A vehicle energy planning method and device, electronic equipment and storage medium

Abstract

The application provides a vehicle energy planning method and device, electronic equipment and storage medium, the method comprises the following steps: obtaining multi-modal travel data of a vehicle in driving; determining a target energy demand curve according to a preset driving resistance relationship and the multi-modal travel data; determining at least one user travel scenario based on the multi-modal travel data; and performing parameter item adjustment reasoning on a preset energy cost relationship according to a preset language model and the at least one user travel scenario to obtain at least one target energy cost relationship, wherein the parameter item adjustment comprises at least one of cost item adjustment and cost weight adjustment, the cost item adjustment comprises at least one of cost item increase / decrease and internal parameter adjustment of the cost item, and global energy planning is performed according to the target energy demand curve and the at least one target energy cost relationship to obtain a current energy planning strategy; and the application adjusts the parameter items of the preset energy cost relationship, takes into account the multi-dimensional requirements of users, and improves the accuracy of vehicle energy management planning.

Description

Technical Field

[0001] This invention relates to the field of vehicle energy planning technology, and in particular to a vehicle energy planning method, apparatus, electronic device and storage medium. Background Technology

[0002] As the global automotive industry transitions to electrification, the market share of battery electric vehicles (BEVs) has increased significantly. To address the range anxiety of some existing BEV owners, a flexible, plug-and-play aftermarket range extender has emerged, providing users with an option to increase range without replacing their vehicles. To improve the efficiency and intelligence of aftermarket range extenders, this paper proposes an energy management scheme based on navigation information, drawing on the predictive energy management approach of original equipment manufacturers (OEMs) and incorporating cloud computing technology. The energy management solution includes an aftermarket range extender, an onboard gateway, and a cloud server. The onboard gateway sends the planned complete route data and vehicle status data to the cloud server. After receiving the vehicle status data, the cloud server performs segmented or point-by-point predictions of the vehicle's driving resistance and power demand on the future route. Guided by the optimization objective of minimizing total equivalent energy consumption cost, it performs calculations to generate an optimal range extender power output sequence or battery state of charge (SOC) curve covering the entire route or a future distance. This yields a control strategy, which is then sent to the onboard gateway. The onboard gateway then manages the energy of the aftermarket range extender according to the control strategy.

[0003] The above energy management solutions still have the following shortcomings: there is a huge discrepancy between the predicted value of the vehicle's future energy consumption and the actual energy consumption value, resulting in a mismatch between the working time and power output of the aftermarket range extender and the actual demand, making it impossible to achieve accurate energy management; there is a lack of global planning capabilities throughout the entire journey, resulting in the overall energy consumption not being optimal under long-distance and complex road conditions; and the optimization goal of minimizing economical energy consumption costs ignores the actual needs of users to optimize energy management in multiple dimensions under different driving situations. Summary of the Invention

[0004] This invention provides a vehicle energy planning method, apparatus, electronic device, and storage medium to solve the aforementioned technical problems of low accuracy and failure to meet user driving needs in energy management planning.

[0005] This invention provides a vehicle energy planning method, the method comprising: acquiring multimodal travel data of a vehicle in motion; determining a target energy demand curve based on a preset driving resistance relationship and the multimodal travel data; determining at least one user travel scenario based on the multimodal travel data, and performing parameter adjustment reasoning on a preset energy cost relationship based on a preset language model and at least one of the user travel scenarios to obtain at least one target energy cost relationship, wherein the parameter adjustment includes at least one of cost item adjustment and cost weight adjustment, and the cost item adjustment includes at least one of cost item increase / decrease and internal parameter adjustment of the cost item; performing global energy planning based on the target energy demand curve and at least one of the target energy cost relationship to obtain a current energy planning strategy, wherein the current energy planning strategy includes at least one of an energy management strategy and a thermal management strategy for controlling the operation of a thermal management system, and the energy management strategy includes at least one of a range extender operation strategy for controlling range extender power generation and an external charging strategy for controlling vehicle battery charging on an external power grid.

