Coupling fuel cell dynamic optimal control method and control system

Abstract

The invention relates to a coupling fuel cell dynamic optimal control method and control system, and relates to the technical field of fuel cell control. The method comprises the following steps: acquiring a first characteristic curve corresponding to the proton exchange membrane fuel cell, a currently detected or updated first temperature and a first pressure; and / or obtaining a second characteristic curve corresponding to the solid oxide fuel cell, a currently detected or updated second temperature and a second pressure; determining a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve; and / or determining a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve; and determining a first optimal power corresponding to the proton exchange membrane fuel cell and / or a second optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve and the first life characteristic curve and / or the second efficiency characteristic curve and the second life characteristic curve.

Description

Technical Field

[0001] The present disclosure relates to the field of fuel cell control technology, and in particular to a coupled fuel cell dynamic optimal control method and control system, in particular to a coupled fuel cell dynamic optimal control method and control system based on intelligent particle swarm optimization. Background Art

[0002] Hydrogen fuel cell systems play an important role in modern energy technology. They convert hydrogen into electricity in an efficient and clean way and can be applied in a variety of scenarios, such as industrial power supply and drone power systems. However, these systems face some key challenges during cold start, which directly affect the reliability and efficiency of their practical applications.

[0003] During cold start, hydrogen fuel cell systems need to face the requirement of rapid temperature rise from ambient temperature to operating temperature. Traditional solutions often rely on external auxiliary heating or directly use hydrogen combustion to release heat, which not only increases energy consumption costs, but may also lead to loss of system efficiency. The operating temperature of PEMFC fuel cells is relatively low, generally 60-100°C, and the startup speed is relatively fast, which is suitable for responding to rapid fluctuations in power demand in a short period of time. In addition, PEMFC can be started within a few seconds to a few minutes, which is suitable for responding to frequent fluctuations in wind and solar power generation systems and short-term power supply needs. And there is a demand for external heat sources. The operating temperature of SOFC fuel cells is 600-1000°C, with high efficiency, suitable for providing continuous power output as a base load power supply, and can improve the overall energy efficiency of the system through waste heat recovery. However, the startup time is long, and it generally takes several hours to reach the operating temperature. But SOFC works at high temperatures.

[0004] In the field of control and optimization of fuel cell systems, existing technologies mainly focus on the operation control and optimization strategies of a single type of fuel cell, especially solid oxide fuel cells (SOFC) and proton exchange membrane fuel cells (PEMFC). These control strategies usually adopt a fixed operating point or dynamic adjustment based on a simple feedback mechanism to achieve stable operation and basic efficiency optimization of the fuel cell. Some existing technologies attempt to combine SOFC and PEMFC in parallel or series in order to combine the advantages of both. However, these combined control strategies often adopt a simple allocation ratio or static allocation based on load demand, which fails to fully utilize the characteristics of the two fuel cells under different operating conditions, resulting in the system being difficult to achieve the best balance between efficiency and response speed. Some studies have attempted to introduce optimization algorithms to optimize the operating parameters of a single fuel cell to improve system efficiency. However, these optimization methods are usually targeted at a single fuel cell, fail to consider the coordinated optimization of a multi-fuel cell system, limit the improvement of the overall system performance, and do not consider the characteristics of different fuel cells in the computing power allocation of the optimization algorithm. Summary of the invention

[0005] The present disclosure proposes a technical solution for a coupled fuel cell dynamic optimal control method and a control system.

[0006] According to one aspect of the present disclosure, a method for dynamically optimizing a coupled fuel cell is provided, comprising:

[0007] Obtaining a first characteristic curve corresponding to a proton exchange membrane fuel cell, a first temperature currently detected or updated, and a first pressure; and / or, obtaining a second characteristic curve corresponding to a solid oxide fuel cell, a second temperature currently detected or updated, and a second pressure;

[0008] Determine a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve at the first temperature and the first pressure; and / or determine a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve at the second temperature and the second pressure;

[0009] Based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure, the first optimal power corresponding to the proton exchange membrane fuel cell and / or the second optimal power corresponding to the solid oxide fuel cell are determined.

[0010] Preferably, the method for determining the first optimal power corresponding to the proton exchange membrane fuel cell and / or the second optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure includes: determining a first group of efficiencies corresponding to the first efficiency characteristic curve and a first group of life corresponding to the first life characteristic curve; constructing a first multi-parameter function corresponding to the first efficiency characteristic curve and the first life characteristic curve in the first characteristic curve; wherein the first group of parameters corresponding to the first multi-parameter function includes: a first voltage to be optimized and a first current to be optimized; constructing a first objective function corresponding to the first multi-parameter function with the goal of maximizing the efficiency and life in the first group of efficiencies and the first group of life; and solving the first objective function to obtain the first optimal power with the first voltage-current characteristic curve in the first characteristic curve as the first constraint.

[0011] Preferably, the first optimal power configuration is a product of a first optimal voltage corresponding to the first voltage to be optimized and a first optimal current corresponding to the first current to be optimized.

[0012] Preferably, a second group of efficiencies corresponding to the second efficiency characteristic curve and a second group of lifespans corresponding to the second life characteristic curve are determined; a second multi-parameter function corresponding to the second efficiency characteristic curve and the second life characteristic curve in the second characteristic curve is constructed; wherein the second group of parameters corresponding to the second multi-parameter function includes: a second voltage to be optimized and a second current to be optimized; with the goal of maximizing the efficiency and lifespan in the second group of efficiencies and the second group of lifespans, a second objective function corresponding to the second multi-parameter function is constructed; with the second voltage and current characteristic curve in the second characteristic curve as the second constraint condition, the second objective function is solved to obtain the second optimal power.

[0013] Preferably, the second optimal power configuration is a product of a second optimal voltage corresponding to the second voltage to be optimized and a second optimal current corresponding to the second current to be optimized.

[0014] Preferably, a first group of efficiencies corresponding to the first efficiency characteristic curve and a first group of lifespans corresponding to the first lifespan characteristic curve are determined, and a second group of efficiencies corresponding to the second efficiency characteristic curve and a second group of lifespans corresponding to the second lifespan characteristic curve are determined; a third multi-parameter function is constructed including the first efficiency characteristic curve and the first lifespan characteristic curve in the first characteristic curve, and the second efficiency characteristic curve and the second lifespan characteristic curve in the second characteristic curve; wherein a third group of parameters corresponding to the third multi-parameter function includes: a first voltage to be optimized, a first current to be optimized, a second voltage to be optimized, and a second current to be optimized; with the goal of maximizing the efficiency and lifespan in the first group of efficiencies and the first group of lifespans, and / or the second group of efficiencies and the second group of lifespans, a third objective function corresponding to the third multi-parameter function is constructed; with the first voltage-current characteristic curve in the first characteristic curve and / or the second voltage-current characteristic curve in the second characteristic curve as the third constraint condition, the third objective function is solved to obtain the first optimal power and / or the second optimal power.

[0015] Preferably, the first optimal power configuration corresponding to the third objective function is the product of the first optimal voltage corresponding to the first voltage to be optimized and the first optimal current corresponding to the first current to be optimized; and / or, the second optimal power configuration corresponding to the third objective function is the product of the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized.

[0016] Preferably, in the method of solving the first objective function to obtain the first optimal power by taking the first voltage-current characteristic curve in the first characteristic curve as the first constraint condition, the first constraint condition also includes at least one or more of: a first voltage to be optimized maximum value and a first voltage to be optimized minimum value corresponding to the first voltage to be optimized, a first current to be optimized maximum value and a first current to be optimized minimum value corresponding to the first current to be optimized, a first temperature maximum value and a first temperature minimum value corresponding to the first temperature, and a first pressure maximum value and a first pressure minimum value corresponding to the first pressure.

[0017] Preferably, in the method of solving the second objective function to obtain the second optimal power by taking the second voltage-current characteristic curve in the second characteristic curve as the second constraint condition, the second constraint condition also includes at least one or more of: a second voltage to be optimized maximum value and a second voltage to be optimized minimum value corresponding to the second voltage to be optimized, a second current to be optimized maximum value and a second current to be optimized minimum value corresponding to the second current to be optimized, a second temperature maximum value and a second temperature minimum value corresponding to the second temperature, and a second pressure maximum value and a second pressure minimum value corresponding to the second pressure.

[0018] Preferably, in the method of solving the third objective function to obtain the first optimal power and / or the second optimal power by taking the first voltage-current characteristic curve in the first characteristic curve and / or the second voltage-current characteristic curve in the second characteristic curve as the third constraint condition, the third constraint condition also includes at least: one or more of the first voltage to be optimized maximum value and the first voltage to be optimized minimum value corresponding to the first voltage to be optimized, the first current to be optimized maximum value and the first current to be optimized minimum value corresponding to the first current to be optimized, the first temperature maximum value and the first temperature minimum value corresponding to the first temperature, the first pressure maximum value and the first pressure minimum value corresponding to the first pressure, the second voltage to be optimized maximum value and the second voltage to be optimized minimum value corresponding to the second voltage to be optimized, the second current to be optimized maximum value and the second current to be optimized minimum value corresponding to the second current to be optimized, the second temperature maximum value and the second temperature minimum value corresponding to the second temperature, and the second pressure maximum value and the second pressure minimum value corresponding to the second pressure.

[0019] Preferably, the method for solving the first objective function to obtain the first optimal power includes: constructing a first vector for each particle corresponding to the first objective function; the first vector includes at least: a first power allocation ratio and one or more of a first voltage to be optimized and a first current to be optimized; under a set variable first neighborhood radius, updating the speed and position corresponding to each particle in the first vector until the first efficiency and first life corresponding to each particle in the first vector are maximized, thereby obtaining one or more of the first power allocation ratio, the first optimal voltage corresponding to the first voltage to be optimized, and the first optimal current corresponding to the first current to be optimized.

[0020] Preferably, the method of updating the speed and position corresponding to each particle in the first vector under the set variable first neighborhood radius until the first efficiency and the first life corresponding to each particle in the first vector are maximized comprises: randomly initializing the first position and the first speed, the first individual optimal position and the first global optimal position of each particle in the first vector; updating the first position, the first speed, the first individual optimal position and the first global optimal position of each particle using the continuously updated inertia weight and the acceleration constant; during the updating process, determining the first fitness corresponding to the first efficiency and the first life based on the updated first position and the updated first speed of each particle; stopping the updating until the first fitness reaches the first set fitness; determining the first position for stopping the updating based on the first individual optimal position and the first global optimal position for stopping the updating; configuring each particle corresponding to the first position for stopping the updating to one or more of the first power allocation ratio, the first optimal voltage corresponding to the first voltage to be optimized and the first optimal current corresponding to the first current to be optimized; wherein the first set fitness is configured as a function corresponding to the first efficiency and the first life; the first set fitness is configured as the first value corresponding to the first efficiency and the first life corresponding to each particle when they are maximized.

[0021] Preferably, the method for solving the second objective function to obtain the second optimal power includes: constructing a second vector for each particle corresponding to the second objective function; the second vector includes at least: a second power allocation ratio and one or more of a second voltage to be optimized and a second current to be optimized; under a set variable second neighborhood radius, updating the speed and position corresponding to each particle in the second vector until the second efficiency and second life corresponding to each particle in the second vector are maximized, thereby obtaining one or more of the second power allocation ratio, the second optimal voltage corresponding to the second voltage to be optimized, and the second optimal current corresponding to the second current to be optimized.

[0022] Preferably, the method of updating the speed and position corresponding to each particle in the second vector under the set variable second neighborhood radius until the second efficiency and second life corresponding to each particle in the second vector are maximized comprises: randomly initializing the second position and second speed, the second individual optimal position and the second global optimal position of each particle in the second vector; updating the second position, the second speed, the second individual optimal position and the second global optimal position of each particle using the continuously updated inertia weight and the acceleration constant; during the updating process, determining the second fitness corresponding to the second efficiency and the second life based on the updated second position and the updated second speed of each particle; stopping the updating until the second fitness reaches the second set fitness; determining the second position where the updating is stopped based on the second individual optimal position and the second global optimal position where the updating is stopped; configuring each particle corresponding to the second position where the updating is stopped to one or more of the second power allocation ratio, the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized; wherein the second set fitness is configured as a function corresponding to the second efficiency and the second life; the second set fitness is configured as the second value corresponding to the second efficiency and the second life corresponding to each particle when they are maximized.

