Multi-motor cooperative driving control system suitable for fuel cell heavy truck

Abstract

The invention relates to the technical field of vehicle mechatronics, in particular to a multi-motor cooperative driving control system suitable for a fuel cell heavy truck, which comprises a data acquisition module used for acquiring GPS positioning data and load data of a vehicle, attitude data of a vehicle body and output power curves of a fuel cell and a power cell; the central decision-making module is used for dividing the total driving route into segmented routes and generating a prediction result of the power demand of each segmented route; the energy management module is used for distributing the output power of the fuel cell and the power cell in advance; the multi-motor driving control module is used for adjusting the rotating speeds and torques of different motors according to the power distributed by the energy management module; and the energy recovery module is used for converting kinetic energy of the vehicle into electric energy and storing the electric energy into the power battery. According to the invention, when the vehicle runs, optimization and adjustment of the power can be completed in advance, and an optimal control strategy is generated.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle mechatronics, and particularly to a multi-motor cooperative drive control system applicable to fuel cell heavy trucks. Background Art

[0002] Fuel cells directly convert the chemical energy of fuel into electrical energy through electrochemical reactions to provide power for vehicles. Compared with traditional fuel engines, they greatly reduce pollutant emissions. However, the driving conditions of heavy trucks are extremely complex, often facing frequent start-stop, climbing, turning, and heavy-load transportation scenarios, which pose extremely high requirements for the vehicle's power system.

[0003] Single-motor drive systems are difficult to meet the diverse power and torque requirements of heavy trucks under complex working conditions. Existing multi-motor cooperative drive control systems can accurately control the output of each motor according to different driving conditions to achieve more efficient power distribution and torque adjustment. However, most multi-motor cooperative drive control systems still have problems in that it is difficult to reasonably distribute the output power of fuel cells and power batteries according to the vehicle's real-time working conditions and energy state to achieve the best energy utilization efficiency and vehicle performance.

[0004] In summary, how to solve the problem that existing multi-motor cooperative drive control systems still have difficulty in reasonably distributing the output power of fuel cells and power batteries according to the vehicle's real-time working conditions and energy state has become an urgent problem to be solved in this field. Therefore, it is necessary to propose a multi-motor cooperative drive control system applicable to fuel cell heavy trucks. Summary of the Invention

[0005] To solve the above problems, the present invention provides a multi-motor cooperative drive control system applicable to fuel cell heavy trucks, which is used to be able to complete the optimization adjustment of power in advance during vehicle driving and generate an optimal control strategy.

[0006] To achieve the above object, the technical solution of the present invention is as follows: A multi-motor cooperative drive control system applicable to fuel cell heavy trucks includes a fuel cell and a power battery; it further includes:

[0007] A data acquisition module, including a GPS sensor, a load sensor, a slope sensor, and an acceleration sensor arranged on the vehicle, and a power sensor arranged in the fuel cell and the power battery.

[0008] The data acquisition module is used to collect the vehicle's GPS positioning data, load data, body attitude data, and the output power curves of the fuel cell and the power battery by using the GPS sensor, the load sensor, the slope sensor, the acceleration sensor, and the power sensors in the fuel cell and the power battery.

[0009] The central decision-making module is used to divide the total driving route into segmented routes of uphill sections, downhill sections, turning sections, and straight sections based on the characteristics of the total driving route, and extract the slopes and curvatures of the segmented routes.

[0010] The slopes and curvatures of the segmented routes and the output power curves of the fuel cell and the power battery are input into the LSTM network. By using the time-series correlation analysis function of the LSTM network, the power requirements for each segmented route are analyzed and generated. The power requirements for each segmented route are used as the input layer and input into the neural network model to generate the prediction results of the power requirements for each segmented route.

[0011] The energy management module is used to, based on the vehicle's GPS positioning data, load data, body attitude data, the output power curves of the fuel cell and the power battery, and the prediction results of the power requirements for each segmented route, when the vehicle is traveling on each segmented route of the total driving route, use a three-stage dynamic collaborative management strategy to allocate the output power of the fuel cell and the power battery in advance.

[0012] The multi-motor drive control module is used to adjust the speeds and torques of different motors according to the power allocated by the energy management module for different segmented routes.

