An overall vehicle drive control system for an amphibious transport vehicle based on pattern recognition

Abstract

The present invention relates to the technical field of drive for multi-functional hybrid transport vehicles, and specifically discloses a vehicle drive control system for an amphibious vehicle based on pattern recognition. The system includes a working mode recognition module that predicts future required power by analyzing historical vehicle data, combines the current required power, the state of charge of the battery, and the state of the vehicle mode switch, and selects the optimal working mode based on the logical gate threshold judgment rule; a drive control management module that calculates the initial required torque according to the driver input information and the vehicle state information, designs a torque coordination filtering adaptive function, performs torque filtering using an adaptive torque change factor, converts the current required torque into required power through a power-torque conversion formula, assists in selecting the optimal working mode, and reasonably distributes power based on this working mode. The present invention can effectively improve the adaptability of amphibious vehicles in various complex operating scenarios of "variable environment - cross-medium".

Description

Technical Field

[0001] The present invention relates to the technical field of drive technologies for multi-functional hybrid transport vehicles, and particularly to an integrated drive control system for an amphibious transport vehicle based on pattern recognition. Background Art

[0002] With the diversified development of modern transportation demands, amphibious vehicles have gradually become important tools for coping with transportation tasks in complex terrains. At the same time, the increasingly severe global energy shortage and environmental pollution problems have further promoted the innovation of transportation vehicle technologies. Hybrid power systems, with the complementary advantages of internal combustion engines and electric motors, not only significantly improve energy utilization efficiency but also greatly extend the vehicle's cruising range, and have become the key technical direction for the development of amphibious heavy-load transport vehicles.

[0003] As a special-purpose vehicle, an amphibious heavy-load transport vehicle needs to perform efficient and stable transportation tasks in complex land and water environments. Therefore, methods for recognizing working modes for various complex operating scenarios, as well as stable, reliable, and highly adaptable drive control technologies, are particularly important. Summary of the Invention

[0004] The purpose of the present invention is to provide an integrated drive control system for an amphibious transport vehicle based on pattern recognition to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] Term Explanation:

[0007] ISG motor (Integrated Starter Generator): A motor that integrates starting and generating functions.

[0008] WHVC condition (World Harmonized Vehicle Cycle): A standardized test condition for heavy vehicles (such as commercial trucks, buses, etc.).

[0009] RDE condition (Real Driving Emissions): On-road driving emission test.

[0010] An integrated drive control system for an amphibious transport vehicle based on pattern recognition, the drive control system includes:

[0011] The working mode recognition module analyzes the collected historical vehicle data to predict the required power for the next N moments, calculates the average required power for these moments, and selects the best working mode suitable for the current moment based on logical gate thresholds by combining the current required power, the state of charge of the battery, and the vehicle mode switch state. Among them, the historical vehicle data includes historical vehicle speed, historical power demand, historical accelerator pedal opening, historical brake pedal switch signal, and historical working mode;

[0012] The drive control management module first calculates the initial required torque of the vehicle according to the driver's input information in combination with the vehicle state information; then, it restricts the initial required torque by the maximum allowable power, the maximum allowable vehicle speed, and the maximum drive performance of the motor and the pump; then, it filters the drive torque to avoid excessive torque fluctuations and ensure the smoothness of vehicle driving; finally, it converts the current required torque into the current required power through the power-torque conversion formula and inputs it into the working mode recognition module to assist in selecting the current best working mode, and reasonably distributes the required power to the engine and the power battery based on this working mode.

[0013] As a further technical solution of the present invention, in the working mode recognition module, the present invention uses a feedforward neural network to predict the required power for the next N moments. The parameter update of the feedforward neural network includes two stages: offline optimization and online fine-tuning. First, the parameters of the feedforward neural network are preliminarily learned using standard driving conditions, and the learned network model is installed in the controller. The parameters of the feedforward neural network are updated by online fine-tuning through collecting the actual operation data of the amphibious vehicle, so that it can better adapt to the actual driving condition requirements of the amphibious vehicle;

[0014] Define the online fine-tuning loss function as follows:

[0015] Among them, is the loss value at the th moment, is the total loss value at the th moment, is the actual required power value at the th moment, is the predicted required power value at the th moment, is the number of data samples, is the update threshold used to prevent the model from changing violently and destroying the prediction ability of the model. The selected number of data samples , that is, whenever samples are collected during the actual operation of the amphibious vehicle, the parameters of the feedforward neural network are updated once, and the fine-tuned model is used to predict the next The required power at a moment, and execute the above steps repeatedly to continuously update and optimize the model online.

