Sodium battery SOC and hydrogen combustion efficiency self-adaptive energy management and control system

Abstract

The invention relates to the technical field of hybrid power system energy management, and discloses a sodium battery SOC and hydrogen combustion efficiency adaptive energy management and control system, which comprises a state sensing module used for constructing a multi-dimensional state vector containing sodium ion battery SOC, temperature and total power requirements; the collaborative decision-making module is used for receiving the state vector based on a reinforcement learning model and outputting a collaborative decision-making instruction containing a power distribution coefficient and a thermal management control factor; the electrothermal coupling management module is used for regulating and controlling the transfer of the waste heat of the hydrogen fuel cell to the sodium ion battery according to the thermal management control factor; and the instruction execution module is used for analyzing the collaborative decision instruction into a bottom layer control signal for each power source and a heat management executor. According to the method, the unified decision framework of the electric energy and the heat energy is constructed, so that global self-optimization of the electric power flow and the heat energy flow is realized, and the overall energy utilization efficiency, the dynamic response capability and the environmental adaptability to wide working temperatures of the hybrid power system can be effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hybrid power system energy management, in particular to a sodium battery SOC and hydrogen fuel efficiency adaptive energy management and control system. BACKGROUND

[0002] Hybrid power systems have attracted extensive attention in various application scenarios, such as transportation, aerospace, and stationary energy storage, due to their ability to integrate the advantages of different energy conversion devices, achieving high efficiency and low emission performance goals. In order to effectively coordinate the work of each power source in the system and meet the dynamically changing power demand, energy management strategy is the core element of hybrid power system design. These strategies usually rely on the perception of system state and the optimized allocation of power flow, including rule-based control, instantaneous optimization method or global optimization method, aiming to improve fuel economy or energy efficiency.

[0003] However, existing energy management strategies fail to fully consider the unique electrochemical characteristics of emerging sodium-ion battery technology, especially the impact of power output capability at different states of charge (SOC). Traditional battery management systems are often based on a relatively uniform power limitation model, but this deviates from the nonlinear characteristics of sodium-ion batteries, which have higher rate discharge capability at high SOC, and significant power output limitations at low SOC or specific temperatures. If the actual performance of sodium-ion batteries in different SOC intervals is not accurately evaluated and utilized, it may lead to the underutilization of battery potential, or accelerate its performance degradation and shorten its lifespan when it exceeds its safe operating boundaries, thereby affecting the dynamic response capability and long-term reliability of the entire hybrid power system.

[0004] In addition, the current energy management strategies of hybrid power systems generally treat electrical energy management and thermal energy management as relatively independent links, lacking in-depth utilization of the electro-thermal coupling effects between different power sources. For example, hydrogen fuel cells generate a large amount of waste heat in the process of converting hydrogen chemical energy into electrical energy. In existing technologies, this part of waste heat is usually directly dissipated to the environment through a cooling system, causing energy waste. At the same time, sodium-ion batteries are temperature-sensitive devices, whose working efficiency, power output capability and lifespan are significantly affected by temperature, especially in low temperature environments, where the internal resistance increases, leading to a sharp drop in available power. If the waste heat of hydrogen fuel cells can be effectively recovered and utilized to actively regulate the temperature of sodium-ion batteries, it not only provides a new approach for active-passive collaborative thermal management, but also effectively alleviates the performance limitations of sodium-ion batteries under harsh temperature conditions, thereby improving the thermal efficiency and environmental adaptability of the entire system. SUMMARY

[0005] In view of the deficiencies of the prior art, the sodium-ion SOC and hydrogen fuel efficiency adaptive energy management and control system is provided, which solves the problem that the prior art fails to fully consider the unique and nonlinear power output characteristics of sodium-ion batteries at different state of charge (SOC) and ignores the potential heat coupling effect between the hydrogen fuel cell and the sodium-ion battery, which leads to the problems of limited overall energy utilization efficiency, insufficient dynamic response capability and difficult to guarantee the operation stability under variable working conditions due to the separate management and insufficient cognition of the electrical and thermal characteristics.

[0006] To achieve the above object, the following technical scheme is adopted: the sodium-ion SOC and hydrogen fuel efficiency adaptive energy management and control system comprises:

[0007] A state perception module configured to acquire state information of the hybrid power system in real time. The state information at least includes the state of charge (SOC) of the sodium-ion battery, the temperature of the sodium-ion battery, and the total power demand of the system. In an embodiment, the state information further includes the current working efficiency or the current heat generation power of the hydrogen fuel cell.

