Energy management method and device, vehicle and storage medium

Abstract

The application relates to an energy management method and device, a vehicle and a storage medium, comprising: acquiring a current vehicle speed, a current state of charge of a lithium battery, a variable load frequency of a fuel cell in a current period and a current road condition; determining a prediction probability of the fuel cell reaching a preset variable load frequency in a preset time length according to the current vehicle speed, the current state of charge, the variable load frequency and the current road condition; and performing energy management on the fuel cell based on a preset two-stage power maintenance strategy in the case that at least one of the variable load frequency, the prediction probability and the current state of charge meets a preset energy management condition. Thus, the problem that, in the related art, the flexibility of energy management on a fuel cell vehicle is insufficient, the variable load frequency and amplitude are not fully considered, the fuel cell needs to frequently adjust the output power under large power fluctuation, the dynamic loss is aggravated, the service life is shortened, the adaptability of the fuel cell to load change and the response timeliness are improved, the dynamic loss is reduced, and the service life is prolonged.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to an energy management method, device, vehicle, and storage medium. Background Technology

[0002] Currently, users have increasingly higher requirements for the durability and energy efficiency of fuel cell vehicles. Under variable load conditions such as frequent acceleration and deceleration and sudden load changes, power control of the power system has become a key technical challenge. Frequent load changes in fuel cells can exacerbate membrane electrode degradation, while excessive participation of lithium batteries in dynamic recharging can shorten their cycle life.

[0003] In related technologies, the energy management of fuel cell vehicles is generally controlled by looking up a table based on a preset range of remaining battery charge (SOC) and a power threshold, such as directly matching the fuel cell output power according to the SOC.

[0004] However, the relevant technologies lack flexibility in energy management for fuel cell vehicles and do not fully consider the impact of load variation frequency and amplitude. Under high power fluctuations, fuel cells need to frequently adjust their output power, which exacerbates dynamic losses and shortens lifespan, and urgently needs to be addressed. Summary of the Invention

[0005] This application provides an energy management method, device, vehicle, and storage medium to address the problems of insufficient flexibility in energy management of fuel cell vehicles, inadequate consideration of the impact of load variation frequency and amplitude, frequent adjustment of fuel cell output power under high power fluctuations, exacerbation of dynamic losses, and shortened lifespan. The application aims to improve the adaptability and responsiveness of fuel cells to load changes, reduce dynamic losses, and extend service life.

[0006] The first aspect of this application provides an energy management method, including the following steps: Obtain the vehicle's current speed, the lithium battery's current state of charge, the number of load changes in the fuel cell during the current cycle, and the current road conditions; The predicted probability of the fuel cell reaching the preset number of load changes within a preset time is determined based on the current vehicle speed, current state of charge, number of load changes, and current road conditions. Under the condition that at least one of the load variation number, prediction probability and current state of charge meets the preset energy management conditions, the fuel cell is energy managed based on a preset two-stage power maintenance strategy.

[0007] Optionally, in some embodiments, energy management of the fuel cell is performed based on a preset two-stage power maintenance strategy, including: Obtain the vehicle's required power before and after load change; During the first energy management period, the fuel cell is controlled to maintain the power output required by the vehicle before the load change, and the first difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the first difference. During the second energy management period, the fuel cell is controlled to maintain the power output required by the vehicle after the load change, and a second difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the second difference. Obtain the post-management state of charge of the lithium battery, determine the energy management exit mechanism based on the post-management state of charge, and exit energy management according to the energy management exit mechanism.

[0008] Optionally, in some embodiments, an energy management exit mechanism is determined based on the managed state of charge, and energy management is exited according to the energy management exit mechanism, including: Determine whether the post-management state of charge is greater than or equal to the first state of charge, and whether the post-management state of charge is less than or equal to the second state of charge; If the post-management state of charge is greater than or equal to the first state of charge, and the post-management state of charge is less than or equal to the second state of charge, then the energy management exit mechanism is determined to be an active exit mechanism, and the steps of obtaining the vehicle's current speed, the lithium battery's current state of charge, the number of load changes of the fuel cell in the current cycle, and the current road conditions are re-executed. Otherwise, the energy management exit mechanism is determined to be a forced exit mechanism, and the basic output power of the fuel cell is determined based on the post-management state of charge, and the fuel cell is controlled based on the basic output power.

[0009] Optionally, in some embodiments, determining the base output power of the fuel cell based on the managed post-charge state includes: Obtain the state of charge-fuel cell power conversion table; Based on the state-of-charge-fuel cell power comparison table, the base output power is determined according to the post-management state of charge.

[0010] Optionally, in some embodiments, obtaining the number of load variations of the fuel cell in the current cycle includes: The number of times that the net power of the fuel cell is greater than the preset power and the difference between adjacent net power values ​​is greater than the preset power difference is obtained within the current period; The number of occurrences is taken as the number of times the fuel cell is subjected to load changes in the current cycle.

[0011] Optionally, in some embodiments, after determining the predicted probability that the fuel cell will reach a preset number of load changes within a preset time period based on the current vehicle speed, current state of charge, number of load changes, and current road conditions, the method further includes: Determine whether the number of load changes is greater than or equal to the preset number, or determine whether the predicted probability is greater than or equal to the preset probability, and whether the current state of charge is greater than or equal to the first state of charge, and whether the current state of charge is less than or equal to the second state of charge. If the number of load changes is greater than or equal to the preset number, then the preset energy management conditions are met. Alternatively, if the predicted probability is greater than or equal to the preset probability, and the current state of charge is greater than or equal to the first state of charge, and the current state of charge is less than or equal to the second state of charge, then the preset energy management conditions are met.