[0016] As a further technical solution of the present invention, in the working mode recognition module, the steps of selecting the best working mode most suitable for the current moment based on the logical gate threshold judgment include:

[0017] After the vehicle enters the drivable state, first select three working modes of land, water, and beach landing ship according to the mode switch pressed by the driver. There is an interlock relationship among the three working modes. Once entering a certain working mode, only after exiting this mode can it be switched to other modes. If two or more switches are turned on simultaneously, the mode selection is carried out in the order of priority. Only after the first switch is turned off can it enter another working mode. After determining that the vehicle is in a certain working mode, perform the corresponding working mode selection;

[0018] The land working mode includes six types: hybrid driving mode 1, hybrid driving mode 2, pure electric driving mode, power generation driving mode, energy recovery mode, and mechanical braking mode; when the required power is greater than 0, first judge whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated quantity a; if the difference is greater than a, enter hybrid driving mode 1 to meet the driver's instantaneous high-power demand; if the difference is less than a, judge whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter hybrid driving mode 2 to make the engine work in the high-efficiency area; if it is less than the average required power, judge whether the current state of charge of the battery is lower than the charging threshold. If it is lower than the charging threshold, enter the power generation driving mode, and the engine works in the high-efficiency area to charge the power battery with the excess power; if it is higher than the charging threshold, enter the pure electric driving mode to avoid the engine running in the low-efficiency area to reduce fuel consumption. When the required power is less than 0, first judge whether the current state of charge of the battery is lower than the maximum state of charge threshold. If it is lower, enter the energy recovery mode. If it is higher, enter the conventional mechanical braking mode.

[0019] As a further technical solution of the present invention, the water operation mode includes four types: hybrid navigation mode 1, hybrid navigation mode 2, pure electric navigation mode, and power generation navigation mode; when the required power is greater than 0, first determine whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value b; if the difference is greater than b, enter the hybrid navigation mode 1 to meet the driver's instantaneous high-power demand; if the difference is less than b, determine whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter the hybrid navigation mode 2 to make the engine operate in the high-efficiency area; if it is less than the average required power, determine whether the current state of charge of the battery is lower than the charging threshold. If it is lower than the charging threshold, enter the power generation navigation mode, and the engine operates in the high-efficiency area to charge the power battery with the excess power; if it is higher than the charging threshold, enter the pure electric navigation mode to avoid the engine operating in the low-efficiency area and reduce fuel consumption.

[0020] As a further technical solution of the present invention, the ship's working mode for going up and down the beach includes three types: hybrid up and down the beach mode 1, hybrid up and down the beach mode 2, and pure electric up and down the beach mode; first, determine whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value c; if the difference is greater than c, enter the hybrid up and down the beach mode 1 to meet the driver's instantaneous high-power demand; if the difference is less than c, determine whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter the hybrid up and down the beach mode 2 to make the engine operate in the high-efficiency area; if it is less than the average required power, enter the pure electric up and down the beach mode to avoid the engine operating in the low-efficiency area and reduce fuel consumption.

[0021] As a further technical solution of the present invention, in the drive control management module, a torque coordination filtering adaptive function is designed, and an adaptive torque change factor is used for torque coordination filtering. On the premise of ensuring the acceleration performance, the torque is prevented from changing violently to ensure the smoothness of vehicle driving; the torque change factor will be adaptively adjusted in real time according to the actual torque value and the required torque value, and the strategy is as follows:

[0022] First, determine an initial value of the torque change factor according to the torque interval where the current actual torque value is located. The division of the torque interval can adopt the conventional S-shaped distribution, and each interval corresponds to an initial value of the torque change factor. Secondly, determine a torque change factor coefficient through the adaptive function according to the difference between the actual torque value and the target torque value. Finally, multiply the initial value of the torque change factor by the torque change factor coefficient to obtain the final torque change factor, and use the torque change factor to adjust the actual torque value;

[0023] The required torque after filtering can be converted into the current required power by the following formula:

[0024] ;

[0025] Among them, is the current required power, is the current required torque, is the rotational speeds of the drive motor, in-wheel motor and jet pump, is the motor working efficiency;

[0026] Input the calculated current required power into the working mode recognition module to assist in identifying the current optimal working mode, and according to the power distribution scheme of this working mode, rationally distribute the required power to the engine and the power battery.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] (1) A working mode recognition method applicable to an amphibious heavy-duty transport vehicle is proposed, which improves the working mode recognition accuracy of the vehicle under various complex operating scenarios on land and water.