[0008] A cooperative decision-making module in data communication with the state perception module. The cooperative decision-making module internally solidifies a pre-trained reinforcement learning model. The function of the module is to receive the state information acquired by the state perception module as input and generate a cooperative decision-making instruction through the reinforcement learning model. The cooperative decision-making instruction includes an electrical power distribution instruction and a thermal management instruction.

[0009] The cooperative decision-making instruction generated by the cooperative decision-making module is a motion vector including a power distribution coefficient and a thermal management control factor . The power distribution coefficient is used to determine the power proportion allocated to the hydrogen fuel cell, and the thermal management control factor is used to determine the intervention intensity of the subsequent heat transfer process.

[0010] The cooperative decision-making module internally integrates an electrical-thermal characteristic model of the system, which includes:

[0011] An electrical-thermal characteristic model that divides the SOC of the sodium-ion battery into at least three intervals of high, medium and low, which is used to evaluate the power output capability of the sodium-ion battery according to the current SOC interval of the sodium-ion battery.

[0012] An efficiency-power model of the hydrogen fuel cell, one of the optimization objectives of the cooperative decision-making module when generating the electrical power distribution instruction is to make the output power of the hydrogen fuel cell approach the optimal efficiency operating point determined by the model.

[0013] The reward function adopted by the reinforcement learning model on which the collaborative decision module is based in the training stage is a function containing multiple weighted terms, and at least includes an energy efficiency reward term and a temperature penalty term. The energy efficiency reward term is calculated in the following manner The temperature penalty term is calculated in the following manner ; wherein is the total efficiency of the system, is the real-time temperature of the sodium-ion battery, is the preset optimal working temperature of the sodium-ion battery, and is a preset positive weight coefficient. In a specific embodiment, the reinforcement learning model is a policy network and a value network constructed based on a proximal policy optimization algorithm.

[0014] An electro-thermal coupling management module in data communication with the collaborative decision module. The function of this module is to receive the thermal management instructions and regulate the heat transfer process from the hydrogen fuel cell to the sodium-ion battery accordingly. In one embodiment, this module uses the waste heat generated by the hydrogen fuel cell during power generation to regulate the temperature of the sodium-ion battery through an active heat exchange loop. The heat transfer power in the heat transfer process is limited by the thermal management control factor , and the relationship satisfies

[0015] ;

[0016] wherein is the maximum heat exchange coefficient of the active heat exchange loop, is the temperature of the hydrogen fuel cell, is the temperature of the sodium-ion battery.

[0017] An instruction execution module in data communication with the collaborative decision module and the electro-thermal coupling management module. The function of this module is to receive the electrical power distribution instructions and the thermal management instructions and parse them into specific control signals applied to the hydrogen fuel cell, the sodium-ion battery, and the thermal management actuators in the active heat exchange loop, respectively.

[0018] The second aspect of the present application provides a sodium electric SOC and hydrogen fuel efficiency adaptive energy management and control method, comprising the following steps:

[0019] S1, real-time perception of the state information of the hybrid power system, the state information at least including the state of charge of the sodium-ion battery, the temperature of the sodium-ion battery, and the total power demand of the system.

[0020] S2, based on a preset reinforcement learning model, and according to the state information perceived in step one, a cooperative decision operation is performed to generate an electric power distribution instruction and a thermal management instruction.

[0021] S3, according to the thermal management instruction generated in step two, an electric-thermal coupling process of actively temperature regulating the sodium ion battery by using the waste heat of the hydrogen fuel cell is regulated.

[0022] S4, according to the electric power distribution instruction generated in step two, corresponding power tasks are distributed to the hydrogen fuel cell and the sodium ion battery.

[0023] The present application provides a sodium electric SOC and hydrogen fuel efficiency adaptive energy management and control system. Has the following beneficial effects:

[0024] 1, the present application through the cooperative decision module, based on the accurate modeling of hydrogen fuel cell efficiency-power characteristics, adaptively distributes electric power. This module can maintain the output power of the hydrogen fuel cell in its optimal efficiency operating interval to undertake the basic load under the premise of meeting the total power demand, while using the high-rate discharge characteristics of the sodium ion battery in the high SOC interval to respond to the instantaneous peak power demand. This distribution method avoids the frequent work of the hydrogen fuel cell in the low efficiency interval, thereby reducing the total hydrogen consumption to complete the same task profile.

