A low-temperature environment fuel cell automobile energy optimization control method

Abstract

The application discloses a low-temperature environment fuel cell vehicle energy optimization control method and relates to the technical field of fuel cell vehicle optimization control. The method comprises the following steps: collecting fuel cell vehicle information in real time, constructing a fuel cell hydrogen consumption model related to temperature, constructing a fuel cell stack heat production ratio data set under a low-temperature environment, constructing a vehicle longitudinal dynamics model, a power battery SoC model and a fuel cell stack heat model, and using the models to predict the dynamic-thermal response description in a prediction time domain in a prediction control; combining the operation characteristic differences of the fuel cell stack in different temperature intervals under the low-temperature environment, designing a phased and differentiated energy optimization control strategy; and solving the fuel cell vehicle energy optimization problem based on the optimization control strategy in a rolling time domain to obtain an optimal control sequence in a future time domain and realize rolling real-time energy optimization control. The application solves the problem of insufficient consideration of the low-temperature environment in the existing energy optimization control method.

Description

Technical Field

[0001] This invention relates to the field of fuel cell vehicle optimization control technology, and more specifically to a method for energy optimization control of fuel cell vehicles in low-temperature environments. Background Technology

[0002] Fuel cell vehicles typically employ a hybrid power system that couples a fuel cell with a power battery. Energy optimization control methods dynamically allocate the output power of both systems to balance vehicle economy, power performance, and fuel cell durability. Current mainstream energy optimization control methods, such as the equivalent hydrogen consumption minimization strategy, model predictive control, and rule-based methods, mostly construct optimization objectives based on the efficiency characteristics of the fuel cell system at room temperature. This assumes that the stack efficiency exhibits a relatively stable mapping relationship under different operating conditions, and optimizes power allocation based on this assumption.

[0003] However, ambient temperature has a significant impact on the internal electrochemical reaction activity, membrane electrode hydrothermal balance, and polarization characteristics of fuel cell stacks. Under low-temperature conditions, ohmic polarization and activation polarization losses in the stack intensify, and the efficiency value at the same power point exhibits a nonlinear decay trend as the temperature decreases. Existing energy management strategies generally do not consider the real-time impact of temperature on stack efficiency when constructing the optimization objective function, and still use the ambient temperature efficiency model for power allocation decisions. This causes the actual operating point of the system under low-temperature conditions to deviate from the optimal efficiency range, resulting in a decrease in hydrogen utilization and a deterioration in the overall vehicle economy.

[0004] With the rapid development of intelligent connected vehicle technology, vehicles can obtain forward-looking traffic flow information through V2X. Connected information can pre-plan vehicle speed and power output trajectory, enabling the energy management system to pre-raise the fuel cell stack temperature and maintain it within its efficient operating range, thereby unlocking the energy-saving potential of the entire vehicle.

[0005] Therefore, how to effectively couple the dynamic impact of low temperature environment on fuel cell stack efficiency in energy optimization control methods, and construct an adaptive optimization target that can reflect the change of stack efficiency with temperature, so as to achieve precise control of power distribution between fuel cell and power battery under low temperature conditions, has become a technical problem that urgently needs to be solved in this field.

[0006] Patent CN108944900A proposes an energy management and control method for fuel cell vehicles. This method formulates a power allocation strategy between the fuel cell and the power battery based on the vehicle's real-time driving power demand and the state of charge of the power battery. However, this method is essentially a rule-based instantaneous energy management strategy, which can only make real-time energy allocation decisions based on the vehicle's current state, making it difficult to achieve global optimization of the working efficiency of the fuel cell and the power battery. In addition, this patent does not consider information interaction and forward prediction in the intelligent connected environment, and fails to fully explore the energy-saving potential of the vehicle under connected conditions.

[0007] Patent CN120621168A discloses an energy management strategy for fuel cell vehicles that integrates traffic perception. This invention provides an energy management strategy for fuel cell vehicles that integrates traffic perception, using a Transformer network to construct a predictive model of future speeds. This model effectively represents the dynamic interaction between the vehicle and surrounding multiple vehicles, offering higher accuracy and stability compared to traditional speed prediction methods based on historical data or single-vehicle information. The energy management strategy, built based on the predicted future speed sequence, enables early perception and rapid response to future traffic changes, and also has significant advantages in fuel economy. However, this patent does not consider the impact of low-temperature environments on the fuel cell stack, and therefore cannot balance stack efficiency and overall vehicle economy during the optimization process.

[0008] In summary, although the currently published patents have covered some solutions for energy management of fuel cell vehicles, they are all based on the efficiency characteristics of fuel cell stacks at room temperature to build optimization targets, assuming that their efficiency is stable under different operating conditions. However, in reality, ambient temperature has a significant impact on stack characteristics. Low temperatures will exacerbate polarization losses and cause nonlinear decay of stack efficiency. Existing strategies do not consider this impact and continue to use room temperature efficiency models, which leads to the system deviating from the optimal efficiency range and deteriorating economic performance under low temperature conditions.

[0009] To address the above challenges, this invention proposes an energy optimization control method for fuel cell vehicles operating in low-temperature environments. This method aims to overcome the difficulties in existing technologies and is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0010] In view of this, the present invention provides an energy optimization control method for fuel cell vehicles in low-temperature environments, which solves the problem of insufficient consideration of low-temperature environments in existing energy optimization control methods and further explores the energy-saving potential of fuel cell vehicles in intelligent connected transportation environments.

[0011] To achieve the above objectives, the present invention provides the following technical solution: A method for energy optimization control of fuel cell vehicles in low-temperature environments includes the following steps: S1. Collect fuel cell vehicle information in real time, preprocess the collected fuel cell vehicle information, and construct a temperature-related fuel cell hydrogen consumption model. S2. Construct a dataset of fuel cell stack heat generation ratio under low temperature conditions; S3. Construct a vehicle longitudinal dynamics model, a power battery SoC model, and a fuel cell stack thermal model for predictive control to describe the dynamic-thermal response in the prediction time domain. S4. Based on the collected fuel cell vehicle information, the constructed stack heat production ratio dataset, and the dynamic-thermal response description, and combined with the differences in the operating characteristics of the fuel cell stack in different temperature ranges under low temperature conditions, a phased and differentiated energy optimization control strategy is designed. S5. Using the system state of the current control cycle as the initial value, and combining the future operating condition information predicted by the network information, based on the vehicle longitudinal dynamics model of S3, the energy optimization problem of the fuel cell vehicle is solved in the rolling time domain based on the optimization control strategy in S4, and the optimal control sequence in the future time domain is obtained; only the optimal control command of the current control cycle is executed, and S1-S4 are repeated in the next control cycle to realize rolling real-time energy optimization control.

[0012] Optionally, S1 can collect real-time information about fuel cell vehicles, including ambient temperature, fuel cell thermal state, power battery system status, vehicle driving status, and network connectivity information to predict operating conditions.

[0013] Optionally, the specific content of the temperature-related fuel cell hydrogen consumption model constructed in S1 is as follows: A proton exchange membrane fuel cell converts chemical energy into electrical energy through the movement of hydrogen ions and electrons during the reaction of hydrogen and oxygen. The output voltage is: (1); in, Indicates the fuel cell voltage. Indicates the open-circuit voltage of the fuel cell. Indicates the active voltage drop. This represents the voltage drop across the resistor. Indicates concentration voltage drop; The open-circuit voltage of a fuel cell is the potential difference resulting from the change in Gibbs free energy of the system during the electrochemical reaction of hydrogen and oxygen. Its value is directly related to the activity of the reactants and is expressed as: (2); in, This represents the open-circuit voltage of the fuel cell reaction under standard atmospheric pressure. It is a universal gas constant. For the types of reactants, For fuel cell temperature, Where is the molar mass of oxygen; The active loss is represented by the active voltage drop, as follows: (3); in, Indicates the speed of the electrode reaction. Represents the exchange current density. Indicates the alternating current. Indicates the fuel cell current; The formulas for calculating the consumption rates of hydrogen and oxygen during the reaction are as follows: (4); in, and These represent the consumption rates of hydrogen and oxygen, respectively. and These represent the molar masses of hydrogen and oxygen, respectively. The formulas for calculating the output power and hydrogen consumption rate of a fuel cell are as follows: (5); in, Indicates the fuel cell power; The relationship between the actual output power and the available output power of a fuel cell is as follows: (6); The hydrogen consumption rate of a fuel cell is related to its output power, and the hydrogen consumption rate is expressed as: (7); in, The hydrogen consumption rate of PEMFC, For PEMFC efficiency, It has the lowest calorific value of hydrogen.