[0013] The energy recovery module is used to, when the vehicle is in either a downhill section or a braking state, control the conversion of the motor's power generation state, and convert the vehicle's kinetic energy into electrical energy and store it in the power battery.

[0014] Furthermore, in the data acquisition module, the body attitude data includes uphill state data, downhill state data, steering state data, acceleration / deceleration state data, and braking state data.

[0015] Furthermore, in the central decision-making module, the slope of the uphill section is ≥5% and the length > 500m, the slope of the downhill section is ≤ -3% and the length > 300m, the radius of curvature of the turning section is ≤ 100m, and the radius of curvature of the straight section is > 100m.

[0016] Furthermore, the central decision-making module includes a dynamic adjustment unit. The dynamic adjustment unit is used to compare in real time the errors between the vehicle's GPS positioning data, load data, body attitude data, and the output power curves of the fuel cell and the power battery and the prediction results, and use the errors to adjust the weight coefficients in the LSTM network.

[0017] Furthermore, the central decision-making module also includes an overtaking alternative route unit. The overtaking alternative route unit includes a lidar sensor and a camera installed on the vehicle.

[0018] The overtaking alternative route unit is used to identify and generate the position information of surrounding vehicles by using the lidar sensor and the camera.

[0019] When the driver issues an overtaking command, real-time GPS positioning data and the position information of surrounding vehicles are used to plan alternative overtaking routes, and the time required for overtaking on the alternative overtaking routes and the power consumption during the overtaking process are evaluated and provided to the driver for reference. At the same time, the overtaking command is sent to the central decision-making module to adjust the output power distribution of the fuel cell and the power battery in advance, and the multi-motor drive control module is used to pre-adjust the speed and torque of the motor.

[0020] Further, in the central decision-making module, using the time-series correlation analysis function of the LSTM network, the steps for analyzing and generating the power requirements of each segmented route include:

[0021] S1, data preprocessing: Preprocess the slope and curvature data of the segmented route input to the LSTM network, as well as the output power curve data of the fuel cell and the power battery, to obtain the preprocessed time-series data.

[0022] S2, dividing the sample sequence: Divide the preprocessed time-series data at a certain time step to generate the completed input sample sequence.

[0023] S3, inputting into the model: Sequentially input the completed input sample sequence into the constructed and trained LSTM network model.

[0024] S4, model prediction: The LSTM network uses the completed input sample sequence and the internal memory mechanism, and through the operation of the hidden layer, outputs the preliminary predicted values of the power requirements corresponding to each segmented route.

[0025] S5, fine-tuning the predicted value: Combine the vehicle's historical driving data, the empirical values of power consumption under different working conditions, and the real-time load data to fine-tune the preliminary predicted values.

[0026] Further, in S1, the preprocessing is to normalize the slope and curvature data and smooth the output power curve data of the fuel cell and the power battery.

[0027] Further, in S2, the sample sequence includes the slope, curvature, and the output power of the fuel cell and the power battery of the segmented route within a certain time period.

[0028] Further, in S4, the predicted value is the power magnitude that the vehicle may need when driving on each segmented route under the current state of the output power curves of the fuel cell and the power battery.

[0029] Further, in the energy management module, the three-stage dynamic collaborative management strategy includes:

[0030] The first stage: At the initial stage of vehicle startup and when driving on a flat and lightly loaded straight section, enable the fuel cell to supply power.

[0031] Second paragraph: When the vehicle is in one or more of the working conditions of driving on an uphill section, accelerating, and overtaking, the fuel cell and the power battery supply power simultaneously.

[0032] Third paragraph: When the vehicle is on a downhill section or braking, the energy recovery module is started.

[0033] The following are the beneficial effects of adopting the above solution:

[0034] 1. The present invention collects multi-dimensional data of the vehicle through the data acquisition module, and the central decision-making module analyzes and predicts the power demand, enabling the energy management module to reasonably allocate the output power of the fuel cell and the power battery according to the three-stage dynamic collaborative management strategy, improving the energy utilization efficiency to a certain extent, reducing energy consumption, reducing the vehicle operation cost, and at the same time extending the service life of the fuel cell and the power battery.