[0023] Preferably, the method for solving the third objective function to obtain the first optimal power and / or the second optimal power includes: constructing a third vector for each particle corresponding to the third objective function; the third vector includes at least: a first vector parameter corresponding to a proton exchange membrane fuel cell and a second vector parameter corresponding to a solid oxide fuel cell; wherein the first vector parameter includes: a first power allocation ratio and one or more of a first voltage to be optimized and a first current to be optimized; wherein the second vector parameter includes: a second power allocation ratio and one or more of a second voltage to be optimized and a second current to be optimized; under a set variable third neighborhood radius, the speed and position corresponding to each particle in the third vector are updated until the third efficiency and third life corresponding to each particle in the third vector are maximized, thereby obtaining one or more of the third power allocation ratio, the third optimal voltage corresponding to the third voltage to be optimized, and the third optimal current corresponding to the third current to be optimized.

[0024] Preferably, the method of updating the speed and position corresponding to each particle in the third vector under the set variable third neighborhood radius until the third efficiency and third life corresponding to each particle in the third vector are maximized comprises: randomly initializing the third position and third speed, the third individual optimal position and the third global optimal position of each particle in the third vector; updating the third position, third speed, the third individual optimal position and the third global optimal position of each particle using the continuously updated inertia weight and acceleration constant; during the updating process, determining the corresponding third efficiency based on the updated third position and the updated third speed of each particle. The third fitness of the third efficiency and the third lifespan; until the third fitness reaches the third set fitness, the updating is stopped; based on the third individual optimal position for stopping the updating and the third global optimal position, the third position for stopping the updating is determined; each particle corresponding to the third position for stopping the updating is configured as one or more of the third power allocation ratio, the third optimal voltage corresponding to the third voltage to be optimized, and the third optimal current corresponding to the third current to be optimized; wherein the third set fitness is configured as a function corresponding to the third efficiency and the third lifespan; the third set fitness is configured as a third value corresponding to when the third efficiency and the third lifespan corresponding to each particle are maximized.

[0025] Preferably, before obtaining the first characteristic curve corresponding to the proton exchange membrane fuel cell, the method for determining the first characteristic curve includes: obtaining a first measurement parameter corresponding to the proton exchange membrane fuel cell under a first set condition; wherein the first measurement parameter includes: a first historical measurement current, a first historical measurement temperature, a first historical measurement pressure and its corresponding first historical measurement voltage-current, a first historical measurement efficiency, and a first historical measurement life; determining the first historical measurement voltage corresponding to the proton exchange membrane fuel cell using the first historical measurement current, the first historical measurement temperature, and the first historical measurement pressure; and training or performing data fitting on a preset neural network using the first historical measurement voltage, the first historical measurement current, the first historical measurement temperature, the first historical measurement pressure and its corresponding first historical measurement voltage-current, the first historical measurement efficiency, and the first historical measurement life, respectively, to obtain a first efficiency characteristic curve, a first life characteristic curve, and a first voltage-current characteristic curve corresponding to the first characteristic curve.

[0026] Preferably, before obtaining the second characteristic curve corresponding to the solid oxide fuel cell, the method for determining the second characteristic curve includes: obtaining a second measurement parameter corresponding to the proton exchange membrane fuel cell under a second set condition; wherein the second measurement parameter includes: a second historical measurement current, a second historical measurement temperature, a second historical measurement pressure and its corresponding second historical measurement voltage-current, a second historical measurement efficiency, and a second historical measurement life; using the second historical measurement current, the second historical measurement temperature, and the second historical measurement pressure to determine the second historical measurement voltage corresponding to the proton exchange membrane fuel cell; using the second historical measurement voltage, the second historical measurement current, the second historical measurement temperature, the second historical measurement pressure and its corresponding second historical measurement voltage-current, the second historical measurement efficiency, and the second historical measurement life to train or perform data fitting on a preset neural network, respectively, to obtain a second efficiency characteristic curve, a second life characteristic curve, and a second voltage-current characteristic curve corresponding to the second characteristic curve.

[0027] Preferably, before obtaining the currently detected or updated first temperature and the first pressure, the method for determining the currently detected or updated first temperature and the first pressure comprises: respectively processing the first set of temperatures and the first set of pressures corresponding to the sub-exchange membrane fuel cell at a time before the current time using a sliding window of a set size and its corresponding weights to obtain the currently detected or updated first temperature and the first pressure.

[0028] Preferably, before obtaining the currently detected or updated second temperature and second pressure, the method for determining the currently detected or updated second temperature and second pressure includes: using a sliding window of a set size and its corresponding weights to process the second group of temperatures and second group of pressures corresponding to the solid oxide fuel cell at a time before the current moment, respectively, to obtain the currently detected or updated second temperature and second pressure.

[0029] Preferably, it also includes: when the power load demand corresponding to the power grid is stable, determining the first optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve at the first temperature and the first pressure; at this time, the proton exchange membrane fuel cell is in a standby or low-power operation state; wherein the low-power operation state is configured to be less than a first set ratio of the rated power.

[0030] Preferably, the first set ratio is set to 10%.

[0031] Preferably, when the fluctuation of the power load corresponding to the power grid is less than a first set fluctuation value, the fuel supply speed of the solid oxide fuel cell is adjusted to maintain the first optimal power within a first set power range.

[0032] Preferably, the first interval value and the second interval value of the first set power interval are configured as 95%*first optimal power and 105%*first optimal power respectively.

[0033] Preferably, it also includes: when the predicted growth rate of the power load corresponding to the power grid is greater than the preset growth rate, increasing the power corresponding to the proton exchange membrane fuel cell to a first set output power within a first set time, and increasing the solid oxide fuel cell to a second set output power within a second set time greater than the first set time; when the second set output power reaches the first set power, gradually reducing the output power of the proton exchange membrane fuel cell at a set speed.

[0034] Preferably, it also includes: within a third set time, when the decrease rate of the power load corresponding to the power grid drops to a preset decrease rate, respectively calculating the first optimal power corresponding to the proton exchange membrane fuel cell and the second optimal power corresponding to the solid oxide fuel cell; within a fourth set time, reducing the power corresponding to the proton exchange membrane fuel cell to a third set output power; calculating the power reduction curve corresponding to the solid oxide fuel cell; and reducing the first optimal power according to the power reduction curve.

[0035] Preferably, after the fifth set time, the first characteristic curve corresponding to the proton exchange membrane fuel cell and / or the second characteristic curve corresponding to the solid oxide fuel cell are updated.

[0036] According to one aspect of the present disclosure, there is provided a coupled fuel cell dynamic optimization system, comprising:

[0037] An acquisition unit, used to acquire a first characteristic curve corresponding to a proton exchange membrane fuel cell, a first temperature and a first pressure currently detected or updated; and / or, to acquire a second characteristic curve corresponding to a solid oxide fuel cell, a second temperature and a second pressure currently detected or updated;

[0038] a determining unit, configured to determine a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve at the first temperature and the first pressure; and / or to determine a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve at the second temperature and the second pressure;

[0039] A control optimization unit is used to determine a first optimal power corresponding to the proton exchange membrane fuel cell and / or a second optimal power corresponding to the solid oxide fuel cell based on a first efficiency characteristic curve and a first life characteristic curve at the first temperature and the first pressure and / or a second efficiency characteristic curve and a second life characteristic curve at the second temperature and the second pressure.

[0040] According to one aspect of the present disclosure, a coupled fuel cell dynamic optimal system is provided, comprising: a processor; a memory for storing processor executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned coupled fuel cell dynamic optimal control method.

[0041] According to one aspect of the present disclosure, a coupled fuel cell dynamic optimization system is provided, comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions implement the above-mentioned coupled fuel cell dynamic optimization control method when executed by a processor.

[0042] According to one aspect of the present disclosure, a coupled fuel cell dynamic optimization system is provided, comprising: a computer program product, including a computer program / instruction, which, when executed by a processor, implements the coupled fuel cell dynamic optimal control method described in any one of claims 1 to 8.

[0043] Preferably, the coupled fuel cell dynamic optimal control system is characterized in that it also includes: a solid oxide fuel cell and a proton exchange membrane fuel cell; wherein the solid oxide fuel cell and the proton exchange membrane fuel cell are respectively connected to a fuel processing module for heating hydrogen to a working temperature suitable for the solid oxide fuel cell and the proton exchange membrane fuel cell through a first hydrogen supply valve and a second hydrogen supply valve, and the fuel processing module is connected to the hydrogen supply module.

[0044] Preferably, the solid oxide fuel cell and the proton exchange membrane fuel cell are respectively connected to a DC / AC inverter; the DC / AC inverter is used to convert the DC power output by the solid oxide fuel cell and the proton exchange membrane fuel cell into AC power.

[0045] Preferably, the solid oxide fuel cell and the proton exchange membrane fuel cell are connected to a thermal management module.

[0046] Preferably, the solid oxide fuel cell and the proton exchange membrane fuel cell are connected to a compressor via a first air supply valve and a second air supply valve, respectively, and the compressor is connected to an air supply module.

[0047] Preferably, a first hydrogen flow sensor is provided on the pipeline between the proton exchange membrane fuel cell and the fuel processing module.

[0048] Preferably, a second hydrogen flow sensor is provided on the pipeline between the solid oxide fuel cell and the fuel processing module.

[0049] Preferably, a first air flow sensor is provided on the pipeline between the proton exchange membrane fuel cell and the compressor.

[0050] Preferably, a second air flow sensor is provided on the pipeline between the solid oxide fuel cell and the compressor.

[0051] Preferably, the fuel processing module connected to the solid oxide fuel cell and the proton exchange membrane fuel cell is configured as a hydrogen preheater.

[0052] Preferably, the thermal management module connected to the solid oxide fuel cell and the proton exchange membrane fuel cell includes: a water cooling system and an air cooling system; the solid oxide fuel cell is connected to the air cooling system, and the proton exchange membrane fuel cell is connected to the water cooling system via a pump.

[0053] Preferably, the water cooling system connected to the proton exchange membrane fuel cell is used to flow cooling water through the proton exchange membrane fuel cell and release heat through a first heat exchanger connected to the proton exchange membrane fuel cell.

[0054] In the disclosed embodiments, a coupled fuel cell dynamic optimal control method and control system are provided to solve the problem that the existing fuel cell system control system cannot always operate in the optimal efficiency region, thereby affecting the optimal balance between the efficiency and response speed of the fuel cell system control system.

[0055] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

[0056] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0058] Figure 1 A flow chart showing a method for dynamic optimal control of a coupled fuel cell according to an embodiment of the present disclosure;

[0059] Figure 2 A system block diagram corresponding to a coupled fuel cell dynamic optimal control system according to an embodiment of the present disclosure is shown;

[0060] Figure 3 is a block diagram of an electronic device 800 according to an exemplary embodiment;

[0061] Figure 4is a block diagram of an electronic device 1900 according to an exemplary embodiment. DETAILED DESCRIPTION

[0062] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0063] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0064] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0065] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0066] It can be understood that the above-mentioned various method embodiments mentioned in the present disclosure can be combined with each other to form a combined embodiment without violating the principle logic. Due to space limitations, the present disclosure will not go into details.

[0067] In addition, the present disclosure also provides a coupled fuel cell dynamic optimal control device, an electronic device, a computer-readable storage medium, and a program product optimal control system, all of which can be used to implement any coupled fuel cell dynamic optimal control method provided by the present disclosure. The corresponding technical solutions and descriptions are referred to the corresponding records in the method section and will not be repeated here.

[0068] Figure 1 A flow chart showing a method for dynamic optimal control of a coupled fuel cell according to an embodiment of the present disclosure is shown as follows: Figure 1As shown, the coupled fuel cell dynamic optimal control method includes: step S101: obtaining a first characteristic curve corresponding to a proton exchange membrane fuel cell, a first temperature and a first pressure currently detected or updated; and / or, obtaining a second characteristic curve corresponding to a solid oxide fuel cell, a second temperature and a second pressure currently detected or updated; step S102: determining a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve at the first temperature and the first pressure; and / or, determining a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve at the second temperature and the second pressure; step S103: determining a first optimal power corresponding to the proton exchange membrane fuel cell and / or a second optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure. This solves the problem that the existing fuel cell system control system cannot always operate in the optimal efficiency region, thereby affecting the optimal balance between the efficiency and response speed of the fuel cell system control system.

[0069] In the embodiments of the present disclosure and other possible embodiments, the intelligent control and coordination module monitors the performance characteristic curves of SOFC and PEMFC in real time and dynamically adjusts the operating points of the two batteries so that the entire system always operates in the optimal efficiency area, thereby achieving an optimal balance between system efficiency and response speed.

[0070] Step S101: obtaining a first characteristic curve corresponding to a proton exchange membrane fuel cell, a first temperature and a first pressure currently detected or updated; and / or obtaining a second characteristic curve corresponding to a solid oxide fuel cell, a second temperature and a second pressure currently detected or updated.