[0014] Optionally, the specific content of constructing the fuel cell stack heat generation ratio dataset under low-temperature conditions in S2 is as follows: S21. Analysis of the efficiency characteristics of fuel cell stack in low-temperature environment; The higher the temperature, the higher the output power and voltage of the fuel cell stack at the same current density, that is: (8); S22. Analysis of the temperature rise characteristics of the fuel cell stack in a low-temperature environment; S23. Based on the analysis of the efficiency characteristics of the fuel cell stack in a low-temperature environment and the analysis of the heating characteristics of the fuel cell stack in a low-temperature environment, a power control strategy for the fuel cell stack is constructed based on the principle of prioritizing heat generation, thereby constructing a dataset of the heat generation ratio of the fuel cell stack in a low-temperature environment. The power control strategy for the fuel cell stack, based on the principle of prioritizing heat generation, is as follows: Based on the heat generation ratio characteristics of the fuel cell stack at various power points under different temperature conditions, the power data corresponding to the maximum heat generation ratio at different temperatures was determined. ; When power demand satisfy At that time, the output power that enables the fuel cell to operate The battery replenishes the remaining power required, that is... ; When power demand At that time, in order to heat up the fuel cell as quickly as possible, it was still necessary to... At this point, the excess power of the fuel cell is supplied to the battery to charge it, and the battery's power... .

[0015] Optionally, the specific content of constructing the vehicle's longitudinal dynamics model in S3 is as follows: The driving state of a fuel cell vehicle is determined by the combined force of various external forces acting on the entire vehicle; The driving force required for a car to move is expressed as: (9); in, The driving force exerted by the FCV on the tires. The rolling resistance experienced by the vehicle, The air resistance experienced by the vehicle, The climbing resistance experienced by the vehicle. The acceleration resistance experienced by the vehicle; Tire rolling resistance is expressed as: (10); Air resistance is expressed as: (11); The climbing resistance is expressed as: (12); Acceleration resistance is expressed as: (13); in, For the overall vehicle weight of FCV, The rolling resistance coefficient of the wheel. The slope angle of the road. For windward area, air density, The air drag coefficient, For car speed, This is the rotational mass conversion factor; Given the FCV's speed under specific operating conditions, the required driving power during operation can be calculated based on the vehicle's driving force and speed. The calculation formula is as follows: (14).

[0016] Optionally, the specific content of building the power battery SoC model in S3 is as follows: The power battery was modeled using the Rint equivalent model. (15); in, This is the open-circuit voltage of the battery. For the battery current, This is the equivalent internal resistance of the battery, i.e., the battery charging resistance and discharging resistance. This refers to the voltage across the battery terminals; The current of the power battery is expressed as: (16); The real-time estimation of the battery state of charge (SOC) based on the ampere-hour integration method is expressed as follows: (17); in, This refers to the rated capacity of the power battery.

[0017] Optionally, the specific content of constructing the thermal model of the fuel cell stack in S3 is as follows: Assuming the coolant temperature when it leaves the radiator is equal to the radiator body temperature, the formula for calculating the coolant inlet temperature is derived as follows: (49); in, It's about the quality of the radiator. It is the heat capacity of the radiator. It is the heat exchange between the radiator and the air. It is the differential of x with respect to y. It is the temperature at the coolant inlet. The coolant carries away the heat from inside the fuel cell. Heat exchange between the radiator and the air The formula is as follows: (50); in, It is the airflow at the radiator. It is the heat capacity of air. It is the temperature of the air before it enters the radiator. It is the temperature at which air is output from the radiator.

[0018] Optionally, S4, based on the collected fuel cell vehicle information, the constructed fuel cell stack heat ratio dataset, and the dynamic-thermal response description, and considering the differences in the operating characteristics of the fuel cell stack in different temperature ranges under low-temperature conditions, the specific content of the phased and differentiated energy optimization control strategy is as follows: The overall strategy comprises two control phases: low-temperature start-up and warm-up, and stable operation under normal conditions. Specifically: Low-temperature start-up temperature control stage: For the low-temperature cold start scenario with an initial stack temperature of 0℃, a rule-based temperature optimization control strategy is constructed to achieve preheating of the stack from 0℃ to 70℃. Normal and stable operation control phase: After the stack temperature rises to 70℃ and enters the stable operating range, an optimization strategy based on model predictive control is constructed to dynamically allocate the fuel cell output power and the power battery charging and discharging power according to the real-time load demand of the vehicle, so as to optimize the overall energy consumption of the system.

[0019] Optionally, the specific content of solving the fuel cell vehicle energy optimization problem in S5 based on forward dynamic programming and the optimization control strategy in S4 in the rolling time domain is as follows: S51. Discretization and State Space Partitioning of Energy Optimization Problems: Optimize the time domain Discretize into N stages, step size For state variables of SoC within a feasible range The internal area is uniformly divided to construct a set of state grid points; simultaneously, the control variable, fuel cell output power, is... Within the range of values Discretization is performed; S52. Define the cost function and state transition matrix: exist At time , when the state variable is And take action At that time, the battery output power is determined by the power demand-power balance relationship. The cost of this stage is the consumption of hydrogen. ; S53. Forward dynamic programming recursive solution: From the initial moment The process involves recursively calculating the minimum cumulative hydrogen consumption and optimal control sequence for each stage and discrete state point. The specific steps are as follows: S531. Initialization: Set the initial state Cumulative cost ; S532. Phase recursion: For For the current Iterate through all feasible For each group : Calculate the corresponding Compared to the current cost of hydrogen consumption Store from the initial state to Temporary cumulative cost: ; S533. After the recursion is completed, the state with the minimum cumulative cost is selected as the optimal terminal state in the terminal stage. Starting from the optimal terminal state, the process backtracks to the initial stage based on the recorded predecessor state and control quantity to obtain the globally optimal fuel cell power sequence and the corresponding power battery SoC trajectory.

[0020] As can be seen from the above technical solution, compared with the prior art, the present invention provides an energy optimization control method for fuel cell vehicles in low-temperature environments, which has the following beneficial effects: (1) This invention solves the core problem that the operating point of the low-temperature system deviates from the optimal range due to the use of the room temperature efficiency model in the existing strategy. By constructing a dynamic model that strongly couples temperature and hydrogen consumption, it accurately reflects the nonlinear decay characteristics of the stack efficiency with temperature. (2) This invention combines network-connected forward-looking information to empower energy optimization. Compared with traditional rule-based instantaneous energy management strategies, it can further explore the hydrogen-saving potential in the intelligent network environment. (3) This invention breaks through the limitation of traditional energy management that only focuses on power distribution, and for the first time incorporates the energy consumption characteristics of the fuel cell thermal management system into the optimization target. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0022] Figure 1 A flowchart of an energy optimization control method for fuel cell vehicles in low-temperature environments provided by the present invention; Figure 2 A schematic diagram of fuel cell output voltage and power curves at different temperatures provided by the present invention; Figure 3 A schematic diagram of hydrogen consumption and power curves of fuel cells at different temperatures provided by the present invention; Figure 4 The heat production ratio fitting diagram of the fuel cell under different temperatures and power conditions provided by the present invention; Figure 5 The circulation branch diagram of the fuel cell cooling system provided by this invention; Figure 6This is a schematic diagram showing the relationship between the open-circuit voltage, internal resistance, and battery SoC of the power battery provided by the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] See Figure 1 As shown, this invention discloses an energy optimization control method for fuel cell vehicles in low-temperature environments, comprising the following steps: S1. Collect fuel cell vehicle information in real time, preprocess the collected fuel cell vehicle information, and construct a temperature-related fuel cell hydrogen consumption model. S2. Construct a dataset of fuel cell stack heat generation ratio under low temperature conditions; S3. Construct a vehicle longitudinal dynamics model, a power battery SoC model, and a fuel cell stack thermal model for predictive control to describe the dynamic-thermal response in the prediction time domain. S4. Based on the collected fuel cell vehicle information, the constructed stack heat production ratio dataset, and the dynamic-thermal response description, and combined with the differences in the operating characteristics of the fuel cell stack in different temperature ranges under low temperature conditions, a phased and differentiated energy optimization control strategy is designed. S5. Using the system state of the current control cycle as the initial value, and combining the future operating condition information predicted by the network information, based on the vehicle longitudinal dynamics model of S3, the energy optimization problem of the fuel cell vehicle is solved in the rolling time domain based on the optimization control strategy in S4, and the optimal control sequence in the future time domain is obtained; only the optimal control command of the current control cycle is executed, and S1-S4 are repeated in the next control cycle to realize rolling real-time energy optimization control.