[0035] 2. The central decision-making module of the present invention divides the total driving route into multiple sections based on the total driving route, and accurately analyzes the power demand of each section of the route by using an algorithm, realizing that before the vehicle travels to different sections of the route, the power optimization adjustment can be completed in advance, generating an optimal control strategy, and further reducing the vehicle operation cost.

[0036] 3. The multi-motor drive control module of the present invention can adjust the motor speed and torque according to the power pre-allocated by the energy management module for different sections of the route. In complex working conditions, with the pre-formulated multi-section route planning and power adjustment strategy, each motor can work in coordination. This pre-layout method improves the vehicle's controllability and driving safety to a certain extent.

[0037] The additional aspects and advantages of the present invention will be partially given in the following description, partially will become obvious from the following description, or will be understood through the practice of the present invention. Description of the Drawings

[0038] Figure 1 It is a structural diagram of a multi-motor collaborative drive control system applicable to a fuel cell heavy truck of the present invention. Detailed Embodiment

[0039] The following is a further detailed description through specific embodiments:

[0040] As shown in the Figure 1 drawing: A multi-motor collaborative drive control system applicable to a fuel cell heavy truck mainly includes the following modules:

[0041] Data acquisition module: Collect vehicle positioning, load, attitude, and battery power curve data; Central decision-making module: Divide the route, analyze and generate power requirements for each section, and make predictions; Energy management module: Allocate the output power of the fuel cell and the power battery according to the strategy; Multi-motor drive control module: Adjust the motor speed and torque according to the allocated power; Energy recovery module: Recover kinetic energy during downhill or braking and store it in the battery; Dynamic adjustment unit: Compare the errors and adjust the weight coefficients of the LSTM network; Overtaking alternative route unit: Identify surrounding vehicles, plan and evaluate overtaking routes, and adjust power and motor parameters in advance.

[0042] The functions of each module are explained in detail as follows:

[0043] The data acquisition module includes a GPS sensor, a load sensor, a slope sensor, and an acceleration sensor arranged on the vehicle, as well as a power sensor arranged in the fuel cell and the power battery.

[0044] The data acquisition module is used to collect the vehicle's GPS positioning data, load data, body attitude data, and the output power curves of the fuel cell and the power battery by using the GPS sensor, the load sensor, the slope sensor, the acceleration sensor, and the power sensor in the fuel cell and the power battery.

[0045] In the data acquisition module, the body attitude data includes uphill state data, downhill state data, steering state data, acceleration / deceleration state data, and braking state data.

[0046] The central decision-making module is used to divide the total driving route into segmented routes of uphill sections, downhill sections, turning sections, and straight sections according to characteristics based on the total driving route, and extract the slope and curvature of the segmented routes.

[0047] Input the slope and curvature of the segmented routes and the output power curves of the fuel cell and the power battery into the LSTM network, and use the time series correlation analysis function of the LSTM network to analyze and generate the power requirements for each segmented route; Use the power requirements for each segmented route as the input layer and input them into the neural network model to generate the prediction results of the power requirements for each segmented route.

[0048] In the central decision-making module, the slope of the uphill section is ≥5% and the length > 500m, the slope of the downhill section is ≤ -3% and the length > 300m, the curvature radius of the turning section is ≤ 100m, and the curvature radius of the straight section is > 100m.

[0049] The central decision-making module includes a dynamic adjustment unit. The dynamic adjustment unit is used to compare the errors between the vehicle's GPS positioning data, load data, body attitude data, and the output power curves of the fuel cell and the power battery and the prediction results in real time, and use the errors to adjust the weight coefficients in the LSTM network.

[0050] The central decision-making module further includes an overtaking alternative route unit, and the overtaking alternative route unit includes a lidar sensor and a camera disposed on the vehicle.

[0051] The overtaking alternative route unit is used to identify and generate the position information of surrounding vehicles by using the lidar sensor and the camera.

[0052] When the driver issues an overtaking instruction, by using the real-time GPS positioning data and the position information of surrounding vehicles, an overtaking alternative route is planned, and the time required for overtaking the alternative route and the power consumption during the overtaking process are evaluated and provided for the driver's reference. At the same time, the overtaking instruction is sent to the central decision-making module to adjust the output power distribution of the fuel cell and the power battery in advance, and the rotation speed and torque of the motor are pre-adjusted by using the multi-motor drive control module.