[0025] 2, the present application introduces an electric-thermal coupling management module, which constructs an active heat transfer path from the hydrogen fuel cell to the sodium ion battery. Through the thermal management instruction generated by the cooperative decision module, the system can actively regulate the temperature of the sodium ion battery using the waste heat of the hydrogen fuel cell, so that its working temperature can be maintained near the preset optimal working temperature. This technical feature ensures that the sodium ion battery can maintain its preset power output capability even in low temperature environment, thereby ensuring the performance consistency and operation reliability of the entire hybrid power system in a wide temperature range.

[0026] 3, the reinforcement learning model used in the present application, whose decision basis is a multi-dimensional real-time state vector containing SOC, temperature, power demand, etc., and its output is a cooperative action vector containing power distribution and thermal management. This method is driven by a reward function that integrates system energy efficiency, SOC stability, temperature health, etc. It can autonomously learn and execute a dynamic balance control strategy to adapt to complex and variable flight conditions, achieving unified, closed-loop and adaptive management of the electrical system and thermal system. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 The system architecture diagram of the present application;

[0028] Figure 2A flowchart of the method of the present application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work are within the protection scope of the present application.

[0030] Embodiment:

[0031] Please refer to the accompanying drawings Figure 1 , the sodium-ion battery SOC and hydrogen combustion efficiency adaptive energy management and control system provided in the embodiments of the present application comprises:

[0032] A state perception module is configured to acquire state information of the hybrid power system in real time, and the state information at least includes a state of charge of the sodium-ion battery, a temperature of the sodium-ion battery, and a total power demand of the system.

[0033] In the embodiment, the state perception module is configured as a perception basis of the adaptive energy management and control system, and its function is to provide a comprehensive, accurate and multi-dimensional system real-time state representation for the subsequent collaborative decision module. In order to enable the collaborative decision module to perform a global optimization decision based on the coupling of electric energy and thermal energy, the state perception module in the embodiment is designed to be able to integrate information from multiple physical domains such as electrical, thermal, system performance and task profile.

[0034] Specifically, the state perception module is connected with a plurality of sensors, estimators and upper controllers (such as flight controllers) arranged in the hybrid power system through data interface connection, and at the beginning of each control cycle , a state vector is collected and constructed. The state vector is carefully designed to ensure that it contains all the necessary information required for executing collaborative control decisions.

[0035] The state vector is defined as:

[0036] ;

[0037] Each component in the vector has its specific physical meaning and acquisition method, and provides support for the specific decision dimension of the collaborative decision module.

[0038] Regarding the state of charge of the sodium-ion battery , which represents the remaining available capacity of the battery and is one of the core bases for power allocation. The state awareness module obtains this value through an embedded SOC estimation unit. Preferably, the estimation unit uses real-time integration method for real-time calculation, and its state update equation is:

[0039] ;

[0040] wherein, is the SOC value of the last control period, is the actual current flowing through the sodium-ion battery measured by the current sensor in the current period (positive for discharging and negative for charging), is the control period length, is the rated capacity of the sodium-ion battery. This SOC value is not only directly related to the judgment of the sodium-ion battery segmented power characteristics by the collaborative decision module, but also serves as a key monitoring parameter for maintaining the battery health state.

[0041] Temperature of the sodium-ion battery , which directly reflects the thermal state of the battery and is a key input for the electric-thermal coupled management of the present application. The state awareness module directly measures and obtains the real-time value of this temperature by connecting with one or more temperature sensors (such as thermistors or thermocouples) arranged inside or on the surface of the sodium-ion battery pack. This temperature information is used by the collaborative decision module to evaluate the current power output capability of the sodium-ion battery (through a temperature correction function ), and is also an important basis for determining the thermal management control factor to judge whether and to what extent heat needs to be introduced from the hydrogen fuel cell.

[0042] Total power demand of the system , which represents the total power required for the current flight and task load of the UAV. The state awareness module establishes a communication connection with the flight control unit of the UAV to read the total demand power value calculated by the flight control unit according to the current flight attitude, speed and task instructions in real time. This parameter is the fundamental constraint for the power allocation of the entire energy management system, and the collaborative decision module must ensure that the sum of the power allocated to the hydrogen fuel cell and the sodium-ion battery can meet this demand.

