around the outer wall of the tank to heat the entire tank, relying on a single temperature sensor (usually located at the center or outer wall of the tank) for feedback control. However, this scheme has several drawbacks: First, the temperature field is uneven. The heating wire conducts heat from the outer wall to the inside, inevitably creating a temperature gradient from the outside to the inside of the tank, resulting in uneven heating of the hydrogen storage material. Second, the temperature control accuracy is low. Relying on a single-point temperature sensor cannot monitor and reflect the complex and dynamic true temperature distribution of the entire hydrogen storage material, resulting in coarse temperature control. Third, hydrogen release is unstable. Temperature differences in the hydrogen storage material lead to asynchronous reactions, resulting in large fluctuations in hydrogen release rate and pressure, unstable hydrogen output, and consequently, reduced material lifespan. Local overheating (near the heating wall) and thermal cycling shocks can accelerate material pulverization, phase separation, and irreversible structural damage, reducing cycle life and hydrogen storage capacity.

[0004] Therefore, there is an urgent need for a thermal management method for solid-state hydrogen storage systems that can achieve efficient, stable, and controllable hydrogen release, in order to solve the problem that existing thermal management methods cannot meet the requirements of solid-state hydrogen storage materials for high-precision, uniform, and dynamic temperature control. Summary of the Invention

[0005] This application provides a thermal management method and system for solid-state hydrogen storage systems based on thermodynamic inversion, which overcomes the local thermal stress damage to materials caused by uneven temperature field in traditional solid-state hydrogen storage systems, and achieves coordinated control of temperature, pressure and flow rate to meet the transient response requirements of fuel cells.

[0006] The specific technical solution is as follows: In a first aspect, embodiments of this application provide a thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion. The solid-state hydrogen storage system has independently controlled heating modules arranged in sections on the outer wall of the hydrogen storage tank, and each heating module integrates a temperature sensor. The thermal management method includes: Using all the temperature sensors deployed in the zones, the temperature distribution data of the outer wall of the hydrogen storage tank is collected in real time, and reconstructed into a continuous virtual temperature field cloud map based on a spatial interpolation algorithm. The system receives the target hydrogen flow rate command, obtains the current internal state parameters of the hydrogen storage tank, calls the preset thermodynamic-kinetic coupling model, reversely calculates the internal target temperature field, and optimizes to obtain the optimal target temperature field. The optimal target temperature field is spatially matched and compared with the virtual temperature field cloud map to generate temperature deviation signals for each heating module. Based on a multi-loop decoupled PID controller, the input power of the heating module in each zone is dynamically and independently adjusted according to the temperature deviation signals and the thermal coupling interference matrix of adjacent zones, and the opening degree of the inlet valve and outlet valve is synchronously controlled.

[0007] This thermal management method uses zone-integrated sensors to collect surface data at the sensing layer, and reconstructs it into a virtual temperature field cloud map through spatial interpolation. At the decision layer, it receives flow commands, combines them with internal states, calls a thermodynamic-dynamic coupling model to inversely solve and optimize the optimal target temperature field, and compares the optimal target temperature field with the virtual temperature field cloud map at the execution layer. It uses a multi-loop decoupled PID controller to adjust the zone heating power and control the intake and exhaust valves in a coordinated manner.

[0008] This application, without damaging the structure of the hydrogen storage material, achieves visualized monitoring of the internal three-dimensional thermal state through temperature field reconstruction, solving the industry problem of unmeasurable internal temperature in solid-state hydrogen storage systems. Compared with existing technologies, this application transforms passive response into active feedforward, directly converting flow commands into temperature field setpoints and using thermodynamic models to predict heat demand, completely overcoming the response lag caused by thermal inertia in traditional temperature control and significantly improving the dynamic response speed of the hydrogen supply system. Simultaneously, this application achieves multi-variable coordinated stable control through decoupled control and valve linkage, not only eliminating thermal interference between zones but also realizing three-dimensional coordination of temperature, pressure, and flow, ensuring the stability and safety of hydrogen output under sudden load changes.

[0009] In some embodiments of this application, the step of reconstructing a continuous virtual temperature field cloud map based on a spatial interpolation algorithm specifically includes: Real-time acquisition of total heating power input data of the solid hydrogen storage system and internal pressure data of the hydrogen storage tank; A thermo-chemical coupled physical model of the hydrogen storage material based on the hydrogen storage tank is established, and the temperature distribution data, the total heating power input data, and the internal pressure data are mapped to the boundary adjustment and driving input of the thermo-chemical coupled physical model. Based on the aforementioned thermo-chemical coupled physical model, an iterative calculation process is performed, including: using the current state estimate to perform forward heat conduction deduction to obtain the predicted surface temperature distribution; calculating the residual between the predicted surface temperature distribution and the temperature distribution data; based on the residual, correcting the heat source term distribution and material thermal property parameters within the thermo-chemical coupled physical model using an optimization algorithm to minimize the residual, thereby solving the inverse heat conduction problem; and integrating the corrected heat source term distribution and material thermal property parameters into the model state variables to complete the data assimilation at the current moment and update the state estimate within the hydrogen storage material. Based on the updated state estimates, the three-dimensional temperature field distribution, chemical reaction front coordinates, and heat accumulation index inside the hydrogen storage material are obtained by analysis. Based on the aforementioned three-dimensional temperature field distribution, a virtual temperature field cloud map of the hydrogen storage material is rendered and generated.

[0010] This thermal management method introduces temperature distribution data, total heating power input data, and internal pressure data as boundary conditions to establish a thermo-chemical coupled physical model. Through an iterative process of "forward deduction – residual calculation – reverse correction", it solves the inverse heat conduction problem, completes data assimilation, and finally outputs the internal three-dimensional temperature field, reaction front location, and heat accumulation index.

[0011] Unlike simple linear interpolation, this application considers the dynamic changes of the chemical reaction heat source term and material properties. Through data assimilation techniques, it continuously corrects model errors, making the reconstructed internal temperature field extremely close to the actual physical state. It not only obtains the temperature distribution but also locates the chemical reaction front and heat accumulation zone, providing crucial decision-making basis for preventing localized overheating dead zones and optimizing reaction pathways—something traditional surface thermometry cannot achieve. Furthermore, this application can dynamically adapt to the nonlinear characteristics of the material's thermal conductivity changing with the reaction progress, ensuring the reliability of the reconstruction results throughout its entire lifecycle and adapting to complex operating conditions.

[0012] In some embodiments of this application, the step of receiving the target hydrogen flow rate command, simultaneously acquiring the current internal state parameters of the hydrogen storage tank, and calling a preset thermodynamic-kinetic coupled model to inversely calculate the internal target temperature field data and optimize to obtain the optimal target temperature field specifically includes: The system receives the target hydrogen flow rate command, converts the target hydrogen flow rate command into a target hydrogen molar flow rate per unit time, and simultaneously acquires the current internal state parameters of the hydrogen storage tank; wherein, the current internal state parameters include the current average pressure inside the tank and the real-time physical property state of the hydrogen storage material in the hydrogen storage tank. The preset thermodynamic-kinetic coupled model is invoked, and the target hydrogen molar flow rate, the current average pressure inside the tank, and the real-time physical properties of the hydrogen storage material are used as input variables. By solving the correlation equation between the hydrogen release reaction rate and temperature and pressure, the internal target temperature field of the hydrogen storage material required to meet the target hydrogen molar flow rate is calculated in reverse. The internal target temperature field includes the theoretical optimal average temperature of the hydrogen storage material and the spatial temperature distribution reference values ​​of each zone. A heat transfer mapping relationship between each partition and the internal temperature field is constructed. Taking the internal target temperature field as the constraint target, the heating parameters of each partition are iteratively solved using an optimization algorithm to calculate the optimal target surface temperature setting value corresponding to each partition, so as to obtain the optimal target temperature field.