[0053] In the central decision-making module, using the time-series correlation analysis function of the LSTM network, the steps for analyzing and generating the power requirements of each segmented route are as follows:

[0054] S1, data preprocessing: Preprocess the slope and curvature data of the segmented route input to the LSTM network, as well as the output power curve data of the fuel cell and the power battery, to obtain the preprocessed time-series data.

[0055] In S1, the preprocessing is to perform normalization processing on the slope and curvature data and smoothing processing on the output power curve data of the fuel cell and the power battery.

[0056] S2, dividing the sample sequence: Divide the preprocessed time-series data at a certain time step to generate the completed input sample sequence.

[0057] In S2, the sample sequence includes the slope, curvature, and output power of the fuel cell and the power battery of the segmented route within a certain time period.

[0058] S3, inputting into the model: Input the completed input sample sequence into the constructed and trained LSTM network model in sequence.

[0059] S4, model prediction: The LSTM network uses the completed input sample sequence and the internal memory mechanism, and through the operation of the hidden layer, outputs the preliminary predicted value of the power requirement corresponding to each segmented route.

[0060] In S4, the predicted value is the magnitude of the power that may be required when the vehicle travels on each segmented route under the current state of the output power curve of the fuel cell and the power battery.

[0061] S5, Predicted value fine-tuning: Combine the historical driving data of the vehicle, the empirical power consumption values under different working conditions, and the real-time load data to fine-tune the preliminary predicted value.

[0062] The energy recovery module is used to control the conversion of the motor's power generation state when the vehicle is in either a downhill section or braking, and convert the vehicle's kinetic energy into electrical energy and store it in the power battery.

[0063] The energy management module is used to, based on the vehicle's GPS positioning data, load data, body attitude data, the output power curves of the fuel cell and the power battery, and the prediction results of the power requirements for each segmented route, when the vehicle is traveling on each segmented route of the total driving route, use a three-stage dynamic collaborative management strategy to allocate the output powers of the fuel cell and the power battery in advance.

[0064] In the energy management module, the three-stage dynamic collaborative management strategy includes:

[0065] The first stage: At the initial stage of vehicle startup and when traveling on a flat and lightly loaded straight section, enable the fuel cell to supply power.

[0066] The second stage: When the vehicle is traveling in one or more of the working conditions of an uphill section, accelerating, and overtaking, the fuel cell and the power battery supply power simultaneously.

[0067] The third stage: When the vehicle is in a downhill section or braking, start the energy recovery module.

[0068] The multi-motor drive control module is used to adjust the rotational speed and torque of different motors for different segmented routes according to the power allocated by the energy management module.

[0069] Specifically, first, when the vehicle is traveling on a total driving route, this embodiment takes the example of starting from a logistics park located at 116.3855° east longitude and 39.9049° north latitude and going to the urban distribution center.

[0070] In terms of the data acquisition module, after the vehicle starts, the GPS sensor continuously acquires the vehicle's real-time positioning data. At this moment, the vehicle is at the position of 116.4012° east longitude and 39.9125° north latitude. The load sensor senses that the weight of the goods loaded on the vehicle is 15 tons. The slope sensor detects that the vehicle is on a section with a slope of 3%. The acceleration sensor captures that the vehicle is accelerating at an acceleration of 2 m / s 2 ². At the same time, the power sensors in the fuel cell and the power battery respectively acquire that the output power of the fuel cell is 80 kW and the output power of the power battery is 40 kW.

[0071] After the central decision-making module receives the data transmitted by the data acquisition module, it starts to divide the total driving route.

[0072] For example, there is a section of road 850m long with a slope of 6% ahead. According to the set standards, this section of road is classified as an uphill section. By preprocessing the slope and curvature of this uphill section and the current output power curve data of the fuel cell and power battery, that is, normalizing the slope data to make it in the range of [0,1], and smoothing the power curve data to remove noise interference. Then, at a certain time step, with every 5 seconds as a time step, the sample sequence is divided. Each sample sequence contains the slope, curvature, and the output power of the fuel cell and power battery of this uphill section within these 5 seconds. These divided input sample sequences are successively input into the constructed and trained LSTM network model. Through the internal memory mechanism and the operations of the hidden layer of the LSTM network, the preliminary predicted value of the power demand for this uphill section is output as 150 kilowatts. Then, combining the historical driving data of the vehicle under similar slopes and loads in the past, as well as the empirical values of power consumption under different working conditions, and considering the current real-time load data of 15 tons, the preliminary predicted value of 150 kilowatts is fine-tuned to obtain a more accurate predicted value of the power demand of 160 kilowatts.

