A hybrid power distribution method for range-extended electric vehicles

Abstract

This invention discloses a hybrid power allocation method for range-extended electric vehicles (REEVs), relating to the field of intelligent power distribution technology. The method includes: determining the current external power demand status of the REEV; determining the current driving status of the REEV based on the engine operating status and real-time motor speed; acquiring the current state of charge of the battery and the fuel tank status of the range extender; determining the power supply mode of the REEV based on the external power demand status, driving status, battery state of charge, and fuel tank status; predicting the short-term power demand of the REEV based on its real-time speed and external power load, inputting the short-term power demand into a preset reinforcement learning model, and determining a hybrid power allocation ratio based on the power supply mode with the goal of minimizing energy consumption; and adjusting the range extender's power generation and the battery's output power through a hierarchical state machine based on the hybrid power allocation ratio.

Description

Technical Field

[0001] This invention relates to the field of intelligent power distribution technology, and in particular to a hybrid power distribution method for range-extended electric vehicles. Background Technology

[0002] With the booming development of the new energy industry, the number of distributed power sources is increasing daily. At the same time, frequent extreme weather events place higher demands on emergency power supply capabilities, requiring power generation equipment with rapid response and flexible deployment. Range-extended electric vehicles (REEVs), as a type of flexibly deployable power generation equipment, have significant application value in distributed generation and emergency power supply scenarios, and are highly compatible with the development direction of the State Grid. REEVs are special engineering vehicles that combine range-extended electric vehicle technology with mobile power supply functionality. They possess the long range and low emissions characteristics of range-extended electric vehicles, while also serving as mobile power stations to provide emergency power support to external equipment or vehicles. They are an ideal choice for emergency response, infrastructure, and environmental protection, and their technological maturity and scenario adaptability are driving them to become a key component of new energy infrastructure.

[0003] Existing range-extended electric vehicles have the following technical problems in terms of power supply:

[0004] 1. Range-extended vehicles need to meet the dual requirements of driving power and external power supply at the same time. However, there are differences in the power output characteristics of batteries and range extenders: batteries have a fast response but limited capacity, while range extenders have high power but slow dynamic response. Traditional rule control is difficult to adapt to speed fluctuations and sudden load changes, which can easily lead to problems such as reduced battery life and increased energy consumption.

[0005] 2. Range-extended electric vehicles require multiple mode switching, which can easily cause power oscillations, and battery protection may conflict with external power supply requirements;

[0006] 3. Range-extended electric vehicles require intelligent networking when coordinating power supply with the power grid and supplementing power supply. Summary of the Invention

[0007] To address the problems mentioned above, the present invention provides a hybrid power distribution method for range-extended electric vehicles, thereby solving the aforementioned problems.

[0008] A hybrid power distribution method for range-extended electric vehicles includes:

[0009] The application scenario and external power demand status of the current range-extended electric vehicle are determined based on the current sensors distributed across the external power supply load interface.

[0010] The current driving status of the range-extended power vehicle is determined based on the engine operating status and the real-time speed of the electric motor.

[0011] Obtain the current state of charge of the power supply battery and the state of the range extender fuel tank of the range extender vehicle;

[0012] The power supply mode of the range-extended electric vehicle is determined based on the external power demand status, the driving status, the state of charge of the power supply battery, and the state of the range extender fuel tank.

[0013] Based on the real-time vehicle speed and external power load of the range-extended electric vehicle, the short-term power demand of the range-extended electric vehicle is predicted. The short-term power demand is input into a preset reinforcement learning model, and the hybrid power allocation ratio is determined based on the power supply mode with the goal of minimizing energy consumption.

[0014] The range extender's power generation and the power supply battery's output power are adjusted by a hierarchical state machine based on the hybrid power allocation ratio.

[0015] Preferably, the current sensor based on the distribution of external power supply load interfaces determines the current application scenario and external power supply demand status of the range-extended electric vehicle, including:

[0016] Check the connection status and connected device type of each external power interface of the range-extended power vehicle;

[0017] The Hall current sensor mounted on the external power interface of the range-extended power vehicle monitors the load current, voltage, and current power data of the external power interface in the on state in real time.

[0018] The application scenario of the range-extended electric vehicle is determined based on the connection status of each external power interface and the load current, voltage and current power data of the external power interface. The application scenarios include: off-grid power generation, grid-connected capacity expansion, grid-connected power supply guarantee, microgrid construction and charging scenarios.

[0019] The current external power supply demand status is determined based on the application scenario and the type of connected device.

[0020] Preferably, determining the current driving status of the range-extended electric vehicle based on the engine operating status and the real-time speed of the electric motor includes:

[0021] The vehicle controller receives the CAN bus signal from the engine control unit, obtains the start-stop flag, and determines the engine start-stop status of the range-extended power vehicle. When the engine status is on, the engine operating power is obtained, and the engine start-stop status and the engine operating power are determined as the engine operating status.

[0022] The real-time speed of the electric motor of the range-extended electric vehicle is determined by the speed sensor mounted on the electric motor, and the driving speed change curve of the range-extended electric vehicle is determined based on the data curve of the real-time speed.

[0023] The operating power of the range extender is determined based on the engine operating status, and the total energy consumption power during current driving is determined based on the driving speed change curve of the range extender power vehicle.

[0024] The current driving status of the range-extended electric vehicle is determined based on the driving speed change curve, the operating power of the range extender, and the total energy consumption.

[0025] Preferably, obtaining the current state of charge of the power supply battery and the state of the range extender fuel tank of the range extender vehicle includes:

[0026] The battery management system collects data from multiple battery packs in real time, including the total battery pack voltage, individual cell voltage, charging and discharging current, and battery surface temperature.

[0027] The state of charge of the power supply battery is determined by a multi-parameter fusion correction method based on the data from the various battery packs.

[0028] Based on the pressure sensor built into the fuel tank and the fuel consumption model, the fuel quantity of the range extender and the current fuel consumption rate are determined, and the fuel quantity of the range extender and the current fuel consumption rate are determined as the fuel tank status of the range extender.

[0029] Preferably, determining the power supply mode of the range-extended electric vehicle based on the external power demand status, the driving status, the state of charge of the power supply battery, and the state of the range extender fuel tank includes:

[0030] The external power demand status, the driving status, the power supply battery charge status, and the range extender fuel tank status are synchronized using timestamp technology.

[0031] The current external power demand is determined based on the external power demand status, and the current internal power demand of the range-extended power vehicle is determined based on the driving status.

[0032] Obtain the state of charge of the power supply battery, determine the current percentage of the power supply battery, obtain the state of the range extender fuel tank, and determine the current fuel level in the range extender fuel tank;

[0033] Based on the external power demand, the internal power demand, the current battery charge percentage, and the current fuel level in the range extender's fuel tank, the corresponding power supply mode is matched according to a preset power supply mode allocation strategy table.

[0034] Preferably, the step of predicting the short-term power demand of the range-extended electric vehicle based on its real-time vehicle speed and external power load, inputting the short-term power demand into a preset reinforcement learning model, and determining the hybrid power allocation ratio based on the power supply mode with the goal of minimizing energy consumption includes:

[0035] The real-time speed of the range-extended electric vehicle is obtained, and a speed curve is generated based on the real-time speed within a preset time window.

[0036] Obtain the external power supply load of the range-extended power vehicle, and generate an external load curve based on the total power of the external power supply load within a preset time window;

[0037] Based on the vehicle speed curve and the external load curve, a short-term power demand sequence for the range-extended electric vehicle within a preset time threshold is generated using LSTM time-series prediction technology.