[0013] This thermal management method converts flow rate into molar flow rate, combines the characteristics of the PCT (Pressure–Composition–Temperature) curve with the reaction rate equation to solve the theoretical temperature field in reverse, and further constructs a heat transfer mapping relationship. Through optimization algorithms, it calculates the optimal target surface temperature setpoint considering heat transfer hysteresis.

[0014] This application actively compensates for temperature decay and heat transfer hysteresis from the outside to the inside of the non-uniform target temperature field generated by an optimized algorithm. This ensures that the reaction rate is consistent between the center and edge of the hydrogen storage bed, avoiding the phenomenon of overheating of the outer shell and insufficient reaction in the center, thus improving uniformity. It accurately maps abstract flow requirements to specific temperature distributions, achieving deep coupling between the thermodynamic model and flow control. This allows the system to meet the maximum flow requirements with minimal energy consumption, precisely matching the kinetic demands. Furthermore, this application eliminates theoretical deviations caused by model simplification by iteratively solving for the optimal surface setpoint, ensuring the accuracy of the actual control target and eliminating steady-state errors.

[0015] In some embodiments of this application, the method of using a multi-loop decoupled PID controller to dynamically and independently adjust the input power of the heating module in each zone based on the temperature deviation signals and the thermal coupling interference matrix of adjacent zones, and synchronously control the opening of the outlet valve, specifically includes: Each of the temperature deviation signals is input to the multi-loop decoupled PID controller. The multi-loop decoupled PID controller calculates the thermal interference components between adjacent partitions based on a preset thermal coupling model and the thermal coupling interference matrix of the adjacent partitions. It then decouples and corrects the initial PID control quantities of each partition according to the thermal interference components, generating a decoupled temperature control command. Simultaneously, it generates a linkage opening command for the inlet valve and the outlet valve based on the current pressure setpoint and flow setpoint, thereby achieving multi-variable coordination of temperature, pressure, and flow. The decoupled temperature control command is converted into a power drive signal for the heating module of each zone, the input power applied to the heating module of each zone is dynamically adjusted, and the linkage opening command is sent to the actuator to adjust the opening of the air inlet valve and the air outlet valve. After executing the electric drive signal and the linkage opening command, the optimal target temperature field is reacquired and spatially matched with the virtual temperature field cloud map to generate new temperature deviation signals for each heating module, forming a closed-loop control loop.

[0016] This thermal management method uses a pre-set thermal coupling model and thermal coupling interference matrix to calculate the thermal interference components of adjacent zones, decouples and corrects the PID control quantity, and generates valve linkage commands based on pressure and flow setpoints to form a closed-loop feedback.

[0017] This application effectively counteracts the thermal penetration effect between adjacent heating zones by introducing a thermal coupling interference matrix, preventing temperature overshoot or oscillation in neighboring areas caused by heating in one zone, achieving truly independent and precise temperature control, and completely eliminating thermal crosstalk. Simultaneously, this application employs a multi-dimensional synergistic approach of temperature-pressure-flow, combining heating power regulation with valve opening control to form a synergistic mechanism of "temperature master control response rate and valve auxiliary control back pressure stability," greatly improving the system's robustness under transient load changes. Furthermore, the decoupling correction makes the control loops of each zone independent, accelerating the system's convergence speed to temperature deviations, achieving rapid convergence, and reducing settling time.

[0018] In some embodiments of this application, the dynamic adjustment of the input power applied to the heating modules of each zone specifically includes: When the temperature of a certain zone in the virtual temperature field cloud map exceeds a preset temperature threshold, the input power of the heating module in that zone is reduced first, while the input power of the heating module in the adjacent zone is increased.

[0019] This application makes judgments based on the reconstructed internal temperature field rather than the surface temperature, enabling earlier detection of potential internal hotspots and preventing hydrogen storage materials from sintering, pulverizing, or separating due to localized high temperatures, thus significantly extending the material's cycle life. Furthermore, while suppressing overheating, it maintains the total heat required for the overall reaction, avoiding hydrogen supply interruptions caused by simply cutting off heating, thus balancing safety and continuity.

[0020] In some embodiments of this application, the thermal management method further includes: The multi-loop decoupled PID controller dynamically adjusts the input power of the heating module in the adjacent zone to the outlet valve based on the opening change rate of the outlet valve.

[0021] Based on the Joule-Thomson effect and convective heat transfer, this application provides feedforward compensation to address the problem of a sudden temperature drop near the outlet caused by the rapid outflow of hydrogen carrying away a large amount of heat. This ensures the stability of the reaction front temperature and offsets convective heat loss. Simultaneously, this application improves transient response capability by increasing heating power in advance at the moment of a sudden increase in flow rate, compensating for the delay due to thermal inertia and ensuring continuous and stable hydrogen output under high flow rate conditions. This is particularly suitable for acceleration scenarios in automotive fuel cells.

[0022] In some embodiments of this application, the multi-loop decoupled PID controller employs a model predictive control (MPC) algorithm, using the power upper limit, heating rate threshold, and valve action range of the heating modules in each zone as constraints, and minimizing the error between the optimal target temperature field and the virtual temperature field cloud map as the objective function to continuously optimize the output control quantity; the output control quantity includes the electric drive signal and the linkage opening command.

[0023] Compared to the localized adjustment of traditional PID control, the MPC algorithm can find the globally optimal control sequence over a future period while satisfying all hardware constraints, significantly improving energy efficiency. Furthermore, this MPC algorithm is better suited for handling the complex coupling relationships between temperature, pressure, flow rate, and multi-zone heating in solid-state hydrogen storage systems, avoiding the difficulties of multi-loop PID tuning. Simultaneously, by incorporating hardware physical limits as constraints into the optimization process, it fundamentally eliminates the risk of actuators operating beyond their limits, ensuring safe operation under constraints.

[0024] In some embodiments of this application, the thermal management method further includes: The maximum temperature difference and local overheated areas in the virtual temperature field cloud map are monitored in real time. When the reconstruction temperature of a certain partition exceeds the material safety threshold, the input power of the heating module in that partition is forcibly reduced.

[0025] This application utilizes temperature field reconstruction technology to detect internal abnormal temperature rises that surface sensors may not yet reflect, providing a deeper level of safety protection than traditional surface monitoring. Furthermore, by strictly controlling the maximum temperature difference, it reduces thermal stress caused by uneven thermal expansion of the hydrogen storage tank and internal materials, lowering the risk of container fatigue failure and preventing thermal stress damage. Simultaneously, this application establishes an intelligent heat treatment mechanism based on internal conditions, enhancing the intrinsic safety level of the system under extreme operating conditions.

[0026] In some embodiments of this application, the thermal management method further includes: Periodically compare the theoretical flow rate derived from the thermodynamic-kinetic coupling model with the actual flow meter reading, calculate the residual, and if the residual exceeds a preset residual threshold, then correct the kinetic parameters in the thermodynamic-kinetic coupling model online.

[0027] Since the performance of solid hydrogen storage materials degrades after multiple hydrogen absorption and desorption cycles, this application periodically compares the theoretical flow rate derived from the model with the actual flow meter readings. If the residual exceeds the standard, the kinetic parameters in the model are corrected online. It has full life cycle adaptability, can adapt to material aging, automatically update model parameters, and maintain long-term control accuracy. It solves the problem of the accuracy of fixed parameter models decreasing over time, ensuring the effectiveness of the flow rate back-calculation strategy throughout the entire system life. This allows the system to achieve automatic optimization without frequent manual calibration, reducing the difficulty and cost of operation and maintenance.