[0012] A second aspect of this application provides an energy management device, comprising: The acquisition module is used to acquire the vehicle's current speed, the current state of charge of the lithium battery, the number of load changes of the fuel cell in the current cycle, and the current road conditions. The prediction module is used to determine the predicted probability that the fuel cell will reach a preset number of load changes within a preset time based on the current vehicle speed, current state of charge, number of load changes, and current road conditions. The control module is used to manage the energy of the fuel cell based on a preset two-stage power maintenance strategy when at least one of the load change frequency, prediction probability and current state of charge meets a preset energy management condition.

[0013] Optionally, in some embodiments, the control module is specifically used for: Obtain the vehicle's required power before and after load change; During the first energy management period, the fuel cell is controlled to maintain the power output required by the vehicle before the load change, and the first difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the first difference. During the second energy management period, the fuel cell is controlled to maintain the power output required by the vehicle after the load change, and a second difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the second difference. Obtain the post-management state of charge of the lithium battery, determine the energy management exit mechanism based on the post-management state of charge, and exit energy management according to the energy management exit mechanism.

[0014] Optionally, in some embodiments, the control module is specifically used for: Determine whether the post-management state of charge is greater than or equal to the first state of charge, and whether the post-management state of charge is less than or equal to the second state of charge; If the post-management state of charge is greater than or equal to the first state of charge, and the post-management state of charge is less than or equal to the second state of charge, then the energy management exit mechanism is determined to be an active exit mechanism, and the steps of obtaining the vehicle's current speed, the lithium battery's current state of charge, the number of load changes of the fuel cell in the current cycle, and the current road conditions are re-executed. Otherwise, the energy management exit mechanism is determined to be a forced exit mechanism, and the basic output power of the fuel cell is determined based on the post-management state of charge, and the fuel cell is controlled based on the basic output power.

[0015] Optionally, in some embodiments, the control module is specifically used for: Obtain the state of charge-fuel cell power conversion table; Based on the state-of-charge-fuel cell power comparison table, the base output power is determined according to the post-management state of charge.

[0016] Optionally, in some embodiments, the acquisition module is specifically used for: The number of times that the net power of the fuel cell is greater than the preset power and the difference between adjacent net power values ​​is greater than the preset power difference is obtained within the current period; The number of occurrences is taken as the number of times the fuel cell is subjected to load changes in the current cycle.

[0017] Optionally, in some embodiments, the prediction module is further configured to: Determine whether the number of load changes is greater than or equal to the preset number, or determine whether the predicted probability is greater than or equal to the preset probability, and whether the current state of charge is greater than or equal to the first state of charge, and whether the current state of charge is less than or equal to the second state of charge. If the number of load changes is greater than or equal to the preset number, then the preset energy management conditions are met. Alternatively, if the predicted probability is greater than or equal to the preset probability, and the current state of charge is greater than or equal to the first state of charge, and the current state of charge is less than or equal to the second state of charge, then the preset energy management conditions are met.

[0018] A third aspect of this application provides a vehicle, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the energy management method described in the first aspect embodiment.

[0019] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the energy management method described in the first aspect embodiment.

[0020] Therefore, the embodiments of this application have at least the following beneficial effects: (1) Extending fuel cell lifespan. Frequent and large-scale load changes in fuel cells can easily lead to frequent wet-dry cycles of the membrane electrode assembly (MEA), resulting in reduced proton conduction efficiency and accelerated catalyst degradation. Related technologies do not perform energy management during the high-frequency load change phase in the first 4 minutes of the current cycle, requiring frequent adjustments to the fuel cell, further exacerbating MEA wear. This application adopts a two-stage power maintenance strategy, changing real-time demand following to fixed-window maintenance, reducing the frequency of fuel cell adjustments under load changes; at the same time, by predictively activating the stabilization module in advance, it covers the high-frequency load change scenario in the early stage of the cycle, avoiding additional losses in this stage, thereby slowing down the MEA degradation rate and extending the fuel cell lifespan.

[0021] (2) Optimize lithium battery operating conditions. Over-discharging of lithium batteries can easily lead to the precipitation of lithium dendrites on the negative electrode, while overcharging can cause the collapse of the positive electrode structure, both of which will shorten the cycle life. In this application, [30%, 60%] is set as the activation range of the stable module to avoid the lithium battery from participating in high-power energy replenishment or energy storage under extreme charge states; at the same time, by predicting and intervening in advance, without changing the original SOC constraint logic, only the activation timing is optimized. Since the polarization voltage is the lowest in the middle SOC range, the charging efficiency can be improved, so that the lithium battery can charge and discharge more in the optimal range, slow down battery degradation, and extend cycle life.

[0022] (3) Improve the energy efficiency of the whole vehicle. In related technologies, frequent fluctuations in the power of fuel cells can easily cause their operating point to deviate from the optimal efficiency range, increasing energy consumption. The additional power adjustment caused by high-frequency load changes in the early stage of the cycle will further aggravate the deviation. This application adopts a two-stage maintenance mechanism to make the fuel cell work stably at a fixed power point within a 2-minute window, ensuring that the operating point is in the optimal efficiency range; early activation can avoid additional power adjustments in the early stage of load changes, reduce the deviation of the operating point, improve the efficiency of the fuel cell system, and reduce the hydrogen consumption per 100 kilometers of the whole vehicle.

[0023] (4) Optimize the timeliness of stabilization module triggering. Related technologies adopt a passive statistical mode, which requires the stabilization module to be activated only after the cycle ends. During this period, the fuel cell needs to adjust its power, which aggravates component wear. This application uses a predictive model to predict the subsequent load change trend at the beginning of the load change. For example, if there are 6 load changes in the first minute, it can predict that the load change will reach the standard in the next 3 minutes. Energy management of the fuel cell can be performed 4 minutes in advance, shortening the trigger lag time from a maximum of 5 minutes to ≤1 minute, reducing component wear in the early stage of the load change, and further extending the life of the fuel cell and lithium battery.