[0025] Furthermore, S1 collects real-time information about fuel cell vehicles, including ambient temperature, fuel cell thermal state, power battery system status, vehicle driving status, and network connectivity information to predict operating conditions.

[0026] Furthermore, the specific details of the temperature-related fuel cell hydrogen consumption model constructed in S1 are as follows: A proton exchange membrane fuel cell converts chemical energy into electrical energy through the movement of hydrogen ions and electrons during the reaction of hydrogen and oxygen. The output voltage is: (1); in, Indicates the fuel cell voltage. Indicates the open-circuit voltage of the fuel cell. Indicates the active voltage drop. This represents the voltage drop across the resistor. Indicates concentration voltage drop; The open-circuit voltage of a fuel cell is the potential difference resulting from the change in Gibbs free energy of the system during the electrochemical reaction of hydrogen and oxygen. Its value is directly related to the activity of the reactants and is expressed as: (2); in, This represents the open-circuit voltage of the fuel cell reaction under standard atmospheric pressure. It is a universal gas constant. For the types of reactants, For fuel cell temperature, Where is the molar mass of oxygen; The active loss is represented by the active voltage drop, as follows: (3); in, Indicates the speed of the electrode reaction. Represents the exchange current density. Indicates the alternating current. Indicates the fuel cell current; The formulas for calculating the consumption rates of hydrogen and oxygen during the reaction are as follows: (4); in, and These represent the consumption rates of hydrogen and oxygen, respectively. and These represent the molar masses of hydrogen and oxygen, respectively. The formulas for calculating the output power and hydrogen consumption rate of a fuel cell are as follows: (5); in, Indicates the fuel cell power; The relationship between the actual output power and the available output power of a fuel cell is as follows: (6); The hydrogen consumption rate of a fuel cell is related to its output power, and the hydrogen consumption rate is expressed as: (7); in, The hydrogen consumption rate of PEMFC, For PEMFC efficiency, It has the lowest calorific value of hydrogen.

[0027] Furthermore, the specific content of constructing the fuel cell stack heat generation ratio dataset under low-temperature conditions in S2 is as follows: S21. Analysis of the efficiency characteristics of fuel cell stack in low-temperature environment; The higher the temperature, the higher the output power and voltage of the fuel cell stack at the same current density, that is: (8); S22. Analysis of the temperature rise characteristics of the fuel cell stack in a low-temperature environment; S23. Based on the analysis of the efficiency characteristics of the fuel cell stack in a low-temperature environment and the analysis of the heating characteristics of the fuel cell stack in a low-temperature environment, a power control strategy for the fuel cell stack is constructed based on the principle of prioritizing heat generation, thereby constructing a dataset of the heat generation ratio of the fuel cell stack in a low-temperature environment. The power control strategy for the fuel cell stack, based on the principle of prioritizing heat generation, is as follows: Based on the heat generation ratio characteristics of the fuel cell stack at various power points under different temperature conditions, the power data corresponding to the maximum heat generation ratio at different temperatures was determined. ; When power demand satisfy At that time, the output power that enables the fuel cell to operate The battery replenishes the remaining power required, that is... ; When power demand At that time, in order to heat up the fuel cell as quickly as possible, it was still necessary to... At this point, the excess power of the fuel cell is supplied to the battery to charge it, and the battery's power... .

[0028] Furthermore, the specific details of constructing the vehicle's longitudinal dynamics model in S3 are as follows: The driving state of a fuel cell vehicle is determined by the combined force of various external forces acting on the entire vehicle; The driving force required for a car to move is expressed as: (9); in, The driving force exerted by the FCV on the tires. The rolling resistance experienced by the vehicle, The air resistance experienced by the vehicle, The climbing resistance experienced by the vehicle. The acceleration resistance experienced by the vehicle; Tire rolling resistance is expressed as: (10); Air resistance is expressed as: (11); The climbing resistance is expressed as: (12); Acceleration resistance is expressed as: (13); in, For the overall vehicle weight of FCV, The rolling resistance coefficient of the wheel. The slope angle of the road. For windward area, air density, The air drag coefficient, For car speed, This is the rotational mass conversion factor; Given the FCV's speed under specific operating conditions, the required driving power during operation can be calculated based on the vehicle's driving force and speed. The calculation formula is as follows: (14).

[0029] Furthermore, the specific details of building the power battery SoC model in S3 are as follows: The power battery was modeled using the Rint equivalent model. (15); in, This is the open-circuit voltage of the battery. For the battery current, This is the equivalent internal resistance of the battery, i.e., the battery charging resistance and discharging resistance. This refers to the voltage across the battery terminals; The current of the power battery is expressed as: (16); The real-time estimation of the battery state of charge (SOC) based on the ampere-hour integration method is expressed as follows: (17); in, This refers to the rated capacity of the power battery.

[0030] Furthermore, the specific details of constructing the thermal model of the fuel cell stack in S3 are as follows: Assuming the coolant temperature when it leaves the radiator is equal to the radiator body temperature, the formula for calculating the coolant inlet temperature is derived as follows: (49); in, It's about the quality of the radiator. It is the heat capacity of the radiator. It is the heat exchange between the radiator and the air. It is the differential of x with respect to y. It is the temperature at the coolant inlet. The coolant carries away the heat from inside the fuel cell. Heat exchange between the radiator and the air The formula is as follows: (50); in, It is the airflow at the radiator. It is the heat capacity of air. It is the temperature of the air before it enters the radiator. It is the temperature at which air is output from the radiator.

[0031] Furthermore, S4, based on the collected fuel cell vehicle information, the constructed fuel cell stack heat ratio dataset, and the dynamic-thermal response description, and considering the differences in the operating characteristics of the fuel cell stack in different temperature ranges under low-temperature conditions, designs a phased and differentiated energy optimization control strategy. The specific details are as follows: The overall strategy comprises two control phases: low-temperature start-up and warm-up, and stable operation under normal conditions. Specifically: Low-temperature start-up temperature control stage: For the low-temperature cold start scenario with an initial stack temperature of 0℃, a rule-based temperature optimization control strategy is constructed to achieve preheating of the stack from 0℃ to 70℃. Normal and stable operation control phase: After the stack temperature rises to 70℃ and enters the stable operating range, an optimization strategy based on model predictive control is constructed to dynamically allocate the fuel cell output power and the power battery charging and discharging power according to the real-time load demand of the vehicle, so as to optimize the overall energy consumption of the system.

[0032] Furthermore, the specific content of solving the fuel cell vehicle energy optimization problem in the rolling time domain based on the optimization control strategy in S4 using forward dynamic programming (FDP) in S5 is as follows: S51. Discretization and State Space Partitioning of Energy Optimization Problems: Optimize the time domain Discretize into N stages, step size For state variables of SoC within a feasible range The internal area is uniformly divided to construct a set of state grid points; simultaneously, the control variable, fuel cell output power, is... Within the range of values Discretization is performed; S52. Define the cost function and state transition matrix: exist At time , when the state variable is And take action At that time, the battery output power is determined by the power demand-power balance relationship. The cost of this stage is the consumption of hydrogen. ; S53. Forward dynamic programming recursive solution: From the initial moment The process involves recursively calculating the minimum cumulative hydrogen consumption and optimal control sequence for each stage and discrete state point. The specific steps are as follows: S531. Initialization: Set the initial state Cumulative cost ; S532. Phase recursion: For For the current Iterate through all feasible For each group : Calculate the corresponding Compared to the current cost of hydrogen consumption Store from the initial state to Temporary cumulative cost: ; S533. After the recursion is completed, the state with the minimum cumulative cost is selected as the optimal terminal state in the terminal stage. Starting from the optimal terminal state, the process backtracks to the initial stage based on the recorded predecessor state and control quantity to obtain the globally optimal fuel cell power sequence and the corresponding power battery SoC trajectory.