[0028] Secondly, embodiments of this application provide a thermal management system for implementing the thermodynamic inversion-based thermal management method for solid-state hydrogen storage systems described in the first aspect, the thermal management system comprising: The real-time temperature sensing and reconstruction module is used to collect the temperature distribution data of the outer wall of the hydrogen storage tank in real time using all the temperature sensors deployed in the partition, and reconstruct it into a continuous virtual temperature field cloud map based on the spatial interpolation algorithm. The thermodynamic model inversion module is used to receive the target hydrogen flow command, obtain the current internal state parameters of the hydrogen storage tank, call the preset thermodynamic-dynamic coupling model, reverse calculate the internal target temperature field, and optimize to obtain the optimal target temperature field. The multi-zone closed-loop collaborative control module is used to spatially match and compare the optimal target temperature field with the virtual temperature field cloud map to generate temperature deviation signals for each heating module. Based on the multi-loop decoupled PID controller, according to each temperature deviation signal and the thermal coupling interference matrix of adjacent zones, the input power of the heating module in each zone is dynamically and independently adjusted, and the opening degree of the intake valve and the exhaust valve are synchronously controlled.

[0029] The innovative aspects of this application's embodiments include, but are not limited to, the following: First, distributed high-precision thermal management: abandoning the traditional overall heating mode of the outer wall, it adopts a multi-zone independent temperature control structure. Through a multi-point temperature sensor network, it realizes real-time monitoring and precise and uniform control of the internal temperature field of the material, fundamentally eliminating local thermal stress and material performance degradation caused by excessive temperature difference.

[0030] Second, a coordinated control strategy: A dynamic coupling model is established between temperature, pressure, and hydrogen release flow rate. The system controller dynamically adjusts the heating power of each zone based on real-time temperature monitoring, achieving closed-loop coordinated control of the three factors to ensure a stable and controllable hydrogen release process.

[0031] Third, rapid response and stable output mechanism: In response to rapid fluctuations in fuel cell load, the above-mentioned collaborative control strategy can prioritize the adjustment of the temperature of the key area most directly related to hydrogen release kinetics, and combine pressure feedback to quickly adjust the output, thereby significantly shortening the system response time and still providing a stable hydrogen source in terms of pressure and flow under transient conditions.

[0032] Fourth, system integration and lifespan improvement: This integrated design not only optimizes transient performance, but also significantly reduces the thermomechanical fatigue of materials during cycling through uniform thermal field management, effectively extending the service life of the core hydrogen storage material and the reliability of the entire system. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 A schematic flowchart illustrating a thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion, provided as an embodiment of this application; Figure 2 This is a block diagram illustrating the components of a thermal management system provided in an embodiment of this application. Detailed Implementation

[0035] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0036] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The terms "comprising" and "having," and any variations thereof, in the embodiments and drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0037] This application discloses a thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion. The method is applied to a solid-state hydrogen storage system where the outer wall of the hydrogen storage tank is divided into sections with independently controlled heating modules, each integrating a temperature sensor. The system also includes a pressure and flow detection device connected to the interior of the storage tank. These details are described below.

[0038] In one specific embodiment, the solid-state hydrogen storage system used in this thermal management method includes a cylindrical solid-state hydrogen storage tank, a heating module, a temperature sensor, a multi-channel PWM power drive circuit, and an electric inlet valve and an electric outlet valve. The metal outer shell of the hydrogen storage tank is divided into N annular regions along the axial direction and M sector regions along the circumference, forming a total of N×M independent heating zones. Each zone is attached with a flexible thin-film heating element (i.e., a heating module), and a high-precision temperature sensor is integrated at the center of the flexible thin-film heating element or at a key thermal node. The independently controlled heating modules are arranged on the outer surface of the metal shell of the hydrogen storage tank, and each zone can operate independently, realizing heating or active cooling of different areas of the tank. Integrating the temperature sensor onto the heating module, rather than the hydrogen storage tank itself, simplifies the replacement and maintenance process, facilitates rapid switching of the hydrogen storage tank, and the sensor network is responsible for real-time acquisition of temperature distribution data on the outer wall of the tank, serving as boundary conditions for inferring the internal state. Furthermore, the integrated design of the temperature sensor and the heating module ensures that the temperature measurement point is the heating point, greatly improving the representativeness of the boundary condition data. In addition, the multi-channel PWM power drive circuit, electric intake valve, and electric exhaust valve serve as the actuators of the solid-state hydrogen storage system. The multi-channel PWM power drive circuit is used to independently adjust the heating power of each zone heating module. Both the electric intake valve and the electric exhaust valve are high-precision valves. The electric exhaust valve is used to control the output of hydrogen, and the electric intake valve is used to control the intake of the tank. It should be noted and understood that during the hydrogen release (supply) process of the solid-state hydrogen storage system, the electric exhaust valve is mainly controlled. However, in some circulation systems, in order to maintain back pressure or perform purging operations, the control of the intake side may also be involved, but the core is the adjustment of the exhaust flow rate.

[0039] This application achieves real-time sensing and dynamic equilibrium control of the internal temperature field of hydrogen storage materials through a distributed temperature sensing network and a multi-zone independent heating array. It fundamentally solves the problem of "hot outer shell and cold center" caused by traditional overall heating methods. It can accurately suppress local hot spots and eliminate reaction cold zones, ensuring that the material works in a more uniform and stable thermal environment, thereby improving the overall reaction efficiency and safety.

[0040] Figure 1 This paper illustrates a thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion, according to an embodiment of this application. For example... Figure 1As shown, the thermal management method includes the following steps: Step S110: Using all the temperature sensors deployed in the partition, collect the temperature distribution data of the outer wall of the hydrogen storage tank in real time, and reconstruct it into a continuous virtual temperature field cloud map based on the spatial interpolation algorithm.

[0041] In step S110, considering the unique application characteristics of solid hydrogen storage materials where sensors cannot be directly embedded inside, this application adopts a strategy of "surface sensing + reverse thermal conduction reconstruction".

[0042] Specifically, after the solid-state hydrogen storage system is started, a distributed temperature sensor array performs high-frequency sampling of the hydrogen storage material to achieve real-time sensing of the entire temperature field. It is important to note that these sensors are deployed at key locations on the outer shell of the solid-state hydrogen storage tank to capture spatial temperature differences. Each temperature sensor transmits the collected real-time temperature signals to the central processing unit (CPU). After data acquisition, a cloud map is generated. The CPU uses spatial interpolation algorithms to reconstruct the discrete point-like temperature data into a continuous, visualized two-dimensional or three-dimensional temperature field cloud map, completing the reverse heat conduction reconstruction. This cloud map intuitively reflects the thermal state of the entire reaction at any given time, serving as the basis for all subsequent decisions. It clearly displays "hot spots," "cold zones," and temperature gradient distribution, providing direct evidence for precise intervention.

[0043] In some embodiments, the specific steps for generating a cloud map are as follows: Step S111, Data Acquisition: Real-time acquisition of total heating power input data of the solid hydrogen storage system and internal pressure data of the hydrogen storage tank.

[0044] Step S112, Model Construction and Initialization: Establish a thermo-chemical coupled physical model of hydrogen storage materials based on hydrogen storage tanks, and map temperature distribution data, total heating power input data, and internal pressure data into boundary adjustment and driving inputs of the thermo-chemical coupled physical model.

[0045] Step S113, Inverse Problem Solving and Data Assimilation: Based on the thermo-chemical coupled physical model, an iterative calculation process is performed, including: using the current state estimate to perform forward heat conduction deduction to obtain the predicted surface temperature distribution; calculating the residual between the predicted surface temperature distribution and the temperature distribution data; based on the residual, correcting the heat source term distribution and material thermal property parameters inside the thermo-chemical coupled physical model through an optimization algorithm to minimize the residual, thereby solving the inverse heat conduction problem; integrating the corrected heat source term distribution and material thermal property parameters into the model state variables to complete the data assimilation at the current moment and update the state estimate inside the hydrogen storage material.

[0046] Step S114, State Inversion and Visualization: Based on the updated state estimates, the three-dimensional temperature field distribution, chemical reaction front coordinates, and heat accumulation index inside the hydrogen storage material are analyzed.