[0024] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0025] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of an energy management method provided according to an embodiment of this application; Figure 2 This is a block diagram of an energy management device according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a vehicle according to an embodiment of this application. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The energy management method, apparatus, vehicle, and storage medium of embodiments of this application are described below with reference to the accompanying drawings. Addressing the issues mentioned in the background art regarding insufficient flexibility in energy management of fuel cell vehicles, inadequate consideration of the impact of load variation frequency and amplitude, and the need for frequent power adjustments by the fuel cell under high power fluctuations, which exacerbates dynamic losses and shortens lifespan, this application provides an energy management method. This method determines the predicted probability of the fuel cell reaching a preset number of load variations within a preset time period based on the current vehicle speed, current state of charge, number of load variations, and current road conditions. When at least one of the number of load variations, predicted probability, and current state of charge meets preset energy management conditions, energy management of the fuel cell is performed based on a preset two-stage power maintenance strategy. This solves the problems of insufficient flexibility in energy management of fuel cell vehicles, inadequate consideration of the impact of load variation frequency and amplitude, and the need for frequent power adjustments by the fuel cell under high power fluctuations, which exacerbates dynamic losses and shortens lifespan. It improves the adaptability and responsiveness of the fuel cell to load changes, reduces dynamic losses, and extends service life.

[0028] Specifically, Figure 1 A flowchart illustrating the energy management method provided in this application embodiment.

[0029] like Figure 1 As shown, this energy management method includes the following steps: In step S101, the vehicle's current speed, the lithium battery's current state of charge, the number of load changes of the fuel cell in the current cycle, and the current road conditions are obtained.

[0030] In some embodiments, obtaining the number of load changes of the fuel cell in the current cycle includes: obtaining the number of times that the net power of the fuel cell is greater than a preset power and the difference between adjacent net power values ​​is greater than a preset power difference; and using the number of occurrences as the number of load changes of the fuel cell in the current cycle.

[0031] The net power of the fuel cell is the effective electrical power actually output by the fuel cell after deducting the consumption of its own auxiliary system. The preset power and the preset power difference can be the power and power difference preset by the user. They can be the power and power difference obtained through a limited number of experiments or the power and power difference obtained through a limited number of computer simulations. No specific limitation is made here.

[0032] Specifically, in this application embodiment, the vehicle's current speed and the current state of charge of the lithium battery can be collected in real time through the Controller Area Network (CAN) bus, the net power of the fuel cell can be collected in real time through the fuel cell controller, and the current road conditions (such as congested urban roads and smooth highways) can be obtained through the vehicle camera, vehicle navigation module or road condition sensor, thereby providing real-time and accurate parameters for energy management.

[0033] Furthermore, in this embodiment of the application, if the net power of the fuel cell is greater than a preset power and the difference between adjacent net power values ​​is greater than a preset power difference, the current operating condition can be determined as a load change condition, and the number of load changes of the fuel cell in the current cycle can be obtained accordingly. The preset power can be 31kW to exclude small fluctuations in low power; the preset power difference can be 25kW to define large load changes.

[0034] In step S102, the predicted probability of the fuel cell reaching the preset number of load changes within a preset time is determined based on the current vehicle speed, current state of charge, number of load changes, and current road conditions.

[0035] The preset number of load variations can be a number of load variations set by the user, a number of load variations obtained through a limited number of experiments, or a number of load variations obtained through a limited number of computer simulations; no specific limitation is made here.

[0036] Specifically, this application embodiment can train a Long Short-Term Memory (LSTM) prediction model based on real-vehicle historical data (covering the frequency, amplitude, duration, and SOC change trends of load variations under different road conditions such as urban areas, highways, and mountainous areas). Input features include: the net power fluctuation sequence over the past 1 minute, current vehicle speed, current state of charge, number of load variations, and current road conditions. The model then outputs in real-time the predicted probability that the fuel cell will reach a preset number of load variations within a preset time period. The preset time period can be 3 minutes; the preset number of load variations can be 8.

[0037] It should be understood that the overall implementation logic of the Long Short-Term Memory Network prediction model mainly consists of four parts: training data preparation, model initialization, iterative model training, and model testing and parameter optimization.

[0038] Furthermore, in the training data preparation stage, this embodiment of the application can first perform data collection and screening. Specifically, this embodiment of the application can collect historical operating data of real vehicles, covering different road conditions such as urban areas, highways, and mountainous areas. The collected data dimensions include: net power fluctuation data (time step 1s), vehicle speed data, SOC data, load variation data (load variation frequency, amplitude, and duration), and road condition data, ensuring the comprehensiveness and representativeness of the data and avoiding insufficient generalization ability of the model due to single road condition data.

[0039] Furthermore, after completing data collection and filtering, this embodiment of the application can perform data preprocessing on the collected raw data. Specifically, firstly, the collected raw data is cleaned to remove outliers (such as sudden changes in net power or data where SOC exceeds a reasonable range (0-100%)), and missing values ​​are supplemented using linear interpolation. Secondly, the road condition data is encoded (city = 1, highway = 2, mountain = 3). Next, the net power fluctuation data is divided into 1-minute units to form a net power fluctuation sequence of nearly 1 minute (60 data points). Finally, the number of load changes corresponding to each time node is counted (the number of load changes within 1 minute before the current time node is counted), and label data corresponding to the input features is constructed.

[0040] Furthermore, after completing the data preprocessing, this embodiment can define labels and divide data according to a preset duration and a preset number of load changes. Specifically, this embodiment can set the preset duration to 3 minutes and the preset number of load changes to 8 times. That is, for each time node corresponding to an input feature, it is determined whether the number of load changes of the fuel cell reaches 8 times within 3 minutes after that time node: if it does, the label is set to 1; if it does not, the label is set to 0. The processed input feature data and label data are divided into a training set (for model training), a validation set (for model parameter adjustment), and a test set (for model performance verification) in a 7:2:1 ratio.