[0038] The short-term power demand sequence, current battery state of charge and fuel tank state of the range-extended electric vehicle are obtained, normalized, and a state vector is generated. The power supply mode is obtained and a power supply mode code is generated.

[0039] With minimum energy consumption as a constraint, the state vector and the power supply mode are encoded and input into a preset reinforcement learning model, and the power supply allocation ratio of the range extender and the power supply battery is output.

[0040] Preferably, the power supply modes include: pure electric mode, range-extended mode, hybrid power supply mode, emergency power supply mode, regenerative braking mode, and battery protection mode.

[0041] Preferably, the step of adjusting the range extender's power generation and the power supply battery's output power based on the power allocation ratio using a hierarchical state machine includes:

[0042] The power allocation ratio is input into the hierarchical state machine, and the hierarchical state machine is used to generate corresponding range extender control commands and power supply battery control commands.

[0043] The range extender control commands are parsed to determine the range extender start / stop status and range extender engine speed, and the range extender power is adjusted accordingly.

[0044] The power supply battery control command is parsed to determine the power supply battery start / stop status, output voltage, and output current, and the power supply battery power is adjusted accordingly.

[0045] Preferably, the method further includes obtaining information from a cloud platform regarding the intelligent networking of all range-extended electric vehicles within a preset area based on current external power supply demands.

[0046] The GPS system collects the vehicle locations of multiple range-extended electric vehicles within a preset area, and the battery charge status, fuel tank level, and available power of the range extender for each range-extended electric vehicle are obtained through the range-extended electric vehicle status monitoring system.

[0047] The vehicle locations of the multiple range-extended electric vehicles, the battery charge status of each range-extended electric vehicle, the fuel tank level, and the available power of the range extender are uploaded to the cloud platform. Range-extended electric vehicles with available power of the range extender greater than a preset value are identified as available devices, and a pool of available devices is constructed.

[0048] Obtain all power supply nodes and the total power demand of the current power supply within the preset area through the cloud platform;

[0049] Obtain the power supply parameters of each of the power supply nodes, obtain the total power demand, allocate power according to the power supply parameters of each node, and determine the ideal power supply of each node.

[0050] The current power supply of each node is determined based on the power supply parameters of each node, the ideal power supply of each node is obtained, and the external power supply requirement of each node is determined based on the difference between the current power supply and the ideal power supply.

[0051] The supplementary power demand of each node is uploaded to the cloud platform to build an external power demand pool;

[0052] The network type of the range-extended power supply vehicle is determined based on the external power demand pool.

[0053] Obtain the external power demand pool and the range-extended power vehicle network type, and allocate power supply nodes and design power supply parameters according to the vehicle location and available power of each available device in the available device pool;

[0054] The cloud platform is used to schedule the corresponding available devices according to the power supply nodes, and the current networking is designed according to the power supply parameters.

[0055] Preferably, the method further includes determining the status of each battery in the range-extended power supply vehicle's battery pack based on real-time parameters of the power supply battery, and performing battery protection on the abnormal battery when the battery status is determined to be abnormal:

[0056] The voltage of each battery in the power supply battery pack of the range-extended power vehicle is collected in real time by a single cell voltage sensor, the current of each battery is collected by a Hall current sensor, the instantaneous power and cumulative energy consumption are calculated, and temperature data is obtained in real time by a patch-type temperature sensor deployed on the battery surface and electrodes to generate a temperature gradient curve.

[0057] Multi-source data timestamps are synchronized via bus, and spatial interpolation is used to fill in the voltage, current, and temperature data acquired by the sensors.

[0058] The internal resistance of each battery is measured by measuring the voltage and current of each battery. The state of charge (SOH) of each battery is determined according to the rate of change of internal resistance within a preset time threshold. The state of charge of each battery is determined by using the Kalman filter method.

[0059] The temperature gradient curve, SOH, and state of charge of each battery are input into a preset SOC-SOH coupled model to perform state diagnosis on each battery in the power supply battery pack of the range-extended electric vehicle. When an anomaly is found, the anomaly type of each battery is determined.

[0060] The abnormality type of each battery is mapped to the three-dimensional battery model, the abnormal battery is located visually, and an alarm is sent to the intelligent control interface of the range-extended power vehicle.

[0061] Based on the anomaly type of each battery, match the corresponding emergency handling strategy from the preset emergency handling strategy library and execute it;

[0062] Real-time parameters of the abnormal battery are detected. Once the real-time parameters enter the normal threshold, the internal resistance of the abnormal battery is tested to determine the SOH of the abnormal battery.

[0063] The parameters of the abnormal battery are adjusted based on the SOH of the abnormal battery.

[0064] Through the above-mentioned technical means, the present invention achieves the following beneficial effects:

[0065] 1) Based on LSTM time series prediction of vehicle speed fluctuation and external load, a short-term power demand sequence is generated. Combined with real-time battery state of charge and fuel tank state normalization processing, a state vector input reinforcement learning model is formed. With the minimum energy consumption as the constraint, the power supply allocation ratio of the range extender and battery is optimized online. The hybrid electric vehicle ratio is dynamically optimized through the reinforcement learning model to achieve a balance between dual demands.

[0066] 2) Based on a multi-dimensional sensor network of voltage, current and temperature, the system diagnoses battery anomalies in real time. By using the SOC-SOH coupling model to fuse temperature gradient and electrochemical parameters, the system accurately identifies fault types and links the emergency strategy library to execute graded power reduction or forced disconnection, thereby eliminating the risk of voltage mismatch during mode switching, resolving the conflict between battery protection and power supply demand, and improving system robustness.

[0067] 3) Based on GPS positioning and vehicle status, a pool of available devices is constructed. An external power demand pool is generated by combining the difference between the ideal power and the actual power of the power supply node. The network type is dynamically determined by matching data from the two pools. Cross-regional elastic power supply is achieved through global optimization of the cloud platform and power line carrier self-organizing network.

[0068] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0069] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0070] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0071] Figure 1 A schematic diagram of a hybrid power distribution method for range-extended electric vehicles provided by the present invention;

[0072] Figure 2 This is another schematic diagram of a hybrid power distribution method for range-extended electric vehicles provided by the present invention.

[0073] Figure 3 Another schematic diagram of a hybrid power distribution method for range-extended electric vehicles provided by the present invention;

[0074] Figure 4 This invention provides a schematic diagram of the working process of a range-extended electric vehicle (REEV) using a hybrid power distribution method. Detailed Implementation

[0075] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure.

[0076] With the booming development of the new energy industry, the number of distributed power sources is increasing daily. At the same time, frequent extreme weather events place higher demands on emergency power supply capabilities, requiring power generation equipment with rapid response and flexible deployment. Range-extended electric vehicles (REEVs), as a type of flexibly deployable power generation equipment, have significant application value in distributed generation and emergency power supply scenarios, and are highly compatible with the development direction of the State Grid. REEVs are special engineering vehicles that combine range-extended electric vehicle technology with mobile power supply functionality. They possess the long range and low emissions characteristics of range-extended electric vehicles, while also serving as mobile power stations to provide emergency power support to external equipment or vehicles. They are an ideal choice for emergency response, infrastructure, and environmental protection, and their technological maturity and scenario adaptability are driving them to become a key component of new energy infrastructure.