[0047] Step S115, Cloud Map Generation: Based on the three-dimensional temperature field distribution, a virtual temperature field cloud map inside the hydrogen storage material is rendered and generated.

[0048] In step S113 above, the "state estimate" refers to a set of vectors characterizing the internal physical state of the solid-state hydrogen storage system, including at least: temperature distribution data of discrete points within the hydrogen storage material and local reaction conversion rate distribution data; optionally, it also includes equivalent thermophysical parameters that change with the state. "Correcting the distribution of heat source terms within the thermo-chemical coupling physical model through optimization algorithms" specifically includes: using the adjoint method or the Levenberg-Marquardt algorithm to calculate the gradient of the objective function with respect to the internal heat source distribution, and updating the heat source distribution along the negative gradient direction until the objective function converges. The objective function is defined as the mean square error between the predicted surface temperature and the actual surface temperature. "Data assimilation" specifically employs the ensemble Kalman filter (EnKF) or four-dimensional variational data assimilation (4D-Var) method. In detail, the Kalman gain matrix is ​​calculated using the state covariance matrix of the previous time step and the observation error covariance matrix of the current time step. The model-predicted state and the measured data are then weighted and fused to update the state estimate and its uncertainty range at the current time step.

[0049] The thermo-chemical coupling physical model in the embodiments of this application considers the latent heat of phase change of hydrogen storage materials, reaction kinetic equations, and enthalpy change effects during gas adsorption / desorption processes.

[0050] In one specific embodiment, the specific implementation process of step S110 is as follows: 1. Data Acquisition: The controller acquires temperature data from all zone-integrated sensors at a frequency of 10Hz. Simultaneously read the total heating power of the system. and tank pressure .

[0051] 2. Physical Modeling: A thermo-chemical coupled physical model based on hydrogen storage materials is established. This model includes energy conservation equations and chemical reaction kinetic equations, and considers the thermal conductivity of the materials. Specific heat capacity With temperature and reaction progress The dynamic changes.

[0052] 3. Inverse problem solving and data assimilation: Collected , , As boundary conditions and driving inputs.

[0053] Perform iterative calculations: Using the current state estimate (predicted internal temperature distribution), perform forward heat conduction extrapolation to predict the surface temperature distribution. .

[0054] Calculate residuals .

[0055] Based on optimization algorithms such as the adjoint method or Kalman filtering, the residuals are utilized. The distribution of heat source terms (i.e., the distribution of reaction heat release) and the equivalent thermophysical parameters of the material are corrected in reverse until the residual is minimized.

[0056] After data assimilation is complete, the state variables are updated, and arbitrary coordinates inside the hydrogen storage material are obtained through parsing. Three-dimensional temperature field Coordinates of the chemical reaction front and indicators of the degree of heat accumulation.

[0057] 4. Cloud Map Rendering: The calculated 3D data is rendered into a visualized virtual temperature field cloud map. Operators can intuitively see whether there are "cold cores" or "hot spots" inside, which provides a deeper basis for subsequent control beyond surface temperature measurement.

[0058] This application adopts real-time sensing and reconstruction of the whole-domain temperature field, which breaks through the limitation that the temperature inside solid materials cannot be directly measured. Through surface data assimilation technology, it "sees through" the internal thermal state.

[0059] Step S120: Receive the target hydrogen flow command, obtain the current internal state parameters of the hydrogen storage tank, call the preset thermodynamic-kinetic coupling model, reverse calculate the internal target temperature field, and optimize to obtain the optimal target temperature field.

[0060] In step S120, this application deeply integrates flow rate back-calculation with thermodynamic models. It adopts thermodynamic model inversion based on target flow rate. The pre-set thermodynamic-kinetic coupled model is based on the hydrogen release reaction rate equation and inversely calculates the internal target temperature field required to meet the target flow rate. Furthermore, considering heat transfer hysteresis and uniformity constraints, it optimizes to obtain the optimal target temperature field, realizing the feedforward mapping from "flow rate requirement" to "temperature setting".

[0061] Specifically, the goal of a solid-state hydrogen storage system is to supply hydrogen on demand. Upon receiving a target hydrogen mass flow rate or power demand signal from the fuel cell system or central control unit, the control process enters the feedforward calculation stage. First, the target flow rate demand is converted into the required molar amount of hydrogen per unit time, completing demand analysis. Then, model calculations are performed, calling the built-in high-precision thermodynamic-kinetic coupled model. This model takes the characteristic parameters of the hydrogen storage material, the current average pressure inside the tank, and the calculated required hydrogen flow rate as input. Finally, the target temperature is solved. The core of the model is a functional relationship: Target temperature = This function calculates the theoretically optimal average temperature and reference values ​​for temperature distribution in key regions required by the entire material to achieve the target flow rate by solving the equation relating the hydrogen release reaction rate to temperature and pressure. Step S120 achieves intelligent conversion from "how much hydrogen is needed" to "how high a temperature is needed".

[0062] In some embodiments, step S120 involves the following steps:

[0063] Step S121, Demand Analysis and Parameter Acquisition: Receive the target hydrogen flow rate instruction, convert the target hydrogen flow rate instruction into the target hydrogen molar flow rate per unit time, and simultaneously acquire the current internal state parameters of the hydrogen storage tank.

[0064] The current internal state parameters include the current average pressure inside the tank and the real-time physical properties of the hydrogen storage material. Furthermore, the real-time physical properties of the hydrogen storage material are obtained by using a state observer or soft sensor model, based on measured temperature data, pressure data, and historical hydrogen charge / discharge cycle data from the surface of the hydrogen storage tank, to invert and estimate the local temperature distribution, remaining hydrogen absorption capacity, and material aging coefficient within the hydrogen storage material.

[0065] Step S122: Solving the target temperature field based on the thermodynamic-kinetic coupling model: Call the preset thermodynamic-kinetic coupling model, take the target hydrogen molar flow rate, the current average pressure inside the tank and the real-time physical state of the hydrogen storage material as input variables, and solve the correlation equation between the hydrogen release reaction rate and temperature and pressure to inversely calculate the internal target temperature field of the hydrogen storage material required to meet the target hydrogen molar flow rate.

[0066] The internal target temperature field includes the theoretically optimal average temperature of the hydrogen storage material and reference values ​​for the spatial temperature distribution of each zone. Furthermore, the thermodynamic-kinetic coupled model is constructed based on: the Van't Hoff equation describing the thermodynamic relationship between equilibrium pressure and temperature of the hydrogen storage material, and the Arrhenius equation describing the kinetic relationship between the hydrogen release reaction rate constant and temperature. This application abandons simple linear lookup tables or empirical formulas, constructing a deeply coupled model based on the Van't Hoff equation describing equilibrium thermodynamic properties and the Arrhenius equation describing non-equilibrium reaction kinetic properties. This model not only considers the pressure-temperature equilibrium relationship but also accurately incorporates the nonlinear kinetic characteristics of the reaction rate changing with temperature. By numerically solving this coupled equation set, the true reaction characteristics of the hydrogen storage material under different aging degrees and remaining capacities can be accurately reflected. This gives the calculated target temperature value a solid physicochemical basis, significantly improves the accuracy of control commands, and avoids temperature overshoot or reaction stagnation caused by model mismatch. Furthermore, the reverse calculation process specifically involves numerically solving the aforementioned coupled equations under known target reaction rate (derived from the target hydrogen molar flow rate) and current pressure conditions to obtain the corresponding target temperature value. Traditional control methods typically rely on feedback deviations in pressure or flow rate for lag-based adjustment, which is insufficient to handle the instantaneous high flow rate demand caused by sudden changes in fuel cell load. This application, through a reverse calculation process, directly calculates the theoretical temperature value required to meet the flow rate based on the target hydrogen molar flow rate and current pressure using a coupled model. This "demand-driven temperature control" feedforward control strategy eliminates the lag inherent in traditional feedback control, enabling the system to adjust the heating strategy in advance the instant the load command is issued. This significantly shortens the dynamic response time of the hydrogen supply system, effectively avoids the "hydrogen starvation" phenomenon in fuel cells caused by insufficient hydrogen supply, and significantly improves the real-time performance and dynamic tracking capability of the hydrogen supply response.

