A self-learning based energy management method, device and medium

Abstract

The application provides a self-learning-based energy management method, device and medium, and relates to the technical field of hydrogen fuel cell vehicle, and the method comprises the following steps: acquiring vehicle main drive power, performing self-learning based on the main drive power historical value and the main drive power current value, and obtaining a main drive power self-learning value; obtaining auxiliary drive power self-learning values according to vehicle auxiliary drive device reference powers and corresponding working condition correction coefficients; generating a fuel cell system power self-learning value based on the main drive power self-learning value and the auxiliary drive power self-learning values; acquiring a battery SOC, performing power compensation on the fuel cell system power self-learning value according to the battery SOC, and generating a fuel cell system power limit value; and limiting the fuel cell system output power through the fuel cell system power limit value. According to the self-learning algorithm, the fuel cell system power limit value is updated in real time, the power demand and the energy consumption are balanced, and the fine management of the energy consumption is realized.

Description

Technical Field

[0001] This invention relates to the field of hydrogen fuel cell vehicle technology, and in particular to a self-learning-based energy management method, device, and medium. Background Technology

[0002] To protect the safe operation of the powertrain, extend battery life, and ensure stable vehicle performance, a maximum limit is set for the vehicle's output power. Since battery capacity is limited, high power output accelerates battery consumption and affects driving range, while limiting power helps balance performance and range requirements.

[0003] Patent CN202410696406.1 discloses a control method, device, equipment, and medium for a distributed electric vehicle, specifically including: acquiring the current SOC value of the power battery; when the SOC value of the power battery is less than a first SOC value, controlling the vehicle to enter an emergency rescue mode, where the first SOC value is a preset lower limit value of the power battery when the vehicle's high voltage is cut off; after the vehicle enters the emergency rescue mode, lowering the lower limit value of the power battery to below the first SOC value so that the vehicle can continue to drive to a safe location in pure electric mode. However, this patent has problems such as a fixed maximum power limit value, poor energy recovery coordination, and a single and passive power protection strategy. Specifically, the strategy of using a fixed maximum power limit value of the fuel cell stack does not consider the different actual needs of driving power due to changes in vehicle load, such as no load, full load, and overload, resulting in wasted energy conversion efficiency and excessive energy consumption; when recovering energy on a downhill slope, the fuel cell system and the energy recovery system lack dynamic coordination. The fuel cell stack is still outputting a large amount of power, which conflicts with the power recovery, resulting in some of the recharged energy being wasted and the energy recovery efficiency being low. After the SOC drops to a certain threshold, a "one-size-fits-all" power protection mode is adopted, lacking an active real-time compensation and adjustment strategy. Summary of the Invention

[0004] This invention aims to at least solve the aforementioned technical problems existing in the prior art. To this end, the first aspect of this invention proposes a self-learning-based energy management method, the method comprising:

[0005] Obtain the vehicle's main drive power, and perform self-learning based on the historical value and the current value of the main drive power to obtain the self-learned value of the main drive power;

[0006] The self-learning value of the auxiliary drive power is obtained based on the reference power of each auxiliary drive device of the vehicle and the corresponding operating condition correction coefficient; wherein, the operating condition correction coefficient is dynamically adjusted according to the load rate and temperature of the auxiliary drive device.

[0007] The power self-learning value of the gas-electric system is generated based on the self-learning value of the main drive power and the self-learning value of the auxiliary drive power.

[0008] Obtain the battery SOC, perform power compensation on the power self-learning value of the gas-electric system based on the battery SOC, and generate the power limit value of the gas-electric system.

[0009] The output power of the gas-fired power system is limited by the power limit value of the gas-fired power system.

[0010] Optionally, it also includes:

[0011] If the vehicle slope is detected to meet a preset first condition and the battery SOC is greater than or equal to a preset first SOC threshold, the energy recovery power and the battery allowable recharge power are determined; wherein, the first condition is that the vehicle slope is less than or equal to a preset first slope threshold and the duration exceeds a first time threshold; the first slope threshold is a negative number.