[0077] Existing range-extended electric vehicles have the following technical problems in terms of power supply:

[0078] 1. Range-extended vehicles need to meet the dual requirements of driving power and external power supply at the same time. However, there are differences in the power output characteristics of batteries and range extenders: batteries have a fast response but limited capacity, while range extenders have high power but slow dynamic response. Traditional rule control is difficult to adapt to speed fluctuations and sudden load changes, which can easily lead to problems such as reduced battery life and increased energy consumption.

[0079] 2. Range-extended electric vehicles require multiple mode switching, which can easily cause power oscillations, and battery protection may conflict with external power supply requirements;

[0080] 3. Range-extended electric vehicles require intelligent networking when coordinating power supply with the power grid and supplementing power supply.

[0081] A hybrid power distribution method for range-extended electric vehicles, such as Figure 1 As shown, it includes:

[0082] Step S101: Determine the current application scenario and external power supply demand status of the range-extended electric vehicle based on the current sensors distributed across the external power supply load interfaces;

[0083] In some embodiments, this embodiment uses a distributed Hall current sensor network deployed at each external power supply interface of the range-extended power vehicle to collect key parameters such as load current, voltage and power factor in real time. By analyzing current fluctuation characteristics, such as abrupt change amplitude, steady-state accuracy, harmonic distortion rate and power direction, combined with interface type and equipment access status, it can accurately identify application scenarios such as off-grid power generation, grid-connected capacity expansion, and microgrid construction, and quantify the external power supply demand status to provide high-precision operating condition input for subsequent power allocation strategies.

[0084] Step S102: Determine the current driving status of the range-extended power vehicle based on the engine operating status and the real-time speed of the electric motor.

[0085] In some embodiments, this embodiment is based on the engine start-stop state of the range-extended electric vehicle, such as idling, high-efficiency power generation or forced shutdown, and the real-time speed of the electric motor is analyzed together. Combined with the current battery state of charge (SOC) threshold logic, the vehicle is dynamically determined to be in pure electric cruise, hybrid drive or other driving states, so as to provide real-time operating condition basis for the vehicle energy distribution strategy.

[0086] Step S103: Obtain the current state of charge of the power supply battery and the state of the range extender fuel tank of the range extender vehicle;

[0087] Step S104: Determine the power supply mode of the range-extended electric vehicle based on the external power demand status, the driving status, the charging status of the power supply battery, and the fuel tank status of the range extender;

[0088] In some embodiments, this embodiment dynamically selects the optimal power supply mode based on the real-time external power supply demand status of the range-extended electric vehicle, such as off-grid voltage stabilization or grid-connected capacity expansion, vehicle driving status, power battery state of charge threshold range, and remaining fuel in the range extender's fuel tank. For example, when the SOC is high and the driving load is low, the pure electric mode is activated to prioritize the consumption of battery energy; when the SOC is medium to low or the external power supply demand is high, the range-extending mode is switched to utilize the engine's high-efficiency range to generate electricity and simultaneously balance the fuel consumption in the fuel tank; under extreme load, the hybrid mode is activated to coordinate the dual energy output of the range extender and the battery, ensuring power supply continuity and avoiding the risk of battery over-discharge.

[0089] Step S105: Predict the short-term power demand of the range-extended electric vehicle based on the real-time vehicle speed and external power supply load, input the short-term power demand into a preset reinforcement learning model, and determine the hybrid power allocation ratio based on the power supply mode with the goal of minimizing energy consumption.

[0090] In some embodiments, this embodiment is based on the real-time speed fluctuation characteristics of the range-extended electric vehicle, such as acceleration, and the load current harmonic distortion rate of the external power supply interface. It uses a temporal convolutional network to predict the short-term power demand sequence within a preset time (e.g., 10s). The sequence is then input into a preset deep reinforcement learning model. With the current power supply mode as the state constraint, the model outputs the mixed allocation ratio of the engine target power and the battery output power to minimize the system's equivalent fuel consumption rate.

[0091] Step S106: Adjust the range extender's power generation and the power supply battery's output power based on the hybrid power allocation ratio using a hierarchical state machine.

[0092] In some embodiments, such as Figure 4 As shown, this embodiment uses a hierarchical management architecture with a layered state machine to parse the hybrid power allocation ratio output by the reinforcement learning model into range extender power generation commands and battery output power parameters. The top-level state machine sets the range extender start-stop logic and battery charging and discharging safety boundaries according to the vehicle's driving conditions. The bottom-level state machine dynamically fine-tunes the range extender speed range and battery output threshold based on the real-time power allocation ratio, thereby achieving smooth switching and steady-state maintenance of dual energy output.

[0093] The working principle of the above technical solution is as follows: First, determine the current application scenario of the range-extended electric vehicle and the external power supply demand status, driving status, battery charge status, and range extender fuel tank status; then, determine the power supply mode of the range-extended electric vehicle based on the external power supply demand status, driving status, battery charge status, and range extender fuel tank status; predict the short-term power demand of the range-extended electric vehicle based on the real-time vehicle speed and external power supply load, input the short-term power demand into a preset reinforcement learning model, and determine the hybrid power allocation ratio with the goal of minimizing energy consumption; finally, adjust the range extender power generation and battery output power based on the hybrid power allocation ratio through a hierarchical state machine.

[0094] The beneficial effects of the above technical solution are as follows: By sensing external load fluctuations in real time through current sensors and combining vehicle driving status and energy status, the power supply mode is accurately matched to avoid inefficient operation; a reinforcement learning model is adopted to dynamically optimize the power distribution ratio between the range extender and the battery with the goal of minimizing energy consumption, and simultaneously balance fuel consumption, battery loss and mode switching penalties; a hierarchical state machine executes power distribution based on priority, and combines short-term power prediction to pre-adjust the range extender speed to achieve fast response and ensure voltage stability and system robustness.

[0095] In one embodiment, such as Figure 2 As shown, the current sensor based on the distribution of external power supply load interfaces determines the current application scenario and external power supply demand status of the range-extended electric vehicle, including:

[0096] Step S201: Detect the connection status and connected device type of each external power interface of the range-extended power vehicle;

[0097] In some embodiments, this embodiment is based on multiple high-voltage external power interfaces configured in the range-extended power vehicle. By collecting electrical signals of auxiliary contacts of each interface in real time, such as voltage / current thresholds and bus communication protocols, and combining voltage comparator circuits and protocol decoding chips for collaborative analysis, the physical connection status of each interface and the type of connected equipment are accurately determined: power supply equipment such as grid charging piles, and power receiving equipment such as emergency rescue tools. This provides key input basis for multi-mode switching of the vehicle power management system.

[0098] Step S202: Monitor the load current, voltage, and current power data of the external power interface when it is in the ON state in real time using the Hall current sensor mounted on the external power interface of the range-extended power vehicle.

[0099] Step S203: Based on the connection status of each external power interface and the load current, voltage and current power data of the external power interface, determine the current application scenario of the range-extended power vehicle. The application scenarios include: off-grid power generation, grid-connected capacity expansion, grid-connected power supply, microgrid construction and charging scenarios.

[0100] In some embodiments, this embodiment uses the physical connection signals of each external power interface of the range-extended power vehicle and real-time load electrical parameters to accurately determine the current application scenario through multi-dimensional data fusion analysis. For example, when an external load is detected to be operating independently and without grid characteristic signals, the off-grid power generation mode is activated; if the load current phase is detected to be synchronized with the grid and the power demand exceeds the baseline, the grid-connected capacity expansion mode is switched; when the grid voltage and frequency are abnormal, the grid-connected power supply mode is automatically switched to provide voltage stabilization support; when multiple interface loads form a power balance closed loop, the microgrid construction function is activated; when the interface is connected to a charging pile and the battery SOC is lower than the threshold, the charging scenario is prioritized.