[0043] Further, in order to enable the collaborative decision module to have predictability for the dynamic performance of the system, the state vector also includes the working efficiency of the hydrogen fuel cell at the last time. This parameter is not directly measured, but is calculated by the state awareness module according to the decision action (i.e. the power allocated to the fuel cell) of the last period and the preset fuel cell efficiency model The calculation is made. The purpose of including this historical information is to provide the reinforcement learning model with time series information about the trend of fuel cell efficiency changes, so that its decision-making is not only based on the current static snapshot, but also takes into account the dynamic evolution process of the system.

[0044] In addition, in order to enable the energy management strategy to adapt to different task stages, a discrete variable representing the current flight stage is also included in the state vector . This information is also provided by the flight control unit, for example, it can specifically indicate that the UAV is in different modes such as take-off, climb, cruise, hover or return. By including this high-level task information in the state vector, the collaborative decision-making module can learn and execute the optimal energy management sub-strategy that matches the specific flight stage, thereby achieving higher-level adaptive control.

[0045] In summary, the state perception module in this embodiment provides a complete state vector that includes electrochemical state, thermodynamic state, power constraints, historical performance trends, and macro task background, providing a solid data foundation for the collaborative decision-making module to perform precise and adaptive electro-thermal collaborative energy management.

[0046] The collaborative decision-making module, connected to the state perception module, generates a collaborative decision-making instruction containing electric power allocation instructions and thermal management instructions based on a pre-set reinforcement learning model according to the state information;

[0047] In this embodiment, the collaborative decision-making module constitutes the core of the adaptive energy management and control system, its function is to receive the real-time state vector provided by the state perception module, and based on the internal solidified, pre-trained reinforcement learning model, autonomously generate a collaborative decision-making instruction that can simultaneously optimize electric power allocation and thermal management.

[0048] Specifically, the collaborative decision-making module integrates a deep neural network inside, which has been trained offline to map the complex, multi-objective energy management problem solving process into a nonlinear function from the state space to the action space . At each control period , after receiving the state vector , the module performs a forward propagation calculation, outputting a multi-dimensional action vector as the collaborative decision-making instruction.

[0049] The action vector is defined as:

[0050] ;

[0051] Each component of the motion vector corresponds to a specific physical control dimension. Power allocation coefficient It is a continuous value between 0 and 1 used to determine the proportion of power allocated to the hydrogen fuel cell. This coefficient directly determines the division of electrical energy flow between the two power sources. Thermal management control factor It is also a continuous value between 0 and 1, used to regulate the intensity of heat transfer in the electrothermal coupling management module. This factor determines the distribution of heat energy flow within the system.

[0052] In order to ensure that the decision-making process can fully take into account the unique physical characteristics of each power source, the collaborative decision-making module embeds an accurate multiphysics model of the system into its internal algorithm.

[0053] On one hand, this module includes an electrothermal characteristic model for sodium-ion batteries. A key feature of this model is that it divides the state of charge (SOC) of sodium-ion batteries into multiple operating ranges, such as a high SOC range. SOC range and low SOC range The design defines different maximum sustainable power output capabilities for each range. The aim is to accurately characterize the nonlinear characteristics of sodium-ion batteries, which exhibit stronger rate discharge capability at high SOC but limited capability at low SOC. The collaborative decision-making module determines the power allocation coefficient. At that time, it will refer to the current situation in real time. The interval in which it is located is determined to ensure that the power task allocated to the sodium-ion battery is within the capability boundary of its current state.

[0054] On the other hand, this module also includes an electrical efficiency model for hydrogen fuel cells. This model describes their power generation efficiency using a polynomial function. With output power The nonlinear relationship between them. When making decisions, one of the optimization objectives of the collaborative decision-making module is to maximize the output power of the hydrogen fuel cell. Guided to its optimal efficiency operating point Nearby. This move aims to maximize the hydrogen conversion efficiency of fuel cells.

[0055] To ensure the decision-making process fully considers the unique physical characteristics of each power source, the collaborative decision-making module embeds a precise multiphysics model of the system into its internal algorithm. On one hand, this module includes an electro-thermal characteristic model for sodium-ion batteries, which divides the state of charge (SOC) of the sodium-ion battery into multiple operating intervals to accurately characterize its nonlinear power release capability. On the other hand, the module also includes an electrical efficiency model for hydrogen fuel cells, which describes the power generation efficiency using a polynomial function. With output power the relationship between them.