[0029] (2) A drive control management method applicable to an amphibious heavy-duty transport vehicle is proposed. Based on the result of pattern recognition, the power distribution scheme of the vehicle under different operating scenarios can be adjusted in real time, which improves the adaptability of the vehicle under various complex operating scenarios on land and water.

[0030] (3) A torque coordination filtering adaptive function is designed. Adaptive torque change factors are used for torque coordination filtering. On the premise of ensuring the acceleration performance, the torque is prevented from changing violently, ensuring the smoothness of the vehicle driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0032] Figure 1 It is the structure diagram of the power system of the amphibious heavy-duty transport vehicle.

[0033] Figure 2 It is the drive control system diagram of the amphibious heavy-duty transport vehicle based on pattern recognition.

[0034] Figure 3 It is the flowchart of the land working mode recognition.

[0035] Figure 4 It is the flowchart of the water working mode recognition.

[0036] Figure 5 It is the flowchart of the up-and-down beach ship working mode recognition. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0038] As Figure 1 shown, the power system of the amphibious heavy-duty transporter adopts an extended-range hybrid power system, which is mainly composed of components such as an engine, a power battery, an ISG motor, a drive motor, a hub motor, a jet pump propeller, a clutch, a transfer case and a transmission. The drive control technology of the amphibious heavy-duty transporter based on pattern recognition disclosed by the present invention aims to select the best working mode according to the different power requirements of the vehicle under various complex operating scenarios on land and water, and reasonably allocate the vehicle power requirements to optimize the power output between the engine and the power battery.

[0039] Please refer to Figure 2 , the embodiment of the present invention provides a vehicle drive control system for an amphibious transporter based on pattern recognition, and the drive control system includes:

[0040] A working mode recognition module, which predicts the required power for the next N moments by analyzing the collected vehicle historical data, calculates the average required power for these moments, and selects the best working mode most suitable for the current moment based on logical gate threshold judgment by combining the current required power, the state of charge of the battery and the state of the vehicle mode switch, wherein the vehicle historical data includes historical vehicle speed, historical power requirement, historical accelerator pedal opening, historical brake pedal switch signal and historical working mode;

[0041] In the working mode recognition module, the present invention uses a feedforward neural network to predict the required power for the next N moments, and utilizes the powerful fitting and feature learning ability of the neural network to achieve accurate prediction of the future required power; the parameter update of the feedforward neural network includes two stages: offline optimization and online fine-tuning. First, the parameters of the feedforward neural network are initially learned using standard driving conditions (such as WHVC, RDE, etc.), and the learned network model is installed in the controller. The parameters of the feedforward neural network are updated by online fine-tuning through collecting the actual operation data of the amphibious transporter to make it more adaptable to the actual driving condition requirements of the amphibious transporter. The actual operation data includes actual vehicle speed, actual power output, actual accelerator pedal opening, actual brake pedal switch and actual working mode;

[0042] Define the online fine-tuning loss function as follows:

[0043] ;

[0044] ;

[0045] Among them, is the loss value at the th moment, is the total loss value at the th moment, is the actual demand power value at the th moment, is the number of data samples, is the update threshold used to prevent the model from changing violently and damaging the prediction ability of the model. By using this loss function for backpropagation calculation, online fine-tuning and parameter update are realized. In the present invention, the selected number of data samples , that is, whenever samples are collected during the actual operation of the amphibious transport vehicle, the parameters of the feedforward neural network are updated once, and the demand power at the next moment is predicted using the fine-tuned model. This is repeatedly executed to continuously update and optimize the model online.