[0067] Step S123: Optimization and allocation of heating power based on heat transfer inversion: Construct the heat transfer mapping relationship between each zone and the internal temperature field, take the internal target temperature field as the constraint target, use the optimization algorithm to iteratively solve the heating parameters of each zone, calculate the optimal target surface temperature setpoint corresponding to each zone, and obtain the optimal target temperature field.

[0068] Furthermore, the optimization algorithm is either a Model Predictive Control (MPC) algorithm or a Quadratic Programming (QP) algorithm. The objective function of the optimization algorithm is configured to minimize the deviation between the internal actual predicted temperature field and the target temperature field, while satisfying the following constraints: Power limit constraints for each external wall heating zone; Safety threshold constraint for the highest surface temperature of hydrogen storage tanks; Thermal coupling interference constraints between adjacent heating zones.

[0069] This application employs MPC or QP algorithms to jointly optimize the heating power of each zone as an overall decision variable. By minimizing the deviation between the "internal actual predicted temperature field" and the "target temperature field," the algorithm can proactively consider thermal coupling effects and automatically coordinate the power allocation of each zone. This global collaborative optimization mechanism effectively eliminates redundant heating, significantly reduces the overall energy consumption of the system while ensuring the hydrogen supply rate, achieves globally optimal control under multivariable coupling, and significantly improves energy efficiency. In the optimization process, this application introduces power upper limit constraints, surface maximum temperature safety threshold constraints, and thermal coupling interference constraints to prevent overload damage to the heating module, ensure electrical system safety, and directly eliminate the possibility of the tank surface temperature exceeding the safety limit from the mathematical programming level. Even when there is an extreme high flow rate demand in the feedforward calculation, the temperature can be forcibly limited to a safe range, realizing the inherent safety design of "control is safety." It also suppresses the diffusion of temperature spikes caused by localized strong heating. Compared with the passive protection of traditional control relying on post-control amplitude limiting or alarm shutdown, this method avoids illegal operations at the source of control command generation, greatly improving the reliability of system operation. This application constructs an objective function centered on "minimizing the deviation between the actual predicted internal temperature field and the target temperature field." This method not only focuses on the temperature at a single measuring point but also aims to reproduce the ideal temperature distribution within the entire hydrogen storage material. The rolling optimization characteristic of the MPC algorithm allows it to predict temperature change trends over a future period using the model, adjusting the heating strategy in advance, thus effectively overcoming the overshoot and oscillation problems common in large-hysteresis systems. The rapid solution capability of the QP algorithm ensures that the optimal solution can be obtained within a millisecond-level control cycle, enabling the internal temperature field to closely follow the dynamically changing target flow rate demand. This significantly improves the stability and consistency of the hydrogen release reaction and avoids hydrogen flow rate fluctuations caused by temperature variations.

[0070] Furthermore, step S120 also includes a feedback correction step, specifically: real-time monitoring of the actual output hydrogen flow rate and tank pressure, calculating the deviation between the actual and target values; when the deviation exceeds a preset threshold, PID compensation correction is performed on the target temperature field calculated in step S122 or the heating power calculated in step S123. This feedback correction step effectively eliminates steady-state errors caused by model mismatch and parameter drift, suppresses interference from unmodeled dynamics and external random disturbances, and balances dynamic response speed and steady-state control accuracy, thereby improving the adaptive maintenance capability of the system throughout its entire lifecycle.

[0071] In some embodiments, step S120 further includes: periodically comparing the theoretical flow rate derived from the thermodynamic-kinetic coupling model with the actual flow meter reading, calculating the residual, and if the residual exceeds a preset residual threshold, then correcting the kinetic parameters in the thermodynamic-kinetic coupling model online. The kinetic parameters include activation energy and pre-exponential factor. This application, by periodically comparing the residual between the theoretical flow rate and the actual flow meter reading, triggers a correction mechanism once the threshold is exceeded, dynamically updating the activation energy and pre-exponential factor parameters. This allows the mathematical model to "learn" and track the actual health status of the hydrogen storage material in real time, ensuring that the model remains highly consistent with the physical object throughout the entire system lifecycle. This avoids the decrease in control accuracy due to material aging, overcomes the time-varying characteristics of materials, and achieves full lifecycle self-adaptation of the model. Using the flow residual as a direct correction signal, the model error is controlled within a preset threshold range through a closed-loop feedback mechanism. This "monitoring-evaluation-correction" self-calibration process eliminates accumulated errors and ensures the long-term stability of hydrogen supply. This application has online self-identification capability, which reduces the dependence on the accuracy of the initial calibration, simplifies system debugging and maintenance, and at the same time, the parameter change trajectory during the correction process (especially the evolution trend of activation energy and pre-exponential factor) directly reflects the performance degradation of hydrogen storage materials, providing a quantitative assessment basis for the health status of materials.

[0072] In one specific embodiment, the specific implementation process of step S120 is as follows: 1. Command Conversion: Receives the target power signal from the downstream fuel cell and converts it into the target hydrogen molar flow rate. .

[0073] 2. Model Invocation and Reverse Solving: A pre-built thermodynamic-kinetic coupled model is invoked, constructed based on the Arrhenius equation and the PCT (pressure-composition-temperature) curve. ; in, Indicates the target hydrogen molar flow rate; Indicates the pre-exponential factor; Represents the Arrhenius exponent. Indicates activation energy. Represents the ideal gas constant. Indicates absolute temperature; Represents the driving force function. This indicates the current hydrogen pressure in the system. Indicates equilibrium pressure (absolute temperature) The function is determined by the PCT curve. This indicates the conversion rate.

[0074] 3. Adaptive Calibration: Periodically compare the theoretical flow rate derived from the model with the actual flow meter readings. If the residual exceeds a threshold (indicating material aging or model drift), the activation energy in the model is corrected online. Or pre-exponential factor This ensures the long-term accuracy of the reverse calculation.

[0075] Step S130: Spatial matching and comparison of the optimal target temperature field and the virtual temperature field cloud map are performed to generate temperature deviation signals for each heating module. Based on the multi-loop decoupled PID controller, the input power of the heating module in each zone is dynamically and independently adjusted according to each temperature deviation signal and the thermal coupling interference matrix of adjacent zones, and the opening degree of the air inlet valve and the air outlet valve are synchronously controlled.

[0076] In step S130, this application achieves multi-zone closed-loop collaborative control by closely integrating the control strategy with specific hardware (heating modules, valves). Based on a multi-loop decoupled PID controller, a thermal coupling interference matrix of adjacent zones is introduced to dynamically and independently adjust the input power of each zone's heating module to offset thermal crosstalk. At the same time, the opening degree of the inlet valve and outlet valve is synchronously linked to achieve multi-variable collaborative stability of temperature, pressure, and flow.

[0077] Specifically, after obtaining the target temperature setpoint, the solid-state hydrogen storage system enters the precise execution control phase. First, deviation calculation: the virtual temperature field cloud map generated in step S110 is compared with the optimal target temperature field calculated in step S120 to generate a temperature deviation signal for each independent heating zone. Second, PID algorithm control: an independent digital PID controller is configured for each heating zone, and the controller receives the deviation signal for that zone. Finally, power output and closed-loop feedback: the PID calculation result is converted into an electrical control signal, dynamically adjusting the electrical power applied to the heating elements of each zone. The temperature sensor continuously feeds back new temperature data to step S110, forming a closed loop.

[0078] In some embodiments, the specific steps of step S130 are as follows: Step S131, Deviation Construction: Spatial matching and comparison of the optimal target temperature field and the virtual temperature field cloud map are performed to generate temperature deviation signals for each heating module.