[0033] In one specific embodiment, the following is included: S1. Collect fuel cell vehicle information in real time, preprocess the collected fuel cell vehicle information, and construct a temperature-related fuel cell hydrogen consumption model. The collected fuel cell vehicle information includes ambient temperature, fuel cell thermal state, power battery system status, vehicle driving status, and network connectivity information predicting operating conditions. All information is transmitted in real-time via the vehicle's CAN bus. Filtering and correction processes are simultaneously performed to address measurement deviations and noise interference in low-temperature environments, as detailed below: (1) Acquisition of ambient temperature and fuel cell thermal state: The ambient temperature outside the vehicle is collected in real time by using a negative temperature coefficient (NTC) type ambient temperature sensor located at the air intake of the front of the vehicle, away from the heat radiation area of ​​the cooling system. The acquisition period is set to 1 second. The Vehicle Control Unit (VCU) performs moving average filtering on the acquired raw temperature signal, with a filtering window width of 5 acquisition periods to eliminate measurement noise caused by airflow disturbances during driving, thus obtaining the filtered effective ambient temperature. The inlet coolant temperature of the fuel cell stack is acquired in real time using PT100 high-precision temperature sensors installed at the inlet and outlet of the fuel cell coolant circuit. outlet coolant temperature The acquisition period is 100ms. The raw signal is acquired by the fuel cell controller (FuelCell Control Unit) and transmitted to the VCU via the CAN bus. It is assumed that the fuel cell stack temperature is the same as the outlet coolant temperature. This temperature is one of the core parameters for subsequent hydrogen consumption model calculation.

[0034] (2) Power battery system state parameter acquisition: The state variables of the power battery are implemented in layers by the dedicated acquisition modules of the battery management system. The single cell voltage acquisition module acquires the terminal voltage of all single cells; the total voltage acquisition module acquires the total DC voltage of the battery pack; the current acquisition module acquires the charging and discharging current of the high-voltage main circuit through the Hall current sensor; the insulation detection module acquires the insulation resistance of the high-voltage bus to ground based on the pulse injection detection principle.

[0035] (3) Vehicle driving status parameter acquisition: Real-time acquisition of vehicle driving status, including vehicle speed Thermal controller energy consumption The vehicle speed is collected by wheel speed sensors and transmitted to the VCU via the Body Control Module (BCM). Based on these parameters, the VCU calculates the current driving power demand of the vehicle in real time using the vehicle's longitudinal dynamics equations. This provides a real-time target input for subsequent power allocation.

[0036] (4) Acquisition and processing of network information prediction working condition information: The future preset prediction time domain is obtained through the in-vehicle intelligent navigation system and the network terminal of the vehicle network control unit (Telematics BOX, T-BOX). The predicted operating condition information within the time domain is set to 20 seconds. Specifically, the network-connected information predicts the operating condition information including: the slope angle of the driving road at each discrete moment within the future time domain. Speed ​​of the vehicle in front Based on the above-mentioned predictive information and combined with the vehicle's longitudinal dynamics model, the VCU pre-calculates the vehicle's power demand sequence in the future prediction time domain. This provides a prerequisite input for subsequent rolling time-domain optimization.

[0037] Construction of a temperature-dependent hydrogen consumption model for fuel cells: Proton exchange membrane fuel cells (PEMFCs) convert chemical energy into electrical energy through the movement of hydrogen ions and electrons during the reaction of hydrogen and oxygen. The output voltage can be characterized by the following formula: The output voltage is: (1); in, Indicates the fuel cell voltage. Indicates the open-circuit voltage of the fuel cell. Indicates the active voltage drop. This represents the voltage drop across the resistor. Indicates concentration voltage drop; The open-circuit voltage of a fuel cell is the potential difference resulting from the change in Gibbs free energy of the system during the electrochemical reaction of hydrogen and oxygen. Its value is directly related to the activity of the reactants and is expressed as: (2); in, This represents the open-circuit voltage of the fuel cell reaction under standard atmospheric pressure. It is a universal gas constant. For the types of reactants, For fuel cell temperature, Where is the molar mass of oxygen; The active loss is represented by the active voltage drop, as follows: (3); in, Indicates the speed of the electrode reaction. Represents the exchange current density. Indicates the alternating current. Indicates the fuel cell current; The formulas for calculating the consumption rates of hydrogen and oxygen during the reaction are as follows: (4); in, and These represent the consumption rates of hydrogen and oxygen, respectively. and These represent the molar masses of hydrogen and oxygen, respectively. The formulas for calculating the output power and hydrogen consumption rate of a fuel cell are as follows: (5); in, Indicates the fuel cell power; The relationship between the actual output power and the available output power of a fuel cell is as follows: (6); During PEMFC operation, the amount of hydrogen consumed is a key factor in evaluating the control effectiveness of the vehicle's Energy Management System (EMS) and fuel economy. The hydrogen consumption rate of a fuel cell is related to its output power, and its hydrogen consumption rate can be expressed as: (7); in, The hydrogen consumption rate of PEMFC, For PEMFC efficiency, It has the lowest calorific value of hydrogen.

[0038] S2. Construct a dataset of fuel cell stack heat generation ratio under low temperature conditions; S21. Analysis of the efficiency characteristics of the fuel cell stack in a low-temperature environment: Using the AVL / Cruise M simulation platform as the core, simulation experiments were conducted from 0℃ to 70℃, testing the output characteristics of the fuel cell stack at 10℃ intervals. By setting different current densities, the output voltage and power of the fuel cell stack were observed, resulting in the output power and voltage curves at different temperatures, as shown below. Figure 2 As shown in the figure. The experimental results show that the higher the temperature, the higher the output power and voltage of the fuel cell stack at the same current density, that is: (8); The curves showing the relationship between hydrogen consumption and power in fuel cells at different temperatures are as follows: Figure 3 As shown, from Figure 3 As can be seen, at lower temperatures, more hydrogen is required to maintain the same power output due to reduced efficiency. Table 1 summarizes the comparison of hydrogen consumption and efficiency at different temperatures, using 70℃ as a baseline.

[0039] In the process of energy optimization, this efficiency characteristic is manifested in the hydrogen consumption rate. Becomes the output power of the fuel cell stack With the temperature of the fuel cell stack The stack is a strongly coupled function. At low temperatures, to provide the same net output power, the stack often needs to consume more hydrogen or is forced to operate in a lower efficiency region. Therefore, the energy optimization strategy for cryogenic environments must fully consider the dynamic characteristics of stack efficiency as a function of temperature.

[0040] Table 1. Comparison of hydrogen consumption and efficiency of fuel cells at different temperatures;

[0041] S22. Analysis of the temperature rise characteristics of the fuel cell stack in a low-temperature environment; Based on the AVL / Cruise M simulation platform, tabular datasets of hydrogen consumption and temperature rise in fuel cell stacks under different temperatures and output power were constructed, as shown in Tables 2 and 3, respectively. The ratio of temperature rise to hydrogen consumption was defined as the heat production ratio. Table 4 shows the relationship between fuel cell stack temperature rise and hydrogen consumption; a larger value in Table 4 indicates more heat production per unit mass of hydrogen consumed. Based on the two-dimensional data tables, the relationship between heat production ratio, temperature, and output power was established through interpolation, allowing the selection of the power corresponding to the maximum heat production ratio based on temperature. The interpolation fitting effect is shown in Table 4. Figure 4 As shown.

[0042] Table 2. Hydrogen consumption of fuel cell stacks at different temperatures and output powers (unit: g);

[0043] Table 3. Temperature rise of fuel cell stack at different temperatures and output powers (unit: °C);

[0044] Table 4. Heat generation ratio of fuel cell stacks at different temperatures and output powers (unit: °C / g);

[0045] S23. Based on the analysis of the efficiency characteristics of the fuel cell stack in a low-temperature environment and the analysis of the heating characteristics of the fuel cell stack in a low-temperature environment, a power control strategy for the fuel cell stack is constructed based on the principle of prioritizing heat generation, thereby constructing a dataset of the heat generation ratio of the fuel cell stack in a low-temperature environment. The power control strategy for the fuel cell stack, based on the principle of prioritizing heat generation, is as follows: The heat generation ratio characteristics of the fuel cell stack at various power points under different temperature conditions have been obtained from the aforementioned experimental data. The power data corresponding to the maximum heat generation ratio at different temperatures has been determined. ; When power demand satisfy At that time, the output power that enables the fuel cell to operate The battery replenishes the remaining power required, that is... ; When power demand At that time, in order to heat up the fuel cell as quickly as possible, it was still necessary to... At this point, the excess power of the fuel cell is supplied to the battery to charge it, and the battery's power... .