[0012] The power limit value of the gas-electric system is determined based on the energy recovery power and the battery's allowable recharge power.

[0013] The output power of the gas-fired power system is limited based on the power limit value of the gas-fired power system.

[0014] Optionally, determining the power limit value of the gas-electric system based on the energy recovery power and the battery's allowable recharge power includes:

[0015] Obtain the difference between the battery's allowable recharge power and the stack's minimum efficient operating point;

[0016] If the energy recovery power is less than or equal to the difference, then the power limit value of the gas-electric system is determined based on the battery's allowable recharge power and the energy recovery power.

[0017] If the energy recovery power is greater than the battery's allowable recharge power, then the minimum high-efficiency operating point of the fuel cell stack will be used as the power limit value of the gas-fired power system.

[0018] Optionally, the minimum efficient operating point of the fuel cell stack is the sum of the idle power and the redundant power.

[0019] Optionally, the step of obtaining the battery SOC and performing power compensation on the power self-learning value of the gas turbine system based on the battery SOC to generate a power limit value for the gas turbine system includes:

[0020] The battery SOC is divided into multiple SOC intervals, and a corresponding compensation coefficient is set for each SOC interval; the compensation coefficient is a positive number.

[0021] The first SOC range is determined based on the current battery SOC, and the power limit value of the gas-electric system is obtained based on the compensation coefficient corresponding to the first SOC range and the power self-learning value of the gas-electric system; the first SOC range is the SOC range corresponding to the current battery SOC.

[0022] Optionally, setting a corresponding compensation coefficient for each SOC interval includes:

[0023] The larger the value of the SOC interval, the smaller the compensation coefficient corresponding to the SOC interval.

[0024] A second aspect of the present invention provides an electronic device comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, the at least one instruction or at least one program being loaded and executed by the processor to implement the self-learning-based energy management method as proposed in the first aspect.

[0025] A third aspect of the present invention provides a computer-readable storage medium storing at least one instruction or at least one program, wherein the at least one instruction or at least one program is loaded and executed by a processor to implement the self-learning-based energy management method as proposed in the first aspect.

[0026] The beneficial effects of a self-learning-based energy management method, device, and medium are as follows: This invention updates the power limit value of the gas-fired power system in real time through a self-learning algorithm, balancing power demand and energy consumption, and realizing refined energy management; it introduces a short-term optimization mechanism based on real-time slope to actively coordinate the power of the gas-fired power system and the recovered power, solving the power conflict problem in the energy recovery process; and it constructs a dynamic power compensation closed loop based on real-time SOC feedback, realizing the leap from "on / off" control to "linear smooth" intelligent adjustment in energy management. Attached Figure Description

[0027] Figure 1 A flowchart of a self-learning-based energy management method provided in an embodiment of the present invention. Detailed Implementation

[0028] 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.

[0029] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of this disclosure, unless otherwise stated, "a plurality of" means two or more. Furthermore, the use of "based on" or "according to" implies openness and inclusiveness, because processes, steps, calculations, or other actions "based on" or "according to" one or more of the stated conditions or values ​​may in practice be based on additional conditions or beyond the stated values.

[0030] This invention provides a self-learning-based energy management method, such as... Figure 1 As shown, the method may include the following steps:

[0031] Step 101: Obtain the vehicle's main drive power. Based on the historical value and current value of the main drive power, perform self-learning to obtain the self-learned value of the main drive power.

[0032] This invention constructs a three-level linked energy management architecture consisting of a "reference power learning layer, a SOC-based power compensation layer, and a road condition-based dynamic optimization layer." This architecture organically integrates long-term learning, SOC-based real-time compensation, and road condition-based dynamic optimization, resulting in a self-learning, multi-level linked intelligent energy management closed-loop control.