[0073] When the vehicle is driving, the driver finds that the vehicle in front is driving slowly and issues an overtaking command. At this time, the lidar sensor and camera of the overtaking alternative route unit start to work, identify and generate the position information of surrounding vehicles. For example, it is detected that the speed of the vehicle in front is 50 km / h, the distance from the vehicle is 30 meters, the speed of the vehicle in the adjacent lane is 60 km / h, and the distance from the vehicle is 25 meters, etc. Using the real-time GPS positioning data and the position information of these surrounding vehicles, multiple overtaking alternative routes are planned. For example, the expected overtaking time of Route A is 30 seconds, and the expected power consumption during overtaking is 180 kilowatts; the expected overtaking time of Route B is 25 seconds, and the power consumption is expected to be 200 kilowatts, and this information is provided for the driver's reference. At the same time, the overtaking command is sent to the central decision-making module. The central decision-making module adjusts the output power distribution of the fuel cell and power battery in advance. For example, the output power of the fuel cell is increased to 100 kilowatts, and the output power of the power battery is adjusted to 70 kilowatts, and notifies the multi-motor drive control module to pre-adjust the motor speed and torque to meet the power demand during overtaking.

[0074] The energy management module executes a three-stage dynamic collaborative management strategy based on various data of the vehicle and the predicted results of the power demand for each section of the route. At the initial stage of vehicle startup, when driving out of the logistics park, since it is in a flat and lightly loaded state, the fuel cell is enabled to supply power alone. When the vehicle drives to the above-mentioned uphill section, the fuel cell and the power battery supply power simultaneously to ensure that the vehicle has enough power to climb the slope. When the vehicle drives to a downhill section 420m long with a slope of 4%, the energy recovery module is started. At this time, the power generation state of the motor is converted, and the kinetic energy of the vehicle is converted into electrical energy and stored in the power battery.

[0075] The multi-motor drive control module operates according to the power allocated by the energy management module. For example, in the uphill section mentioned above, the total power allocated by the energy management module to the motors is 220 kW. The multi-motor drive control module reasonably distributes the power to each motor according to different motor characteristics, and adjusts the rotational speed and torque of different motors.

[0076] Suppose the vehicle has three motors.

[0077] Motor 1 is allocated 80 kW of power, the rotational speed is adjusted to 1800 rpm, and the torque is adjusted to 350 N·m.

[0078] Motor 2 is allocated 70 kW of power, the rotational speed is adjusted to 1700 rpm, and the torque is adjusted to 320 N·m.

[0079] Motor 3 is allocated 70 kW of power, the rotational speed is adjusted to 1600 rpm, and the torque is adjusted to 300 N·m.

[0080] Obviously, the above embodiments are merely examples given for clear illustration and are not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or variations can be made based on the above description. It is not necessary and impossible to list all implementation manners here. And the obvious changes or variations derived therefrom are still within the protection scope of the present invention.