[0071] In an embodiment of the present disclosure, before obtaining the first characteristic curve corresponding to the proton exchange membrane fuel cell, a method for determining the first characteristic curve includes: obtaining a first measurement parameter corresponding to the proton exchange membrane fuel cell under a first set condition; wherein the first measurement parameter includes: a first historical measurement current, a first historical measurement temperature, a first historical measurement pressure and its corresponding first historical measurement voltage-current, a first historical measurement efficiency, and a first historical measurement life; determining the first historical measurement voltage corresponding to the proton exchange membrane fuel cell using the first historical measurement current, the first historical measurement temperature, and the first historical measurement pressure; and training or performing data fitting on a preset neural network using the first historical measurement voltage, the first historical measurement current, the first historical measurement temperature, the first historical measurement pressure and its corresponding first historical measurement voltage-current, the first historical measurement efficiency, and the first historical measurement life, respectively, to obtain a first efficiency characteristic curve, a first life characteristic curve, and a first voltage-current characteristic curve corresponding to the first characteristic curve.

[0072] In an embodiment of the present disclosure, before obtaining the second characteristic curve corresponding to the solid oxide fuel cell, a method for determining the second characteristic curve includes: obtaining a second measurement parameter corresponding to the proton exchange membrane fuel cell under a second set condition; wherein the second measurement parameter includes: a second historical measurement current, a second historical measurement temperature, a second historical measurement pressure and its corresponding second historical measurement voltage-current, a second historical measurement efficiency, and a second historical measurement life; determining the second historical measurement voltage corresponding to the proton exchange membrane fuel cell using the second historical measurement current, the second historical measurement temperature, and the second historical measurement pressure; and training or performing data fitting on a preset neural network using the second historical measurement voltage, the second historical measurement current, the second historical measurement temperature, the second historical measurement pressure and its corresponding second historical measurement voltage-current, the second historical measurement efficiency, and the second historical measurement life, respectively, to obtain a second efficiency characteristic curve, a second life characteristic curve, and a second voltage-current characteristic curve corresponding to the second characteristic curve.

[0073] (1) Establishment of characteristic curve database

[0074] In the embodiments of the present disclosure and other possible embodiments, first, the characteristic curves (first characteristic curve and second characteristic curve) corresponding to SOFC and PEMFC are measured and modeled respectively. For SOFC, the temperature range is set to 200°C to 1000°C, the pressure range is 1atm to 10atm, and the load range is from 0A to the maximum operating current; for PEMFC, the temperature range is 20°C to 80°C, the pressure range is 1atm to 10atm, and the load range is also 0A to the maximum operating current. Under the temperature range, pressure range and load range corresponding to SOFC and PEMFC, the measurement parameters corresponding to SOFC and PEMFC include: measured voltage-current (VI) characteristics, measured efficiency η (expressed as a percentage), measured life 1 / L (in hours) and other parameters. In the modeling process, polynomial fitting and machine learning algorithms (such as neural networks) are used to establish a characteristic curve parameter estimation model. For example, the voltage V corresponding to SOFC and PEMFC is fitted using a cubic polynomial as the dependent variable, and the measured current I, measured temperature T, and pressure P corresponding to SOFC and PEMFC are used as independent variables. The expression of the voltage fitting curve is:

[0075] V=β0+β1I+β2T+β3P+β4I 2 +β5T 2 +β6P 2 +β7IT+β8IP+β9TP+

[0076] β 10 I 3 +β 11 T 3 +β 12 P 3 +∈.

[0077] Among them, β0,β1,…,β 12 are the polynomial coefficients, ∈ is the error term, and the coefficients of the polynomial are determined by the least squares method, and the voltage fitting curves corresponding to SOFC and PEMFC can be obtained respectively.

[0078] In the embodiments of the present disclosure and other possible embodiments, when the neural network algorithm is used respectively, a neural network containing three layers corresponding to SOFC and PEMFC is constructed, the input layer includes four parameters of voltage V, current I, temperature T, pressure P corresponding to SOFC or voltage V, current I, temperature T, pressure P corresponding to PEMFC, the hidden layer has 8 neurons, and the output layer includes efficiency η, life 1 / L and VI (VI characteristics) corresponding to SOFC or efficiency η, life 1 / L and VI (VI characteristics) corresponding to PEMFC. Furthermore, the efficiency η value, life 1 / L value and VI value corresponding to SOFC are used to form the first characteristic curve corresponding to SOFC, including: a first efficiency characteristic curve, a first life characteristic curve and a first VI characteristic curve; at the same time, the efficiency η value, life 1 / L value and VI value corresponding to PEMFC are used to form the second characteristic curve corresponding to PEMFC, including: a second efficiency characteristic curve, a second life characteristic curve and a second VI characteristic curve.

[0079] In the embodiments of the present disclosure and other possible embodiments, in the process of training the neural network corresponding to SOFC and PEMFC respectively, the stochastic gradient descent algorithm is used to divide the data into a training set, a validation set and a test set in a ratio of 7:2:1. The measured parameters (true label values) corresponding to SOFC and PEMFC and the characteristic curve parameter estimation model are calculated respectively. The corresponding efficiency error, life error and VI error (VI characteristic error) are output during the test or verification process until the error (for example, mean square error) of the characteristic curve parameter estimation model on the test set or verification set is less than 0.01. When establishing a dynamic characteristic database, a table containing parameters such as temperature T, pressure P, voltage V, current I, efficiency η, etc. is constructed, and a SQL or NoSQL database is used for data storage and management.

[0080] In the embodiment of the present disclosure and other possible embodiments, for PEMFC, the first setting condition is configured as a temperature range of 20°C to 80°C, a pressure range of 1atm to 10atm, and a load range corresponding to 0A to the maximum operating current. Meanwhile, for SOFC, the second setting condition is configured as a temperature range of 200°C to 1000°C, a pressure range of 1atm to 10atm, and a load range corresponding to 0A to the maximum operating current.

[0081] In the embodiments of the present disclosure and other possible embodiments, the first historical measured current, the first historical measured temperature, and the first historical measured pressure are configured as the measured current I, measured temperature T, and measured pressure P corresponding to the PEMFC; the corresponding first historical measured voltage-current, the first historical measured efficiency, and the first historical measured life force are configured as the measured voltage-current (VI) characteristics (which can be simply referred to as voltage-current or voltage-current / current-voltage), the measured efficiency η (expressed as a percentage), and the measured life 1 / L (in hours) corresponding to the PEMFC. The first historical measured current, the first historical measured temperature, the first historical measured pressure, and their corresponding first historical measured voltage are fitted using a polynomial fitting method to obtain a corresponding first voltage fitting curve;

[0082] In the embodiments of the present disclosure and other possible embodiments, the second historical measured current, the second historical measured temperature, and the second historical measured pressure are configured as the measured current I, the measured temperature T, and the measured pressure P corresponding to the SOFC; the corresponding second historical measured voltage-current, the second historical measured efficiency, and the second historical measured life are configured as the voltage-current (VI) characteristics (which can be referred to as voltage-current or voltage-current / current-voltage) corresponding to the SOFC, the measured efficiency η (expressed as a percentage), and the measured life 1 / L (in hours). The second historical measured current, the second historical measured temperature, the second historical measured pressure, and their corresponding second historical measured voltage are fitted using a polynomial fitting method to obtain a corresponding second voltage fitting curve.

[0083] In an embodiment of the present disclosure, before obtaining the currently detected or updated first temperature and the first pressure, a method for determining the currently detected or updated first temperature and the first pressure includes: using a sliding window of a set size and its corresponding weights to process the first set of temperatures and the first set of pressures corresponding to the sub-exchange membrane fuel cell at a time before the current moment, respectively, to obtain the currently detected or updated first temperature and the first pressure.

[0084] In an embodiment of the present disclosure, before obtaining the currently detected or updated second temperature and second pressure, a method for determining the currently detected or updated second temperature and second pressure includes: using a sliding window of a set size and its corresponding weights to process the second set of temperatures and second set of pressures corresponding to the solid oxide fuel cell at a time before the current moment, respectively, to obtain the currently detected or updated second temperature and second pressure.

[0085] (2) Real-time monitoring and data update

[0086] In the embodiments of the present disclosure and other possible embodiments, temperature sensors, pressure sensors, voltage sensors and current sensors are arranged in the system and installed at key parts of SOFC and PEMFC, such as inlet, outlet or reaction chamber, to ensure that relevant parameters (temperature T, pressure P, voltage V, current I) can be accurately obtained. For example, for SOFC, the temperature range is set to 200°C to 1000°C, the pressure range is 1atm to 10atm, and the load range corresponds to 0A to the maximum operating current; for PEMFC, the temperature range is 20°C to 80°C, the pressure range is 1atm to 10atm, and the load range is also 0A to the maximum operating current. The update mechanism of relevant parameters includes: sliding window algorithm. Among them, the sliding window algorithm (SlidingWindow Algorithm) is a method for real-time data processing, which selects and processes data through a fixed-size time window. The sliding window algorithm selects a window size of 10 seconds and uses a linear weighted method to calculate the updated data. The formula is:

[0087]

[0088] Among them, w k is the data of the kth second (related parameter D k ) weight, satisfying D 更新 Indicates the updated related parameters; D t-k is the relevant parameter corresponding to the current time t k seconds ago, such as the first temperature or the first pressure or the second temperature or the second pressure.

[0089] In the embodiments of the present disclosure and other possible embodiments, D t-k is the relevant parameter corresponding to the current time t k seconds ago, for example, the first temperature or the first pressure corresponding to the sub-exchange membrane fuel cell at the time before the current time or the second set of temperatures or the second set of pressures corresponding to the solid oxide fuel cell at the time before the current time; when D t-k When configured as the first temperature or the first pressure corresponding to the sub-exchange membrane fuel cell at the time before the current moment or the second set of temperatures or the second set of pressures corresponding to the solid oxide fuel cell at the time before the current moment, the corresponding D 更新 It corresponds to the first temperature or first pressure currently detected or updated, or the second temperature and second pressure currently detected or updated.

[0090] Step S102: determining a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve at the first temperature and the first pressure; and / or determining a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve at the second temperature and the second pressure.

[0091] In the embodiments of the present disclosure and other possible embodiments, after the first set of temperatures and the first set of pressures corresponding to the sub-exchange membrane fuel cell at the time before the current time are processed using a sliding window of a set size and its corresponding weights to obtain the currently detected or updated first temperature and the first pressure, based on the currently detected or updated first temperature and the first pressure, the first efficiency characteristic curve and the first life characteristic curve in the first characteristic curve at the first temperature and the first pressure are determined to obtain the first efficiency characteristic curve and the first life characteristic curve corresponding to the currently detected or updated first temperature and the first pressure in the first characteristic curve.

[0092] In the embodiments of the present disclosure and other possible embodiments, after processing the second set of temperatures and the second set of pressures corresponding to the solid oxide fuel cell at the time before the current time using a sliding window of a set size and its corresponding weights to obtain the second temperature and the second pressure currently detected or updated, the second efficiency characteristic curve and the second life characteristic curve in the second characteristic curve at the second temperature and the second pressure are determined to obtain the second efficiency characteristic curve and the second life characteristic curve corresponding to the second temperature and the second pressure currently detected or updated in the second characteristic curve.

[0093] Step S103: Determine the first optimal power corresponding to the proton exchange membrane fuel cell and / or the second optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure.

[0094] In an embodiment of the present disclosure, the method for determining the first optimal power corresponding to the proton exchange membrane fuel cell and / or the second optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure includes: determining a first group of efficiencies corresponding to the first efficiency characteristic curve and a first group of life corresponding to the first life characteristic curve; constructing a first multi-parameter function corresponding to the first efficiency characteristic curve and the first life characteristic curve in the first characteristic curve; wherein the first group of parameters corresponding to the first multi-parameter function includes: a first voltage to be optimized and a first current to be optimized; constructing a first objective function corresponding to the first multi-parameter function with the goal of maximizing the efficiency and life in the first group of efficiencies and the first group of life; solving the first objective function with the first voltage-current characteristic curve in the first characteristic curve as the first constraint condition to obtain the first optimal power; and / or, wherein the first optimal power is configured as the product of the first optimal voltage corresponding to the first voltage to be optimized and the first optimal current corresponding to the first current to be optimized.