[0056] The core decision logic of the collaborative decision module is carried by a pre-trained reinforcement learning agent. Preferably, the agent is built based on a framework of the proximal policy optimization algorithm. The framework includes a policy network and a value network . The policy network is responsible for directly outputting an action vector according to the input state , while the value network is responsible for evaluating the long-term value of taking a certain action in the current state, which is used to guide the training of the policy network.

[0057] The behavior of the agent is guided by a carefully designed reward function , which converts high-level control objectives into a quantifiable scalar signal. In the present embodiment, the reward function is constructed in the form of a multi-objective weighted sum:

[0058] ;

[0059] wherein,

[0060] represents the total reward value calculated in the current cycle ;

[0061] , , , are preset positive weight coefficients for balancing the importance of different optimization objectives;

[0062] is the energy efficiency reward term, the value of which is positively correlated with the total efficiency of the system; is the hydrogen consumption penalty term, the value of which is negatively correlated with the hydrogen consumption rate ;

[0063] is the SOC stability penalty term, which is used to penalize the deviation of the SOC of the sodium-ion battery from the center of its healthy working interval ; is the temperature health penalty term, the value of which is inversely proportional to the absolute value of the deviation of the temperature of the sodium-ion battery from its optimal working point ;

[0064] is the boundary penalty term, which applies a large negative value when the working parameters of any component exceed its preset safety threshold, to ensure the safe operation of the system.

[0065] In summary, the cooperative decision module in this embodiment can receive multi-dimensional system state information and autonomously output a cooperative decision instruction that takes into account multiple objectives such as system energy efficiency, fuel consumption, energy storage unit health, temperature stability, and operation safety, by deeply integrating multi-physical field models with advanced reinforcement learning algorithms. This module realizes a control paradigm shift from passive response to active optimization.

[0066] An electro-thermal coupling management module is connected to the cooperative decision module and is configured to regulate heat transfer from the hydrogen fuel cell to the sodium-ion battery according to the thermal management instruction.

[0067] In this embodiment, the electro-thermal coupling management module is an important component of the adaptive energy management and control system, and its function is to realize heat transfer and management between the hydrogen fuel cell and the sodium-ion battery. The module communicates data with the cooperative decision module and actively regulates the heat flow within the system according to the thermal management instruction output by the cooperative decision module, to maintain the sodium-ion battery operating within the preset optimal temperature range.

[0068] Specifically, the core of the design of the electro-thermal coupling management module is to utilize the waste heat generated by the hydrogen fuel cell during power generation. When the hydrogen fuel cell converts hydrogen chemical energy into electrical energy, it will inevitably generate a portion of waste heat. The temperature of this waste heat source is usually higher than the temperature of the sodium-ion battery under certain operating conditions, especially when the ambient temperature is low or the sodium-ion battery is under high-rate discharge resulting in supercooling. The electro-thermal coupling management module establishes an active heat exchange circuit to direct and transfer this portion of waste heat that would otherwise be directly dissipated to the sodium-ion battery.

[0069] The active heat exchange circuit typically includes, but is not limited to, a heat exchanger, a circulating pump, and a flow control valve. The circuit can accurately control the intensity and direction of heat transfer according to the thermal management instruction received from the cooperative decision module. The cooperative decision instruction output by the cooperative decision module contains a thermal management control factor , which is a dimensionless continuous value between , used to represent the intervention intensity of the thermal management system. When approaches 1, it indicates that the system needs to maximize heat transfer, for example when the sodium-ion battery is in a low temperature state; when approaches 0, it indicates that heat transfer is minimized or stopped.

[0070] The thermal management control factor is converted into specific control signals for physical actuators in the active heat exchange circuit through the instruction execution module. Preferably, this can be the adjustment of the circulating pump speed, or the adjustment of the flow control valve opening, to accurately control the flow rate of the cooling liquid flowing through the heat exchanger, and thus regulate the actual heat transfer capacity of the heat exchanger.