[0079] Step S132, Decoupling Calculation and Cooperative Control: Input each temperature deviation signal to the multi-loop decoupled PID controller. Based on the preset thermal coupling model and the thermal coupling interference matrix of adjacent partitions, the multi-loop decoupled PID controller calculates the thermal interference components between adjacent partitions, and decouples and corrects the initial PID control quantities of each partition according to the thermal interference components, generating decoupled temperature control commands. At the same time, based on the current pressure setpoint and flow setpoint, it generates linkage opening commands for the inlet valve and outlet valve to achieve multi-variable coordination of temperature, pressure and flow.

[0080] Step S133, Execution Drive: Convert the decoupled temperature control command into the power drive signal of each zone heating module, dynamically adjust the input power applied to each zone heating module, and send the linkage opening command to the actuator to adjust the opening of the intake valve and the exhaust valve.

[0081] Step S134, Closed-loop feedback: After executing the electric drive signal and linkage opening command, the optimal target temperature field and the virtual temperature field cloud map are reacquired and spatially matched and compared to generate new temperature deviation signals for each heating module, forming a closed-loop control loop.

[0082] In step S132 above, the thermal coupling model is either a transfer function matrix established through system identification experiments or a thermal resistance network model obtained based on finite element analysis. The system identification transfer function matrix transforms complex partial differential equations into algebraic operations, greatly reducing the computational load and meeting the requirements for millisecond-level real-time control. The thermal resistance network model preserves the complex spatial geometry and non-uniform heat transfer characteristics inside the hydrogen storage tank, ensuring high fidelity in the prediction of the temperature field distribution. Both modeling methods achieve accurate description of the multi-zone thermal coupling effect under limited computing power, balancing computational real-time performance and physical accuracy. The specific algorithm for decoupling correction is feedforward decoupling control or state-space decoupling control, which completely eliminates thermal coupling interference between multiple heating zones, achieving independent, rapid, and oscillatory precise control of the temperature of each channel, significantly reducing the difficulty of system debugging. Furthermore, the multivariate coordination specifically includes: taking temperature control as the main loop, and superimposing the calculated heating power change rate as a feedforward signal onto the pressure and flow control loop. The coordination mechanism of this application actively offsets the gas thermal expansion effect caused by temperature change by introducing the heating power change rate feedforward, eliminates false fluctuations in pressure and flow, and realizes the non-disruptive dynamic switching and high-precision stable output of the hydrogen supply system under variable temperature conditions.

[0083] In step S133 above, the input power applied to each zone heating module is dynamically adjusted. Specifically, when the temperature of a zone in the virtual temperature field cloud map exceeds a preset temperature threshold, the input power of the heating module in that zone is reduced first, while the input power of the heating modules in adjacent zones is increased. This application adopts a "peak shaving and valley filling" strategy. By monitoring the virtual temperature field cloud map in real time, the system can accurately identify local overheating areas. Once a zone exceeds the temperature, its heating power is immediately reduced or even the heat source is cut off. At the same time, the power of adjacent zones is increased. This not only quickly eliminates the risk of local high temperature, but also smoothly transfers the heat load originally concentrated at one point to the surrounding area. The thermal conductivity of the hydrogen storage material is used to achieve uniform heat diffusion, effectively preventing material sintering or performance degradation caused by local overheating, ensuring a constant total hydrogen supply capacity, maintaining stable system output, extending the life of the hydrogen storage material, and improving the uniformity of the temperature field.

[0084] Furthermore, the thermal management method also includes: a multi-loop decoupled PID controller dynamically adjusts the input power of the heating modules in the adjacent zones based on the opening change rate of the outlet valve. This application utilizes the opening change rate as a feedforward signal to increase the heating power of adjacent zones before the actual temperature drop occurs due to valve action. This predictive compensation effectively overcomes the lag inherent in traditional PID control that relies solely on temperature difference feedback, achieving zero-delay temperature tracking, accurately predicting and compensating for the heat absorption and cooling effect at the outlet, and realizing spatiotemporal synchronization between the heat source and cold source, completely eliminating temperature lag and system oscillation under variable load conditions.

[0085] Furthermore, the multi-loop decoupled PID controller employs a model predictive control algorithm, using the power upper limit, heating rate threshold, and valve operating range of each zone's heating module as constraints. The objective function is to minimize the error between the optimal target temperature field and the virtual temperature field cloud map, and the output control quantity is continuously optimized. This output control quantity includes the electric drive signal and the linkage opening command. This application utilizes the multi-step prediction and continuous optimization capabilities of model predictive control to achieve globally coordinated optimal control of electric heating and valve flow under strict adherence to power, heating rate, and valve physical constraints. This ensures both the safe lifespan of the hydrogen storage material and high-precision uniform tracking of the temperature field.

[0086] In other embodiments, the thermal management method further includes: real-time monitoring of the maximum temperature difference and local overheating areas in the virtual temperature field cloud map; and forcibly reducing the input power of the heating module of a certain zone when the reconstruction temperature of a certain zone exceeds the material safety threshold. Based on the virtual temperature field cloud map, this application can accurately capture internal local overheating that cannot be directly measured by sensors. Once the reconstruction temperature of a certain zone reaches the material safety threshold, it immediately triggers a forced power reduction, constructing an active safety defense line based on the "real internal state," eliminating potential safety hazards at the outset, preventing catastrophic consequences caused by undetectable internal overheating, eliminating thermal damage to hydrogen storage materials, extending the lifespan of core components, and achieving a control upgrade from "passive alarm" to "active suppression."

[0087] In one specific embodiment, the specific implementation process of step S130 is as follows: 1. Deviation generation: This involves generating the optimal target temperature field. Mapped onto the surface of each partition, and compared with the measured values ​​of the sensors in each partition, a partition temperature deviation vector is generated. .

[0088] 2. Thermal coupling decoupling: This application introduces a thermal coupling interference matrix. Matrix elements Indicates the first Zone heating for the first The influence coefficient of temperature in the region (obtained through experimental identification or simulation).

[0089] Calculate the decoupling term for a multi-loop decoupled PID controller: .in, This represents the final control output vector after decoupling; This represents the original PID control output vector; Represents the thermal coupling interference matrix; This represents the predicted value of the disturbance to this area caused by control actions in other areas.

[0090] For example, when the left partition needs a significant temperature increase, the controller will automatically slightly reduce the output of the adjacent right partition to offset the overshoot caused by heat penetration, thus achieving truly independent and precise temperature control.

[0091] 3. Valve linkage and coordination: The controller not only outputs heating power commands, but also generates linkage opening commands for the outlet valve and inlet valve based on the current pressure and flow setpoints.

[0092] Dynamic compensation strategy: When a rapid increase in the outlet valve opening (a sudden surge in flow demand) is detected, the airflow carries away a significant amount of heat. The controller immediately adjusts the response based on the rate of change in opening. The heating power of the zone near the air outlet is dynamically increased and feedforward compensation is performed to prevent a sudden drop in local temperature.

[0093] Safety protection strategy: If the virtual temperature field cloud map shows that the internal reconstruction temperature of a certain zone exceeds the safety threshold (such as 160℃), even if the surface temperature of the zone is normal, the controller will prioritize cutting off or significantly reducing the heating power of the zone and increase the power of the adjacent low temperature zone to balance the thermal field and prevent the material from sintering or pulverizing.

[0094] 4. Advanced Control Algorithm: In high-end applications, Model Predictive Control (MPC) can replace traditional PID control. By using the upper limit of heating power, maximum heating rate, and valve operating range as hard constraints, and minimizing the overall temperature error as the objective function, the control sequence is continuously optimized over a future period to achieve global optimum for the multivariable, strongly coupled system.