[0041] Furthermore, in the model initialization stage, this embodiment can first initialize the model layer parameters. Specifically, this embodiment can initialize the relevant parameters of the input layer, feature processing layer, LSTM core layer, and output layer according to the above-described model core architecture: the moving average window of the feature processing layer is set to 3, and the min-max normalization interval is set to [0,1]; the number of units in the three layers of the LSTM core layer is set to 64, 32, and 16 respectively, the dropout probability is set to 0.2, and the activation functions of each layer are set according to the above architecture; the number of fully connected neurons in the output layer is set to 1, and the activation function is sigmoid.

[0042] Furthermore, after initializing the model hierarchical parameters, this embodiment can initialize the optimizer and loss function. Specifically, this embodiment can select the Adam optimizer as the optimizer for model training, with an initial learning rate of 0.001, a decay coefficient (beta1) of 0.9, a decay coefficient (beta2) of 0.999, and epsilon of 1e-8, to optimize model parameters and accelerate training convergence. The cross-entropy loss function is selected as the model's loss function to measure the error between the model's predicted values ​​and the true labels; a smaller loss function value indicates higher model prediction accuracy.

[0043] Furthermore, during the model iterative training phase, this embodiment can first perform forward propagation computation. Specifically, this embodiment can input the input feature data of the training set into the model, which then sequentially undergoes input layer standardization, feature processing layer preprocessing and fusion, LSTM core layer temporal feature extraction, and output layer fully connected computation to obtain the model's predicted probability output value.

[0044] Furthermore, after completing the forward propagation calculation, this embodiment of the application can perform loss calculation and backpropagation. Specifically, this embodiment of the application can use the cross-entropy loss function to calculate the loss value between the model's predicted output value and the true label of the training set; based on the Adam optimizer, the loss value is backpropagated to each layer of the model, and the relevant parameters of the input layer, feature processing layer, LSTM core layer, and output layer are updated sequentially to minimize the loss value.

[0045] Furthermore, after completing loss calculation and backpropagation, this embodiment of the application can perform validation set verification and parameter adjustment. Specifically, after each training epoch, this embodiment of the application can input validation set data into the model, calculate the validation set loss value and prediction accuracy; if the validation set loss value does not decrease for three consecutive epochs (i.e., an overfitting trend appears), then the model parameters are adjusted: reduce the learning rate (decay by 50% each time), adjust the dropout probability (fine-tuning between 0.1 and 0.3), or reduce the number of units in the LSTM core layer, until the validation set loss value tends to stabilize.

[0046] Furthermore, after completing validation set verification and parameter adjustment, this embodiment of the application can determine the termination of the iteration. Specifically, this embodiment of the application can set the maximum number of training rounds to 100 rounds. If, during the training process, the validation set loss value stabilizes for 5 consecutive rounds (fluctuation range less than 1e-4), and the validation set prediction accuracy reaches 90% or higher, then the iterative training is terminated. If the above conditions are not met even after reaching the maximum number of training rounds, then the data preprocessing process or the initial setting of the model parameters needs to be re-examined, adjusted, and iterative training is performed again.

[0047] Furthermore, in the model testing and parameter optimization stage, this embodiment of the application can first perform model testing. Specifically, this embodiment of the application can input test set data into the trained model, calculate the prediction accuracy, precision, recall, and F1 score of the test set, and verify the generalization ability and prediction accuracy of the model; if the prediction accuracy of the test set is lower than 85%, the process returns to the model iterative training step, adjusts the parameters, and retrains.

[0048] Furthermore, after completing the model test, the embodiments of this application can perform parameter optimization. Specifically, in response to problems such as overfitting, slow convergence speed, and insufficient prediction accuracy that occur during model training, the embodiments of this application can adopt the following optimization methods: (1) Regularization optimization: Add L2 regularization terms to the LSTM core layer and output layer, with the regularization coefficient set to 0.001, to limit the absolute value of model parameters and reduce overfitting; (2) Learning rate optimization: Adopt a dynamic learning rate strategy, using an initial learning rate of 0.001 for the first 30 rounds of training, adjusting the learning rate to 0.0005 for rounds 30-60, and adjusting it to 0.0001 after round 60, to avoid training oscillations caused by excessively high learning rates and slow convergence caused by excessively low learning rates; (3) Feature optimization: Perform correlation analysis on the input features, remove features with a correlation of less than 0.3 with the label, and if there are redundant features, use principal component analysis (PCA). (3) Dimensionality reduction using the PCA method to improve training efficiency; (4) LSTM structure optimization: Through comparative experiments, adjust the number of layers and units in the core layer of LSTM. For example, LSTM can be set to 3 layers (64, 32, 16 units) to ensure feature extraction capability and avoid overfitting due to excessive model complexity; (4) Model saving: Save the optimized model parameters (weights, biases, etc.) to form the final fuel cell load prediction model for subsequent real-time prediction.

[0049] In step S103, if at least one of the load change number, prediction probability and current state of charge satisfies a preset energy management condition, the fuel cell is energy managed based on a preset two-stage power maintenance strategy.

[0050] In some embodiments, energy management of the fuel cell is based on a preset two-stage power maintenance strategy, including: acquiring the vehicle's required power before and after the load change; controlling the fuel cell to maintain the vehicle's required power output before the load change during a first energy management period, calculating a first difference between the vehicle's required power output before and after the load change, and controlling the vehicle's lithium battery based on the first difference; controlling the fuel cell to maintain the vehicle's required power output after the load change during a second energy management period, calculating a second difference between the vehicle's required power output before and after the load change, and controlling the vehicle's lithium battery based on the second difference; acquiring the post-management state of charge of the lithium battery, determining an energy management exit mechanism based on the post-management state of charge, and exiting energy management based on the energy management exit mechanism.