[0095] In the embodiment of the present disclosure and other possible embodiments, the construction includes configuring the first multi-parameter function corresponding to the first efficiency characteristic curve and the first life characteristic curve in the first characteristic curve as η, 1 / L(V, I, T, P), that is, η(V, I, T, P) and 1 / L(V, I, T, P); wherein T and P represent the first temperature and the first pressure, respectively; and V and I represent the first voltage to be optimized and the first current to be optimized in the first group of parameters corresponding to the first multi-parameter function, respectively. With the goal of maximizing the efficiency and life in the first group of efficiencies and the first group of life, the first objective function corresponding to the first multi-parameter function is configured as maxη, 1 / L(V, I, T, P).

[0096] In an embodiment of the present disclosure, in the method of solving the first objective function to obtain the first optimal power by taking the first voltage-current characteristic curve in the first characteristic curve as the first constraint condition, the first constraint condition further includes at least one or more of: a first voltage to be optimized maximum value and a first voltage to be optimized minimum value corresponding to the first voltage to be optimized, a first current to be optimized maximum value and a first current to be optimized minimum value corresponding to the first current to be optimized, a first temperature maximum value and a first temperature minimum value corresponding to the first temperature, and a first pressure maximum value and a first pressure minimum value corresponding to the first pressure.

[0097] In the embodiment of the present disclosure and other possible embodiments, the constraint conditions corresponding to the maximum value of the first voltage to be optimized and the minimum value of the first voltage to be optimized corresponding to the first voltage to be optimized are configured as V min ≤V≤V max , the constraint conditions corresponding to the first maximum value of the current to be optimized and the first minimum value of the current to be optimized corresponding to the first current to be optimized are configured as I min ≤I≤I max , the constraint conditions corresponding to the first temperature maximum value and the first temperature minimum value corresponding to the first temperature are T min ≤T≤T max , the constraint conditions corresponding to the first pressure maximum value and the first pressure minimum value configuration corresponding to the first pressure are P min ≤P≤P max .

[0098] In an embodiment of the present disclosure, the method for solving the first objective function to obtain the first optimal power includes: constructing a first vector for each particle corresponding to the first objective function; the first vector includes at least: a first power allocation ratio and a first voltage to be optimized, and one or more of a first current to be optimized; under a set variable first neighborhood radius, updating the speed and position corresponding to each particle in the first vector until the first efficiency and first life corresponding to each particle in the first vector are maximized, and obtaining the first power allocation ratio, the first optimal voltage corresponding to the first voltage to be optimized, and the first optimal current corresponding to the first current to be optimized.

[0099] In the embodiment of the present disclosure and other possible embodiments, the first vector may be configured as: x=[α PEMFC ,V PEMFC ,I PEMFC ]. Among them, α PEMFC ,V PEMFC ,I PEMFC They respectively represent the first power allocation ratio, the first voltage to be optimized, and the first current to be optimized.

[0100] In the embodiments of the present disclosure, the method of updating the speed and position corresponding to each particle in the first vector under the first neighborhood radius of the setting change until the first efficiency and the first life corresponding to each particle in the first vector are maximized includes: randomly initializing the first position and the first speed, the first individual optimal position and the first global optimal position of each particle in the first vector; updating the first position, the first speed, the first individual optimal position and the first global optimal position of each particle using the continuously updated inertia weight and the acceleration constant; during the updating process, determining the first fitness corresponding to the first efficiency and the first life based on the updated first position and the updated first speed of each particle; stopping the updating until the first fitness reaches the first set fitness; determining the first position to stop updating based on the first individual optimal position and the first global optimal position to stop updating; configuring each particle corresponding to the first position to stop updating to one or more of the first power allocation ratio, the first optimal voltage corresponding to the first voltage to be optimized and the first optimal current corresponding to the first current to be optimized; wherein the first set fitness is configured as a function corresponding to the first efficiency and the first life; the first set fitness is configured as the first value corresponding to the first efficiency and the first life corresponding to each particle when they are maximized.

[0101] In the embodiment of the present disclosure and other possible embodiments, the mathematical expressions of the continuously updated inertia weight and acceleration constant or their corresponding adaptive formulas are:

[0102] Among them, w max and w min are the maximum and minimum values ​​of the first inertia weight, c1,max and c 1,min is the maximum and minimum value of the first cognitive acceleration constant, c 2,max and c 2,min is the maximum and minimum value of the first social acceleration constant, and Tp is the first maximum number of iterations. The multi-level search mechanism introduces a double-layer particle swarm structure, including a local particle swarm and a global particle swarm, which optimize the high efficiency of SOFC and the fast response characteristics of PEMFC respectively, and coordinate at the global level to improve the performance of the overall system. The dynamic neighborhood strategy defines the first neighborhood radius r(t) that changes over time: r(t) = r max ×e -λt .

[0103] Among them, r maxis the initial first neighborhood radius, and λ is the first attenuation coefficient. The speed and position update formulas are combined with adaptive parameters and multi-level search mechanisms. Under the setting of the variable first neighborhood radius r(t), the i-th particle x in each particle x corresponding to time t+1 is i (The i-th element in the vector corresponding to each particle x above) Update speed v i (t+1) and update position x i (t+1), expressed as:

[0104] v i (t+1)=w(t)·v i (t)+c1(t)·rand1·(p best,i -x i (t))+c2(t)·rand2·(g best -x i (t));

[0105] x i (t+1)=x i (t)+v i (t+1).

[0106] Among them, v i (t) represents the speed corresponding to time t; rand1 represents the first random parameter; rand2 represents the second random parameter; p best,i Indicates that at time t x i (t) The corresponding first optimal position (individual optimal solution / individual optimal position); g best represents the i-th particle x in each particle x i The corresponding second optimal position (global optimal solution / global optimal position).

[0107] The function corresponding to the first fitness is defined as:

[0108] Among them, η is the efficiency of the system, τ is the response speed, and L is the equipment life loss rate (fuel cell life loss rate). The algorithm process includes the initialization phase, the iterative optimization phase, and the output phase. In the initialization phase, the position and speed of the particle swarm are randomly initialized, the fitness of each particle is evaluated, and p is initialized. best,i and g best In the iterative optimization phase, the inertia weight w(t) and acceleration constants c1(t), c2(t) are updated according to the adaptive formula, the speed and position are updated, the boundary check and adjustment are performed, the new fitness is evaluated, and p is updated. best,i and g best , adjust the neighborhood radius r(t) until the termination condition is met. In the output stage, the power allocation strategy of the global optimal particle is output The optimal voltage V corresponding to PEMFC* (The first optimal voltage V PEMFC ) and optimal current I * (The first optimal current I PEMFC ); The set power configuration is the power load corresponding to the power grid.

[0109] In an embodiment of the present disclosure, a second group of efficiencies corresponding to the second efficiency characteristic curve and a second group of lifespans corresponding to the second life characteristic curve are determined; a second multi-parameter function corresponding to the second efficiency characteristic curve and the second life characteristic curve in the second characteristic curve is constructed; wherein the second group of parameters corresponding to the second multi-parameter function include: a second voltage to be optimized and a second current to be optimized; with the goal of maximizing the efficiency and lifespan in the second group of efficiencies and the second group of lifespans, a second objective function corresponding to the second multi-parameter function is constructed; with the second voltage-current characteristic curve in the second characteristic curve as the second constraint condition, the second objective function is solved to obtain the second optimal power; and / or, wherein the second optimal power is configured as the product of the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized.

[0110] In the embodiment of the present disclosure and other possible embodiments, the construction includes that the second multi-parameter function corresponding to the second efficiency characteristic curve and the second life characteristic curve in the second characteristic curve is configured as η, 1 / L (V, I, T, P), that is, η (V, I, T, P) and 1 / L (V, I, T, P); wherein T and P at this time represent the second temperature and the second pressure respectively; and V and I at this time represent the second voltage to be optimized and the second current to be optimized in the second group of parameters corresponding to the second multi-parameter function respectively. At this time, with the goal of maximizing the efficiency and life in the second group of efficiencies and the second group of life, the second objective function corresponding to the second multi-parameter function is configured as maxη, 1 / L (V, I, T, P).

[0111] Similarly, in an embodiment of the present disclosure, in the method of solving the second objective function to obtain the second optimal power using the second voltage-current characteristic curve in the second characteristic curve as the second constraint condition, the second constraint condition also includes at least one or more of: a second voltage to be optimized maximum value and a second voltage to be optimized minimum value corresponding to the second voltage to be optimized, a second current to be optimized maximum value and a second current to be optimized minimum value corresponding to the second current to be optimized, a second temperature maximum value and a second temperature minimum value corresponding to the second temperature, and a second pressure maximum value and a second pressure minimum value corresponding to the second pressure.

[0112] In an embodiment of the present disclosure, the method for solving the second objective function to obtain the second optimal power includes: constructing a second vector for each particle corresponding to the second objective function; the second vector includes at least: a second power allocation ratio and one or more of a second voltage to be optimized and a second current to be optimized; under a set variable second neighborhood radius, updating the speed and position corresponding to each particle in the second vector until the second efficiency and second life corresponding to each particle in the second vector are maximized, and obtaining one or more of the second power allocation ratio, the second optimal voltage corresponding to the second voltage to be optimized, and the second optimal current corresponding to the second current to be optimized.

[0113] In the embodiment of the present disclosure and other possible embodiments, the second vector at this time can be configured as: x=[α SOFC ,V SOFC ,I SOFC ]. Among them, α SOFC ,V SOFC ,I SOFC They respectively represent the second power allocation ratio, the second voltage to be optimized, and the second current to be optimized.

[0114] The method of updating the speed and position corresponding to each particle in the second vector under the second neighborhood radius of the set change until the second efficiency and the second life corresponding to each particle in the second vector are maximized comprises: randomly initializing the second position and the second speed, the second individual optimal position and the second global optimal position of each particle in the second vector; updating the second position, the second speed, the second individual optimal position and the second global optimal position of each particle by using the continuously updated inertia weight and the acceleration constant; during the updating process, determining the second fitness corresponding to the second efficiency and the second life of each particle based on the updated second position and the updated second speed of each particle; stopping the updating until the second fitness reaches the second set fitness; determining the second position where the updating is stopped based on the second individual optimal position and the second global optimal position where the updating is stopped; configuring each particle corresponding to the second position where the updating is stopped to one or more of the second power allocation ratio, the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized; wherein the second set fitness is configured as a function corresponding to the second efficiency and the second life; and the second set fitness is configured as a second value corresponding to when the second efficiency and the second life corresponding to each particle are maximized.

[0115] In the embodiment of the present disclosure and other possible embodiments, the mathematical expressions of the continuously updated inertia weight and acceleration constant or their corresponding adaptive formulas are:

[0116] Among them, w max and w min are the maximum and minimum values ​​of the second inertia weight, c1,max and c 1,min is the maximum and minimum value of the second cognitive acceleration constant, c 2,max and c 2,min is the maximum and minimum value of the second social acceleration constant, and Tp is the second maximum number of iterations. The multi-level search mechanism introduces a double-layer particle swarm structure, including a local particle swarm and a global particle swarm, which optimize the high efficiency of SOFC and the fast response characteristics of PEMFC respectively, and coordinate at the global level to improve the performance of the overall system. The dynamic neighborhood strategy defines the second neighborhood radius r(t) that changes over time: r(t) = r max ×e -λt .

[0117] Among them, r max is the initial second neighborhood radius, and λ is the second attenuation coefficient. The speed and position update formulas are combined with adaptive parameters and multi-level search mechanisms. Under the set variable second neighborhood radius r(t), the i-th particle x in each particle x corresponding to time t+1 is i (The i-th element in the vector corresponding to each particle x above) Update speed v i (t+1) and update position x i (t+1), expressed as:

[0118] v i (t+1)=w(t)·v i (t)+c1(t)·rand1·(p best,i -x i (t))+c2(t)·rand2·(g best -x i (t));

[0119] x i (t+1)=x i (t)+v i (t+1).

[0120] Among them, v i (t) represents the speed corresponding to time t; rand1 represents the first random parameter; rand2 represents the second random parameter; p best,i Indicates that at time t x i (t) The corresponding first optimal position (individual optimal solution / individual optimal position); g best represents the i-th particle x in each particle x i The corresponding second optimal position (global optimal solution / global optimal position).

[0121] The function corresponding to the second fitness is defined as:

[0122] Among them, η is the efficiency of the system, τ is the response speed, and L is the equipment life loss rate (fuel cell life loss rate). The algorithm process includes the initialization phase, the iterative optimization phase, and the output phase. In the initialization phase, the position and speed of the particle swarm are randomly initialized, the fitness of each particle is evaluated, and p is initialized. best,i and g best In the iterative optimization phase, the inertia weight w(t) and acceleration constants c1(t), c2(t) are updated according to the adaptive formula, the speed and position are updated, the boundary check and adjustment are performed, the new fitness is evaluated, and p is updated. best,i and g best , adjust the neighborhood radius r(t) until the termination condition is met. In the output stage, the power allocation strategy of the global optimal particle is output The optimal voltage V corresponding to SOFC * (The second optimal voltage V SODC ) and optimal current I * (The second optimal current I SOFC ); The set power configuration is the power load corresponding to the power grid.