[0006] In one embodiment of the present invention, determining a target energy demand curve based on a preset driving resistance relationship and the multimodal travel data includes: determining reference driving resistance for multiple driving path segments based on path planning data and the preset driving resistance relationship, and determining a reference energy demand curve based on a preset motor efficiency operation mapping relationship and each of the reference driving resistances; inputting external environment data into a large language model for energy demand calibration inference to obtain energy demand correction information, wherein the energy demand correction information includes at least one of additional load and resistance correction coefficient; correcting the reference energy demand curve based on the energy demand correction information and the corresponding reference driving resistance to obtain the target energy demand curve; wherein the multimodal travel data includes the path planning data and the external environment data, and the preset language model includes the large language model running in the cloud.

[0007] In one embodiment of the present invention, after determining the target energy demand curve based on the preset driving resistance relationship and the multimodal travel data, the method further includes: determining at least one target segment based on the target energy demand curve, wherein the target segment is used to characterize a physical key segment; inputting the geographical location information of at least one target segment into the preset language model for event reasoning to obtain corresponding event attribute tags; wherein the event attribute tags are used to obtain the user travel scenario, and the geographical location information is obtained based on the multimodal travel data.

[0008] In one embodiment of the present invention, global energy planning is performed based on the target energy demand curve and at least one target energy cost relationship to obtain a current energy planning strategy. This includes: determining the stage state and decision action for each travel stage, and calculating the stage cost after executing each decision action in different stage states for the same travel stage; using dynamic programming to perform reverse optimization on the stage state, decision action, and stage cost corresponding to each travel stage, with minimizing the total travel cost as the optimization objective; selecting one decision action from multiple decision actions corresponding to each travel stage to obtain the current energy planning strategy from the starting point to the destination of the current travel route; wherein, the stage state of each travel stage... The state is obtained based on the discretization of the vehicle's current travel distance and / or the execution of corresponding decision actions under the energy demand constraints of the corresponding stage. The stage state includes at least one of battery state of charge and battery temperature. Each decision action includes at least one of energy management decision and thermal management decision. The energy management decision includes at least one of range extender operation decision and external charging decision. The stage energy demand is obtained based on the target energy demand curve and the travel distance corresponding to the travel stage. The stage cost is determined based on the target energy cost relationship corresponding to the travel stage. The total travel cost is obtained based on the sum of the stage costs corresponding to the selected decision actions executed in each travel stage.

[0009] In one embodiment of the present invention, after obtaining the current energy planning strategy, the method further includes: acquiring the network status of the vehicle and new multimodal travel data, wherein the new multimodal travel data is obtained based on a new travel route; determining the cloud or vehicle as the energy planning endpoint based on the network status; if the energy planning endpoint is the vehicle endpoint, and the lowest route matching degree between the new travel route and at least one historical travel route is greater than or equal to a preset matching threshold, then determining the historical energy planning strategy corresponding to the highest route matching degree as the travel energy planning strategy for the new travel route; if the energy planning endpoint is the vehicle endpoint, and the lowest route matching degree between the new travel route and at least one historical travel route is less than a preset matching threshold, then performing global energy planning based on a local language model and the new multimodal travel data to obtain the travel energy planning strategy for the new travel route; wherein the preset language model includes the local language model built into the vehicle, and the local language model is used to infer the target energy demand curve and at least one target energy cost relationship corresponding to the new travel route.

[0010] In one embodiment of the present invention, after obtaining the current energy planning strategy, the method further includes: smoothing the initial power curve in the range extender's operating strategy to obtain a target power curve; dividing the vehicle's current travel distance into multiple task segments; sampling the target power curve according to a preset first distance interval to obtain a power command mapping relationship corresponding to each task segment, wherein the power command mapping relationship is used to characterize the mapping relationship between distance and target power within a task segment; and determining energy consumption reference strategy information for each task segment based on each power command mapping relationship and the geofence of each task segment; wherein the geofence is obtained based on the multimodal travel data.

[0011] In one embodiment of the present invention, after determining the energy consumption reference strategy information for each task segment, the method further includes: obtaining the vehicle's current location information and the distance traveled; determining the task segment to which the current location information belongs as the target segment based on the geofence, and determining the distance traveled by the vehicle on the target segment based on the distance traveled and the current location information; determining target data pairs on both sides of the distance traveled on the target segment based on the power command mapping relationship of the target segment, wherein the target data pairs include the target power and the distance within the task segment; interpolating the target data pairs on both sides based on the distance traveled on the target segment to obtain the power to be controlled by the vehicle, and sending it to the range extender controller to control the range extender to generate electricity.