[0051] In some embodiments, the energy management exit mechanism is determined based on the post-management state of charge, and energy management is exited according to the energy management exit mechanism. This includes: determining whether the post-management state of charge is greater than or equal to a first state of charge and whether the post-management state of charge is less than or equal to a second state of charge; if the post-management state of charge is greater than or equal to the first state of charge and less than or equal to the second state of charge, then the energy management exit mechanism is determined to be an active exit mechanism, and the steps of obtaining the vehicle's current speed, the current state of charge of the lithium battery, the number of load changes of the fuel cell in the current cycle, and the current road conditions are re-executed; otherwise, the energy management exit mechanism is determined to be a forced exit mechanism, and the basic output power of the fuel cell is determined based on the post-management state of charge, and the fuel cell is controlled based on the basic output power.

[0052] In some embodiments, determining the base output power of the fuel cell based on the post-managed state of charge includes: obtaining a state of charge-fuel cell power lookup table; and determining the base output power based on the post-managed state of charge based on the state of charge-fuel cell power lookup table.

[0053] Wherein, the first energy management duration and the second energy management duration are the durations of the preceding and following stages of the dual-stage power maintenance strategy, respectively; the first difference is the difference between the vehicle's required power after load change and the vehicle's required power before load change within the first energy management duration; the second difference is the difference between the vehicle's required power after load change and the vehicle's required power before load change within the second energy management duration; the post-management state of charge is the fuel cell's state of charge after adjustment by the dual-stage power maintenance strategy; the first state of charge and the second state of charge are the lower and upper limits of the SOC range corresponding to the stable activation module trigger threshold, respectively; the basic output power is the power obtained by querying the state of charge-fuel cell power lookup table.

[0054] Specifically, in this embodiment of the application, after the vehicle is started, the electronic control unit (ECU) can call a preset SOC-fuel cell power lookup table to directly determine the vehicle's required power before the fuel cell load changes.

[0055] For example, when SOC=50%, the corresponding vehicle power requirement before load change is 40kW; when SOC=30%, the corresponding vehicle power requirement before load change is 35kW. The lithium battery only provides a small amount of supplementary power when the vehicle power requirement before load change cannot meet the actual power requirement of the vehicle.

[0056] Furthermore, in the embodiments of this application, during the first energy management period, the fuel cell is controlled to maintain the power output required by the vehicle before the load change. At the same time, a first difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated. The portion of the first difference that is insufficient is supplemented by the lithium battery, and the portion of the first difference that exceeds the first difference is used to charge the lithium battery. During the second energy management period, the fuel cell is controlled to maintain the power output required by the vehicle after the load change. At the same time, a second difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated. The processing method for the second difference is the same as that for the first difference. The first energy management duration and the second energy management duration can both be in the range of 1.5 min to 2.5 min. However, based on the results of actual vehicle tests, the duration of vehicle load changes is mostly 1.8 to 2.2 min. A 1.5 min duration is too short and cannot cover most load change cycles, while a 2.5 min duration is too long and can easily lead to excessive lithium battery replenishment or storage. Therefore, as a preferred solution, the embodiments of this application can set both the first energy management duration and the second energy management duration to 2 min, thereby minimizing the excessive loss rate of the lithium battery.

[0057] For example, during the first energy management period, when the vehicle's power demand changes abruptly from 40kW before the load change to 50kW after the load change, the fuel cell is controlled to maintain an output of 40kW for 2 minutes. During this period, the first difference of 10kW is made up by the lithium battery to avoid membrane electrode stress fluctuations caused by the instantaneous power increase of the fuel cell. After 2 minutes, during the second energy management period, the vehicle's power demand changes to 50kW after the load change and is maintained for 2 minutes. If the vehicle's power demand changes to 45kW again during this period, the fuel cell is still controlled to maintain an output of 50kW after the load change. The excess second difference of 5kW is used to charge the lithium battery to ensure stable fuel cell output.

[0058] Furthermore, after the first and second energy management periods end, this embodiment of the application can obtain the post-management state of charge (SBC) of the lithium battery. If the post-management SBC is greater than or equal to the first SBC, and the post-management SBC is less than or equal to the second SBC, then the energy management exit mechanism is determined to be an active exit mechanism. The steps of obtaining the vehicle's current speed, the current SBC of the lithium battery, the number of load changes in the fuel cell during the current cycle, and the current road conditions are then re-executed to determine whether energy management needs to be performed again. The first SBC can range from 28% to 30%, and the second SBC can range from 60% to 62%. However, 28% is close to the 25% over-discharge threshold of the lithium battery, and 62% is close to the 65% overcharge threshold, resulting in insufficient safety redundancy. Therefore, as a preferred solution, this embodiment of the application can set the first SBC to 30% and the second SBC to 60%, thereby minimizing the safety failure rate of the lithium battery.

[0059] It should be understood that, without considering cost and vehicle weight, the embodiments of this application can use a hybrid energy storage system of "lithium battery + supercapacitor" to manage the energy of fuel cells. The supercapacitor is used to make up for the instantaneous power difference, and the lithium battery is used to handle the continuous power difference. The supercapacitor response time is <1ms, which can further improve the power response speed.

[0060] Furthermore, if the post-management state of charge is less than the first state of charge, and / or the post-management state of charge is greater than the second state of charge, the energy management exit mechanism is determined to be a forced exit mechanism. The system immediately exits the stabilization module and re-executes the step of determining the basic output power of the fuel cell based on the post-management state of charge, prioritizing the safety of the lithium battery.

[0061] Therefore, the embodiments of this application can design a collaborative rule for load change frequency statistics and model prediction. By setting load change judgment criteria, it integrates passive statistical logic and the active prediction logic of the LSTM prediction model, and combines the stable module triggering conditions, the dual-stage power maintenance mechanism, the dynamic correction logic of the LSTM prediction model, and the dynamic exit logic to ensure battery safety and control accuracy. This solves the problems in related technologies, such as adjusting fuel cell power only based on the SOC range (25%-35% low SOC, 35%-65% medium SOC), lacking clear load change frequency statistics rules, relying only on single load change amplitude to trigger stable control, and lacking early high-frequency load change prediction and early intervention capabilities, which lead to increased component losses and decreased energy distribution efficiency.