[0123] In an embodiment of the present disclosure, a first group of efficiencies corresponding to the first efficiency characteristic curve and a first group of lifespans corresponding to the first lifespan characteristic curve are determined, and a second group of efficiencies corresponding to the second efficiency characteristic curve and a second group of lifespans corresponding to the second lifespan characteristic curve are determined; a third multi-parameter function including the first efficiency characteristic curve and the first lifespan characteristic curve in the first characteristic curve, and the second efficiency characteristic curve and the second lifespan characteristic curve in the second characteristic curve is constructed; wherein a third group of parameters corresponding to the third multi-parameter function includes: a first voltage to be optimized, a first current to be optimized, a second voltage to be optimized, and a second current to be optimized; with the goal of maximizing the efficiency and lifespan in the first group of efficiencies and the first group of lifespans, and / or the second group of efficiencies and the second group of lifespans, a third objective function corresponding to the third multi-parameter function is constructed; with the first voltage-current characteristic curve in the first characteristic curve and / or the second voltage-current characteristic curve in the second characteristic curve as the third constraint condition, the third objective function is solved to obtain the first optimal power and / or the second optimal power.

[0124] In an embodiment of the present disclosure, the first optimal power configuration corresponding to the third objective function is the product of the first optimal voltage corresponding to the first voltage to be optimized and the first optimal current corresponding to the first current to be optimized; and / or, the second optimal power configuration corresponding to the third objective function is the product of the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized.

[0125] In the embodiment of the present disclosure and other possible embodiments, the construction includes the third multi-parameter function corresponding to the first efficiency characteristic curve and the first life characteristic curve in the first characteristic curve, and the second efficiency characteristic curve and the second life characteristic curve in the second characteristic curve, which is configured as η, 1 / L (V, I, T, P), that is, η (V, I, T, P) and 1 / L (V, I, T, P). Wherein, with the maximization of the efficiency and life in the first group of efficiencies and the first group of life as the goal, the first objective function corresponding to the first multi-parameter function is configured as maxη, 1 / L (V, I, T, P).

[0126] In an embodiment of the present disclosure, the method for solving the third objective function to obtain the first optimal power and / or the second optimal power includes: constructing a third vector for each particle corresponding to the third objective function; the third vector includes at least: a first vector parameter corresponding to a proton exchange membrane fuel cell and a second vector parameter corresponding to a solid oxide fuel cell; wherein the first vector parameter includes: a first power allocation ratio and one or more of a first voltage to be optimized and a first current to be optimized; wherein the second vector parameter includes: a second power allocation ratio and one or more of a second voltage to be optimized and a second current to be optimized; under a set variable third neighborhood radius, the speed and position corresponding to each particle in the third vector are updated until the third efficiency and third life corresponding to each particle in the third vector are maximized, thereby obtaining one or more of the third power allocation ratio, the third optimal voltage corresponding to the third voltage to be optimized, and the third optimal current corresponding to the third current to be optimized.

[0127] In the embodiment of the present disclosure and other possible embodiments, the first vector may be configured as: x=[α SOFC ,α PEMFC ,V SOFC ,I SOFC ,V PEMFC ,I PEMFC ].

[0128] Among them, α SOFC ,α PEMFC ,V SOFC ,I SOFC ,V PEMFC ,IPEMFC They respectively represent the first power allocation ratio corresponding to SOFC, the second power allocation ratio corresponding to PEMFC, the second voltage to be optimized, the second current to be optimized, the first voltage to be optimized, and the first current to be optimized.

[0129] In an embodiment of the present disclosure, the method of updating the speed and position corresponding to each particle in the third vector under a set variable third neighborhood radius until the third efficiency and third life corresponding to each particle in the third vector are maximized includes: randomly initializing the third position and third speed, the third individual optimal position and the third global optimal position of each particle in the third vector; updating the third position, third speed, the third individual optimal position and the third global optimal position of each particle using continuously updated inertia weights and acceleration constants; during the updating process, determining the corresponding third position and third speed of each particle based on the updated third position and updated third speed of each particle a third fitness of the third efficiency and the third lifespan; stopping updating until the third fitness reaches a third set fitness; determining the third position for stopping updating based on the third individual optimal position for stopping updating and the third global optimal position; configuring each particle corresponding to the third position for stopping updating to one or more of the third power allocation ratio, the third optimal voltage corresponding to the third voltage to be optimized, and the third optimal current corresponding to the third current to be optimized; wherein the third set fitness is configured as a function of the third efficiency and the third lifespan; the third set fitness is configured as a third value corresponding to when the third efficiency and the third lifespan corresponding to each particle are maximized.

[0130] Similarly, in an embodiment of the present disclosure, in the method of solving the third objective function to obtain the first optimal power and / or the second optimal power by using the first voltage-current characteristic curve in the first characteristic curve and / or the second voltage-current characteristic curve in the second characteristic curve as the third constraint condition, the third constraint condition also includes at least: one or more of the first voltage to be optimized maximum value and the first voltage to be optimized minimum value corresponding to the first voltage to be optimized, the first current to be optimized maximum value and the first current to be optimized minimum value corresponding to the first current to be optimized, the first temperature maximum value and the first temperature minimum value corresponding to the first temperature, the first pressure maximum value and the first pressure minimum value corresponding to the first pressure, the second voltage to be optimized maximum value and the second voltage to be optimized minimum value corresponding to the second voltage to be optimized, the second current to be optimized maximum value and the second current to be optimized minimum value corresponding to the second current to be optimized, the second temperature maximum value and the second temperature minimum value corresponding to the second temperature, and the second pressure maximum value and the second pressure minimum value corresponding to the second pressure.

[0131] In the embodiment of the present disclosure and other possible embodiments, the mathematical expressions of the continuously updated inertia weight and acceleration constant or their corresponding adaptive formulas are:

[0132] Among them, w max and w min are the maximum and minimum values ​​of the third inertia weight, c1,max and c 1,min is the maximum and minimum value of the third cognitive acceleration constant, c 2,max and c 2,min is the maximum and minimum value of the third social acceleration constant, and Tp is the third maximum number of iterations. The multi-level search mechanism introduces a double-layer particle swarm structure, including a local particle swarm and a global particle swarm, which optimize the high efficiency of SOFC and the fast response characteristics of PEMFC respectively, and coordinate at the global level to improve the performance of the overall system. The dynamic neighborhood strategy defines the third neighborhood radius r(t) that is set to change over time: r(t) = r max ×e -λt .

[0133] Among them, r max is the initial third neighborhood radius, and λ is the third attenuation coefficient. The speed and position update formulas are combined with adaptive parameters and multi-level search mechanisms. Under the set variable third neighborhood radius r(t), the i-th particle x in each particle x corresponding to time t+1 is i (The i-th element in the vector corresponding to each particle x above) Update speed v i (t+1) and update position x i (t+1), expressed as:

[0134] v i (t+1)=w(t)·v i (t)+c1(t)·rand1·(p best,i -x i (t))+c2(t)·rand2·(g best -x i (t));

[0135] x i (t+1)=x i (t)+v i (t+1).

[0136] Among them, v i (t) represents the speed corresponding to time t; rand1 represents the first random parameter; rand2 represents the second random parameter; p best,i Indicates that at time t x i (t) The corresponding first optimal position (individual optimal solution / individual optimal position); gbest represents the i-th particle x in each particle x i The corresponding second optimal position (global optimal solution / global optimal position).

[0137] The function corresponding to the second fitness is defined as:

[0138] Among them, η is the efficiency of the system, τ is the response speed, and L is the equipment life loss rate (fuel cell life loss rate). The algorithm process includes the initialization phase, the iterative optimization phase, and the output phase. In the initialization phase, the position and speed of the particle swarm are randomly initialized, the fitness of each particle is evaluated, and p is initialized. best,i and g best In the iterative optimization phase, the inertia weight w(t) and acceleration constants c1(t), c2(t) are updated according to the adaptive formula, the speed and position are updated, the boundary check and adjustment are performed, the new fitness is evaluated, and p is updated. best,i and g best , adjust the neighborhood radius r(t) until the termination condition is met.

[0139] In the output stage, the power allocation strategy of the global optimal particle is output The optimal voltage V for SOFC and PEMFC * (The first optimal voltage V PEMFC and the second optimal voltage V SOFC ) and optimal current I * (The first optimal current I PEMFC and the second optimal current I SOFC ); The set power configuration is the power load corresponding to the power grid.

[0140] (3) Dynamic optimal curve tracking control strategy

[0141] In the embodiment of the present disclosure and other possible embodiments, the calculation of the optimal operating point is based on the characteristic curve database, and the optimal voltage V is determined using an interpolation or optimization algorithm. * and the optimal current I * .

[0142] Among them, the interpolation algorithm adopts a linear interpolation method, and finds adjacent data points (voltage V, current I) of the first characteristic curve (first efficiency characteristic curve, first life characteristic curve) and the second characteristic curve (second efficiency characteristic curve, second life characteristic curve) corresponding to SOFC and PEMFC in the characteristic curve database according to the currently monitored (updated) temperature T and the previously monitored (updated) pressure P, and calculates the corresponding optimal operating point (optimal voltage V*, optimal current I*) through a linear relationship.

[0143] The optimization algorithm takes the efficiency η and the life 1 / L corresponding to the fuel cell life loss rate L as the objective function, considers the constraints such as the voltage V, current I, updated temperature T, and updated pressure P corresponding to SOFC and PEMFC, and solves the optimal operating point (optimal voltage V * , optimal current I * ).

[0144] The formula corresponding to the objective function is expressed as: maxη,1 / L(V,I,T,P).

[0145] Subject to (constraints): V min ≤V≤V max ;I min ≤I≤I max ; T min ≤T≤T max ;P min ≤P≤P max ;VI characteristic curve (first VI characteristic curve and / or second VI characteristic curve).

[0146] The above formula is used to obtain the optimal voltage V* and the optimal current I*, and the optimal operating point (optimal voltage V * , optimal current I * ); wherein the first optimal operating point of SOFC is configured as the first optimal voltage V * , the first optimal current I * At the same time, the second optimal operating point of PEMFC is configured as the second optimal voltage V * , the second optimal current I * Furthermore, based on the optimal voltage V corresponding to SOFC and PEMFC, * And the optimal current I * Determine the corresponding first optimal power and second optimal power. Based on the first optimal power, second optimal power and set power corresponding to SOFC and PEMFC; wherein the sum of the first optimal power and the second optimal power is the set power. Specifically, by adjusting the fuel supply and output current of SOFC and PEMFC corresponding to the fuel cell respectively, the adjustment accuracy of the first optimal power and the second optimal power reaches ±1%.

[0147] When the load fluctuates, PEMFC responds first and SOFC adjusts gradually, with the response time being less than 1 second for PEMFC and less than 10 seconds for SOFC. The power configuration is set to the power load corresponding to the grid.

[0148] (4) Collaborative control algorithm (determine the optimal voltage V* and optimal current I* based on maxη,1 / L(V,I,T,P) and its corresponding constraints)

[0149] In the embodiments of the present disclosure and other possible embodiments, in order to better adapt to the different characteristics of solid oxide fuel cells (SOFC) and proton exchange membrane fuel cells (PEMFC), the present invention proposes an intelligent particle swarm optimization algorithm. Based on the traditional particle swarm optimization algorithm (PSO), the algorithm combines the high efficiency and response speed of the two fuel cells, and realizes more efficient power allocation strategy optimization by adaptively adjusting parameters and introducing a multi-level search mechanism. The design concept (method) of the intelligent particle swarm optimization algorithm includes: adaptively adjusting the inertia weight and acceleration constant, adaptively adjusting the inertia weight w and acceleration constants c1, c2 of the PSO according to the current load change rate of the system and the working state of the fuel cell, so as to balance the global search capability and the local search capability. At the same time, the algorithm introduces a multi-level search mechanism, constructs a hierarchical particle swarm, independently optimizes SOFC and PEMFC respectively, and coordinates at the global level to make full use of the characteristics of the two batteries, solid oxide fuel cells (SOFC) and proton exchange membrane fuel cells (PEMFC). In addition, the dynamic neighborhood strategy adjusts the search range of particles according to real-time monitoring data to improve the convergence speed and optimization accuracy of the algorithm.