[0101] Step S204: Determine the current external power supply demand status based on the application scenario and the type of connected device.

[0102] In some embodiments, this embodiment, based on the application scenario currently identified by the range-extended power vehicle and the types of devices connected to each interface, analyzes the electrical characteristics of the interfaces through multi-source data fusion to determine the status parameters of external power supply demand in real time, such as voltage and current, thereby dynamically adapting the power supply quality and energy scheduling strategies under different operating conditions.

[0103] The beneficial effects of the above technical solution are as follows: By collecting dynamic data of current, voltage and power from external interfaces in real time through Hall current sensors, and combining it with equipment type identification, it can automatically distinguish complex scenarios such as off-grid power generation, grid-connected capacity expansion and microgrid construction, avoiding the risk of misjudgment of traditional single threshold strategies; Based on scenario and equipment type analysis, it can dynamically generate differentiated power demand states, providing high-precision input for subsequent power allocation and significantly improving the accuracy of hybrid power allocation methods.

[0104] In one embodiment, such as Figure 3 As shown, the current driving status of the range-extended electric vehicle is determined based on the engine operating status and the real-time speed of the electric motor, including:

[0105] Step S301: Receive the CAN bus signal from the engine control unit through the vehicle controller, obtain the start-stop flag bit, determine the engine start-stop status of the range-extended power vehicle, and when the engine status is on, obtain the engine operating power, and determine the engine start-stop status and the engine operating power as the engine operating status.

[0106] In some embodiments, this embodiment is based on the vehicle controller of the range-extended electric vehicle to analyze the start-stop flag signal transmitted by the engine control unit through the CAN bus in real time, and combine it with auxiliary parameters such as engine speed to determine the engine start-stop status, such as shutdown / idle / high-efficiency power generation; when the flag indicates the operating status, the real-time output power data of the engine is collected synchronously, and the start-stop status and operating power are merged to construct the engine operating status;

[0107] Step S302: Determine the real-time speed of the electric motor of the range-extended power vehicle using the speed sensor mounted on the electric motor, and determine the driving speed change curve of the range-extended power vehicle based on the data curve of the real-time speed.

[0108] In some embodiments, this embodiment uses a Hall effect speed sensor built into the drive motor of the range-extended electric vehicle to collect the rotor angular velocity signal of the motor in real time, and transmits it to the vehicle controller via the CAN bus; combined with the transmission ratio of the motor reducer and the wheel rolling radius parameters, the angular velocity sequence is mapped into linear vehicle speed data points to generate a continuous vehicle speed change curve, which dynamically reflects the driving conditions such as vehicle acceleration, constant speed or braking coasting.

[0109] Step S303: Determine the operating power of the range extender based on the engine operating status, and determine the total energy consumption power of the current driving based on the driving speed change curve of the range extender power vehicle;

[0110] In some embodiments, this embodiment dynamically matches the range extender's operating power based on the real-time operating status of the range extender's engine, including start / stop flag and output power. Based on the vehicle speed change curve, it integrates parameters such as drag coefficient to calculate the total energy consumption under the current driving conditions in real time, providing a collaborative control benchmark for range extender power distribution and battery charging and discharging strategies.

[0111] Step S304: Determine the current driving status of the range-extended power vehicle based on the driving speed change curve, the operating power of the range extender, and the total energy consumption.

[0112] In some embodiments, this embodiment uses the real-time speed change curve of the range-extended electric vehicle, dynamic characteristics such as acceleration, constant speed or braking coasting, range extender operating power and total energy consumption to accurately determine the current driving state through time-series feature analysis and power.

[0113] The beneficial effects of the above technical solution are as follows: by acquiring the engine start / stop flag and power data and the dynamic information of the electric motor speed sensor in real time through the vehicle controller, the driving speed change curve and total energy consumption power are analyzed in a coordinated manner to accurately determine the vehicle driving status and achieve the comprehensive optimization goal of improving energy efficiency, enhancing safety and increasing equipment reliability of the range-extended electric vehicle in complex scenarios.

[0114] In one embodiment, obtaining the current state of charge of the power supply battery and the state of the range extender fuel tank of the range-extended electric vehicle includes:

[0115] The battery management system collects data from multiple battery packs in real time, including the total battery pack voltage, individual cell voltage, charging and discharging current, and battery surface temperature.

[0116] In some embodiments, this embodiment collects multi-dimensional battery pack operating parameters in real time through a distributed sensor network, including the total battery pack voltage (reflecting the overall energy state), individual cell voltage (monitoring cell consistency through an isolation operational amplifier circuit), bidirectional charge and discharge current (capturing the energy flow direction and rate using a Hall sensor or shunt), and multi-node battery surface temperature (detecting abnormal heat distribution using a thermistor or digital temperature sensor), and transmits these data to the control unit via a high-speed data bus.

[0117] The state of charge of the power supply battery is determined by a multi-parameter fusion correction method based on the data from the various battery packs.

[0118] Based on the pressure sensor built into the fuel tank and the fuel consumption model, the fuel quantity of the range extender and the current fuel consumption rate are determined, and the fuel quantity of the range extender and the current fuel consumption rate are determined as the fuel tank status of the range extender.

[0119] In some embodiments, this embodiment uses a pressure sensor built into the fuel tank of the range-extended electric vehicle to monitor the fuel level in real time. Combined with the dynamic calculation of the remaining fuel volume in the fuel tank and the fuel consumption rate per unit time, the state of the range extender fuel tank is defined as the state of the range extender fuel tank, providing the core decision-making basis for the range extender start-stop strategy and power distribution.

[0120] The beneficial effects of the above technical solution are as follows: By fusing multi-source battery data and dynamically modeling fuel levels, the energy state monitoring accuracy and system collaborative control capability of the range-extended electric vehicle are significantly improved. Based on multi-dimensional parameters such as total battery voltage, individual cell voltage, charging and discharging current, and surface temperature collected in real time by the battery management system, the estimated state of charge is dynamically corrected using a multi-parameter fusion correction method, effectively overcoming the cumulative deviation problem of the traditional single ampere-hour integration method and providing a highly reliable input for energy distribution strategy. Simultaneously, the fuel tank pressure sensor senses fuel level changes in real time, and the current fuel consumption rate is accurately calculated by combining the fuel consumption model, constructing a fuel level decay trajectory and remaining range prediction model. This not only avoids the mechanical lag of traditional liquid level gauges but also provides transient response basis for the range extender start-stop logic and power output.

[0121] In one embodiment, determining the power supply mode of the range-extended electric vehicle based on the external power demand status, the driving status, the state of charge of the power supply battery, and the state of the range extender fuel tank includes:

[0122] The external power demand status, the driving status, the power supply battery charge status, and the range extender fuel tank status are synchronized using timestamp technology.

[0123] The current external power demand is determined based on the external power demand status, and the current internal power demand of the range-extended power vehicle is determined based on the driving status.

[0124] In some embodiments, this embodiment accurately calculates the current external power supply demand based on the real-time status of the external power supply interface of the range-extended power vehicle and the type of connected equipment, such as industrial equipment requiring voltage and frequency stabilization and charging piles requiring bidirectional energy interaction; at the same time, it dynamically analyzes the internal power supply demand based on the vehicle's driving status (such as a sharp increase in drive power during rapid acceleration and negative power output during regenerative braking), providing a power reference input for the vehicle's energy scheduling.