[0062] Furthermore, to enable those skilled in the art to better understand how this application determines whether the preset energy management conditions are currently met, the following description is provided in conjunction with specific embodiments.

[0063] As one possible implementation, in some embodiments, after determining the predicted probability that the fuel cell will reach a preset number of load changes within a preset time based on the current vehicle speed, current state of charge, number of load changes, and current road conditions, the method further includes: determining whether the number of load changes is greater than or equal to a preset number, or determining whether the predicted probability is greater than or equal to a preset probability, and whether the current state of charge is greater than or equal to a first state of charge, and whether the current state of charge is less than or equal to a second state of charge; if the number of load changes is greater than or equal to the preset number, then it is determined that the preset energy management conditions are met; or, if the predicted probability is greater than or equal to the preset probability, and the current state of charge is greater than or equal to the first state of charge, and the current state of charge is less than or equal to the second state of charge, then it is determined that the preset energy management conditions are met.

[0064] Among them, the preset number of times is a pre-set lower limit threshold for determining the number of load changes that meet the preset energy management conditions; the preset probability is a pre-set lower limit threshold for the predicted probability that meets the preset energy management conditions.

[0065] Specifically, in this embodiment, a fixed 5-minute cycle can be used. If the number of load changes of the fuel cell in the current cycle is greater than or equal to a preset number, the preset energy management conditions are deemed met. If the number of load changes of the fuel cell in the current cycle is less than the preset number, the preset energy management conditions are deemed not met, the count is reset, and the steps of obtaining the vehicle's current speed, the current state of charge of the lithium battery, the number of load changes of the fuel cell in the current cycle, and the current road conditions are re-executed. The current cycle can range from 4 to 6 minutes, and the preset number of times can range from 8 to 12 times. However, based on real-vehicle test results, the average daily load changes in urban road conditions are 20-30 times. A cycle that is too short is prone to false triggering, while a cycle that is too long is prone to missed triggering. Therefore, as a preferred solution, this embodiment can set the current cycle to 5 minutes and the preset number of times to 10 times, thereby minimizing the false triggering rate.

[0066] For example, in this embodiment of the application, the current cycle can be set to 5 minutes, the preset power can be set to 31 kW, the preset power difference can be set to 25 kW, and the preset number of times can be set to 10. Under these conditions, if the fuel cell experiences 12 instances within 5 minutes where the net power of the fuel cell is 35 kW and the adjacent net power difference is 28 kW, that is, the net power of the fuel cell is greater than the preset power, and the actual number of times the adjacent net power difference of the fuel cell is greater than the preset power difference is greater than the preset number of times, then it is determined that the preset energy management conditions are met.

[0067] Furthermore, embodiments of this application can determine whether preset energy management conditions are met by using predicted probability and the current state of charge (SBC). Specifically, if the predicted probability is greater than or equal to the preset probability, and the current SBC is greater than or equal to the first SBC and less than or equal to the second SBC, then the preset energy management conditions are met, and the stabilization module is activated ahead of time without waiting for the 5-minute cycle to end. If the predicted probability is less than the preset probability, and / or the current SBC is less than the first SBC, and / or the current SBC is greater than the second SBC, then the preset energy management conditions are not met, and the step of determining the predicted probability of the fuel cell reaching the preset number of load changes within a preset time based on the current vehicle speed, current SBC, number of load changes, and current road conditions is re-executed. The preset probability can be 80% to ensure prediction accuracy; the first SBC can be 30%, and the second SBC can be 60% to meet safety constraints.

[0068] Furthermore, in the embodiments of this application, rule-based prediction can be used to replace the LSTM prediction model. For example, when the current vehicle speed is <20km / h and the current acceleration fluctuation is >±1m / s², it is determined to be a high-frequency variable load condition. It can achieve basic prediction function, but rule-based prediction only relies on a single or a few features and cannot cover complex road conditions such as sudden traffic jams on highways or when the current vehicle speed is >60km / h but the current acceleration fluctuation is still large. The prediction accuracy is low.

[0069] It should be understood that regardless of whether the process is achieved through "variable load statistical cycle compliance" or "predictive early triggering," the SOC constraint must be met before proceeding to the subsequent dual-stage power maintenance phase. The SOC constraint is: if and only if 30% ≤ SOC ≤ 60%, the preset energy management conditions are considered met. The reasoning is as follows: when SOC < 30%, the remaining charge of the lithium battery is low, and continued charging will lead to over-discharge (voltage below 2.5V / cell); when SOC > 60%, the lithium battery is close to full charge, and continued charging will lead to overcharging (voltage above 3.65V / cell), and the fuel cell's redundant power cannot be stored, resulting in energy waste.

[0070] For example, in this embodiment of the application, the preset probability can be set to 80%, the first state of charge can be set to 30%, and the second state of charge can be set to 60%. Under these conditions, when the vehicle is traveling on a congested urban road, if the LSTM prediction model outputs a probability of "≥8 load changes in the next 3 minutes" based on features such as "6 net power fluctuations in the last 1 minute, all with a difference ≥ 25kW, current vehicle speed ≤ 20km / h, and acceleration fluctuation ± 1.5m / s²", and the current state of charge is 45%, that is, the predicted probability is greater than the preset probability, and the current state of charge is between the first and second states of charge, then it is determined that the preset energy management conditions are met, and the stabilization module is activated in advance.

[0071] Furthermore, after the stabilization module is activated in advance, the actual number of load changes is counted every 30 seconds. If the actual number of load changes deviates from the preset number of load changes by more than 3 times, for example, if the preset number of load changes is 8 times, but the actual number of load changes is only 4 times, it is determined to be a "prediction deviation". The system immediately exits the pre-activation state and re-executes the step of determining the predicted probability of the fuel cell reaching the preset number of load changes within a preset time based on the current vehicle speed, current state of charge, number of load changes, and current road conditions, so as to avoid the stabilization module being triggered erroneously.