[0150] In the algorithm design, each particle x is represented as a vector containing system parameters (voltage V, current I, updated temperature T, and updated pressure P corresponding to SOFC and PEMFC): x = [α SOFC ,α PEMFC ,V SOFC ,I SOFC ,V PEMFC ,I PEMFC ].

[0151] First, the corresponding efficiency characteristic curve and life characteristic curve are determined using the currently detected (updated) operating temperature and pressure corresponding to SOFC and PEMFC, respectively. Among them, the efficiency characteristic curve and life characteristic curve corresponding to SOFC are the first efficiency characteristic curve and the first life characteristic curve; the efficiency characteristic curve and life characteristic curve corresponding to PEMFC are the second efficiency characteristic curve and the second life characteristic curve. Then, the first efficiency characteristic curve and the first life characteristic curve corresponding to SOFC are used to determine the first optimal voltage V* and the first optimal current I* corresponding to the maximization of the efficiency and life 1 / L corresponding to SOFC; at the same time, the second efficiency characteristic curve and the second life characteristic curve corresponding to PEMFC are used to determine the second optimal voltage V* and the second optimal current I* corresponding to the maximization of the efficiency and life 1 / L corresponding to PEMFC.

[0152] Among them, α SOFC and α PEMFCRepresent the power allocation ratio of SOFC and PEMFC respectively, V SOFC、 I SOFC Represent the operating voltage and current of SOFC, V PEMFC、 I PEMFC Respectively represent the operating voltage and current of PEMFC, T SOFC ,P SOFC is the current (new) SOFC operating temperature and pressure, T PEMFC ,P PEMFC is the current (new) PEMFC operating temperature and pressure. The inertia weight w and acceleration constants c1, c2 (the first acceleration constant C1 and the second acceleration constant C2) are dynamically adjusted with the number of iterations t, and the corresponding mathematical expression of the adaptive formula is:

[0153] Among them, w max and w min are the maximum and minimum values ​​of the inertia weight, c1,max and c 1,min is the maximum and minimum value of the cognitive acceleration constant, c 2,max and c 2,min is the maximum and minimum value of the social acceleration constant, and Tp is the maximum number of iterations. The multi-level search mechanism introduces a double-layer particle swarm structure, including a local particle swarm and a global particle swarm, which optimize the high efficiency of SOFC and the fast response characteristics of PEMFC respectively, and coordinate at the global level to improve the performance of the overall system. The dynamic neighborhood strategy defines that the neighborhood radius r(t) is adjusted over time: r(t) = r max ×e -λt .

[0154] Among them, r max is the initial neighborhood radius, and λ is the attenuation coefficient. The speed and position update formulas are combined with adaptive parameters and multi-level search mechanisms. Under the set neighborhood radius r(t), the i-th particle x in each particle x corresponding to time t+1 is i (The i-th element in the vector corresponding to each particle x above) Update speed v i (t+1) and update position x i (t+1), expressed as:

[0155] v i (t+1)=w(t)·v i (t)+c1(t)·rand1·(p best,i -x i (t))+c2(t)·rand2·(g best -x i (t));

[0156] xi (t+1)=x i (t)+v i (t+1).

[0157] Among them, v i (t) represents the speed corresponding to time t; rand1 represents the first random parameter; rand2 represents the second random parameter; p best,i Indicates that at time t x i (t) The corresponding first optimal position (individual optimal solution / individual optimal position); g best represents the i-th particle x in each particle x i The corresponding second optimal position (global optimal solution / global optimal position).

[0158] The function corresponding to fitness is defined as:

[0159] Among them, η is the efficiency of the system, τ is the response speed, and L is the equipment life loss rate (fuel cell life loss rate). The algorithm process includes the initialization phase, the iterative optimization phase, and the output phase. In the initialization phase, the position and speed of the particle swarm are randomly initialized, the fitness of each particle is evaluated, and p is initialized. best,i and g best In the iterative optimization phase, the inertia weight w(t) and acceleration constants c1(t), c2(t) are updated according to the adaptive formula, the speed and position are updated, the boundary check and adjustment are performed, the new fitness is evaluated, and p is updated. best,i and g best , adjust the neighborhood radius r(t) until the termination condition is met. In the output stage, the power allocation strategy of the global optimal particle is output The optimal voltage V for SOFC and PEMFC * (The first optimal voltage and the second optimal voltage V SOFC、 V PEMFC ) and optimal current I * (The first optimal current and the second optimal current I PEMFC ,I SOFC ); The set power configuration is the power load corresponding to the power grid.

[0160] (5) Forecasting and pre-scheduling

[0161] A load forecasting model is established based on historical data (voltage V, current I, updated temperature T, updated pressure P and other data corresponding to SOFC and PEMFC) and machine learning algorithms (such as support vector machine SVM). The load data (electricity load corresponding to the power grid) of the first historical time period in the past (such as the past 1 hour) is sampled at set time intervals (such as 5 minutes), and the electricity load corresponding to the power grid in the second historical time period in the future (5 minutes to 1 hour) is predicted.

[0162] The pre-scheduling strategy adjusts the SOFC working state in advance of the second historical time period (5 minutes to 1 hour), and according to the corresponding changes in the predicted power load, if the predicted power load increases by more than a set value (for example, 10%) compared with the current power load, the fuel supply of the SOFC is increased in advance to gradually adjust its power to a higher power state; if the predicted load decreases by more than a set value (for example, 10%), the fuel supply of the SOFC is reduced in advance to gradually adjust it to a lower power state.

[0163] To further illustrate the practical application of the present invention, the following describes in detail the working mode of the system under different load conditions through several operation examples:

[0164] In an embodiment of the present disclosure, it also includes: when the power load demand corresponding to the power grid is stable, the first optimal power corresponding to the solid oxide fuel cell is determined based on the first efficiency characteristic curve at the first temperature and the first pressure; at this time, the proton exchange membrane fuel cell is in a standby or low-power operation state; wherein the low-power operation state is configured to be less than the rated power of the first set ratio. Wherein, the first set ratio is set to 10%. Wherein, when the fluctuation of the power load corresponding to the power grid is less than the first set fluctuation value, the fuel supply speed of the solid oxide fuel cell is adjusted to maintain the first optimal power within the first set power interval. Wherein, the first interval value and the second interval value of the first set power interval are configured as 95%*first optimal power and 105%*first optimal power, respectively.

[0165] Example 1: The power load corresponding to the power grid operates smoothly

[0166] In the embodiments of the present disclosure and other possible embodiments, the power load demand corresponding to the power grid is stable during nighttime operation. The first optimal operating point corresponding to the SOFC is determined, and is set to the first optimal voltage V* and the first optimal current I* corresponding to the first optimal operating point. The first output power P1 of the SOFC = first optimal voltage * first optimal current. At this time, the PEMFC is in standby or low-power operation (P2 <10% rated power). When the power load corresponding to the power grid fluctuates slightly (the fluctuation amplitude is less than 5%), the fuel supply speed of the SOFC is fine-tuned (adjusted), and its output power is adjusted to maintain the high efficiency area within a range of ±5%*P1.

[0167] In an embodiment of the present disclosure, it also includes: when the predicted growth rate of the power load corresponding to the power grid is greater than the preset growth rate, the power corresponding to the proton exchange membrane fuel cell is increased to a first set output power within a first set time, and the solid oxide fuel cell is increased to a second set output power within a second set time greater than the first set time; when the second set output power reaches the first set power, the output power of the proton exchange membrane fuel cell is gradually reduced at a set speed.

[0168] Example 2: The power load corresponding to the power grid increases rapidly

[0169] In the embodiments of the present disclosure and other possible embodiments, when the power load demand corresponding to the power grid increases rapidly in the morning, an upward trend in the power load corresponding to the power grid is detected (the predicted growth rate>10% / minute). Since SOFC cannot respond quickly, PEMFC switches to a high-efficiency fast response mode and quickly increases the output power P3 (the power increases from 0 to 80% of the instantaneous load demand within 1 second, the first set output power) to meet the instantaneous load demand; at the same time, SOFC gradually increases the output power to P4 (at a rate of 5% of the rated power per second, the second set output power). When SOFC reaches the corresponding optimal operating point (efficiency reaches more than 95% and fluctuates less than 1% within 1 minute), the system gradually reduces the output of PEMFC at a set rate of 10%*P3 per second, so that the system returns to a high-efficiency operating state.

[0170] In an embodiment of the present disclosure, it also includes: within a third set time, when the decrease rate of the power load corresponding to the power grid drops to a preset decrease rate, respectively calculating the first optimal power corresponding to the proton exchange membrane fuel cell and the second optimal power corresponding to the solid oxide fuel cell; within a fourth set time, reducing the power corresponding to the proton exchange membrane fuel cell to a third set output power; calculating the power reduction curve corresponding to the solid oxide fuel cell; and reducing the first optimal power according to the power reduction curve.

[0171] Example 3: The power load corresponding to the power grid drops rapidly

[0172] In the case of a sudden drop in the power load demand corresponding to the power grid (such as a factory shutdown), the first optimal operating point and the second optimal operating point corresponding to SOFC and PEMFC are calculated respectively, and determined according to the current temperature and pressure parameters using an optimization algorithm. Rapidly reduce the output power of PEMFC (the power is reduced to less than 10% of the rated power within 1 second, the third set output power) to avoid energy waste. Since SOFC responds slowly, calculate the best power reduction curve, and reduce the first optimal power corresponding to the first optimal operating point according to the power reduction curve; wherein, a polynomial fitting method is used to calculate the optimal power reduction curve from the current value (the power P(t) corresponding to the current time t) to the target value P(t+a), a=1,2,3,…,N, N is a positive integer greater than 1, to ensure that the temperature change rate is less than the set temperature change rate To avoid thermal shock to the battery, the power reduction curve is expressed as: P(t+1)=P(t)-k(t)·P(t), where k(t) is the power reduction rate corresponding to time t.

[0173] In an embodiment of the present disclosure, after a fifth set time, the first characteristic curve corresponding to the proton exchange membrane fuel cell and / or the second characteristic curve corresponding to the solid oxide fuel cell are updated.

[0174] Example 4: Dealing with equipment aging

[0175] After SOFC runs for a period of time (fifth set time), the performance begins to decline. The system updates the first characteristic curve corresponding to SOFC according to the real-time monitoring data (temperature T, pressure P, voltage V, current I corresponding to SOFC), re-measures the first characteristic curve of SOFC (including the first V-1 characteristic curve, the first efficiency curve, and the first life curve) every 10 hours, and integrates the new first characteristic curve into the characteristic curve database. At the same time, after PEMFC runs for a period of time, the performance begins to decline. The system updates the second characteristic curve corresponding to PEMFC according to the real-time monitoring data (temperature T, pressure P, voltage V, current I corresponding to PEMFC), re-measures the second characteristic curve of PEMFC (including the second V-1 characteristic curve, the second efficiency curve, and the second life curve) every 10 hours, and integrates the new second characteristic curve into the characteristic curve database. The first optimal operating point corresponding to SOFC is calculated according to the new first characteristic curve, and the same optimal operating point calculation algorithm as before is used. At the same time, the second optimal operating point corresponding to PEMFC is calculated according to the new second characteristic curve, and the same optimal operating point calculation algorithm as before is used. In terms of power allocation adjustment, the output of PEMFC is appropriately increased according to the set ratio (for example, 1:1.2). For example, if the SOFC power decreases by 10%, the PEMFC power increases by 12% to make up for the power gap caused by the decline in SOFC performance.

[0176] The execution subject of the coupled fuel cell dynamic optimal control method may be an image processing device, for example, the coupled fuel cell dynamic optimal control method may be executed by a terminal device or a server or other processing device, wherein the terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the coupled fuel cell dynamic optimal control method may be implemented by a processor calling a computer-readable instruction stored in a memory.

[0177] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0178] Figure 2 A system block diagram corresponding to the coupled fuel cell dynamic optimal control system according to an embodiment of the present disclosure is shown, as shown in FIG. Figure 2As shown, the coupled fuel cell dynamic optimal control system includes: an acquisition unit, used to acquire a first characteristic curve corresponding to the proton exchange membrane fuel cell, a first temperature and a first pressure currently detected or updated; and / or, to acquire a second characteristic curve corresponding to the solid oxide fuel cell, a second temperature and a second pressure currently detected or updated; a determination unit, used to determine a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve at the first temperature and the first pressure; and / or, to determine a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve at the second temperature and the second pressure; a control optimization unit, used to determine the first optimal power corresponding to the proton exchange membrane fuel cell and / or the second optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure.