[0125] Obtain the state of charge of the power supply battery, determine the current percentage of the power supply battery, obtain the state of the range extender fuel tank, and determine the current fuel level in the range extender fuel tank;

[0126] In some embodiments, this embodiment uses multi-dimensional parameters such as total battery voltage, individual cell voltage, charging and discharging current, and node surface temperature collected by the battery management system of the range-extended electric vehicle to accurately calculate the state of charge (SOC) of the power supply battery through a multi-parameter fusion correction method and convert it into an intuitive current charge percentage; at the same time, based on the fuel static pressure data monitored in real time by the pressure sensor built into the fuel tank, and combined with the dynamic analysis of the remaining fuel volume in the range extender's fuel tank by the range extender's operating power, the accurate value of the current fuel quantity is output.

[0127] Based on the external power demand, the internal power demand, the current battery charge percentage, and the current fuel level in the range extender's fuel tank, the corresponding power supply mode is matched according to a preset power supply mode allocation strategy table.

[0128] In some embodiments, this embodiment dynamically matches the optimal power supply mode based on the real-time external power supply demand, internal power supply demand, battery charge percentage (SOC), and current fuel level of the range-extended electric vehicle through a preset power supply mode allocation strategy table (mapping relationship such as: high SOC + low power demand pure electric mode; low SOC + high external demand range-extended dominant mode). Under extreme conditions (such as dual energy shortage, high load output), a hybrid power supply strategy is triggered, and finally, a power supply mode command adapted to the current energy status and demand intensity is output.

[0129] The beneficial effects of the above technical solution are as follows: By using a multi-source state coordination and dynamic power matching mechanism with timestamp synchronization, the energy utilization efficiency and power supply reliability of the range-extended electric vehicle are significantly improved. The timestamp technology is used to achieve accurate synchronization of external power demand status, driving status, battery state of charge, and fuel tank level, eliminating the timing deviation caused by independent data collection from multiple systems in the traditional way, and providing a millisecond-level consistent data foundation for power supply mode decision-making. Based on the real-time analysis of external power demand and internal power demand during vehicle operation, combined with the battery SOC threshold and fuel tank level decay model, the preset power supply strategy table is dynamically matched, improving the intelligence of the power supply strategy.

[0130] In one embodiment, the step of predicting the short-term power demand of the range-extended electric vehicle based on its real-time vehicle speed and external power load, inputting the short-term power demand into a preset reinforcement learning model, and determining the hybrid power allocation ratio based on the power supply mode with the goal of minimizing energy consumption includes:

[0131] The real-time speed of the range-extended electric vehicle is obtained, and a speed curve is generated based on the real-time speed within a preset time window.

[0132] In some embodiments, the preset time window in this embodiment is 1 minute;

[0133] Obtain the external power supply load of the range-extended power vehicle, and generate an external load curve based on the total power of the external power supply load within a preset time window;

[0134] In some embodiments, the preset time window in this embodiment is 1 minute;

[0135] Based on the vehicle speed curve and the external load curve, a short-term power demand sequence for the range-extended electric vehicle within a preset time threshold is generated using LSTM time-series prediction technology.

[0136] In some embodiments, this embodiment constructs a multivariable input sequence based on the vehicle speed curve and external load curve collected in real time by the range-extended power vehicle using LSTM time-series prediction technology: by using the gating mechanism of LSTM, the relationship between the vehicle speed inertia change and the load step response is captured synchronously, and the short-term power demand sequence within a fixed time period, such as 30 seconds, is predicted in a rolling manner.

[0137] The short-term power demand sequence, current battery state of charge and fuel tank state of the range-extended electric vehicle are obtained, normalized, and a state vector is generated. The power supply mode is obtained and a power supply mode code is generated.

[0138] In some embodiments, this embodiment uses the short-term power demand sequence of the range-extended electric vehicle (predicted by an LSTM model), the current battery state of charge (SOC percentage), and the fuel tank status (remaining fuel and consumption rate) to scale the multi-source heterogeneous data to the [0,1] interval through a normalization algorithm to construct a state vector with unified dimensions. At the same time, according to the current power supply mode (e.g., pure electric mode is encoded as 01, range-extended mode as 10, and hybrid mode as 11), the discrete mode labels are converted into encoded vectors to form the standardized input feature matrix of the reinforcement learning model.

[0139] With minimum energy consumption as a constraint, the state vector and the power supply mode are encoded and input into a preset reinforcement learning model, and the power supply allocation ratio of the range extender and the power supply battery is output.

[0140] In some embodiments, this embodiment is based on the normalized state vector and power supply mode encoding of the range-extended electric vehicle, with the minimum equivalent fuel consumption rate of the system as the constraint objective. The input is a preset deep reinforcement learning model: the relationship between state features and mode encoding is analyzed, the policy gradient is iteratively optimized in the continuous action space, and the optimal action value is output, namely the target power ratio of the range extender and the output power ratio of the power supply battery.

[0141] The beneficial effects of the above technical solution are as follows: A short-term power demand sequence is generated by LSTM time-series prediction of real-time vehicle speed and external power supply load. This sequence is then combined with normalized vectors of battery state of charge and fuel tank state, along with power supply mode encoding, and input into a reinforcement learning model to dynamically optimize the power allocation ratio between the range extender and the power supply battery. This improves system energy efficiency and economy. Based on the collaborative prediction of vehicle speed curves and external load curves, transient demand fluctuations in driving and power supply scenarios are accurately captured, avoiding power response lag or redundancy caused by traditional static threshold control. The reinforcement learning model uses minimum energy consumption as a constraint for online optimization, adaptively adjusting the hybrid power ratio to ensure the range extender continuously operates within its high-efficiency range. The fusion of power supply mode encoding enhances operational adaptability, enabling rapid switching of power allocation strategies under sudden load changes or acceleration demands, eliminating power interruptions or sharp increases in energy consumption caused by mode switching deadlocks in traditional rule-based control.

[0142] In one embodiment, the power supply mode includes: pure electric mode, range-extended mode, hybrid power supply mode, emergency power supply mode, regenerative braking mode, and battery protection mode.

[0143] In some embodiments, the pure electric mode is used to supply power independently using the battery; the range-extended mode is used to supply power using the range extender and supplement the peak power supply using the battery; the hybrid power supply mode is used to supply power jointly using the battery and the range extender; the emergency power supply mode is used to supply power at full load using the range extender and the battery in emergency power supply scenarios; the regenerative braking mode is used to recover kinetic energy by reversing the motor in the braking scenario of the range-extended electric vehicle and charge the battery in reverse; and the battery protection mode is used to charge the battery using the range extender when the battery state of charge is lower than a preset threshold.

[0144] In one embodiment, adjusting the range extender's power generation and the battery's output power via a hierarchical state machine based on the power allocation ratio includes:

[0145] The power allocation ratio is input into the hierarchical state machine, and the hierarchical state machine is used to generate corresponding range extender control commands and power supply battery control commands.

[0146] In some embodiments, this embodiment uses a hierarchical processing architecture of a hierarchical state machine to parse the power allocation ratio output by the reinforcement learning model into executable instructions: the top-level state sets the range extender start-stop logic and battery charging and discharging safety boundary according to the driving conditions; the bottom-level state dynamically generates range extender control instructions and power supply battery control instructions based on the ratio value, and intercepts the oscillation risk caused by power mutation through state guard conditions, so as to achieve steady-state maintenance of dual energy coordinated output.