[0012] This invention provides a vehicle energy planning device, comprising: a data acquisition module for acquiring multimodal travel data of a vehicle during driving; an energy demand determination module for determining a target energy demand curve based on a preset driving resistance relationship and the multimodal travel data; a cost determination module for determining at least one user travel scenario based on the multimodal travel data, and performing parameter adjustment reasoning on a preset energy cost relationship based on a preset language model and at least one of the user travel scenarios to obtain at least one target energy cost relationship, wherein the parameter adjustment includes at least one of cost adjustment and cost weight adjustment, and the cost adjustment includes at least one of cost increase / decrease and internal parameter adjustment of the cost item; and an energy planning module for performing global energy planning based on the target energy demand curve and at least one of the target energy cost relationship to obtain a current energy planning strategy, wherein the current energy planning strategy includes at least one of an energy management strategy and a thermal management strategy for controlling the operation of a thermal management system, and the energy management strategy includes at least one of a range extender operation strategy for controlling the generation of the range extender and an external charging strategy for controlling the charging of the vehicle battery on the external power grid.

[0013] The present invention provides an electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device enables the vehicle energy planning method as described in any of the above embodiments.

[0014] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a computer processor, causes the computer to perform the vehicle energy planning method described in any of the above embodiments.

[0015] The beneficial effects of the present invention are as follows: The vehicle energy planning method, device, electronic device and storage medium proposed in this invention determine the user's travel scenario through multimodal travel data, and adjust the parameters of the preset energy cost relationship based on the user's travel scenario through a preset language model. In this way, while ensuring high economic efficiency, it can also take into account the user's multi-dimensional needs, provide a more humanized and intelligent driving experience, and thus improve the accuracy of vehicle energy management planning. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0017] In the attached diagram:

[0018] Figure 1 A schematic diagram of an exemplary system architecture provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating a vehicle energy planning method provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of decision optimization based on dynamic programming provided in one embodiment of the present invention; Figure 4 This is a flowchart illustrating global energy planning based on dynamically configured target energy cost relationships, provided in one embodiment of the present invention. Figure 5 This is a flowchart illustrating the vehicle-side strategy execution phase provided in one embodiment of the present invention; Figure 6 This is a block diagram of a vehicle energy planning device provided in one embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of a computer system for an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0019] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0020] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. The drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0021] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0022] Please see Figure 1 , Figure 1 This is a schematic diagram of an exemplary system architecture provided in an embodiment of the present invention. Figure 1 As shown, the system architecture can include a cloud platform 1100 and a vehicle 1200. The vehicle 1200 acquires multimodal travel data and uploads it to the cloud platform 1100 for global energy planning.

[0023] For example, the system acquires multimodal travel data of vehicle 1200 while it is in motion; determines a target energy demand curve based on a preset driving resistance relationship and the multimodal travel data; determines at least one user travel scenario based on the multimodal travel data, and performs parameter adjustment reasoning on a preset energy cost relationship based on a preset language model and at least one user travel scenario to obtain at least one target energy cost relationship. The parameter adjustment includes at least one of cost item adjustment and cost weight adjustment. The cost item adjustment includes at least one of cost item increase / decrease and internal parameter adjustment of the cost item. The system performs global energy planning based on the target energy demand curve and at least one target energy cost relationship to obtain a current energy planning strategy. The current energy planning strategy includes at least one of an energy management strategy and a thermal management strategy for controlling the operation of the thermal management system. The energy management strategy includes at least one of a range extender operation strategy for controlling the generation of the range extender and an external charging strategy for controlling the charging of the vehicle battery on the external power grid.

[0024] Among the related technologies, there are technical problems such as low accuracy in energy management planning and failure to meet the needs of users' driving.

[0025] To address the aforementioned technical problems, this invention provides a vehicle energy planning method, apparatus, electronic device, and storage medium. The implementation details of the technical solutions of the embodiments of this invention are described in detail below.

[0026] Please see Figure 2 , Figure 2 This is a flowchart illustrating a vehicle energy planning method provided in one embodiment of the present invention. Figure 2 As shown, in an exemplary embodiment, the vehicle energy planning method includes at least steps S210 to S240, which are described in detail below: Step S210: Obtain multimodal travel data of the vehicle.