[0179] In an embodiment of the present disclosure, the coupled fuel cell dynamic optimal control system comprises: a processor; a memory for storing processor executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned coupled fuel cell dynamic optimal control method; or, comprises: a computer-readable storage medium having computer program instructions stored thereon, characterized in that the computer program instructions implement the above-mentioned coupled fuel cell dynamic optimal control method when executed by the processor; or, comprises: a computer program product, comprising a computer program / instructions, which implements the above-mentioned coupled fuel cell dynamic optimal control method when executed by the processor.

[0180] In an embodiment of the present disclosure, the coupled fuel cell dynamic optimal control system is characterized in that it also includes: a solid oxide fuel cell 3 and a proton exchange membrane fuel cell 4; wherein the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 are respectively connected to a fuel processing module for heating hydrogen to a working temperature suitable for the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 through a first hydrogen supply valve 2 and a second hydrogen supply valve, and the fuel processing module is connected to the hydrogen supply module 17.

[0181] In the embodiment of the present disclosure, the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 are respectively connected to a DC / AC inverter 10; the DC / AC inverter 10 is used to convert the DC power output by the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 into AC power.

[0182] In the embodiment of the present disclosure, the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 are connected to a thermal management module; and / or, the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 are connected to a compressor 7 through a first air supply valve 6 and a second air supply valve, respectively, and the compressor 7 is connected to an air supply module 18; and / or,

[0183] In an embodiment of the present disclosure, a first hydrogen flow sensor 13 is provided on the pipeline between the proton exchange membrane fuel cell 4 and the fuel processing module; and / or, a second hydrogen flow sensor 12 is provided on the pipeline between the solid oxide fuel cell 3 and the fuel processing module; and / or, a first air flow sensor 14 is provided on the pipeline between the proton exchange membrane fuel cell 4 and the compressor 7; and / or, a second air flow sensor 15 is provided on the pipeline between the solid oxide fuel cell 3 and the compressor 7.

[0184] In an embodiment of the present disclosure, the fuel processing module connected to the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 is configured as a hydrogen preheater 1; and / or, the thermal management module connected to the solid oxide fuel cell 3 and the proton exchange membrane fuel cell 4 includes: a water cooling system 20, an air cooling system 19; the solid oxide fuel cell 3 is connected to the air cooling system 19, and the proton exchange membrane fuel cell 4 is connected to the water cooling system 20 via a pump 9; and / or, the water cooling system 20 connected to the proton exchange membrane fuel cell 4 is used to pass cooling water through the proton exchange membrane fuel cell 4, and release heat through a first heat exchanger 8 connected to the proton exchange membrane fuel cell 4.

[0185] In the embodiments of the present disclosure and other possible embodiments, a power generation system based on the coupling of PEMFC (proton exchange membrane fuel cell) and SOFC (solid oxide fuel cell) is provided. The specific embodiments involve a hydrogen supply module 17, an air supply module 18, a fuel cell stack (PEMFC4 and SOFC3), a fuel processing module (hydrogen preheater 1), a thermal management module (water cooling system 20, air cooling system 19), a power output adjustment module, and an intelligent control and coordination module. The fuel cell stack module includes two parts: PEMFC4 and SOFC3.

[0186] PEMFC4: It uses a proton exchange membrane cell stack with an operating temperature of 60-100°C and is responsible for quickly responding to changes in power demand. The PEMFC cell stack converts hydrogen and oxygen in the air into electrical energy through electrochemical reactions between electrodes. The cell stack is directly connected to the hydrogen supply module 17 and the air supply module 18, with hydrogen flowing into the anode side and oxygen flowing into the cathode side to generate direct current.

[0187] SOFC3: It uses a solid oxide fuel cell stack with an operating temperature of 600-1000°C and can generate electricity stably for a long time. SOFC uses hydrogen as fuel and generates current through electrochemical reactions on both sides of the electrolyte membrane. SOFC is connected to the fuel processing module through a gas pipeline. Hydrogen enters the anode after pretreatment, and air enters the cathode after compression.

[0188] In the embodiment of the present disclosure and other possible embodiments, the DC power of the two battery stacks is output to the DC / AC inverter 10 through a cable, and after being converted into AC power by the inverter, the power is output to the power grid or energy storage system. The inverter is regulated by the power control module to control the voltage, current and power output in real time.

[0189] The fuel processing module is used to pre-process the fuel of PEMFC and SOFC to ensure that the purity and temperature of the fuel gas meet the power generation requirements. It includes:

[0190] Hydrogen preheater 1: used to heat hydrogen to a temperature suitable for the working of PEMFC and SOFC. Hydrogen is supplied through a gas tank. This device connects the hydrogen supply pipeline and the gas inlet on the anode side of the cell stack.

[0191] Thermal management module: Water cooling system 20 (for PEMFC): The water cooling system uses a pump 9 to flow cooling water through the PEMFC cell stack 4 to take away the generated medium and low temperature waste heat. The cooled water releases heat through the first heat exchanger 8, and this part of the waste heat can be used for building heating or hot water supply. Air cooling system 19 (for SOFC): The high-temperature gas of SOFC3 takes away part of the waste heat through the air cooling system 19, and the waste heat is recovered through the second heat exchanger 5. The high-temperature waste heat can be used to preheat the fuel entering the SOFC, or to drive other high-temperature processes (such as electrolysis of water to produce hydrogen). The cooled exhaust gas enters the cooling water system of the PEMFC through the heat exchanger, providing low-temperature hot water (such as 60-80°C) for building heating or other low-temperature applications.

[0192] In the embodiment of the present disclosure and other possible embodiments, in the power output module, direct current (DC) generated by the fuel cell is converted into alternating current (AC) to supply the power grid or energy storage system.

[0193] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0194] The embodiment of the present disclosure also provides a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above method is implemented. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

[0195] The embodiment of the present disclosure also provides an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to perform the above method. The electronic device can be provided as a terminal, a server or other forms of equipment.

[0196] Figure 3 8 is a block diagram of an electronic device 800 according to an exemplary embodiment. For example, the electronic device 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, or the like.

[0197] Reference Figure 3 , the electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816. In an exemplary embodiment, the electronic device 800 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0198] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 804 including computer program instructions, which can be executed by a processor 820 of an electronic device 800 to perform the above method.

[0199] Figure 4 1 is a block diagram of an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 may be provided as a server. Figure 4 , the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

[0200] The electronic device 1900 may also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output (I / O) interface 1958. The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™ or the like.

[0201] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0202] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for dynamic optimal control of a coupled fuel cell, characterized in that: include: Obtaining a first characteristic curve, a currently detected or updated first temperature, and a first pressure corresponding to a proton exchange membrane fuel cell; and / or, obtaining a second characteristic curve corresponding to the solid oxide fuel cell, a second temperature and a second pressure currently detected or updated; Determine a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve at the first temperature and the first pressure; and / or determine a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve at the second temperature and the second pressure; Based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure, the first optimal power corresponding to the proton exchange membrane fuel cell and / or the second optimal power corresponding to the solid oxide fuel cell are determined.

2. The coupled fuel cell dynamic optimal control method according to claim 1, characterized in that: The method for determining the first optimal power corresponding to the proton exchange membrane fuel cell and / or the second optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve and the first life characteristic curve at the first temperature and the first pressure and / or the second efficiency characteristic curve and the second life characteristic curve at the second temperature and the second pressure comprises: Determine a first group of efficiencies corresponding to the first efficiency characteristic curve and a first group of lifespans corresponding to the first lifespan characteristic curve; construct a first multi-parameter function corresponding to the first efficiency characteristic curve and the first lifespan characteristic curve in the first characteristic curve; wherein the first group of parameters corresponding to the first multi-parameter function include: a first voltage to be optimized and a first current to be optimized; with the goal of maximizing the efficiency and lifespan in the first group of efficiencies and the first group of lifespans, construct a first objective function corresponding to the first multi-parameter function; with the first voltage-current characteristic curve in the first characteristic curve as the first constraint condition, solve the first objective function to obtain the first optimal power; and / or, wherein the first optimal power is configured as the product of a first optimal voltage corresponding to the first voltage to be optimized and a first optimal current corresponding to the first current to be optimized; and / or, Determine a second group of efficiencies corresponding to the second efficiency characteristic curve and a second group of lifespans corresponding to the second lifespan characteristic curve; construct a second multi-parameter function including the second efficiency characteristic curve and the second lifespan characteristic curve in the second characteristic curve; wherein the second group of parameters corresponding to the second multi-parameter function include: a second voltage to be optimized and a second current to be optimized; with the goal of maximizing the efficiency and lifespan in the second group of efficiencies and the second group of lifespans, construct a second objective function corresponding to the second multi-parameter function; with the second voltage-current characteristic curve in the second characteristic curve as the second constraint condition, solve the second objective function to obtain the second optimal power; and / or, wherein the second optimal power is configured as the product of the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized; and / or, Determine a first group of efficiencies corresponding to the first efficiency characteristic curve and a first group of lifespans corresponding to the first lifespan characteristic curve, determine a second group of efficiencies corresponding to the second efficiency characteristic curve and a second group of lifespans corresponding to the second lifespan characteristic curve; construct a third multi-parameter function including the first efficiency characteristic curve and the first lifespan characteristic curve in the first characteristic curve, and the second efficiency characteristic curve and the second lifespan characteristic curve in the second characteristic curve; wherein a third group of parameters corresponding to the third multi-parameter function includes: a first voltage to be optimized, a first current to be optimized, a second voltage to be optimized, and a second current to be optimized; with the first group of efficiencies and the first group of lifespans, and / or the second group of efficiencies and the second group of lifespans in the efficiency and lifespans. Taking maximizing the life as the goal, construct a third objective function corresponding to the third multi-parameter function; taking the first voltage-current characteristic curve in the first characteristic curve and / or the second voltage-current characteristic curve in the second characteristic curve as the third constraint condition, solving the third objective function to obtain the first optimal power and / or the second optimal power; and / or, wherein the first optimal power corresponding to the third objective function is configured as the product of the first optimal voltage corresponding to the first voltage to be optimized and the first optimal current corresponding to the first current to be optimized; and / or, the second optimal power corresponding to the third objective function is configured as the product of the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized.

3. The coupled fuel cell dynamic optimal control method according to claim 2, characterized in that: In the method of solving the first objective function to obtain the first optimal power by taking the first voltage-current characteristic curve in the first characteristic curve as the first constraint condition, the first constraint condition further includes at least one or more of: a first voltage maximum value and a first voltage minimum value corresponding to the first voltage to be optimized, a first current maximum value and a first current minimum value corresponding to the first current to be optimized, a first temperature maximum value and a first temperature minimum value corresponding to the first temperature, and a first pressure maximum value and a first pressure minimum value corresponding to the first pressure; and / or, In the method for solving the second objective function to obtain the second optimal power by taking the second voltage-current characteristic curve in the second characteristic curve as the second constraint condition, the second constraint condition further includes at least one or more of: a second voltage maximum value and a second voltage minimum value corresponding to the second voltage to be optimized, a second current maximum value and a second current minimum value corresponding to the second current to be optimized, a second temperature maximum value and a second temperature minimum value corresponding to the second temperature, and a second pressure maximum value and a second pressure minimum value corresponding to the second pressure; and / or, In the method for solving the third objective function to obtain the first optimal power and / or the second optimal power by using the first voltage-current characteristic curve in the first characteristic curve and / or the second voltage-current characteristic curve in the second characteristic curve as the third constraint condition, the third constraint condition further includes at least: one or more of the first voltage to be optimized maximum value and the first voltage to be optimized minimum value corresponding to the first voltage to be optimized, the first current to be optimized maximum value and the first current to be optimized minimum value corresponding to the first current to be optimized, the first temperature maximum value and the first temperature minimum value corresponding to the first temperature, the first pressure maximum value and the first pressure minimum value corresponding to the first pressure, the second voltage to be optimized maximum value and the second voltage to be optimized minimum value corresponding to the second voltage to be optimized, the second current to be optimized maximum value and the second current to be optimized minimum value corresponding to the second current to be optimized, the second temperature maximum value and the second temperature minimum value corresponding to the second temperature, and the second pressure maximum value and the second pressure minimum value corresponding to the second pressure.