[0046] Under this strategy, fuel cells can increase their own temperature at the highest cost-effectiveness in low-temperature environments.

[0047] S3. Construct a vehicle longitudinal dynamics model, a power battery SoC model, and a fuel cell stack thermal model for predictive control to describe the dynamic-thermal response in the prediction time domain. Vehicle longitudinal dynamics model: The vehicle dynamics model is the prerequisite for solving the vehicle's power demand and achieving optimal energy allocation. The driving state of a fuel cell vehicle is determined by the resultant force of various external forces acting on the vehicle. When performing power calculations, the interference effects of various driving resistances must be fully considered. During operation, the vehicle needs to overcome various driving resistances, such as tire rolling resistance, air resistance, climbing resistance, and inertial resistance generated during acceleration. The driving force required for a car to move is expressed as: (9); in, The driving force exerted by the FCV on the tires. The rolling resistance experienced by the vehicle, The air resistance experienced by the vehicle, The climbing resistance experienced by the vehicle. The acceleration resistance experienced by the vehicle; Tire rolling resistance is expressed as: (10); Air resistance is expressed as: (11); The climbing resistance is expressed as: (12); Acceleration resistance is expressed as: (13); in, For the overall vehicle weight of FCV, The rolling resistance coefficient of the wheel. The slope angle of the road. For windward area, air density, The air drag coefficient, For car speed, This is the rotational mass conversion factor; Given the FCV's speed under specific operating conditions, the required driving power during operation can be calculated based on the vehicle's driving force and speed. The calculation formula is as follows: (14).

[0048] The FCV model parameters are selected from the high-precision commercial simulation software AVL / Cruise M. The model parameters are shown in Table 5.

[0049] Table 5. FCV Model Parameter Table;

[0050] Power Battery SoC Model: Fuel cell vehicles are generally equipped with a power battery as an auxiliary power source to compensate for the performance shortcomings of a single fuel cell system. On the one hand, under high-load conditions such as rapid acceleration and hill climbing, the power battery can release its stored electrical energy to provide sufficient power support for the entire vehicle and meet the power performance requirements; on the other hand, when the vehicle brakes and decelerates, the power battery can recover and store regenerative braking energy, significantly improving the vehicle's fuel economy.

[0051] In light of the research needs of FCV energy management, the modeling of power batteries can disregard complex characteristics such as electrochemical reactions and aging during charging and discharging. Therefore, the Rint equivalent model is used to complete the power battery modeling. (15); in, This is the open-circuit voltage of the battery. For the battery current, This is the equivalent internal resistance of the battery, i.e., the battery charging resistance and discharging resistance. This refers to the voltage across the battery terminals; The current of the power battery is expressed as: (16); The real-time estimation of the battery state of charge (SOC) based on the ampere-hour integration method is expressed as follows: (17); in, This refers to the rated capacity of the power battery.

[0052] Fuel cell stack thermal model: The circulation branch of the fuel cell cooling system is as follows Figure 5 As shown in the diagram, this system relies on coolant temperature to achieve intelligent loop switching. When the coolant temperature exceeds a preset threshold, the three-way valve opens, and the coolant flows through the radiator for forced heat dissipation, which can quickly complete the cooling process. When the coolant temperature is within the normal range, the three-way valve remains closed, and the coolant only flows along the pipes for heat dissipation. Relying solely on pipe heat dissipation is sufficient to meet the system's heat dissipation requirements.

[0053] The temperature of the fuel cell stack is calculated as follows: (18); in, , , These are the specific heat capacity, mass, and temperature of the fuel cell. It is the total energy produced by the fuel cell. It is the output power. , and These are the heat exchange amounts for the coolant, the reacting gases, and the ambient air, respectively.

[0054] Under the dynamic operating conditions of fuel cell vehicles, the stack operating current changes in real time with the load. Based on this, the consumption of hydrogen and oxygen and the amount of water generated in the electrochemical reaction can be dynamically calculated. The calculation formula is as follows: (19); in, , These refer to the oxygen that participates in the reaction and the amount of oxygen. It is the amount of water produced by the electrochemical reaction. This is the operating current of the fuel cell. It is the Faraday constant, which characterizes the amount of electric charge carried by a unit mass of substance.

[0055] The output electrical energy of a fuel cell is determined by its output voltage and current. To establish a complete dynamic model of the fuel cell stack, the mathematical expression for its output voltage is as follows: (20); Based on the stoichiometric constraints of the electrochemical reactions in fuel cells, the partial pressure of hydrogen on the anode side and the partial pressure of oxygen on the cathode side are respectively correlated with the total cathode pressure and the total anode pressure of the fuel cell stack. The specific quantitative relationships are as follows: (twenty one); (twenty two); in, and These are the anode pressure and the cathode pressure, respectively.

[0056] Based on the charge conservation principle of the electrochemical reaction in a fuel cell, the consumption of reactant gases is linearly related to the operating current, thus causing the anode and cathode pressures to change dynamically with the current. Combining experimentally measured current and pressure data, the quantitative expression for the cathode pressure is as follows: (twenty three); The pressure at the anode can be calculated from the pressure at the cathode, and the calculation formula is as follows: (twenty four); in, It is the voltage drop when the current density is zero. It is the current density of the fuel cell. It is an empirical constant. The definition of current density is as follows: (25); in, It is the activation area of ​​the fuel cell. and The values ​​are all related to temperature, the pressure of the reacting gas, and the water saturation pressure, as shown in the following formulas: (26); The ohmic polarization voltage is a comprehensive reflection of the ohmic losses in proton and electron conduction during charge transport in a fuel cell: the ionic resistance of the proton exchange membrane hinders proton migration, while the electronic resistance of the electrodes and current collectors hinders electron flow; both together form the ohmic polarization voltage. According to Ohm's law, its value is linearly related to the operating current density, and the corresponding mathematical expression is as follows: (27); in, This refers to the internal resistance of the fuel cell, which can be expressed by the membrane conductivity, as shown in the following formula: (28); in, and It refers to the thickness and conductivity of the proton exchange membrane.

[0057] The proton conduction performance of proton exchange membranes is significantly affected by temperature. The conductivity changes systematically with temperature, and its relationship with the operating temperature of the fuel cell is expressed as follows: (29); in, It is a parameter related to the water content of the membrane. These are empirical parameters.

[0058] The calculation formula is as follows: (30); in, These are all empirical parameters. It is the water content of the proton exchange membrane.

[0059] Water content of proton exchange membrane The value ranges from 0 to 14, and it corresponds one-to-one with the water content of the membrane. =0 corresponds to a water content of 0%, when When the water content of the membrane is 14, the water content is 100%.

[0060] The core cause of concentration polarization voltage is the limited mass transfer of reactants: When a fuel cell is operating, if the reaction consumption rate at the electrode surface exceeds the diffusion rate of gas into the catalyst layer, the concentration of reactants at the interface will be lower than the concentration in the bulk channel, forming a concentration difference and generating a polarization voltage. The higher the current density and the faster the reaction rate, the more significant the mass transfer limitation effect, and the greater the concentration polarization voltage. The calculation formula is as follows: (31); in, It is the current density value that causes a sudden drop in voltage. These are empirical parameters, related to fuel cell temperature. The calculation formula is as follows: (32); in, This is the saturated water vapor pressure, and its calculation formula is as follows: (33); Coolant cooling is the core heat dissipation method of a fuel cell cooling system. The coolant flows through the fuel cell stack to complete heat exchange, promptly removing excess heat generated by the reaction and effectively regulating the stack's operating temperature. Under the influence of heat transfer, the coolant's own temperature rises, and it then flows into the pipeline to the three-way valve for loop diversion and regulation. When the coolant temperature is below the preset temperature control threshold, the fluid flows directly through the bypass without needing to enter the radiator for cooling; when the coolant temperature exceeds the set limit, the three-way valve automatically switches the flow path, forcing the coolant to flow through the radiator for forced cooling. The formula for calculating the heat removed by the coolant from the fuel cell is as follows: (34); in, , , These are the coolant flow rate, heat capacity, and inlet temperature.