[0027] In one embodiment of the present invention, the vehicle type of the present invention includes at least one of a car with a factory-integrated range extender, a car with a pluggable aftermarket range extender, a plug-in hybrid electric vehicle, and a pure electric vehicle.

[0028] In one embodiment of the present invention, please continue reading. Figure 1 ,like Figure 1As shown, the cloud 1100 includes an energy planning engine 1110, and the vehicle 1200 includes a task gateway 1210 and a target controller 1220. The energy planning engine 1110 includes a data fusion and modeling module 1111, a large model interface module 1112, an event recognition module 1113, a dynamic target configuration module 1114, an optimization solution module 1115, and a strategy generation and translation module 1116. The task gateway 1210 includes a data acquisition and reporting module 1211, a strategy caching and management module 1212, a task scheduling and execution module 1213, and a human-computer interaction interface module 1214. The data fusion and modeling module 1111 is responsible for receiving and processing multimodal travel data, and performing energy consumption model calculations to obtain the target energy demand curve; the large model interface module 1112 is used to construct inference requests and interact with the artificial intelligence services of the large model; the event recognition module 1113 is used for physical event-based recognition to obtain target segments, and calls the large model for semantic enrichment to obtain event attribute labels; the dynamic target configuration module 1114 is used to call the large model to generate dynamic target energy cost relationships; the optimization solution module 1115 is used for global energy planning through the execution of optimization algorithms; and the policy... The generation and translation module 1116 is used to convert the current energy planning strategy into an executable instruction format to obtain the target energy planning strategy; the data acquisition and reporting module 1211 is used to collect multimodal travel data and upload it to the cloud; the strategy caching and management module 1212 is used to store, manage and reuse historical energy planning strategies; the task scheduling and execution module 1213 is used to parse the target energy planning strategy at high frequency and in real time and issue hardware instructions to the corresponding target controller, such as the range extender controller; the human-machine interface module 1214 is used to display the humanized explanation text corresponding to the target energy planning strategy on the vehicle screen.

[0029] In one embodiment of the present invention, the cloud receives energy planning requests from vehicles via the Internet through security protocols such as Hypertext Transfer Protocol Secure (HTTPS), and returns the target energy planning strategy in the form of JSON (a data exchange format) data packets.

[0030] In one embodiment of the present invention, the task gateway is located at the vehicle terminal or the edge terminal. The vehicle terminal can be the vehicle's native infotainment system (IVI), and the edge terminal includes an aftermarket intelligent network management system (T-Box) connected to the vehicle via an on-board diagnostic system (OBD) and a controller area network (CAN) bus. The vehicle terminal or the edge terminal reads data from the vehicle's underlying electronic control unit (Electronic Control Unit), such as the battery management system (BMS) or the vehicle control unit (VCU), through the vehicle bus interface, such as an in-vehicle Ethernet or a flexible data rate controller area network (CAN FD), and issues hardware instructions corresponding to the target energy planning strategy to the target controller through the vehicle bus interface.

[0031] In one embodiment of the present invention, the range extender controller is a highly reliable and safe embedded controller. As the final execution unit, the range extender controller is only responsible for receiving simple and explicit hardware instructions from the vehicle terminal or edge terminal, such as a CAN message containing the target power of the range extender, so as to accurately control the hardware parameters of the range extender, such as the throttle opening of the internal combustion engine and the excitation current of the generator, so as to stably achieve the target power, and report the working status of the range extender itself to the upper level through heartbeat packets or status messages, forming a high-frequency control closed loop.

[0032] In one embodiment of the present invention, the preset language model includes a large language model running in the cloud. The large language model is referred to as the large model in this invention.

[0033] This invention can decouple complex cognitive computing from real-time reliable control by using the functional boundaries and information flow of a three-layer collaborative architecture, with the cloud as the cognitive brain, the vehicle terminal or edge terminal as the intelligent gateway, and the target controller as the execution unit.

[0034] In one embodiment of the present invention, the data acquisition and reporting module is triggered by a navigation event initiated by the user, and it acquires and constructs a JSON data packet including multimodal travel data.

[0035] In one embodiment of the present invention, multimodal travel data includes vehicle static data, vehicle dynamic data, route planning data, and external environment data.

[0036] In one embodiment of the present invention, multimodal travel data includes vehicle static data, vehicle dynamic data, and route planning data.