4. The method for dynamic optimal control of a coupled fuel cell according to any one of claims 2 to 3, characterized in that: The method for solving the first objective function to obtain the first optimal power includes: Constructing a first vector for each particle corresponding to the first objective function; the first vector at least includes: a first power allocation ratio and one or more of a first voltage to be optimized and a first current to be optimized; updating the speed and position corresponding to each particle in the first vector under a set variable first neighborhood radius until the first efficiency and the first life corresponding to each particle in the first vector are maximized, and obtaining one or more of the first power allocation ratio, the first optimal voltage corresponding to the first voltage to be optimized, and the first optimal current corresponding to the first current to be optimized; and / or, The method of updating the speed and position corresponding to each particle in the first vector under the set variable first neighborhood radius until the first efficiency and the first life corresponding to each particle in the first vector are maximized, comprises: randomly initializing the first position and the first speed, the first individual optimal position and the first global optimal position of each particle in the first vector; updating the first position, the first speed, the first individual optimal position and the first global optimal position of each particle using the continuously updated inertia weight and the acceleration constant; during the updating process, determining the first fitness corresponding to the first efficiency and the first life based on the updated first position and the updated first speed of each particle; stopping the updating until the first fitness reaches the first set fitness; determining the first position where the updating is stopped based on the first individual optimal position and the first global optimal position where the updating is stopped; configuring each particle corresponding to the first position where the updating is stopped to one or more of the first power allocation ratio, the first optimal voltage corresponding to the first voltage to be optimized and the first optimal current corresponding to the first current to be optimized; Among them, the first set fitness configuration is a function corresponding to the first efficiency and the first lifespan; the first set fitness configuration is a first value corresponding to when the first efficiency and the first lifespan corresponding to each particle are maximized.

5. The method for dynamic optimal control of a coupled fuel cell according to any one of claims 2 to 4, characterized in that: The method for solving the second objective function to obtain the second optimal power includes: Constructing a second vector for each particle corresponding to the second objective function; the second vector at least includes: a second power allocation ratio and one or more of a second voltage to be optimized and a second current to be optimized; updating the speed and position corresponding to each particle in the second vector under a set variable second neighborhood radius until the second efficiency and second life corresponding to each particle in the second vector are maximized, and obtaining one or more of the second power allocation ratio, the second optimal voltage corresponding to the second voltage to be optimized, and the second optimal current corresponding to the second current to be optimized; and / or, The method of updating the speed and position corresponding to each particle in the second vector under the set variable second neighborhood radius until the second efficiency and second life corresponding to each particle in the second vector are maximized, comprising: randomly initializing the second position and second speed, the second individual optimal position and the second global optimal position of each particle in the second vector; updating the second position, the second speed, the second individual optimal position and the second global optimal position of each particle using the continuously updated inertia weight and the acceleration constant; during the updating process, determining the second fitness corresponding to the second efficiency and the second life based on the updated second position and the updated second speed of each particle; stopping the updating until the second fitness reaches the second set fitness; determining the second position where the updating is stopped based on the second individual optimal position and the second global optimal position where the updating is stopped; configuring each particle corresponding to the second position where the updating is stopped to one or more of the second power allocation ratio, the second optimal voltage corresponding to the second voltage to be optimized and the second optimal current corresponding to the second current to be optimized; Among them, the second set fitness configuration is a function corresponding to the second efficiency and the second life; the second set fitness configuration is a second value corresponding to when the second efficiency and the second life corresponding to each particle are maximized.

6. The method for dynamic optimal control of a coupled fuel cell according to any one of claims 2 to 5, characterized in that: The method of solving the third objective function to obtain the first optimal power and / or the second optimal power includes: Construct a third vector for each particle corresponding to the third objective function; the third vector includes at least: a first vector parameter corresponding to a proton exchange membrane fuel cell and a second vector parameter corresponding to a solid oxide fuel cell; wherein the first vector parameter includes: a first power allocation ratio and a first voltage to be optimized, a first current to be optimized, or several thereof; wherein the second vector parameter includes: a second power allocation ratio and a second voltage to be optimized, a second current to be optimized, or several thereof; under a set variable third neighborhood radius, update the speed and position corresponding to each particle in the third vector until the third efficiency and third life corresponding to each particle in the third vector are maximized, and obtain one or several of the third power allocation ratio, the third optimal voltage corresponding to the third voltage to be optimized, and the third optimal current corresponding to the third current to be optimized; and / or, The method of updating the speed and position corresponding to each particle in the third vector under the set variable third neighborhood radius until the third efficiency and third life corresponding to each particle in the third vector are maximized comprises: randomly initializing the third position and third speed, the third individual optimal position and the third global optimal position of each particle in the third vector; updating the third position, the third speed, the third individual optimal position and the third global optimal position of each particle using the continuously updated inertia weight and acceleration constant; during the updating process, determining the third fitness of the third efficiency and the third life corresponding to each particle based on the updated third position and the updated third speed of each particle; stopping the updating until the third fitness reaches the third set fitness; determining the third position where the updating is stopped based on the third individual optimal position and the third global optimal position where the updating is stopped; configuring each particle corresponding to the third position where the updating is stopped to be one or more of the third power allocation ratio, the third optimal voltage corresponding to the third voltage to be optimized and the third optimal current corresponding to the third current to be optimized; Among them, the third set fitness configuration is a function corresponding to the third efficiency and the third lifespan; the third set fitness configuration is a third value corresponding to when the third efficiency and the third lifespan corresponding to each particle are maximized.

7. The method for dynamic optimal control of a coupled fuel cell according to any one of claims 1 to 6, characterized in that: Before obtaining the first characteristic curve corresponding to the proton exchange membrane fuel cell, the method for determining the first characteristic curve includes: obtaining a first measurement parameter corresponding to the proton exchange membrane fuel cell under a first set condition; wherein the first measurement parameter includes: a first historical measurement current, a first historical measurement temperature, a first historical measurement pressure and its corresponding first historical measurement voltage-current, a first historical measurement efficiency, and a first historical measurement life; determining the first historical measurement voltage corresponding to the proton exchange membrane fuel cell by using the first historical measurement current, the first historical measurement temperature, and the first historical measurement pressure; training or performing data fitting on a preset neural network by using the first historical measurement voltage, the first historical measurement current, the first historical measurement temperature, the first historical measurement pressure and its corresponding first historical measurement voltage-current, the first historical measurement efficiency, and the first historical measurement life respectively, to obtain a first efficiency characteristic curve, a first life characteristic curve, and a first voltage-current characteristic curve corresponding to the first characteristic curve; and / or, Before obtaining the second characteristic curve corresponding to the solid oxide fuel cell, the method for determining the second characteristic curve includes: obtaining a second measurement parameter corresponding to the proton exchange membrane fuel cell under a second set condition; wherein the second measurement parameter includes: a second historical measurement current, a second historical measurement temperature, a second historical measurement pressure and its corresponding second historical measurement voltage-current, a second historical measurement efficiency, and a second historical measurement life; determining the second historical measurement voltage corresponding to the proton exchange membrane fuel cell using the second historical measurement current, the second historical measurement temperature, and the second historical measurement pressure; and training or performing data fitting on a preset neural network using the second historical measurement voltage, the second historical measurement current, the second historical measurement temperature, the second historical measurement pressure and its corresponding second historical measurement voltage-current, the second historical measurement efficiency, and the second historical measurement life, respectively, to obtain a second efficiency characteristic curve, a second life characteristic curve, and a second voltage-current characteristic curve corresponding to the second characteristic curve.

8. The method for dynamic optimal control of a coupled fuel cell according to any one of claims 1 to 7, characterized in that: Before obtaining the first temperature and the first pressure detected or updated, the method for determining the first temperature and the first pressure detected or updated comprises: using a sliding window of a set size and its corresponding weight to process the first temperature and the first pressure corresponding to the sub-exchange membrane fuel cell at a time before the current time, respectively, to obtain the first temperature and the first pressure detected or updated; and / or, Before obtaining the second temperature and the second pressure currently detected or updated, the method for determining the second temperature and the second pressure currently detected or updated comprises: using a sliding window of a set size and its corresponding weight to process the second group of temperatures and the second group of pressures corresponding to the solid oxide fuel cell at a time before the current time, respectively, to obtain the second temperature and the second pressure currently detected or updated; and / or, It also includes: when the power load demand corresponding to the power grid is stable, determining the first optimal power corresponding to the solid oxide fuel cell based on the first efficiency characteristic curve at the first temperature and the first pressure; at this time, the proton exchange membrane fuel cell is in a standby or low-power operation state; wherein the low-power operation state is configured to be less than a first set ratio of the rated power; and / or, when the fluctuation of the power load corresponding to the power grid is less than the first set fluctuation value, adjusting the fuel supply speed of the solid oxide fuel cell to maintain the first optimal power within a first set power range; and / or, It also includes: when the predicted growth rate of the power load corresponding to the power grid is greater than the preset growth rate, increasing the power corresponding to the proton exchange membrane fuel cell to a first set output power within a first set time, and increasing the solid oxide fuel cell to a second set output power within a second set time greater than the first set time; when the second set output power reaches the first set power, gradually reducing the output power of the proton exchange membrane fuel cell at a set speed; and / or, The method further includes: when the decrease rate of the power load corresponding to the power grid drops to a preset decrease rate within a third set time, respectively calculating the first optimal power corresponding to the proton exchange membrane fuel cell and the second optimal power corresponding to the solid oxide fuel cell; within a fourth set time, reducing the power corresponding to the proton exchange membrane fuel cell to a third set output power; calculating a power reduction curve corresponding to the solid oxide fuel cell; reducing the first optimal power according to the power reduction curve; and / or, After a fifth set time, the first characteristic curve corresponding to the proton exchange membrane fuel cell and / or the second characteristic curve corresponding to the solid oxide fuel cell are updated.

9. A coupled fuel cell dynamic optimal control system, characterized in that: include: An acquisition unit, used to acquire a first characteristic curve corresponding to the proton exchange membrane fuel cell, a first temperature and a first pressure currently detected or updated; and / or, obtaining a second characteristic curve corresponding to the solid oxide fuel cell, a second temperature and a second pressure currently detected or updated; a determining unit, configured to determine a first efficiency characteristic curve and a first life characteristic curve in the first characteristic curve at the first temperature and the first pressure; and / or to determine a second efficiency characteristic curve and a second life characteristic curve in the second characteristic curve at the second temperature and the second pressure; a control optimization unit, configured to determine a first optimal power corresponding to the proton exchange membrane fuel cell and / or a second optimal power corresponding to the solid oxide fuel cell based on a first efficiency characteristic curve and a first life characteristic curve at the first temperature and the first pressure and / or a second efficiency characteristic curve and a second life characteristic curve at the second temperature and the second pressure; or Comprising: a processor; a memory for storing processor executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the coupled fuel cell dynamic optimal control method according to any one of claims 1 to 8; or, Comprising: a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the method for dynamic optimal control of a coupled fuel cell according to any one of claims 1 to 8; or, The invention comprises: a computer program product, comprising a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the method for dynamic optimal control of a coupled fuel cell as claimed in any one of claims 1 to 8 is implemented.

10. The coupled fuel cell dynamic optimal control system according to claim 9, characterized in that: Also includes: A solid oxide fuel cell (3) and a proton exchange membrane fuel cell (4); wherein the solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) are connected to a fuel processing module for heating hydrogen to a working temperature suitable for the solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) through a first hydrogen supply valve (2) and a second hydrogen supply valve, respectively, and the fuel processing module is connected to the hydrogen supply module (17); and / or, The solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) are respectively connected to a DC / AC inverter (10); the DC / AC inverter (10) is used to convert the DC power output by the solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) into AC power; and / or, the solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) are connected to a thermal management module; and / or, the solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) are respectively connected to a compressor (7) via a first air supply valve (6) and a second air supply valve, and the compressor (7) is connected to an air supply module (18); and / or, A first hydrogen flow sensor (13) is provided on the pipeline between the proton exchange membrane fuel cell (4) and the fuel processing module; and / or, a second hydrogen flow sensor (12) is provided on the pipeline between the solid oxide fuel cell (3) and the fuel processing module; and / or, a first air flow sensor (14) is provided on the pipeline between the proton exchange membrane fuel cell (4) and the compressor (7); and / or, a second air flow sensor (15) is provided on the pipeline between the solid oxide fuel cell (3) and the compressor (7); and / or, The fuel processing module connected to the solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) is configured as a hydrogen preheater (1); and / or the thermal management module connected to the solid oxide fuel cell (3) and the proton exchange membrane fuel cell (4) comprises: a water cooling system (20) and an air cooling system (19); the solid oxide fuel cell (3) is connected to the air cooling system (19), and the proton exchange membrane fuel cell (4) is connected to the water cooling system (20) via a pump (9); and / or the water cooling system (20) connected to the proton exchange membrane fuel cell (4) is used to pass cooling water through the proton exchange membrane fuel cell (4) and release heat through a first heat exchanger (8) connected to the proton exchange membrane fuel cell (4).

Images