[0147] The range extender control commands are parsed to determine the range extender start / stop status and range extender engine speed, and the range extender power is adjusted accordingly.

[0148] In some embodiments, the range extender controller in this embodiment determines the start-stop state of the range extender by deconstructing the start-stop logic signal and target speed command output by the hierarchical state machine and combining it with the real-time engine speed feedback; at the same time, it dynamically adjusts the generator speed through the engine controller.

[0149] The power supply battery control command is parsed to determine the power supply battery start / stop status, output voltage, and output current, and the power supply battery power is adjusted accordingly.

[0150] In some embodiments, this embodiment decodes the energy parameter encoding in the battery management system instruction and dynamically determines the start-stop state based on the real-time state of charge (SOC) of the battery pack. The bidirectional DC / DC converter then uses voltage / current dual closed-loop control to adjust the switching frequency of the power devices by analyzing the parsed target voltage / current parameters.

[0151] The beneficial effects of the above technical solution are as follows: by utilizing the hierarchical structure of the hierarchical state machine, the power allocation ratio is converted into range extender start / stop commands and speed parameters, battery start / stop commands and voltage and current parameters, and redundant judgments are reduced through the state inheritance mechanism to ensure the real-time performance of command parsing and execution; at the same time, by predicting and intercepting system conflict risks through state transitions, the response delay or command oscillation problems caused by logic dispersion in traditional control are eliminated.

[0152] In one embodiment, the method further includes obtaining information from a cloud platform regarding the intelligent networking of all range-extended electric vehicles within a preset area based on current external power supply demand.

[0153] The GPS system collects the vehicle locations of multiple range-extended electric vehicles within a preset area, and the battery charge status, fuel tank level, and available power of the range extender for each range-extended electric vehicle are obtained through the range-extended electric vehicle status monitoring system.

[0154] The vehicle locations of the multiple range-extended electric vehicles, the battery charge status of each range-extended electric vehicle, the fuel tank level, and the available power of the range extender are uploaded to the cloud platform. Range-extended electric vehicles with available power of the range extender greater than a preset value are identified as available devices, and a pool of available devices is constructed.

[0155] In some embodiments, this embodiment uses dynamic parameters such as vehicle location (GPS coordinates), battery state of charge (SOC percentage), fuel tank level (remaining fuel capacity), and range extender available power (maximum output capacity under current operating conditions) collected in real time by the on-board terminal of the range extender vehicle to aggregate multi-source heterogeneous data to the cloud platform in real time through a wireless communication module; the cloud service dynamically filters the uploaded data according to a preset range extender power threshold (e.g., ≥50kW), marks the range extender vehicles that meet the conditions as available devices, and builds a pool of available devices;

[0156] Obtain all power supply nodes and the total power demand of the current power supply within the preset area through the cloud platform;

[0157] In some embodiments, this embodiment is based on a cloud platform-based regional energy dispatch system, which collects the topology of all power supply nodes in a preset area in real time through the power grid interface, synchronously obtains the real-time load data of each node, and dynamically calculates the total power demand for the current power supply by combining historical load curves.

[0158] Obtain the power supply parameters of each of the power supply nodes, obtain the total power demand, allocate power according to the power supply parameters of each node, and determine the ideal power supply of each node.

[0159] In some embodiments, this embodiment obtains the real-time operating parameters of all power supply nodes in a preset area based on a cloud platform and integrates the total demand power synchronously; by combining the equal incremental rate criterion with the upper limit of node power supply capacity and voltage stability margin constraints, the ideal power supply power of each node is dynamically calculated to achieve the global optimal allocation of power gap;

[0160] The current power supply of each node is determined based on the power supply parameters of each node, the ideal power supply of each node is obtained, and the external power supply requirement of each node is determined based on the difference between the current power supply and the ideal power supply.

[0161] In some embodiments, this embodiment dynamically calculates the current power supply of each node based on the real-time power supply node operating parameters collected by the cloud platform, combined with network loss correction and load forecast data, synchronously obtains the ideal power supply, accurately quantifies the external power supply demand of each node, and provides a dynamic target value for power gap compensation of the range-extended electric vehicle.

[0162] The supplementary power demand of each node is uploaded to the cloud platform to build an external power demand pool;

[0163] In some embodiments, this embodiment uses a regional energy dispatching system based on a cloud platform to collect the supplementary power demand (including node ID, power gap value and timestamp) of each power supply node in real time and construct a structured external power demand pool.

[0164] The network type of the range-extended power supply vehicle is determined based on the external power demand pool.

[0165] In some embodiments, this embodiment is based on the spatial distribution characteristics, time series fluctuation characteristics and power constraints (voltage / frequency compensation requirements) of the node power gap recorded in the external power demand pool, combined with the real-time location distribution and power output capability of the range-extended power vehicles in the available equipment pool, and dynamically determines the optimal network type through a topology optimization algorithm—including centralized star topology (single-point centralized power compensation), distributed mesh topology (multi-node collaborative power supply) or hybrid hierarchical topology.

[0166] Obtain the external power demand pool and the range-extended power vehicle network type, and allocate power supply nodes and design power supply parameters according to the vehicle location and available power of each available device in the available device pool;

[0167] In some embodiments, this embodiment is based on the external power demand pool and the range-extended power vehicle networking type, such as star / mesh topology, combined with the real-time location of each vehicle in the available equipment pool and the available power of the range extender, and dynamically allocates power supply nodes and designs power supply parameters through a spatial matching algorithm.

[0168] The cloud platform is used to schedule the corresponding available devices according to the power supply nodes, and the current networking is designed according to the power supply parameters.

[0169] In some embodiments, this embodiment uses a cloud-based intelligent scheduling system to dynamically schedule range-extended power vehicles in the available equipment pool based on the spatial distribution and power gap of power supply nodes, matching the real-time location of the vehicles with the available power of the range extenders, and designing a current networking scheme according to the node requirements.

[0170] The beneficial effects of the above technical solution are as follows: By dynamically integrating the real-time status of range-extended electric vehicles with regional power supply needs through a cloud platform, a globally optimized mobile microgrid network system is constructed. An available device pool is built based on GPS positioning and vehicle status monitoring systems. An external power supply demand pool is dynamically generated by combining the difference between the current power and ideal power of power supply nodes. The network type is accurately determined through dual-pool data matching. Relying on the global optimization algorithm of the cloud platform, power supply nodes are dynamically allocated and power supply parameters are designed according to the vehicle location distribution and the available power of the range extender. The optimal power supply topology is generated by scheduling available devices through current networking commands, enabling remote scheduling and grid connection of cross-regional electric vehicles in disaster environments.

[0171] In one embodiment, the method further includes determining the status of each battery in the range-extended electric vehicle's battery pack based on real-time parameters of the power supply battery, and performing battery protection on the abnormal battery when the battery status is determined to be abnormal:

[0172] The voltage of each battery in the power supply battery pack of the range-extended power vehicle is collected in real time by a single cell voltage sensor, the current of each battery is collected by a Hall current sensor, the instantaneous power and cumulative energy consumption are calculated, and temperature data is obtained in real time by a patch-type temperature sensor deployed on the battery surface and electrodes to generate a temperature gradient curve.