[0072] Therefore, this application's embodiments, by establishing a "threshold + cycle" variable load statistical rule, can accurately identify high-frequency, large-amplitude variable load conditions, avoiding false or missed triggering of the stabilization module; using 30%≤SOC≤60% as a stability control premise, it prevents over-discharge and over-charge of the lithium battery, balancing the losses of the fuel cell and the lithium battery; designing a dual-stage power maintenance mechanism with a fixed duration of 2 minutes each, it reduces the frequency of dynamic adjustments to the fuel cell, improving energy distribution efficiency and component lifespan; introducing an LSTM prediction model, it enables early identification of high-frequency variable load conditions and early activation of the stabilization module, solving the problem of delayed intervention of the stabilization module during the early high-frequency variable load within the cycle, further reducing component losses, and effectively improving the practicality, safety, and prediction accuracy of this application.

[0073] The energy management method proposed in this application can determine the predicted probability of a fuel cell reaching a preset number of load changes within a preset time period based on the current vehicle speed, current state of charge, number of load changes, and current road conditions. If at least one of the load change number, predicted probability, and current state of charge meets a preset energy management condition, energy management of the fuel cell is performed based on a preset two-stage power maintenance strategy. This solves the problems of insufficient flexibility in energy management of fuel cell vehicles in related technologies, insufficient consideration of the impact of load change frequency and amplitude, and the need for frequent power adjustments by the fuel cell under high power fluctuations, which exacerbates dynamic losses and shortens lifespan. The method improves the adaptability and responsiveness of the fuel cell to load changes, reduces dynamic losses, and extends its service life.

[0074] Next, the energy management device proposed in the embodiments of this application is described with reference to the accompanying drawings.

[0075] Figure 2 This is a block diagram of an energy management device proposed in an embodiment of this application.

[0076] like Figure 2 As shown, the energy management device 10 includes: an acquisition module 100, a prediction module 200, and a control module 300.

[0077] The acquisition module 100 is used to acquire the vehicle's current speed, the current state of charge of the lithium battery, the number of load changes of the fuel cell in the current cycle, and the current road conditions; the prediction module 200 is used to determine the predicted probability that the fuel cell will reach a preset number of load changes within a preset time period based on the current vehicle speed, the current state of charge, the number of load changes, and the current road conditions; and the control module 300 is used to perform energy management on the fuel cell based on a preset two-stage power maintenance strategy when at least one of the load changes, the predicted probability, and the current state of charge meets a preset energy management condition.

[0078] Optionally, in some embodiments, the control module 300 is specifically configured to: acquire the vehicle's required power before and after the load change; during a first energy management period, control the fuel cell to maintain the vehicle's required power output before the load change, calculate a first difference between the vehicle's required power output after and before the load change, and control the vehicle's lithium battery based on the first difference; during a second energy management period, control the fuel cell to maintain the vehicle's required power output after the load change, calculate a second difference between the vehicle's required power output after and before the load change, and control the vehicle's lithium battery based on the second difference; acquire the post-management state of charge of the lithium battery, determine an energy management exit mechanism based on the post-management state of charge, and exit energy management based on the energy management exit mechanism.

[0079] Optionally, in some embodiments, the control module 300 is specifically used to: determine whether the post-managed state of charge is greater than or equal to the first state of charge and whether the post-managed state of charge is less than or equal to the second state of charge; if the post-managed state of charge is greater than or equal to the first state of charge and less than or equal to the second state of charge, then determine that the energy management exit mechanism is an active exit mechanism, and re-execute the steps of obtaining the vehicle's current speed, the current state of charge of the lithium battery, the number of load changes of the fuel cell in the current cycle, and the current road conditions; otherwise, determine that the energy management exit mechanism is a forced exit mechanism, determine the basic output power of the fuel cell based on the post-managed state of charge, and control the fuel cell based on the basic output power.

[0080] Optionally, in some embodiments, the control module 300 is specifically used to: obtain a state of charge-fuel cell power lookup table; and determine the basic output power based on the state of charge-fuel cell power lookup table and the managed state of charge.

[0081] Optionally, in some embodiments, the acquisition module 100 is specifically used to: acquire the number of times that the net power of the fuel cell is greater than a preset power and the difference between adjacent net power values ​​is greater than a preset power difference within the current cycle; and use the number of occurrences as the number of times the fuel cell is subjected to load changes in the current cycle.

[0082] Optionally, in some embodiments, the prediction module is further configured to: determine whether the number of load changes is greater than or equal to a preset number, or determine whether the prediction probability is greater than or equal to a preset probability, and whether the current state of charge is greater than or equal to a first state of charge, and whether the current state of charge is less than or equal to a second state of charge; if the number of load changes is greater than or equal to the preset number, then it is determined that the preset energy management conditions are met; or, if the prediction probability is greater than or equal to the preset probability, and the current state of charge is greater than or equal to the first state of charge, and the current state of charge is less than or equal to the second state of charge, then it is determined that the preset energy management conditions are met.

[0083] It should be noted that the foregoing explanation of the energy management method embodiment also applies to the energy management device of this embodiment, and will not be repeated here.

[0084] The energy management device proposed in this application can determine the predicted probability of a fuel cell reaching a preset number of load changes within a preset time period based on the current vehicle speed, current state of charge, number of load changes, and current road conditions. If at least one of the load change number, predicted probability, and current state of charge meets a preset energy management condition, energy management of the fuel cell is performed based on a preset two-stage power maintenance strategy. This solves the problems of insufficient flexibility in energy management of fuel cell vehicles in related technologies, insufficient consideration of the impact of load change frequency and amplitude, and the need for frequent power adjustments by the fuel cell under high power fluctuations, which exacerbates dynamic losses and shortens lifespan. It improves the adaptability and responsiveness of the fuel cell to load changes, reduces dynamic losses, and extends its service life.