[0071] In any control cycle Heat power transferred from hydrogen fuel cell to sodium-ion battery The calculation relationship satisfies:

[0072] ;

[0073] Wherein, represents the maximum heat transfer coefficient of the active heat exchange circuit, which is determined by the geometric structure and material properties of the heat exchanger;

[0074] represents the heat source temperature of the hydrogen fuel cell at the moment, which can usually be obtained by the internal temperature sensor of the fuel cell;

[0075] represents the real-time temperature of the sodium-ion battery at the moment, which is provided by the state perception module. This formula clearly shows that the heat transfer power is directly affected by the dynamic adjustment of the thermal management control factor and the driving of the temperature difference between the two heat sources.

[0076] Through the above mechanism, the electro-thermal coupling management module can effectively recover and utilize the waste heat of the hydrogen fuel cell when needed according to the intelligent judgment of the collaborative decision-making module, providing a technical means for the main control module to actively adjust the temperature of the sodium-ion battery. This has a direct contribution to maintaining the expected power output capability of the sodium-ion battery in harsh environments (such as low temperature environments) and ensuring the dynamic response performance of the system. At the same time, this step-by-step utilization of thermal energy also indirectly reflects the optimization of overall energy efficiency, avoiding direct energy waste.

[0077] The instruction execution module is connected with the collaborative decision-making module and the electro-thermal coupling management module, and is used for parsing the electric power distribution instructions and the thermal management instructions into specific control signals for the hydrogen fuel cell, the sodium-ion battery and the thermal management actuator.

[0078] In this embodiment, the instruction execution module is the final execution end of the adaptive energy management and control system, and its function is to receive the high-level collaborative decision-making instructions generated by the collaborative decision-making module and accurately parse and convert them into bottom-layer executable specific control signals for each physical component in the hybrid power system. This module is a bridge between the upper-layer intelligent decision-making and the bottom-layer physical execution, ensuring the integrity and effectiveness of the entire closed-loop control link.

[0079] Specifically, the instruction execution module communicates data with the collaborative decision-making module and the electro-thermal coupling management module. In each control cycle , which receives the action vector output by the co-decision module . The core task of this module is to decouple and transform the two components of the action vector.

[0080] On one hand, for the power allocation coefficient in the action vector , the instruction execution module performs the following operations to generate the electrical power allocation instruction:

[0081] Firstly, the module obtains the total power demand of the system at the current time from the state perception module .

[0082] Secondly, based on the received power allocation coefficient , the module calculates the target power instruction allocated to the hydrogen fuel cell and the sodium-ion battery. Its calculation method is as follows:

[0083] The target power instruction allocated to the hydrogen fuel cell is:

[0084] ;

[0085] The target power instruction allocated to the sodium-ion battery is:

[0086] ;

[0087] where and represent the target power set value issued to the hydrogen fuel cell controller and the sodium-ion battery power converter, respectively;

[0088] is the power allocation coefficient provided by the co-decision module; is the total power demand provided by the state perception module.

[0089] On the other hand, for the thermal management control factor in the action vector , the instruction execution module performs the following operations to generate the thermal management instruction:

[0090] The module contains a pre-set mapping relationship or function inside, which is used to convert the dimensionless thermal management control factor into specific control parameters for physical actuators in the electro-thermal coupling management module.

[0091] Preferably, if the physical actuator is a variable-speed circulating pump, its target rotational speed can be set as:

[0092] ;

[0093] where is the maximum rated rotational speed of the circulating pump.

[0094] Or, if the physical actuator is a proportional control valve, its target opening Can be set as:

[0095] ;

[0096] Where, Is the maximum opening of the valve.

[0097] The instruction execution module sends the calculated specific control parameters (such as target speed or target opening) to the corresponding actuator drive circuit, to realize accurate regulation of the cooling liquid flow in the heat exchange circuit, and thus realize control of the heat transfer process.

[0098] In summary, the instruction execution module in this embodiment converts the abstract decision instructions output by the collaborative decision module into specific, real-time control signals for power electronic devices and electromechanical actuators through precise mathematical analysis and physical mapping. This module ensures that the upper-layer adaptive energy management strategy can be implemented without deviation and efficiently at the physical level, and is a necessary link to realize the collaborative control function of the entire system.

[0099] Please refer to the attached Figure 2 , the sodium SOC and hydrogen fuel efficiency adaptive energy management and control method, comprising the following steps:

[0100] S1, real-time sensing of the state information of the hybrid power system, the state information at least including the state of charge (SOC) of the sodium ion battery, the temperature of the sodium ion battery and the total power demand of the system;

[0101] S2, based on a preset reinforcement learning model, and according to the sensed state information, collaboratively decide to generate an electric power distribution instruction and a thermal management instruction;

[0102] S3, according to the thermal management instruction, regulating an electric-thermal coupling process of actively temperature regulating the sodium ion battery using the waste heat of the hydrogen fuel cell;

[0103] S4, and according to the electric power distribution instruction, distributing power tasks to the hydrogen fuel cell and the sodium ion battery.