[0061] The operating temperature of a fuel cell is always higher than the ambient temperature. This temperature difference promotes natural heat exchange between the fuel cell stack and the external environment. The corresponding formula for calculating the heat exchange is as follows: (35); in, It is the temperature of the environment surrounding the fuel cell. It is the thermal resistance of the fuel cell.

[0062] Fuel cells utilize three heat dissipation methods: coolant cooling, air cooling, and reactant gas cooling. The temperatures and compositions of the reactant gases entering and exiting the fuel cell stack differ, creating a heat difference that affects the stack temperature. The expression for this gas heat change is as follows: (36); in, and These are the heat of the output gas and the heat of the input gas, respectively.

[0063] The system supplies two types of reactant gases to the fuel cell stack: hydrogen and oxygen-containing air. Due to economic constraints, air is generally used directly as the oxygen source, and oxygen accounts for approximately 21% of the internal volume. The theoretical hydrogen consumption is directly related to the operating current. Insufficient hydrogen supply can lead to fuel starvation. Therefore, the actual hydrogen flow rate is set to a fixed multiple of the theoretical reaction rate, as calculated in the following formula: (37); in, This refers to the amount of hydrogen gas input. It is the excess coefficient of hydrogen. It is the amount of hydrogen gas that participates in the reaction, which is calculated from equation (37).

[0064] The actual total oxygen supply can be calculated from the oxygen consumption of the fuel cell reactor. Considering that oxygen accounts for approximately 21% of the volume of air, the formula for calculating the air input flow rate is derived as follows: (38); in, It refers to the amount of air input. It is the excess oxygen coefficient. It is the amount of air that participates in the reaction, calculated by equation (38).

[0065] The water content of the proton exchange membrane (PEM) is a key parameter affecting the output performance of a fuel cell, and the humidity of the reactant gas directly determines the hydration state of the membrane. When the reactant gas entering the stack is too dry, the water content of the PEM decreases, leading to a decline in stack performance. To ensure the output performance and lifespan of the fuel cell, the reactant gas must be thoroughly humidified before entering the stack. Therefore, the reactant gas entering the stack is in a humid state, and its water content is calculated using the following formula: (39); (40); in, and These are the temperatures of the reactant gases input at the anode and cathode, respectively. and These are the water content of the reaction gases input at the anode and cathode, respectively.

[0066] The heat of the input reactant gas is expressed as follows: (41); in, and These are the heat capacities of hydrogen gas and gaseous water molecules, respectively. It is the standard pressure temperature. It is the heat capacity of air.

[0067] Fuel cells directly convert the chemical energy of hydrogen and oxygen into electrical energy. Nitrogen from the air does not participate in the electrochemical reaction, therefore the nitrogen flow rate at the inlet and outlet of the fuel cell stack remains constant. To ensure the reaction rate and power supply stability, the amount of hydrogen and oxygen supplied to the system is greater than the theoretical reaction consumption. Unreacted hydrogen and oxygen will remain in the exhaust gas. The relevant parameter calculation formulas are as follows: (42); (43); (44); in, and These represent the amount of hydrogen gas output from the anode and cathode, respectively. This refers to the amount of oxygen output from the cathode. The formula for calculating water vapor in the fuel cell outlet gas is as follows: (45); (46); in, It is the water vapor output from the anode. It is the water vapor output from the cathode.

[0068] The water produced by the fuel cell exists in both gaseous and liquid states. The content of gaseous water can be calculated using equations (45) and (46); this invention assumes that all liquid water is discharged from the cathode, and the corresponding calculation formula is as follows: (47); in, This refers to the liquid water output from the cathode of the fuel cell. The formula for the heat of the output gas from the fuel cell is as follows: (48); A fuel cell system consists of multiple stacked cells forming a stack module, not an independent single-cell structure. The temperature of a fuel cell, as commonly referred to, is the overall temperature of the entire stack, while the internal temperature distribution is not uniform. The coolant enters the stack at a relatively low temperature and continuously absorbs heat generated by the reaction during its flow, causing its own temperature to gradually rise and the heat exchange temperature difference between it and the stack to continuously decrease. Therefore, the coolant outlet temperature is significantly higher than the inlet temperature. When the inlet and outlet temperature difference exceeds the allowable range, it can easily lead to localized thermal runaway, resulting in proton exchange membrane dehydration and performance failure, severely impacting the stack's lifespan and operational safety. To control the temperature difference within a reasonable range and prevent thermal failures, precise control of the coolant inlet temperature is required, thus establishing a corresponding inlet temperature calculation model. This invention assumes that the coolant temperature leaving the radiator is equal to the radiator body temperature. Based on this condition, the coolant inlet temperature calculation formula is derived as follows: (49); in, It's about the quality of the radiator. It is the heat capacity of the radiator. It refers to the heat exchange between the radiator and the air. The formula is as follows: (50); in, It is the airflow at the radiator. It is the heat capacity of air. It is the temperature of the air before it enters the radiator. It is the temperature at which air is output from the radiator.

[0069] S4. Based on the collected fuel cell vehicle information, the constructed stack heat production ratio dataset, and the dynamic-thermal response description, and combined with the differences in the operating characteristics of the fuel cell stack in different temperature ranges under low temperature conditions, a phased and differentiated energy optimization control strategy is designed. Based on S1-S3, a phased and differentiated energy optimization control strategy is designed to achieve efficient energy utilization throughout the entire process of low-temperature startup, heating up and stable operation of the fuel cell stack.

[0070] The overall strategy comprises two interconnected and coordinated control phases: low-temperature start-up and warm-up, and stable operation under normal conditions. The specific design approach is as follows: First, for the low-temperature cold start scenario with an initial stack temperature of 0℃, a rule-based precise temperature rise optimization control method is constructed to achieve rapid preheating of the stack from 0℃ to 70℃, effectively solving the problems of slow heating rate and high hydrogen consumption during low-temperature start-up. Second, once the stack temperature reaches 70℃ and enters the stable operating range, a vehicle-wide thermal-electric synergistic energy optimization control strategy adapted to the low-temperature environment is established. This dynamically allocates the fuel cell output power and the power battery charging and discharging power according to the real-time load requirements of the vehicle, achieving overall system energy consumption optimization and reduction while ensuring stable stack operating temperature and safe and reliable operation.

[0071] (1) Low-temperature start-up and heating: In step S2, the output efficiency and heat generation of the fuel cell stack under low-temperature conditions were systematically investigated. In response to the prominent problems of high vehicle heating demand and high system energy loss in low-temperature environments, this step established the core control principle of prioritizing heat generation. By coordinating and dynamically matching the coolant circulation pump and radiator fan, and adopting differentiated control schemes according to different operating temperature ranges of the fuel cell stack, the comprehensive optimization of fuel cell stack heat generation and energy consumption under low-temperature conditions was achieved.

[0072] The control rules for different actuators are as follows: The coolant pump is the core actuator for fuel cell stack temperature regulation. It adjusts its operating speed to change the circulation flow rate of the coolant in the stack channels, cooling pipes, and radiator circuit, thereby controlling the total amount of heat removed from the stack per unit time. The difference between the actual coolant temperature at the stack outlet and the set desired temperature (70℃) is used as the core criterion for pump control. The optimal operating temperature range for the stack is set to 60℃~80℃. Phased pump output control rules are formulated: when the stack temperature is below 60℃, i.e. When the system determines that heat dissipation needs to be reduced and the fuel cell stack temperature needs to be accelerated, the water pump maintains basic circulation at the lowest speed to prevent heat loss due to excessively rapid coolant circulation, while also maintaining uniform fuel cell stack temperature to prevent localized overheating. At this time, the water pump operates at only 10% of its rated output power. The fuel cell stack operates within its optimal range and needs to maintain steady-state conditions. The water pump operates at low speed to maintain flow, only compensating for natural heat dissipation and pipeline heat loss. When the coolant temperature at the fuel cell stack outlet is below 68°C, the water pump maintains 30%~50% of its output power. When the temperature approaches the upper limit, the output power is linearly adjusted according to the difference between the desired temperature and the outlet temperature, expressed as: When the temperature is between 68℃ and 80℃, that is... The system is in a state of mild heat dissipation, requiring a slight increase in flow rate to enhance heat dissipation and prevent the temperature from rising continuously. At this time, the water pump output power is within the range of 60% to 80%, and is linearly adjusted according to the coolant outlet temperature, expressed as: When the coolant temperature at the fuel cell stack outlet exceeds 80°C, that is... At this point, the temperature of the fuel cell stack significantly exceeds the optimal range, requiring maximum flow rate to enhance heat dissipation and quickly bring the temperature back to a safe range. The water pump operates at its rated power during this time. .