[0085] Figure 3 A schematic diagram of the structure of a vehicle provided in an embodiment of this application. The vehicle may include: The memory 301, the processor 302, and the computer program stored on the memory 301 and capable of running on the processor 302.

[0086] When the processor 302 executes the program, it implements the energy management method provided in the above embodiments.

[0087] Furthermore, the vehicle also includes: Communication interface 303 is used for communication between memory 301 and processor 302.

[0088] The memory 301 is used to store computer programs that can run on the processor 302.

[0089] The memory 301 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0090] If the memory 301, processor 302, and communication interface 303 are implemented independently, then the communication interface 303, memory 301, and processor 302 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0091] Optionally, in a specific implementation, if the memory 301, processor 302, and communication interface 303 are integrated on a single chip, then the memory 301, processor 302, and communication interface 303 can communicate with each other through an internal interface.

[0092] Processor 302 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0093] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following... Figure 1 The energy management method shown.

[0094] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0095] Furthermore, 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 technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0096] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0097] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0098] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0099] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0100] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0101] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. An energy management method, characterized in that, include: Obtain the vehicle's current speed, the lithium battery's current state of charge, the number of load changes in the fuel cell during the current cycle, and the current road conditions; The predicted probability of the fuel cell reaching a preset number of load changes within a preset time is determined based on the current vehicle speed, the current state of charge, the number of load changes, and the current road conditions. If at least one of the load variation number, the predicted probability, and the current state of charge satisfies the preset energy management conditions, the fuel cell is energy managed based on a preset two-stage power maintenance strategy.

2. The method according to claim 1, characterized in that, The energy management of the fuel cell based on the preset two-stage power maintenance strategy includes: Obtain the vehicle's required power before and after load change; During the first energy management period, the fuel cell is controlled to maintain the power output required by the vehicle before the load change, and a first difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the first difference. During the second energy management period, the fuel cell is controlled to maintain the power output required by the vehicle after the load change, and a second difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the second difference. The state of charge of the lithium battery after management is obtained, and an energy management exit mechanism is determined based on the state of charge after management. Energy management is then exited based on the energy management exit mechanism.

3. The method according to claim 2, characterized in that, The step of determining the energy management exit mechanism based on the post-management state of charge, and exiting energy management according to the energy management exit mechanism, includes: Determine whether the post-management charge state is greater than or equal to the first charge state, and whether the post-management charge state is less than or equal to the second charge state; If the post-management state of charge is greater than or equal to the first state of charge, and the post-management state of charge is less than or equal to the second state of charge, then the energy management exit mechanism is determined to be an active exit mechanism, and the steps of obtaining the vehicle's current speed, the lithium battery's current state of charge, the number of load changes of the fuel cell in the current cycle, and the current road conditions are re-executed. Otherwise, the energy management exit mechanism is determined to be a forced exit mechanism, and the basic output power of the fuel cell is determined based on the post-management state of charge, and the fuel cell is controlled based on the basic output power.

4. The method according to claim 3, characterized in that, Determining the base output power of the fuel cell based on the post-management state of charge includes: Obtain the state of charge-fuel cell power conversion table; Based on the state of charge-fuel cell power lookup table, the base output power is determined according to the post-managed state of charge.

5. The method according to claim 1, characterized in that, The step of obtaining the number of load changes of the fuel cell in the current cycle includes: The number of times that the net power of the fuel cell is greater than a preset power and the difference between adjacent net power values ​​is greater than a preset power difference is obtained within the current period; The number of occurrences is taken as the number of load variations of the fuel cell in the current cycle.

6. The method according to claim 1, characterized in that, After determining the predicted probability that the fuel cell will reach a preset number of load changes within a preset time period based on the current vehicle speed, the current state of charge, the number of load changes, and the current road conditions, the method further includes: Determine whether the number of load changes is greater than or equal to a preset number, or determine whether the predicted probability is greater than or equal to a preset probability, and whether the current state of charge is greater than or equal to a first state of charge, and whether the current state of charge is less than or equal to a second state of charge; If the number of load changes is greater than or equal to the preset number, then the preset energy management conditions are satisfied. Alternatively, if the predicted probability is greater than or equal to the preset probability, and the current state of charge is greater than or equal to the first state of charge, and the current state of charge is less than or equal to the second state of charge, then the preset energy management conditions are satisfied.

7. An energy management device, characterized in that, include: The acquisition module is used to acquire the vehicle's current speed, the current state of charge of the lithium battery, the number of load changes of the fuel cell in the current cycle, and the current road conditions. The prediction module is used to determine the predicted probability that the fuel cell will reach a preset number of load changes within a preset time based on the current vehicle speed, the current state of charge, the number of load changes, and the current road conditions. The control module is configured to manage the energy of the fuel cell based on a preset two-stage power maintenance strategy when at least one of the load variation number, the predicted probability, and the current state of charge satisfies the preset energy management conditions.

8. The apparatus according to claim 7, characterized in that, The control module is specifically used for: Obtain the vehicle's required power before and after load change; During the first energy management period, the fuel cell is controlled to maintain the power output required by the vehicle before the load change, and a first difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the first difference. During the second energy management period, the fuel cell is controlled to maintain the power output required by the vehicle after the load change, and a second difference between the power required by the vehicle after the load change and the power required by the vehicle before the load change is calculated, and the lithium battery of the vehicle is controlled according to the second difference. The state of charge of the lithium battery after management is obtained, and an energy management exit mechanism is determined based on the state of charge after management. Energy management is then exited based on the energy management exit mechanism.

9. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the energy management method as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the program is executed by the processor, it implements the energy management method as described in any one of claims 1-7.

Images