[0046] In the working mode recognition module, the steps of selecting the best working mode most suitable for the current moment based on the logical gate threshold judgment include:

[0047] After the vehicle enters the drivable state (Ready), first select three working modes: land, water, and beach landing ship according to the mode switch pressed by the driver. There is an interlock relationship among the three working modes. Once entering a certain working mode, only after exiting this mode can it be switched to other modes. If two or more switches are turned on simultaneously, the mode selection is carried out in the order of priority. Only after the first switch is turned off can it enter another working mode. After determining that the vehicle is in a certain working mode, the corresponding working mode selection is carried out;

[0048] For example, Figure 3As shown in the figure, the onshore working modes include six types: hybrid driving mode 1 (mainly driven by the power battery to meet the instantaneous response ability of the vehicle), hybrid driving mode 2 (mainly driven by the engine, and the engine operates in the high-efficiency region), pure electric driving mode (driven by the motor alone to reduce the fuel consumption of the engine), power generation driving mode (the engine operates in the high-efficiency region, and the excess power is used to charge the power battery), energy recovery mode, and mechanical braking mode; when the required power is greater than 0, first judge whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value a; if the difference is greater than a, enter hybrid driving mode 1 to meet the driver's instantaneous high-power demand; if the difference is less than a, judge whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter hybrid driving mode 2 to make the engine operate in the high-efficiency region; if it is less than the average required power, judge whether the current state of charge of the battery is lower than the charging threshold. If it is lower than the charging threshold, enter the power generation driving mode, and the engine operates in the high-efficiency region to use the excess power to charge the power battery; if it is higher than the charging threshold, enter the pure electric driving mode to avoid the engine operating in the low-efficiency region to reduce fuel consumption. When the required power is less than 0, first judge whether the current state of charge of the battery is lower than the maximum state of charge threshold. If it is lower, enter the energy recovery mode; if it is higher, enter the conventional mechanical braking mode.

[0049] As Figure 4 shown in the figure, the offshore working modes include four types: hybrid sailing mode 1 (mainly driven by the power battery to meet the instantaneous response ability of the vehicle), hybrid sailing mode 2 (mainly driven by the engine, and the engine operates in the high-efficiency region), pure electric sailing mode (driven by the motor alone to reduce the fuel consumption of the engine), power generation sailing mode (the engine operates in the high-efficiency region, and the excess power is used to charge the power battery); when the required power is greater than 0, first judge whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value b; if the difference is greater than b, enter hybrid sailing mode 1 to meet the driver's instantaneous high-power demand; if the difference is less than b, judge whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter hybrid sailing mode 2 to make the engine operate in the high-efficiency region; if it is less than the average required power, judge whether the current state of charge of the battery is lower than the charging threshold. If it is lower than the charging threshold, enter the power generation sailing mode, and the engine operates in the high-efficiency region to use the excess power to charge the power battery; if it is higher than the charging threshold, enter the pure electric sailing mode to avoid the engine operating in the low-efficiency region to reduce fuel consumption.

[0050] As Figure 5As shown in the figure, the working modes of the up-and-down beach ship include three types: hybrid up-and-down beach mode 1 (mainly driven by the power battery to meet the instantaneous response ability of the vehicle), hybrid up-and-down beach mode 2 (mainly driven by the engine, and the engine works in the efficient area), and pure electric up-and-down beach mode (driven by the motor alone to reduce the fuel consumption of the engine); First, judge whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value c; if the difference is greater than c, enter the hybrid up-and-down beach mode 1 to meet the driver's instantaneous high-power demand; if the difference is less than c, judge whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter the hybrid up-and-down beach mode 2 to make the engine work in the efficient area; if it is less than the average required power, enter the pure electric up-and-down beach mode to avoid the engine running in the inefficient area to reduce fuel consumption;

[0051] The drive control management module first calculates the initial required torque of the vehicle according to the driver's input information (such as accelerator pedal, brake pedal, gear, etc.) and combines with the vehicle state information (such as vehicle speed, motor speed, spray pump speed, battery state of charge, etc.); Then, limit the initial required torque through the maximum allowable power, the highest allowable vehicle speed, and the maximum drive performance of the motor and the spray pump; Next, filter the drive torque to avoid excessive torque fluctuations and ensure the smoothness of vehicle driving; Finally, through the power-torque conversion formula, convert the current required torque into the current required power, input it into the working mode recognition module, assist in selecting the current best working mode, and reasonably distribute the required power to the engine and the power battery based on this working mode;