[0073] The control rules for the radiator fan are as follows: The fan enhances the convective heat transfer between the coolant and the external environment by adjusting the airflow velocity through the radiator, thereby accelerating the coolant cooling rate. This study employs rule-based control logic, using the water pump adjustment effect and the coolant temperature at the fuel cell outlet as the core judgment criteria. The fan speed is activated and adjusted only when the water pump flow rate adjustment cannot meet the heat dissipation requirements, ensuring both the response speed of temperature control and minimizing fan power consumption. At the operating condition judgment level, the fan's operating condition classification is completely consistent with that of the water pump, but its activation has a lag: the corresponding fan adjustment phase only begins after the water pump has completed its maximum speed adjustment for the corresponding operating condition and the fuel cell temperature has not yet reached the desired level. The specific control rules are as follows: when the coolant temperature at the fuel cell outlet is less than 60℃, i.e. The fan is completely turned off to prevent air cooling from exacerbating heat loss from the fuel cell stack, and a water pump is used to rapidly heat the stack. When the fuel cell stack is at 60℃~70℃, that is... The fuel cell stack is operating within its optimal range and needs to maintain steady-state conditions. If the water pump's cooling capacity is insufficient, and the temperature exceeds 65°C and shows no decreasing trend for 5 seconds, the fan should be started and linearly adjusted within 40% to 70% of its rated output power to assist in cooling the fuel cell stack. The expression is: When the temperature is between 70℃ and 80℃, that is... The system is in a state of mild cooling, with the fan operating at 70%~80% of its rated power. This balances cooling efficiency and energy consumption while quickly reducing the fuel cell stack temperature to 70℃. The expression is: When the coolant temperature at the fuel cell stack outlet exceeds 80°C, the fuel cell stack needs to be rapidly cooled back to its optimal temperature range, with the fan operating at full power. .

[0074] (2) Normal and stable operation: When the temperature of the fuel cell stack rises to the appropriate operating range, the energy consumption characteristics corresponding to the aforementioned thermal actuator control rules are combined with the vehicle driving status and network information collected by S1 to predict the operating conditions and carry out the vehicle economic optimization control.

[0075] At low temperatures, the internal reaction kinetics, proton exchange membrane conductivity, and reactant mass transfer processes of fuel cell stacks all undergo significant changes, resulting in efficiency characteristics that differ markedly from those at room temperature. Traditional energy optimization methods based on room-temperature efficiency models are often constrained by the decrease in stack efficiency at low temperatures, making it difficult to output net power sufficient to meet the vehicle's requirements. For the low-temperature operating characteristics of fuel cell vehicles (FCVs), the core of this optimization problem is: while simultaneously satisfying the vehicle's driving power requirements and the power output of the thermal actuator, fully considering the differences in stack efficiency at different temperatures, and through a reasonable allocation of fuel cell output power and battery output power, minimizing hydrogen consumption throughout the vehicle's driving process while maintaining the battery's state of charge (SOC) within a reasonable range. This problem belongs to a multi-objective collaborative optimization problem under multiple constraints; therefore, this invention constructs an optimization framework based on model predictive control as follows: (51); in, To predict the total hydrogen consumption in the time domain while satisfying system terminal constraints, It controls the input variables. It controls the feasible region of the input variables. The representative system is The amount of hydrogen consumed at any given time is a function of output power and temperature. It predicts the time domain. It is the terminal constraint of the state variable. The purpose of setting this terminal constraint is to balance the impact of changes in battery SoC energy consumption assessment. This represents the discrete form of the power battery SoC model formula (17), with the initial value of the battery SoC set to 50%. and These represent the upper and lower limits of the constraints on the power battery SoC, with values ​​of 0.2 and 0.8 respectively. Figure 6 It can be seen that within this range, the battery's internal resistance is the lowest, and its operating efficiency is relatively high. Fuel cell output power. With battery output power As a control variable and These are the upper and lower limits of the battery's output power. and These represent the upper and lower limits of the fuel cell's output power, corresponding to the output capacity ranges of the two types of power sources, respectively. The fuel cell and the power battery together provide the power required for vehicle operation. Compared to the energy consumption of all thermal actuators Energy supply.

[0076] S5. Using the system state of the current control cycle as the initial value, combined with the future operating condition information predicted by the network information, and based on the vehicle longitudinal dynamics model of S3, the energy optimization problem of the fuel cell vehicle is solved in the rolling time domain based on the optimization control strategy in S4, and the optimal control sequence in the future time domain is obtained; only the optimal control command of the current control cycle is executed, and S1-S4 are repeated in the next control cycle to realize rolling real-time energy optimization control.

[0077] The prediction time domain is 20s, the sampling time is 1s, and the optimization problem is solved using FDP, including the following specific steps: The specific content of solving the fuel cell vehicle energy optimization problem in S5 based on forward dynamic programming (FDP) and the optimization control strategy in S4 in the rolling time domain is as follows: S51. Discretization and State Space Partitioning of Energy Optimization Problems: Optimize the time domain Discretize into N stages, step size For state variables of SoC within a feasible range The internal area is uniformly divided to construct a set of state grid points; simultaneously, the control variable, fuel cell output power, is... Within the range of values Discretization is performed; S52. Define the cost function and state transition matrix: exist At time , when the state variable is And take action At that time, the battery output power is determined by the power demand-power balance relationship. The cost of this stage is the consumption of hydrogen. The calculation incorporates equation (8) from S21; S53. Forward dynamic programming recursive solution: From the initial moment The process involves recursively calculating the minimum cumulative hydrogen consumption and optimal control sequence for each stage and discrete state point. The specific steps are as follows: S531. Initialization: Set the initial state Cumulative cost ; S532. Phase recursion: For For the current Iterate through all feasible For each group : Calculate the corresponding Compared to the current cost of hydrogen consumption Store from the initial state to Temporary cumulative cost: ; S533. After the recursion is completed, the state with the minimum cumulative cost is selected as the optimal terminal state in the terminal stage. Starting from the optimal terminal state, the process backtracks to the initial stage based on the recorded predecessor state and control quantity to obtain the globally optimal fuel cell power sequence and the corresponding power battery SoC trajectory.

[0078] Through the above steps, the FDP method makes full use of long-term network information to plan the optimal SoC reference trajectory in the prediction time domain and transmits it to the lower-level controller, providing a reference for real-time energy optimization control.

[0079] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0080] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for energy optimization control of fuel cell vehicles in low-temperature environments, characterized in that, Includes the following steps: S1. Collect fuel cell vehicle information in real time, preprocess the collected fuel cell vehicle information, and construct a temperature-related fuel cell hydrogen consumption model. S2. Construct a dataset of heat generation ratio for fuel cell stacks under low-temperature conditions; where the ratio of temperature rise to hydrogen consumption is defined as the heat generation ratio. S3. Construct a vehicle longitudinal dynamics model, a power battery SoC model, and a fuel cell stack thermal model for predictive control to describe the dynamic-thermal response in the prediction time domain. S4. Based on the collected fuel cell vehicle information, the constructed stack heat production ratio dataset, and the dynamic-thermal response description, and combined with the differences in the operating characteristics of the fuel cell stack in different temperature ranges under low temperature conditions, a phased and differentiated energy optimization control strategy is designed. S5. Using the system state of the current control cycle as the initial value, combined with the future operating condition information predicted by the network information, and based on the vehicle longitudinal dynamics model in S3, the energy optimization problem of the fuel cell vehicle is solved in the rolling time domain based on the optimization control strategy in S4, and the optimal control sequence in the future time domain is obtained; only the optimal control command of the current control cycle is executed, and S1-S4 are repeated in the next control cycle to realize rolling real-time energy optimization control. S4. Based on the collected fuel cell vehicle information, the constructed fuel cell stack heat production ratio dataset, and the dynamic-thermal response description, and considering the differences in the operating characteristics of the fuel cell stack in different temperature ranges under low-temperature conditions, the specific content of the phased and differentiated energy optimization control strategy is as follows: The overall strategy comprises two control phases: low-temperature start-up and warm-up, and stable operation under normal conditions. Specifically: Low-temperature start-up temperature control stage: For the low-temperature cold start scenario with an initial stack temperature of 0℃, a rule-based temperature optimization control strategy is constructed to achieve preheating of the stack from 0℃ to 70℃. Normal and stable operation control phase: After the stack temperature rises to 70℃ and enters the stable operating range, an optimization strategy based on model predictive control is constructed to dynamically allocate the fuel cell output power and the power battery charging and discharging power according to the real-time load demand of the vehicle, so as to optimize the overall energy consumption of the system.