[0173] Multi-source data timestamps are synchronized via bus, and spatial interpolation is used to fill in the voltage, current, and temperature data acquired by the sensors.

[0174] In some embodiments, this embodiment uses bus synchronization technology to realize the timestamp alignment of multi-source sensors, such as voltage, current, and temperature acquisition units, to eliminate timing deviations caused by hardware clock drift or transmission delay. On this basis, for missing data points in the sensor network, spatial interpolation is used to reconstruct complete data by combining spatial correlation characteristics (such as voltage gradient and temperature field distribution of adjacent nodes).

[0175] The internal resistance of each battery is measured by measuring the voltage and current of each battery. The state of charge (SOH) of each battery is determined according to the rate of change of internal resistance within a preset time threshold. The state of charge of each battery is determined by using the Kalman filter method.

[0176] In some embodiments, this embodiment measures the battery internal resistance by DC discharge method based on the real-time collected voltage and current of each battery, combines a temperature compensation model to eliminate environmental interference, maps the battery aging curve according to the internal resistance change rate within a preset time threshold (5 days in this case), quantitatively evaluates the SOH of each battery, constructs a first-order equivalent circuit model of the battery using Kalman filtering, uses voltage and current as observation inputs, and estimates the SOC of each battery in real time through state equation iteration and covariance adaptive correction.

[0177] The temperature gradient curve, SOH, and state of charge of each battery are input into a preset SOC-SOH coupled model to perform state diagnosis on each battery in the power supply battery pack of the range-extended electric vehicle. When an anomaly is found, the anomaly type of each battery is determined.

[0178] In some embodiments, this embodiment synchronously inputs the temperature gradient curve, SOH, and state of charge (SOC) of each battery in the range-extended electric vehicle power supply battery pack into a preset SOC-SOH coupled model, and performs state diagnosis of the battery through multi-dimensional residual analysis; if an anomaly is detected, such as a sudden change in temperature gradient, accelerated decay of SOH, or a jump in SOC, the anomaly type of each battery is accurately located by combining the fault tree model.

[0179] The abnormality type of each battery is mapped to the three-dimensional battery model, the abnormal battery is located visually, and an alarm is sent to the intelligent control interface of the range-extended power vehicle.

[0180] In some embodiments, this embodiment uses the real-time diagnostic results of the range-extended electric vehicle battery management system to bind the abnormal type of each battery, such as thermal runaway precursor, internal short circuit or electrode aging, to the three-dimensional mesh model of the battery pack through a coordinate mapping algorithm, visualizes and locates the abnormal individual cells in the form of a heat map, and transmits alarm signals and heat maps to the intelligent control interface.

[0181] Based on the anomaly type of each battery, match the corresponding emergency handling strategy from the preset emergency handling strategy library and execute it;

[0182] In some embodiments, this embodiment is based on the abnormality types diagnosed in real time by the battery management system of the range-extended electric vehicle, such as thermal runaway precursors, internal short circuits, or electrode aging. It dynamically matches graded response measures through a preset emergency handling strategy library. For example, in response to thermal runaway risk, it immediately triggers full-power cooling of the liquid cooling system and cuts off the charging and discharging circuit; in response to internal short circuit faults, it isolates the corresponding module and limits the output power; in response to electrode aging abnormalities, it automatically activates the pulse repair mode and adjusts the charging current limiting threshold.

[0183] Real-time parameters of the abnormal battery are detected. Once the real-time parameters enter the normal threshold, the internal resistance of the abnormal battery is tested to determine the SOH of the abnormal battery.

[0184] In some embodiments, this embodiment is based on the real-time monitoring of the battery management system of the range-extended electric vehicle. When the real-time parameters of the abnormal battery (such as voltage mutation or abnormal temperature rise) return to the preset normal threshold range, the system automatically triggers the internal resistance test process.

[0185] The parameters of the abnormal battery are adjusted based on the SOH of the abnormal battery.

[0186] In some embodiments, this embodiment dynamically adjusts the charging and discharging parameters of abnormal batteries based on the SOH value diagnosed by the battery management system of the range-extended electric vehicle. For example, for batteries with low SOH values, the charging and discharging cutoff voltage threshold is reduced and the SOC working window is narrowed; for batteries with medium SOH values ​​but significantly increased internal resistance, the power limiting threshold is increased to avoid local overload; at the same time, the energy distribution weight of the battery pack is reconstructed according to the SOH distribution, and high SOH battery cells are prioritized to achieve flexible load control of aging batteries.

[0187] The beneficial effects of the above technical solution are as follows: It achieves refined safety management of the battery pack in range-extended electric vehicles through multi-source data fusion and SOC-SOH coupled diagnostics. Based on a multi-dimensional sensor network of voltage, current, and temperature, it collects individual cell-level data in real time, fills data gaps using spatial interpolation, and synchronizes timestamps to ensure the spatiotemporal consistency of monitoring data. It combines internal resistance change rate dynamic tracking of battery health (SOH) and uses a Kalman filter algorithm to correct SOC estimation errors in real time, effectively solving the problem of missed detection of early anomalies in traditional single-parameter monitoring. By fusing temperature gradient curves and electrochemical parameters through the SOC-SOH coupled model, it achieves accurate identification of anomaly types and simultaneously maps the anomaly type to the battery's three-dimensional model for spatial location visualization, significantly shortening fault location time. It relies on a preset emergency strategy library to dynamically match protection measures and performs internal resistance retesting to verify SOH decay after parameters return to normal.

[0188] Those skilled in the art should understand that the "first" and "second" in this invention simply refer to different application stages.

[0189] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0190] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for hybrid power distribution in range-extended electric vehicles, characterized in that, include: The application scenario and external power demand status of the current range-extended electric vehicle are determined based on the current sensors distributed across the external power supply load interface. The current driving status of the range-extended power vehicle is determined based on the engine operating status and the real-time speed of the electric motor. Obtain the current state of charge of the power supply battery and the state of the range extender fuel tank of the range extender vehicle; The power supply mode of the range-extended electric vehicle is determined based on the external power demand status, the driving status, the state of charge of the power supply battery, and the state of the range extender fuel tank. Based on the real-time vehicle speed and external power load of the range-extended electric vehicle, the short-term power demand of the range-extended electric vehicle is predicted. The short-term power demand is input into a preset reinforcement learning model, and the hybrid power allocation ratio is determined based on the power supply mode with the goal of minimizing energy consumption. The range extender's power generation and the power supply battery's output power are adjusted by a hierarchical state machine based on the hybrid power allocation ratio. The current sensor based on the distribution of external power supply load interfaces determines the current application scenario and external power supply demand status of the range-extended electric vehicle, including: Check the connection status and connected device type of each external power interface of the range-extended power vehicle; The Hall current sensor mounted on the external power interface of the range-extended power vehicle monitors the load current, voltage, and current power data of the external power interface in the on state in real time. The application scenario of the range-extended electric vehicle is determined based on the connection status of each external power interface and the load current, voltage and current power data of the external power interface. The application scenarios include: off-grid power generation, grid-connected capacity expansion, grid-connected power supply guarantee, microgrid construction and charging scenarios. Determine the current external power supply demand status based on the application scenario and the type of connected device; Determining the current driving status of the range-extended electric vehicle based on the engine operating status and the real-time speed of the electric motor includes: The vehicle controller receives the CAN bus signal from the engine control unit, obtains the start-stop flag, and determines the engine start-stop status of the range-extended power vehicle. When the engine status is on, the engine operating power is obtained, and the engine start-stop status and the engine operating power are determined as the engine operating status. The real-time speed of the electric motor of the range-extended electric vehicle is determined by the speed sensor mounted on the electric motor, and the driving speed change curve of the range-extended electric vehicle is determined based on the data curve of the real-time speed. The operating power of the range extender is determined based on the engine operating status, and the total energy consumption power during current driving is determined based on the driving speed change curve of the range extender power vehicle. The current driving status of the range-extended electric vehicle is determined based on the driving speed change curve, the operating power of the range extender, and the total energy consumption.