[0033] To implement this solution, the first step is to obtain the self-learning power value of the fuel cell electric system. This requires identifying the vehicle's load state through the onboard system and calculating the self-learning power values ​​of the main drive and auxiliary drive within a single charge / discharge cycle. Based on these two values, and combined with a power margin parameter, these three values ​​are summed to obtain and store the vehicle's fuel cell electric system power self-learning value. This value then constrains the operating range of the fuel cell electric system's power, ensuring it always operates within its high-efficiency range. The power margin parameter is a preset calibration constant used to provide safety redundancy. Specifically, self-learning is performed based on the historical and current values ​​of the main drive power. Combining historical and current weights, a current self-learned value of the main drive power is obtained. This learned value is stored as the historical value for the next learning iteration. In the next learning iteration, the same method is used, based on the next learning iteration's historical value, current value, historical weights, and current weights, to obtain the next self-learned value of the main drive power. This process is repeated until the learning is complete, yielding the final self-learned value of the main drive power. It can be seen that the historical value of the main drive power is also the average value of the historical self-learned main drive power, which is stored in the vehicle controller. It should also be noted that when calculating the main drive power, the value at that moment is only averaged when the drive power is positive and the vehicle speed is greater than or equal to 0 km / h. The historical and current weights are automatically adjusted based on the vehicle's effective operating time within a single charge / discharge cycle. The sum of the historical and current weights is 1. The longer the single charge / discharge cycle, the smaller the historical weight and the larger the current weight.

[0034] Step 102: Obtain the self-learning value of the auxiliary drive power based on the reference power of each auxiliary drive device of the vehicle and the corresponding operating condition correction coefficient; wherein, the operating condition correction coefficient is dynamically adjusted according to the load rate and temperature of the auxiliary drive device.

[0035] The power self-learning value of each auxiliary drive device is obtained by multiplying its base power by its corresponding operating condition correction factor. The power self-learning values ​​of all auxiliary drive devices are then summed to obtain the vehicle's auxiliary drive power self-learning value. The auxiliary drive devices on the vehicle include the air compressor, air conditioning compressor, power steering pump, and other auxiliary drive equipment. The operating condition correction factor is dynamically adjusted based on the temperature and load rate of the auxiliary drive devices. For example, when the air conditioning temperature is greater than or equal to 35°C, the operating condition correction factor of the air conditioning compressor is adjusted to 1.2.

[0036] Step 103: Generate the power self-learning value of the gas-electric system based on the main drive power self-learning value and the auxiliary drive power self-learning value.

[0037] Specifically, the self-learning values ​​of the main drive power, auxiliary drive power, and power margin parameters are added together to obtain the power self-learning value of the gas turbine electric system. The power margin parameter is a constant used to provide safety redundancy.

[0038] Step 104: Obtain the battery SOC, perform power compensation on the power self-learning value of the gas-electric system based on the battery SOC, and generate the power limit value of the gas-electric system.

[0039] In one possible implementation, the step of obtaining the battery SOC and performing power compensation on the power self-learning value of the gas-electric system based on the battery SOC to generate a power limit value for the gas-electric system includes:

[0040] The battery SOC is divided into multiple SOC intervals, and a corresponding compensation coefficient is set for each SOC interval; the compensation coefficient is a positive number.

[0041] The first SOC range is determined based on the current battery SOC, and the power limit value of the gas-electric system is obtained based on the compensation coefficient corresponding to the first SOC range and the power self-learning value of the gas-electric system; the first SOC range is the SOC range corresponding to the current battery SOC.

[0042] In one possible implementation, setting a corresponding compensation coefficient for each SOC interval includes:

[0043] The larger the value of the SOC interval, the smaller the compensation coefficient corresponding to the SOC interval.