[0052] In the drive control management module, a torque coordination filtering adaptive function is designed, and an adaptive torque change factor is used for torque coordination filtering. On the premise of ensuring the acceleration performance, avoid drastic changes in torque and ensure the smoothness of vehicle driving; The torque change factor will be adaptively adjusted in real time according to the actual torque value and the required torque value, and the strategy is as follows:

[0053] First, determine an initial value of the torque change factor according to the torque interval where the current actual torque value is located. The division of the torque interval can adopt the conventional S-shaped distribution, and each interval corresponds to an initial value of the torque change factor. Secondly, determine a torque change factor coefficient through the adaptive function according to the difference between the actual torque value and the target torque value. Finally, multiply the initial value of the torque change factor by the torque change factor coefficient to obtain the final torque change factor, and use the torque change factor to adjust the actual torque value;

[0054] ;

[0055] Among them, x represents the torque change factor coefficient, and y represents the difference between the actual torque and the target torque. During actual use, the value of x at this moment, that is, the torque change factor coefficient, is determined by looking up the above adaptive function based on the torque difference y at this moment. It should be noted that the value of x ranges from 1 to 2.

[0056] The required torque after filtering can be converted into the current required power through the following formula:

[0057] ;

[0058] Among them, is the current required power, is the current required torque, is the rotational speeds of the drive motor, hub motor and jet pump, is the motor working efficiency;

[0059] The calculated current required power is input into the working mode recognition module to assist in identifying the current optimal working mode, and according to the power distribution scheme of this working mode, the required power is reasonably distributed to the engine and the power battery.

[0060] It should be noted that in this article, the term "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to this process, method, article or system. Without more limitations, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or system including this element.

[0061] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. An overall vehicle drive control system for an amphibious transport vehicle based on pattern recognition, characterized in that, The described drive control system includes: A working mode recognition module that predicts the required power for the next N moments by analyzing the collected vehicle historical data, calculates the average required power for these moments, and selects the most suitable best working mode for the current moment based on a logical gate threshold judgment by combining the current required power, the state of charge of the battery, and the vehicle mode switch state. Among them, the vehicle historical data includes historical vehicle speed, historical power demand, historical accelerator pedal opening, historical brake pedal switch signal, and historical working mode; A drive control management module that first calculates the initial required torque of the vehicle according to the driver's input information in combination with the vehicle state information; then limits the initial required torque through the maximum allowable power, the maximum allowable vehicle speed, and the maximum driving performance of the motor and the injection pump; then filters the driving torque to avoid excessive torque fluctuations and ensure the smoothness of vehicle driving; finally, converts the current required torque into the current required power through a power-torque conversion formula, inputs it into the working mode recognition module to assist in selecting the current best working mode, and rationally distributes the required power to the engine and the power battery based on this working mode; In the drive control management module, a torque coordination filtering adaptive function is designed, and an adaptive torque change factor is used for torque coordination filtering to avoid drastic torque changes on the premise of ensuring the acceleration performance and ensure the smoothness of vehicle driving; the torque change factor will be adaptively adjusted in real time according to the actual torque value and the required torque value, and the strategy is as follows: First, determine an initial value of the torque change factor according to the torque interval where the current actual torque value is located. The division of the torque interval can adopt a conventional S-shaped distribution, and each interval corresponds to an initial value of the torque change factor; secondly, determine a torque change factor coefficient through an adaptive function according to the difference between the actual torque value and the target torque value; finally, multiply the initial value of the torque change factor by the torque change factor coefficient to obtain the final torque change factor, and use the torque change factor to adjust the actual torque value; ； Among them, x represents the torque change factor coefficient, and y represents the difference between the actual torque and the target torque; in actual use, the x value, that is, the torque change factor coefficient, at this moment is determined by looking up the above adaptive function through the torque difference y at this moment; The filtered required torque can be converted into the current required power through the following formula: ； Among them, is the current required power, is the current required torque, is the rotational speeds of the drive motor, the in-wheel motor and the jet pump, is the motor working efficiency; Input the calculated current required power into the working mode recognition module to assist in identifying the current best working mode, and rationally distribute the required power to the engine and the power battery according to the power distribution plan of this working mode.