[0095] This application integrates a thermodynamic-kinetic coupled model for feedforward predictive control. Upon receiving a flow demand command, the model can pre-calculate the optimal temperature target, driving the heating module to proactively operate before the actual hydrogen flow rate decreases. Combined with a more direct heat application method, this significantly reduces the system's thermal inertia, enabling the hydrogen supply flow rate to respond quickly and smoothly to dynamic demands. Precise and uniform temperature control avoids crystal structure damage caused by localized overheating of the material. Simultaneously, the rapid response capability reduces the need for severe and frequent temperature shocks to the material. By maintaining a suitable and stable reaction temperature and reducing thermal cycling stress, the system effectively slows down material pulverization and capacity decay, significantly extending its cycle life and overall economic efficiency.

[0096] Corresponding to the above embodiments of the thermal management method for solid-state hydrogen storage systems based on thermodynamic inversion, another embodiment of this application provides a thermal management system. For example... Figure 2 As shown, the thermal management system mainly includes: a real-time temperature sensing and reconstruction module 210, a thermodynamic model inversion module 220, and a multi-zone closed-loop collaborative control module 230.

[0097] Specifically, the real-time temperature sensing and reconstruction module 210 uses all temperature sensors deployed in the zones to collect the temperature distribution data of the outer wall of the hydrogen storage tank in real time, and reconstructs it into a continuous virtual temperature field cloud map based on a spatial interpolation algorithm; the thermodynamic model inversion module 220 receives the target hydrogen flow command, obtains the current internal state parameters of the hydrogen storage tank, and calls the preset thermodynamic-dynamic coupling model to inversely calculate the internal target temperature field and optimize it to obtain the optimal target temperature field; the multi-zone closed-loop collaborative control module 230 performs spatial matching and comparison between the optimal target temperature field and the virtual temperature field cloud map, generates temperature deviation signals for each heating module, and dynamically and independently adjusts the input power of the heating modules in each zone according to each temperature deviation signal and the thermal coupling interference matrix of adjacent zones based on a multi-loop decoupled PID controller, and synchronously controls the opening of the inlet valve and outlet valve.

[0098] Among them, the real-time temperature sensing and reconstruction module 210 and the thermodynamic model inversion module 220 are the core modules of the thermal management system. The thermodynamic model inversion module 220 adopts a high-fidelity algorithm model pre-established based on the thermal structure fluid multiphysics coupling algorithm, which includes the geometric structure of the hydrogen storage tank, the thermophysical parameters (thermal conductivity, specific heat capacity, density) of each layer of materials (shell, hydrogen storage material), and the reaction heat effect. During system operation, the real-time temperature sensing and reconstruction module 210 uses a real-time state inversion algorithm to take the collected shell surface temperature distribution, total heating power input, and internal pressure as the boundary conditions and inputs of the model. By solving the inverse heat conduction problem and assimilating the model with real-time data, it inverts the estimated values ​​of the temperature field of the hydrogen storage material, the reaction front position, and the heat accumulation state, thereby generating a virtual temperature field cloud map inside the hydrogen storage material. The thermodynamic model inversion module 220 receives the target hydrogen flow command and, combined with the currently inverted internal material state and pressure, calculates the target temperature field that the hydrogen storage material needs to achieve to meet future flow requirements through model prediction. Then, through an optimization algorithm, it solves for the optimal heating power or target surface temperature setpoint required for each outer wall heating zone to achieve this internal target. The multi-zone closed-loop collaborative control module 230 employs a multi-loop decoupled PID controller, which not only controls the deviation between the setpoint and measured values ​​of the surface temperature of each zone but also specifically considers the thermal coupling interference between zones. The controller's output signal is converted into independent electrical control for the heating elements of each zone, and includes linked control of the opening of the inlet and outlet valves to achieve coordinated pressure and flow.

[0099] This thermal management system is an intelligent system integrating real-time sensing, model prediction, and closed-loop feedback. By precisely regulating the temperature field inside the solid hydrogen storage material, it dynamically matches the hydrogen demand of downstream fuel cells, achieving a stable and efficient hydrogen supply. The entire workflow—"real-time sensing and visualization reconstruction of the entire temperature field → calculation of the feedforward temperature setpoint based on the target flow rate → precise temperature tracking and control based on the PID algorithm"—forms a complete control closed loop. This integrated "sensing-prediction-control" workflow upgrades the extensive overall heating to precise on-demand heating. It not only diagnoses the system status in real time through temperature cloud maps but also predictively sets targets through model feedforward. Finally, through precise PID closed-loop control, it synergistically achieves three major goals: temperature uniformity, pressure stability, and rapid flow response, fundamentally overcoming the shortcomings of traditional technical solutions.

[0100] It should be noted that the above-described thermal management system embodiment corresponds to the thermal management method embodiment for solid-state hydrogen storage systems based on thermodynamic inversion, and has the same technical effects as that thermal management method embodiment. For details, please refer to the thermal management method embodiment. Furthermore, the thermal management system embodiment is derived from the thermal management method embodiment; for details, please refer to the thermal management method section, which will not be repeated here.

[0101] The workflow of a thermal management system applied to vehicle power scenarios is described in detail below.

[0102] After the solid-state hydrogen storage system is powered on, it first performs a self-test and loads initial thermophysical parameters.

[0103] 1. Sensing stage: The temperature sensor array samples at high frequency, and the reconstruction module generates an internal virtual temperature field cloud map in real time, identifying that the current reaction is mainly concentrated in the upper half of the tank.

[0104] 2. Decision-making phase: The vehicle accelerates, and the target flow command jumps from 10 NL / min to 50 NL / min. The thermodynamic model inversion module immediately calculates that in order to meet this flow rate, the temperature in the central region of the tank needs to be increased from 60°C to 85°C within the next 5 seconds, and generates the optimal target temperature field with a radial gradient.

[0105] 3. Execution Phase: The decoupled controller calculates the required PWM duty cycle for each zone. For the heating zone corresponding to the central area, a high-power command is output; at the same time, considering the thermal interference of adjacent zones, fine-tuning is performed on the edge zones.

[0106] Simultaneously fine-tune the outlet valve, initially closing it slightly to maintain back pressure, and gradually opening it after the temperature field is properly tracked to ensure a steady increase in flow rate without overshoot.

[0107] If an abnormal increase in reconfiguration temperature is detected at a certain location, local power limiting protection will be triggered immediately.

[0108] 4. Closed-loop iteration: In the next cycle, data is re-acquired, a new temperature field is reconstructed, and comparisons and corrections are made again to form a high-speed closed loop at the millisecond level.

[0109] In summary, the thermodynamic inversion-based thermal management method and system for solid-state hydrogen storage systems provided in this application solves the "black box" problem of solid-state hydrogen storage systems by using temperature field reconstruction technology and surface sparse sensor data to invert the internal three-dimensional temperature field and reaction front position. This provides reliable state feedback for precise control, overcomes the limitations of perception, and realizes visualization of the internal state. Furthermore, this application innovatively proposes a flow rate inversion mechanism, using a thermodynamic-dynamic coupling model to directly invert the required temperature field from the target flow rate. This allows the system to act in advance before load changes, overcoming thermal inertia, transforming passive response into active response, significantly improving response speed, and meeting the stringent requirements for transient flow rates in automotive and other scenarios. Furthermore, this application effectively suppresses thermal crosstalk between adjacent heating zones by introducing a multi-loop decoupling control strategy with a thermal coupling interference matrix, avoiding local overheating and temperature oscillation, improving temperature control uniformity, ensuring that the hydrogen storage material works in a uniform thermal environment, and significantly extending the material's cycle life. Moreover, this application features deep software and hardware synergy, with the control strategy closely tied to the specific heating modules, integrated sensors, and valve hardware characteristics, achieving seamless integration between theoretical models and engineering execution, improving the system's adaptability and stability under complex operating conditions, and enhancing the system's robustness.