[0044] After obtaining the self-learning value of the fuel cell power system, the self-learning value is compensated according to the range of the power battery's State of Charge (SOC) to maintain the power battery's SOC within the range where both charging and discharging energy are optimal. Specifically, when the power battery SOC is low, the fuel cell power output is appropriately increased to charge the battery; when the power battery SOC is high, the power is appropriately reduced to prioritize the use of battery energy. For example, the power battery SOC can be divided into four ranges, numbered from smallest to largest as range 1, range 2, range 3, and range 4. As shown in Table 1, if the SOC is in range 1, the compensation coefficient is a1; if the SOC is in range 2, the compensation coefficient is a2; if the SOC is in range 3, the compensation coefficient is a3; and if the SOC is in range 4, the compensation coefficient is a4. The determined compensation coefficient is multiplied by the fuel cell power self-learning value P to obtain the fuel cell power limit value. Among them, a1 > a2 > a3 > a4, and a1, a2, a3, and a4 are all numbers greater than 0 and less than 1.

[0045] Table 1

[0046] Power Battery SOC Interval 1 Interval 2 Interval 3 Interval 4 Power limit values ​​for gas-fired power systems P*a1 P*a2 P*a3 P*a4

[0047] Step 105: Limit the output power of the gas-electric system using the power limit value of the gas-electric system.

[0048] In one possible implementation, it also includes:

[0049] If the vehicle slope is detected to meet a preset first condition and the battery SOC is greater than or equal to a preset first SOC threshold, the energy recovery power and the battery allowable recharge power are determined; wherein, the first condition is that the vehicle slope is less than or equal to a preset first slope threshold and the duration exceeds a first time threshold; the first slope threshold is a negative number.

[0050] The power limit value of the gas-electric system is determined based on the energy recovery power and the battery's allowable recharge power.

[0051] The output power of the gas-fired power system is limited based on the power limit value of the gas-fired power system.

[0052] In one possible implementation, determining the power limit value of the gas-electric system based on the energy recovery power and the battery's permissible recharge power includes:

[0053] Obtain the difference between the battery's allowable recharge power and the stack's minimum efficient operating point;

[0054] If the energy recovery power is less than or equal to the difference, then the power limit value of the gas-electric system is determined based on the battery's allowable recharge power and the energy recovery power.

[0055] If the energy recovery power is greater than the battery's allowable recharge power, then the minimum high-efficiency operating point of the fuel cell stack will be used as the power limit value of the gas-fired power system.

[0056] In one possible implementation, the minimum efficient operating point of the fuel cell stack is the sum of the idle power and the redundant power.

[0057] Specifically, when the vehicle is in energy recovery mode, the system actively identifies the road gradient and ensures that the battery's State of Charge (SOC) meets preset conditions (i.e., SOC is greater than or equal to a preset first SOC threshold). It then calculates and compares the energy recovery power with the battery's allowable recharge power, dynamically optimizing the power value of the fuel cell system based on the comparison results to ensure maximum energy recovery utilization. For example, the first gradient threshold can be set to -2, the first time threshold to 15 seconds, and the first SOC threshold to 30%.

[0058] Calculate the difference between the battery's allowable recharge power and the stack's minimum efficient operating point. If the energy recovery power is less than or equal to this difference, then the difference between the battery's allowable recharge power and the energy recovery power is used as the power limit value for the gas-fired power system.

[0059] Wherein, energy recovery power = recovery torque * motor speed / 9550, battery allowable recharge power = Min[P1, Psoc] * η, P1 is the battery thermal safety power, which is negatively correlated with temperature; Psoc is the upper limit power of SOC recharge, which is negatively correlated with SOC; η is the battery charge and discharge efficiency, which is taken as 0.92–0.95 in this embodiment of the invention.

[0060] In summary, in this embodiment of the invention, the power limit value of the gas-fired power system is updated in real time through a self-learning algorithm, balancing power demand and energy consumption, and achieving refined energy management; a short-term optimization mechanism based on real-time slope is introduced to actively coordinate the power of the fuel cell stack and the power recovery, solving the power conflict problem in the energy recovery process; and a dynamic power compensation closed loop based on real-time SOC feedback is constructed, realizing the leap from "on / off" control to "linear smooth" intelligent adjustment in energy management.