Claims

1. A multi-motor coordinated drive control system suitable for fuel cell heavy trucks, including a fuel cell and a power battery, characterized in that: Also includes: A data acquisition module, including a GPS sensor, a load sensor, a slope sensor and an acceleration sensor arranged on the vehicle, and a power sensor arranged in the fuel cell and the power battery; The data acquisition module is used to collect GPS positioning data, load data, vehicle posture data and output power curves of the fuel cell and power battery using GPS sensors, load sensors, slope sensors, acceleration sensors and power sensors in the fuel cell and power battery; A central decision module is used for dividing the total driving route into segmented routes of uphill segments, downhill segments, turning segments and straight segments according to characteristics based on the total driving route, and extracting the slope and curvature of the segmented routes; The slope and curvature of the segmented route and the output power curves of the fuel cell and the power battery are input into the LSTM network, and the time series association analysis function of the LSTM network is used to analyze and generate the power demand of each segmented route; the power demand of each segmented route is input into the neural network model as the input layer to generate the prediction result of the power demand of each segmented route; The energy management module is used to allocate the output power of the fuel cell and the power battery in advance using a three-stage dynamic collaborative management strategy when the vehicle is traveling on each segment of the total driving route based on the vehicle's GPS positioning data, load data, vehicle body posture data, output power curves of the fuel cell and the power battery, and the predicted results of the power demand of each segment route; A multi-motor drive control module is used to adjust the speed and torque of different motors for different segmented routes according to the power allocated by the energy management module; The energy recovery module is used to control the conversion of the power generation state of the motor when the vehicle is in a downhill section or braking, so as to convert the kinetic energy of the vehicle into electrical energy and store it in the power battery.

2. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 1 is characterized in that: In the data acquisition module, the posture data of the vehicle body includes uphill state data, downhill state data, steering state data, acceleration and deceleration state data and braking state data.

3. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 2 is characterized in that: In the central decision module, the slope of the uphill section is ≥5% and the length is >500m, the slope of the downhill section is ≤-3% and the length is >300m, the curvature radius of the turning section is ≤100m, and the curvature radius of the straight section is >100m.

4. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 3 is characterized in that: The central decision module includes a dynamic adjustment unit, which is used to compare the vehicle's GPS positioning data, load data, vehicle body posture data, and the output power curves of the fuel cell and power battery with the errors of the predicted results in real time, and use the errors to adjust the weight coefficients in the LSTM network.

5. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 4 is characterized in that: The central decision module also includes an alternative overtaking route unit, which includes a laser radar sensor and a camera arranged on the vehicle; The overtaking alternative route unit is used to generate surrounding vehicle position information using laser radar sensors and camera recognition; When the driver issues an overtaking command, the system uses real-time GPS positioning data and surrounding vehicle location information to plan an alternative overtaking route, and evaluates the time for overtaking on the alternative overtaking route and the power consumption during the overtaking process, which are provided to the driver for reference. At the same time, the overtaking command is sent to the central decision-making module to adjust the output power distribution of the fuel cell and power battery in advance, and use the multi-motor drive control module to pre-adjust the speed and torque of the motor.

6. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 5 is characterized in that: In the central decision-making module, the time series correlation analysis function of the LSTM network is used to analyze and generate the power requirements of each segment route, including the following steps: S1, data preprocessing: preprocessing the slope and curvature data of the segmented route input to the LSTM network, as well as the output power curve data of the fuel cell and the power battery, to obtain preprocessed time series data; S2, divide the sample sequence: divide the preprocessed time series data according to a certain time step to generate a divided input sample sequence; S3, input model: input the divided input sample sequence into the constructed and trained LSTM network model in sequence; S4, model prediction: The LSTM network uses the divided input sample sequence and internal memory mechanism to output the preliminary predicted value of power demand corresponding to each segment route through hidden layer operations; S5, fine-tuning of predicted values: fine-tuning of the preliminary predicted values ​​based on the historical driving data of the vehicle, the empirical values ​​of power consumption under different working conditions, and the real-time load data.

7. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 6 is characterized in that: In S1, the preprocessing is to normalize the slope and curvature data and smooth the output power curve data of the fuel cell and the power battery.

8. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 7 is characterized in that: In S2, the sample sequence includes the slope and curvature of the segmented route within a certain period of time, as well as the output power of the fuel cell and the power battery.

9. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 8 is characterized in that: In S4, the predicted value is the power that the vehicle may need when traveling on each segmented route under the current state of the fuel cell and power battery output power curve.

10. The multi-motor coordinated drive control system suitable for fuel cell heavy trucks according to claim 9, characterized in that: In the energy management module, the three-stage dynamic collaborative management strategy includes: The first stage: when the vehicle is initially started and driving on a flat and lightly loaded straight section, the fuel cell is enabled to supply power; The second stage: When the vehicle is driving uphill, accelerating or overtaking, the fuel cell and the power battery supply power at the same time; The third stage: When the vehicle is on a downhill slope or braking, the energy recovery module is activated.

Images