Claims

1. A sodium-ion battery SOC and hydrogen fuel efficiency adaptive energy management and control system, characterized in that, include: The state perception module is used to acquire the state information of the hybrid power system in real time. The state information includes at least the state of charge of the sodium-ion battery, the temperature of the sodium-ion battery, and the total power demand of the system. The collaborative decision-making module, connected to the state perception module, generates collaborative decision-making instructions including power allocation instructions and thermal management instructions based on the state information according to a preset reinforcement learning model. An electrothermal coupling management module, connected to the collaborative decision-making module, is used to regulate the heat transfer process from the hydrogen fuel cell to the sodium-ion battery according to the thermal management command. The instruction execution module, connected to the collaborative decision-making module and the electrothermal coupling management module, is used to parse the power distribution instruction and the thermal management instruction into specific control signals for the hydrogen fuel cell, the sodium-ion battery and the thermal management actuator.

2. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 1, characterized in that, The state information acquired by the state sensing module further includes: the current operating efficiency or current heat production power of the hydrogen fuel cell.

3. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 1, characterized in that, The collaborative decision-making module generates a collaborative decision-making instruction that is an action vector containing a power allocation coefficient and a thermal management control factor. The power allocation coefficient The thermal management control factor is used to determine the proportion of power allocated to the hydrogen fuel cell. Used to determine the intensity of intervention in the heat transfer process.

4. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 1, characterized in that, The electrothermal coupling management module regulates the heat transfer process in the following way: The waste heat generated during the power generation of the hydrogen fuel cell is used to regulate the temperature of the sodium-ion battery through an active heat exchange loop.

5. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 3, characterized in that, The heat transfer power in the heat transfer process By the thermal management control factor The relationship is subject to the following constraints: ； in, For the maximum heat exchange coefficient, The temperature of the hydrogen fuel cell, The temperature of the sodium-ion battery is [temperature value missing].

6. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 1, characterized in that, The reinforcement learning model upon which the collaborative decision-making module bases its decisions includes an electrothermal characteristic model that divides the SOC of the sodium-ion battery into at least three intervals: high, medium, and low. The model also adaptively adjusts the evaluation of the power output capability based on the current SOC interval of the sodium-ion battery.

7. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 1, characterized in that, When generating the power allocation command, one of the optimization objectives of the collaborative decision-making module is to maintain the output power of the hydrogen fuel cell near its preset optimal efficiency operating range.

8. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 1, characterized in that, The reinforcement learning model upon which the collaborative decision-making module is based uses a reward function during training. It is a function containing multiple weighted terms, and includes at least an energy efficiency bonus term and a temperature penalty term, where: The calculation method for the energy efficiency bonus is as follows: , The temperature penalty term is calculated as follows: ; in, For the overall system efficiency, This represents the real-time temperature of the sodium-ion battery. This is the optimal operating temperature for sodium-ion batteries. and These are preset positive weighting coefficients.

9. The sodium-electric SOC and hydrogen fuel efficiency adaptive energy management and control system according to claim 1, characterized in that, The reinforcement learning model used in the collaborative decision-making module is a policy network and a value network constructed based on the proximal policy optimization algorithm.

10. A method for adaptive energy management and control of sodium-ion battery SOC and hydrogen fuel efficiency, comprising the sodium-ion battery SOC and hydrogen fuel efficiency adaptive energy management and control system according to any one of claims 1-9, characterized in that, Includes the following steps: S1. Real-time sensing of the status information of the hybrid power system, the status information including at least the state of charge of the sodium-ion battery, the temperature of the sodium-ion battery, and the total power demand of the system; S2. Based on a preset reinforcement learning model and according to the perceived state information, collaborative decision-making generates power allocation instructions and thermal management instructions. S3. According to the thermal management command, regulate the electrothermal coupling process of actively regulating the temperature of the sodium-ion battery by utilizing the waste heat of the hydrogen fuel cell; S4, and according to the power allocation instruction, allocate power to the hydrogen fuel cell and the sodium-ion battery.

Images