[0110] It will be understood by those skilled in the art that the accompanying drawings are merely schematic diagrams of one embodiment, and the components shown in the drawings are not necessarily essential for implementing the invention. It should also be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0111] In the description of the embodiments of this application, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. Furthermore, in the description of the embodiments of this application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the above embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion, characterized in that, The solid-state hydrogen storage system has independently controlled heating modules arranged in sections on the outer wall of the hydrogen storage tank, and each heating module integrates a temperature sensor; the thermal management method includes: Using all the temperature sensors deployed in the zones, the temperature distribution data of the outer wall of the hydrogen storage tank is collected in real time, and reconstructed into a continuous virtual temperature field cloud map based on a spatial interpolation algorithm. The system receives the target hydrogen flow rate command, obtains the current internal state parameters of the hydrogen storage tank, calls the preset thermodynamic-kinetic coupling model, reversely calculates the internal target temperature field, and optimizes to obtain the optimal target temperature field. The optimal target temperature field is spatially matched and compared with the virtual temperature field cloud map to generate temperature deviation signals for each heating module. Based on a multi-loop decoupled PID controller, the input power of the heating module in each zone is dynamically and independently adjusted according to the temperature deviation signals and the thermal coupling interference matrix of adjacent zones, and the opening degree of the inlet valve and outlet valve is synchronously controlled.

2. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 1, characterized in that, The method of reconstructing a continuous virtual temperature field cloud map based on spatial interpolation algorithm specifically includes: Real-time acquisition of total heating power input data of the solid hydrogen storage system and internal pressure data of the hydrogen storage tank; A thermo-chemical coupled physical model of the hydrogen storage material based on the hydrogen storage tank is established, and the temperature distribution data, the total heating power input data, and the internal pressure data are mapped to the boundary adjustment and driving input of the thermo-chemical coupled physical model. Based on the aforementioned thermo-chemical coupled physical model, an iterative calculation process is performed, including: using the current state estimate to perform forward heat conduction deduction to obtain the predicted surface temperature distribution; calculating the residual between the predicted surface temperature distribution and the temperature distribution data; based on the residual, correcting the heat source term distribution and material thermal property parameters within the thermo-chemical coupled physical model using an optimization algorithm to minimize the residual, thereby solving the inverse heat conduction problem; and integrating the corrected heat source term distribution and material thermal property parameters into the model state variables to complete the data assimilation at the current moment and update the state estimate within the hydrogen storage material. Based on the updated state estimates, the three-dimensional temperature field distribution, chemical reaction front coordinates, and heat accumulation index inside the hydrogen storage material are obtained by analysis. Based on the aforementioned three-dimensional temperature field distribution, a virtual temperature field cloud map of the hydrogen storage material is rendered and generated.

3. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 1, characterized in that, The process of receiving the target hydrogen flow rate command, simultaneously acquiring the current internal state parameters of the hydrogen storage tank, and invoking a preset thermodynamic-kinetic coupled model to inversely calculate the internal target temperature field data, and optimizing to obtain the optimal target temperature field, specifically includes: The system receives the target hydrogen flow rate command, converts the target hydrogen flow rate command into a target hydrogen molar flow rate per unit time, and simultaneously acquires the current internal state parameters of the hydrogen storage tank; wherein, the current internal state parameters include the current average pressure inside the tank and the real-time physical property state of the hydrogen storage material in the hydrogen storage tank. The preset thermodynamic-kinetic coupled model is invoked, and the target hydrogen molar flow rate, the current average pressure inside the tank, and the real-time physical properties of the hydrogen storage material are used as input variables. By solving the correlation equation between the hydrogen release reaction rate and temperature and pressure, the internal target temperature field of the hydrogen storage material required to meet the target hydrogen molar flow rate is calculated in reverse. The internal target temperature field includes the theoretical optimal average temperature of the hydrogen storage material and the spatial temperature distribution reference values ​​of each zone. A heat transfer mapping relationship between each partition and the internal temperature field is constructed. Taking the internal target temperature field as the constraint target, the heating parameters of each partition are iteratively solved using an optimization algorithm to calculate the optimal target surface temperature setting value corresponding to each partition, so as to obtain the optimal target temperature field.

4. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 1, characterized in that, The multi-loop decoupled PID controller dynamically and independently adjusts the input power of the heating module in each zone based on the temperature deviation signals and the thermal coupling interference matrix of adjacent zones, and synchronously controls the opening of the outlet valve, specifically including: Each of the temperature deviation signals is input to the multi-loop decoupled PID controller. The multi-loop decoupled PID controller calculates the thermal interference components between adjacent partitions based on a preset thermal coupling model and the thermal coupling interference matrix of the adjacent partitions. It then decouples and corrects the initial PID control quantities of each partition according to the thermal interference components, generating a decoupled temperature control command. Simultaneously, it generates a linkage opening command for the inlet valve and the outlet valve based on the current pressure setpoint and flow setpoint, thereby achieving multi-variable coordination of temperature, pressure, and flow. The decoupled temperature control command is converted into a power drive signal for the heating module of each zone, the input power applied to the heating module of each zone is dynamically adjusted, and the linkage opening command is sent to the actuator to adjust the opening of the air inlet valve and the air outlet valve. After executing the electric drive signal and the linkage opening command, the optimal target temperature field is reacquired and spatially matched with the virtual temperature field cloud map to generate new temperature deviation signals for each heating module, forming a closed-loop control loop.

5. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 4, characterized in that, The dynamic adjustment of the input power applied to the heating modules in each zone specifically includes: When the temperature of a certain zone in the virtual temperature field cloud map exceeds a preset temperature threshold, the input power of the heating module in that zone is reduced first, while the input power of the heating module in the adjacent zone is increased.

6. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 4, characterized in that, The thermal management method further includes: The multi-loop decoupled PID controller dynamically adjusts the input power of the heating module in the adjacent zone to the outlet valve based on the opening change rate of the outlet valve.

7. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 4, characterized in that, The multi-loop decoupled PID controller employs a model predictive control algorithm, using the power limit, heating rate threshold, and valve action range of the heating modules in each zone as constraints, and minimizing the error between the optimal target temperature field and the virtual temperature field cloud map as the objective function to continuously optimize the output control quantity; the output control quantity includes the electric drive signal and the linkage opening command.

8. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 1, characterized in that, The thermal management method further includes: The maximum temperature difference and local overheated areas in the virtual temperature field cloud map are monitored in real time. When the reconstruction temperature of a certain partition exceeds the material safety threshold, the input power of the heating module in that partition is forcibly reduced.

9. The thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion according to claim 1, characterized in that, The thermal management method further includes: Periodically compare the theoretical flow rate derived from the thermodynamic-kinetic coupling model with the actual flow meter reading, calculate the residual, and if the residual exceeds a preset residual threshold, then correct the kinetic parameters in the thermodynamic-kinetic coupling model online.

10. A thermal management system for implementing the thermal management method for a solid-state hydrogen storage system based on thermodynamic inversion as described in any one of claims 1-9, characterized in that, The thermal management system includes: The real-time temperature sensing and reconstruction module is used to collect the temperature distribution data of the outer wall of the hydrogen storage tank in real time using all the temperature sensors deployed in the partition, and reconstruct it into a continuous virtual temperature field cloud map based on the spatial interpolation algorithm. The thermodynamic model inversion module is used to receive the target hydrogen flow command, obtain the current internal state parameters of the hydrogen storage tank, call the preset thermodynamic-dynamic coupling model, reverse calculate the internal target temperature field, and optimize to obtain the optimal target temperature field. The multi-zone closed-loop collaborative control module is used to spatially match and compare the optimal target temperature field with the virtual temperature field cloud map to generate temperature deviation signals for each heating module. Based on the multi-loop decoupled PID controller, according to each temperature deviation signal and the thermal coupling interference matrix of adjacent zones, the input power of the heating module in each zone is dynamically and independently adjusted, and the opening degree of the intake valve and the exhaust valve are synchronously controlled.

Images