[0061] In another embodiment of the present invention, an electronic device is also provided, the electronic device including a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or at least one program being loaded and executed by the processor to implement the self-learning-based energy management method proposed in the embodiments of the present invention.

[0062] In another embodiment of the present invention, a computer-readable storage medium is also provided, wherein at least one instruction or at least one program is stored in the storage medium, and the at least one instruction or at least one program is loaded and executed by a processor to implement the self-learning-based energy management method proposed in the embodiments of the present invention.

[0063] The foregoing primarily describes the solutions provided by the embodiments of the present invention from the perspective of the device. It is understood that, in order to achieve the above functions, the device includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the algorithmic steps of the various examples described in the embodiments disclosed herein, the present invention can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0064] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions within the technical scope disclosed in the present invention should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A self-learning-based energy management method, characterized in that, include: The vehicle's main drive power is obtained, and self-learning is performed based on the historical value and the current value of the main drive power. The self-learned value of the main drive power is obtained by combining the historical weight and the current weight. The historical value of the main drive power is the self-learned value of the main drive power obtained from the previous self-learning; The self-learning value of the auxiliary drive power is obtained based on the base power of each auxiliary drive device in the vehicle and the corresponding operating condition correction coefficient; wherein, the auxiliary drive devices include air compressor, air conditioning compressor, and power steering pump; the operating condition correction coefficient is dynamically adjusted according to the load rate and temperature of the auxiliary drive devices; The power self-learning value of the gas-electric system is generated based on the self-learning value of the main drive power and the self-learning value of the auxiliary drive power. Obtain the battery SOC, perform power compensation on the power self-learning value of the gas-electric system based on the battery SOC, and generate the power limit value of the gas-electric system. Also includes: If the vehicle slope is detected to meet a preset first condition and the battery SOC is greater than or equal to a preset first SOC threshold, the energy recovery power and the battery allowable recharge power are determined; wherein, the first condition is that the vehicle slope is less than or equal to a preset first slope threshold and the duration exceeds a first time threshold; the first slope threshold is a negative number. The power limit value of the gas-electric system is determined based on the energy recovery power and the battery's allowable recharge power. The step of determining the power limit value of the gas-electric system based on the energy recovery power and the battery's allowable recharge power includes: Obtain the difference between the battery's allowable recharge power and the stack's minimum efficient operating point; the stack's minimum efficient operating point is the sum of idle power and redundant power; If the energy recovery power is less than or equal to the difference, then the power limit value of the gas-electric system is determined based on the battery's allowable recharge power and the energy recovery power. If the energy recovery power is greater than the battery's allowable recharge power, then the minimum high-efficiency operating point of the fuel cell stack will be used as the power limit value of the gas-fired power system. The output power of the gas-fired power system is limited by the power limit value of the gas-fired power system.

2. The energy management method based on self-learning according to claim 1, characterized in that, The process of obtaining the battery SOC and performing power compensation on the power self-learning value of the gas turbine system based on the battery SOC to generate a power limit value for the gas turbine system includes: The battery SOC is divided into multiple SOC intervals, and a corresponding compensation coefficient is set for each SOC interval; the compensation coefficient is a positive number. The first SOC range is determined based on the current battery SOC, and the power limit value of the gas-electric system is obtained based on the compensation coefficient corresponding to the first SOC range and the power self-learning value of the gas-electric system; the first SOC range is the SOC range corresponding to the current battery SOC.

3. The energy management method based on self-learning according to claim 2, characterized in that, The step of setting a corresponding compensation coefficient for each SOC interval includes: The larger the value of the SOC interval, the smaller the compensation coefficient corresponding to the SOC interval.

4. An electronic device, characterized in that, The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the self-learning-based energy management method as described in any one of claims 1-3.

5. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction or at least one program, which is loaded and executed by a processor to implement the self-learning-based energy management method as described in any one of claims 1-3.

Images