2. The whole vehicle drive control system of an amphibious transport vehicle based on pattern recognition according to claim 1, characterized in that, In the working mode recognition module, the present invention uses a feedforward neural network to predict the required power for the next N moments. The parameter update of the feedforward neural network includes two stages: offline optimization and online fine-tuning. First, the parameters of the feedforward neural network are initially learned using a standard driving cycle. After the learning is completed, the network model is installed in the controller, and the parameters of the feedforward neural network are updated through online fine-tuning by collecting the actual operation data of the amphibious vehicle, so that it can better meet the actual driving condition requirements of the amphibious vehicle. The online fine-tuning loss function is defined as follows: ； ； Among them, is the loss value at the -th moment, is the total loss value at moments, is the actual demand power value at the -th moment, is the predicted demand power value at the -th moment, is the number of data samples, is the update threshold used to prevent the model from changing violently and damaging the prediction ability of the model. The selected number of data samples , that is, whenever samples are collected during the actual operation of the amphibious transport vehicle, the parameters of the feedforward neural network are updated once, and the demand power at the next moment is predicted using the fine-tuned model. This process is repeated continuously to update and optimize the model online.

3. The vehicle drive control system of an amphibious transport vehicle based on pattern recognition according to claim 1, wherein In the working mode recognition module, the steps of selecting the best working mode most suitable for the current moment based on the logic gate threshold judgment include: After the vehicle enters the drivable state, first select three working modes: on-land, on-water, and beach-landing and beaching based on the mode switch pressed by the driver. There is an interlock relationship between the three working modes. Once entering a certain working mode, only after exiting this mode can it be switched to other modes. If two or more switches are turned on simultaneously, the mode selection is made in the order of precedence. Only after the first switch is turned off can it enter another working mode. After determining that the vehicle is in a certain working mode, the corresponding working mode selection is made. The on-land working mode includes six types: hybrid driving mode 1, hybrid driving mode 2, pure electric driving mode, power generation driving mode, energy recovery mode, and mechanical braking mode. When the required power is greater than 0, first judge whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value a. If the difference is greater than a, enter hybrid driving mode 1 to meet the driver's instantaneous high-power demand. If the difference is less than a, judge whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter hybrid driving mode 2 to make the engine work in the high-efficiency region. If it is less than the average required power, judge whether the current state of charge of the battery is lower than the charging threshold. If it is lower than the charging threshold, enter the power generation driving mode, and the engine works in the high-efficiency region to charge the power battery with the excess power. If it is higher than the charging threshold, enter the pure electric driving mode to avoid the engine running in the low-efficiency region to reduce fuel consumption. When the required power is less than 0, first judge whether the current state of charge of the battery is lower than the maximum state of charge threshold. If it is lower, enter the energy recovery mode. If it is higher, enter the conventional mechanical braking mode.

4. The vehicle drive control system of an amphibious vehicle based on pattern recognition according to claim 3, characterized in that The on-water working mode includes four types: hybrid navigation mode 1, hybrid navigation mode 2, pure electric navigation mode, and power generation navigation mode; When the required power is greater than 0, first determine whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value b; if the difference is greater than b, enter the hybrid navigation mode 1 to meet the driver's instantaneous high-power demand; if the difference is less than b, then determine whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter the hybrid navigation mode 2 to make the engine operate in the high-efficiency region; if it is less than the average required power, then determine whether the current state of charge of the battery is lower than the charging threshold. If it is lower than the charging threshold, enter the power generation navigation mode, and the engine operates in the high-efficiency region to charge the power battery with the excess power; if it is higher than the charging threshold, enter the pure electric navigation mode to avoid the engine operating in the low-efficiency region to reduce fuel consumption.

5. The overall driving control system of an amphibious transport vehicle based on pattern recognition according to claim 3, wherein, The working modes of the ship going up and down the shoal include three types: hybrid mode for going up and down the shoal 1, hybrid mode for going up and down the shoal 2, and pure electric mode for going up and down the shoal; first determine whether the difference between the required power at the current moment and the required power at the previous moment is greater than the calibrated value c; if the difference is greater than c, enter the hybrid mode for going up and down the shoal 1 to meet the driver's instantaneous high-power demand; if the difference is less than c, then determine whether the current required power is greater than the predicted average required power. If it is greater than the average required power, enter the hybrid mode for going up and down the shoal 2 to make the engine operate in the high-efficiency region; If it is less than the average required power, enter the pure electric mode for going up and down the shoal to avoid the engine operating in the low-efficiency region to reduce fuel consumption.

Images