2. The hybrid power distribution method for range-extended electric vehicles according to claim 1, characterized in that, The process of obtaining the current state of charge of the power supply battery and the state of the range extender fuel tank of the range extender vehicle includes: The battery management system collects data from multiple battery packs in real time, including the total battery pack voltage, individual cell voltage, charging and discharging current, and battery surface temperature. The state of charge of the power supply battery is determined by a multi-parameter fusion correction method based on the data from the various battery packs. Based on the pressure sensor built into the fuel tank and the fuel consumption model, the fuel quantity of the range extender and the current fuel consumption rate are determined, and the fuel quantity of the range extender and the current fuel consumption rate are determined as the fuel tank status of the range extender.

3. The hybrid power distribution method for range-extended electric vehicles according to claim 1, characterized in that, The process of determining the power supply mode of the range-extended electric vehicle based on the external power demand status, the driving status, the state of charge of the power supply battery, and the state of the range extender fuel tank includes: The external power demand status, the driving status, the power supply battery charge status, and the range extender fuel tank status are synchronized using timestamp technology. The current external power demand is determined based on the external power demand status, and the current internal power demand of the range-extended power vehicle is determined based on the driving status. Obtain the state of charge of the power supply battery, determine the current percentage of the power supply battery, obtain the state of the range extender fuel tank, and determine the current fuel level in the range extender fuel tank; Based on the external power demand, the internal power demand, the current battery charge percentage, and the current fuel level in the range extender's fuel tank, the corresponding power supply mode is matched according to a preset power supply mode allocation strategy table.

4. The hybrid power distribution method for range-extended electric vehicles according to claim 1, characterized in that, The step of predicting the short-term power demand of the extended-range electric vehicle based on its real-time vehicle speed and external power load, inputting the short-term power demand into a preset reinforcement learning model, and determining the hybrid power allocation ratio based on the power supply mode with the goal of minimizing energy consumption includes: The real-time speed of the range-extended electric vehicle is obtained, and a speed curve is generated based on the real-time speed within a preset time window. Obtain the external power supply load of the range-extended power vehicle, and generate an external load curve based on the total power of the external power supply load within a preset time window; Based on the vehicle speed curve and the external load curve, a short-term power demand sequence for the range-extended electric vehicle within a preset time threshold is generated using LSTM time-series prediction technology. The short-term power demand sequence, current battery state of charge and fuel tank state of the range-extended electric vehicle are obtained, normalized, and a state vector is generated. The power supply mode is obtained and a power supply mode code is generated. With minimum energy consumption as a constraint, the state vector and the power supply mode are encoded and input into a preset reinforcement learning model, and the power supply allocation ratio of the range extender and the power supply battery is output.

5. The hybrid power distribution method for range-extended electric vehicles according to claim 1, characterized in that, The power supply modes include: pure electric mode, range-extended mode, hybrid power supply mode, emergency power supply mode, regenerative braking mode, and battery protection mode.

6. The hybrid power distribution method for range-extended electric vehicles according to claim 1, characterized in that, The adjustment of the range extender's power generation and the battery's output power via a hierarchical state machine based on the power allocation ratio includes: The power allocation ratio is input into the hierarchical state machine, and the hierarchical state machine is used to generate corresponding range extender control commands and power supply battery control commands. The range extender control commands are parsed to determine the range extender start / stop status and range extender engine speed, and the range extender power is adjusted accordingly. The power supply battery control command is parsed to determine the power supply battery start / stop status, output voltage, and output current, and the power supply battery power is adjusted accordingly.

7. The hybrid power distribution method for range-extended electric vehicles according to claim 1, characterized in that, The method also includes using a cloud platform to obtain information on all range-extended electric vehicles within a preset area and intelligently network them with the power grid based on current external power demand. The GPS system collects the vehicle locations of multiple range-extended electric vehicles within a preset area, and the battery charge status, fuel tank level, and available power of the range extender for each range-extended electric vehicle are obtained through the range-extended electric vehicle status monitoring system. The vehicle locations of the multiple range-extended electric vehicles, the battery charge status of each range-extended electric vehicle, the fuel tank level, and the available power of the range extender are uploaded to the cloud platform. Range-extended electric vehicles with available power of the range extender greater than a preset value are identified as available devices, and a pool of available devices is constructed. Obtain all power supply nodes and the total power demand of the current power supply within the preset area through the cloud platform; Obtain the power supply parameters of each of the power supply nodes, obtain the total power demand, allocate power according to the power supply parameters of each node, and determine the ideal power supply of each node. The current power supply of each node is determined based on the power supply parameters of each node, the ideal power supply of each node is obtained, and the external power supply requirement of each node is determined based on the difference between the current power supply and the ideal power supply. The supplementary power demand of each node is uploaded to the cloud platform to build an external power demand pool; The network type of the range-extended power supply vehicle is determined based on the external power demand pool. Obtain the external power demand pool and the range-extended power vehicle network type, and allocate power supply nodes and design power supply parameters according to the vehicle location and available power of each available device in the available device pool; The cloud platform is used to schedule the corresponding available devices according to the power supply nodes, and the current networking is designed according to the power supply parameters.

8. The hybrid power distribution method for range-extended electric vehicles according to claim 1, characterized in that, The method further includes determining the status of each battery in the range-extended power supply vehicle's battery pack based on real-time parameters of the power supply battery, and performing battery protection for the abnormal battery when the battery status is determined to be abnormal: The voltage of each battery in the power supply battery pack of the range-extended power vehicle is collected in real time by a single cell voltage sensor, the current of each battery is collected by a Hall current sensor, the instantaneous power and cumulative energy consumption are calculated, and temperature data is obtained in real time by a patch-type temperature sensor deployed on the battery surface and electrodes to generate a temperature gradient curve. Multi-source data timestamps are synchronized via bus, and spatial interpolation is used to fill in the voltage, current, and temperature data acquired by the sensors. The internal resistance of each battery is measured by measuring the voltage and current of each battery. The state of charge (SOH) of each battery is determined according to the rate of change of internal resistance within a preset time threshold. The state of charge of each battery is determined by using the Kalman filter method. The temperature gradient curve, SOH, and state of charge of each battery are input into a preset SOC-SOH coupled model to perform state diagnosis on each battery in the power supply battery pack of the range-extended electric vehicle. When an anomaly is found, the anomaly type of each battery is determined. The abnormality type of each battery is mapped to the three-dimensional battery model, the abnormal battery is located visually, and an alarm is sent to the intelligent control interface of the range-extended power vehicle. Based on the anomaly type of each battery, match the corresponding emergency handling strategy from the preset emergency handling strategy library and execute it; Real-time parameters of the abnormal battery are detected. Once the real-time parameters enter the normal threshold, the internal resistance of the abnormal battery is tested to determine the SOH of the abnormal battery. The parameters of the abnormal battery are adjusted based on the SOH of the abnormal battery.

Images