2. The energy optimization control method for a fuel cell vehicle in a low-temperature environment according to claim 1, characterized in that, The S1 collects real-time information about fuel cell vehicles, including ambient temperature, fuel cell thermal state, power battery system status, vehicle driving status, and network connectivity information for predicting operating conditions.

3. The energy optimization control method for a fuel cell vehicle in a low-temperature environment according to claim 1, characterized in that, The specific content of the temperature-related fuel cell hydrogen consumption model constructed in S1 is as follows: A proton exchange membrane fuel cell converts chemical energy into electrical energy through the movement of hydrogen ions and electrons during the reaction of hydrogen and oxygen. The output voltage is: （1）； in, Indicates the fuel cell voltage. Indicates the open-circuit voltage of the fuel cell. Indicates the active voltage drop. This represents the voltage drop across the resistor. Indicates concentration voltage drop; The open-circuit voltage of a fuel cell is the potential difference resulting from the change in Gibbs free energy of the system during the electrochemical reaction of hydrogen and oxygen. Its value is directly related to the activity of the reactants and is expressed as: （2）； in, This represents the open-circuit voltage of the fuel cell reaction under standard atmospheric pressure. It is a universal gas constant. For the types of reactants, For fuel cell temperature, Here is the molar mass of oxygen; The active loss is represented by the active voltage drop, as follows: （3）； in, Indicates the speed of the electrode reaction. Represents the exchange current density. Indicates alternating current. Indicates the fuel cell current; The formulas for calculating the consumption rates of hydrogen and oxygen during the reaction are as follows: （4）； in, and These represent the consumption rates of hydrogen and oxygen, respectively. and These represent the molar masses of hydrogen and oxygen, respectively. The formulas for calculating the output power and hydrogen consumption rate of a fuel cell are as follows: （5）； in, Indicates the fuel cell power; The relationship between the actual output power and the available output power of a fuel cell is as follows: （6）； The hydrogen consumption rate of a fuel cell is related to its output power, and the hydrogen consumption rate is expressed as: （7）； in, The hydrogen consumption rate of PEMFC, For PEMFC efficiency, It has the lowest calorific value of hydrogen.

4. The energy optimization control method for a fuel cell vehicle in a low-temperature environment according to claim 1, characterized in that, The specific content of constructing the fuel cell stack heat generation ratio dataset under low-temperature conditions in S2 is as follows: S21. Analysis of the efficiency characteristics of fuel cell stack in low-temperature environment; The higher the temperature, the higher the output power and voltage of the fuel cell stack at the same current density, that is: （8）； S22. Analysis of the temperature rise characteristics of the fuel cell stack in a low-temperature environment; S23. Based on the analysis of the efficiency characteristics of the fuel cell stack in a low-temperature environment and the analysis of the heating characteristics of the fuel cell stack in a low-temperature environment, a power control strategy for the fuel cell stack is constructed based on the principle of prioritizing heat generation, thereby constructing a dataset of the heat generation ratio of the fuel cell stack in a low-temperature environment. The power control strategy for the fuel cell stack, based on the principle of prioritizing heat generation, is as follows: Based on the heat generation ratio characteristics of the fuel cell stack at various power points under different temperature conditions, the power data corresponding to the maximum heat generation ratio at different temperatures was determined. ; When power demand satisfy At that time, the output power that enables the fuel cell to operate The battery replenishes the remaining power required, that is... ; When power demand At that time, in order to heat up the fuel cell as quickly as possible, it was still necessary to... At this point, the excess power of the fuel cell is supplied to the battery to charge it, and the battery's power... .

5. The energy optimization control method for a fuel cell vehicle in a low-temperature environment according to claim 1, characterized in that, The specific content of constructing the vehicle longitudinal dynamics model in S3 is as follows: The driving state of a fuel cell vehicle is determined by the combined force of various external forces acting on the entire vehicle; The driving force required for a car to move is expressed as: （9）； in, The driving force exerted by the FCV on the tires. The rolling resistance experienced by the vehicle, The air resistance experienced by the vehicle, The climbing resistance experienced by the vehicle, The acceleration resistance experienced by the vehicle; Tire rolling resistance is expressed as: （10）； Air resistance is expressed as: （11）； The climbing resistance is expressed as: （12）； Acceleration resistance is expressed as: （13）； in, For the overall vehicle weight of FCV, The rolling resistance coefficient of the wheel. The slope angle of the road. For windward area, air density, The air drag coefficient, For car speed, This is the rotational mass conversion factor; Given the FCV's speed under specific operating conditions, the required driving power during operation can be calculated based on the vehicle's driving force and speed. The calculation formula is as follows: （14）。 6. The energy optimization control method for a fuel cell vehicle in a low-temperature environment according to claim 1, characterized in that, The specific content of building the power battery SoC model in S3 is as follows: The power battery was modeled using the Rint equivalent model. （15）； in, This is the open-circuit voltage of the battery. For the battery current, This is the equivalent internal resistance of the battery, i.e., the battery charging resistance and discharging resistance. This is the voltage across the battery terminals; The current of the power battery is expressed as: （16）； The real-time estimation of the battery state of charge (SOC) based on the ampere-hour integration method is expressed as follows: （17）； in, This refers to the rated capacity of the power battery.

7. The energy optimization control method for a fuel cell vehicle in a low-temperature environment according to claim 1, characterized in that, The specific content of constructing the thermal model of the fuel cell stack in S3 is as follows: Assuming the coolant temperature when it leaves the radiator is equal to the radiator body temperature, the formula for calculating the coolant inlet temperature is derived as follows: （49）； in, It's about the quality of the radiator. It is the heat capacity of the radiator. It is the heat exchange between the radiator and the air. It is the differential of x with respect to y. It is the temperature at the coolant inlet. The coolant carries away the heat from inside the fuel cell. Heat exchange between the radiator and the air The formula is as follows: （50）； in, It is the airflow at the radiator. It is the heat capacity of air. It is the temperature of the air before it enters the radiator. It is the temperature at which air is output from the radiator.

8. The energy optimization control method for a fuel cell vehicle in a low-temperature environment according to claim 1, characterized in that, The specific content of S5, which uses forward dynamic programming to solve the energy optimization problem of fuel cell vehicles based on the optimization control strategy in S4 within the rolling time domain, is as follows: S51. Discretization and State Space Partitioning of Energy Optimization Problems: Optimize the time domain Discretize into N stages, step size For state variables of SoC within a feasible range The internal area is uniformly divided to construct a set of state grid points; simultaneously, the control variable, fuel cell output power, is... Within the range of values Discretize the data; S52. Define the cost function and state transition matrix: exist At time , when the state variable is And take action At that time, the battery output power is determined by the power demand-power balance relationship. The cost of this stage is the amount of hydrogen consumed. ; S53. Forward dynamic programming recursive solution: From the initial moment The process involves recursively calculating the minimum cumulative hydrogen consumption and optimal control sequence for each stage and discrete state point. The specific steps are as follows: S531. Initialization: Set the initial state Cumulative cost ; S532. Phase recursion: For For the current Iterate through all feasible For each group : Calculate the corresponding Compared to the current cost of hydrogen consumption Store from the initial state to Temporary cumulative cost: ; S533. After the recursion is completed, the state with the minimum cumulative cost is selected as the optimal terminal state in the terminal stage. Starting from the optimal terminal state, the process backtracks to the initial stage based on the recorded predecessor state and control quantity to obtain the globally optimal fuel cell power sequence and the corresponding power battery SoC trajectory.

Images