[0037] In one embodiment of the present invention, the energy planning terminal is determined according to the network status; if the energy planning terminal is the cloud, the multimodal travel data is uploaded to the cloud; if the energy planning terminal is the vehicle terminal, global energy planning is performed on the vehicle terminal based on the multimodal travel data.

[0038] In one embodiment of the present invention, determining the energy planning endpoint based on the network status includes: if the network status is no network or weak network, and data interaction with the external network is impossible or difficult, then the vehicle endpoint is determined as the energy planning endpoint; if the network status is not no network or weak network, then the cloud is determined as the energy planning endpoint, thereby performing global energy planning on the cloud based on the multimodal data uploaded by the vehicle endpoint.

[0039] In one embodiment of the present invention, the vehicle static data includes vehicle identification number (ID); battery rated capacity, which may be in kilowatt-hours (kWh); motor rated power / torque, which may be in kilowatts / newton-meters (kW / Nm); range extender rated power, which may be in kilowatts (kW); curb weight, which may be in kilograms (kg); frontal area, which may be in square meters (m²); and drag coefficient.

[0040] In one embodiment of the present invention, the vehicle dynamic data includes: the state of charge (SOC) of the battery at the time of departure (in terms of %), the average temperature of the battery pack (in degrees Celsius (°C)); the total mileage of the vehicle (in terms of km); the current tire pressure (in bar); the status of the air conditioning system; and the set temperature.

[0041] In one embodiment of the present invention, the route planning data includes a high-precision path point sequence provided by vehicle navigation. Each path point is a data object, including longitude; latitude; altitude (in meters (m); road type, such as highway, national highway, city, etc.); road gradient (in percentage); road curvature radius (in meters (m)); and legal speed limit (in kilometers per hour (km / h).

[0042] In one embodiment of the present invention, external environmental data is used to characterize external weather data, specifically including: departure point name, destination name, current timestamp, and detailed weather forecast along the route. The detailed weather forecast along the route is obtained through a third-party weather interface and specifically includes: temperature (in °C); wind speed (in meters per second, m / s); wind direction; precipitation probability (in %); and descriptions of weather phenomena, such as light snow or moderate rain.

[0043] In one embodiment of the present invention, in the absence of network or in the presence of weak network, route planning data can be obtained through offline maps; however, detailed weather forecasts along the route cannot be obtained.

[0044] In one embodiment of the present invention, the route planning data may also include the origin name, the destination name, and the current timestamp.

[0045] In one embodiment of the present invention, the route planning data also includes at least one waypoint data for temporary parking, thereby allowing users to rest, pick up or drop off other users, conduct business, and independently plan at least one of charging and refueling.

[0046] Step S220: Determine the target energy demand curve based on the preset driving resistance relationship and multimodal travel data.

[0047] In one embodiment of the present invention, determining the target energy demand curve based on a preset driving resistance relationship and multimodal travel data includes: determining the reference driving resistance of multiple driving path segments based on path planning data and the preset driving resistance relationship, and determining the reference energy demand curve based on a preset motor efficiency operation mapping relationship and each reference driving resistance; inputting external environment data into a large language model for energy demand calibration inference to obtain energy demand correction information, the energy demand correction information including at least one of additional load and resistance correction coefficient; correcting the reference energy demand curve based on the energy demand correction information and the corresponding reference driving resistance to obtain the target energy demand curve; wherein, the multimodal travel data includes path planning data and external environment data, and the preset language model includes a large language model running in the cloud.

[0048] In one embodiment of the present invention, the preset driving resistance relationship is as follows: Equation (1) in, The total driving resistance, For tire rolling resistance, For slope resistance, For air resistance, To increase resistance.

[0049] In one embodiment of the present invention, a preset motor efficiency operation mapping relationship is used to characterize the mapping relationship between motor speed, motor torque, and motor efficiency. This can be represented by a motor efficiency MAP diagram.

[0050] In one embodiment of the present invention, the number of drag correction factors is at least one.

[0051] In one embodiment of the invention, key external environment data is passed to the large model interface module. The large model interface module constructs a structured inference request (Prompt) and calls the large model to perform calibration inference.

[0052] In one embodiment of the present invention, the construction of the reasoning request is based on modular splicing, such as splicing roles, tasks, scenarios, output requirements, road segments, etc.

[0053] In one embodiment of the present invention, an example of a reasoning request is: Role: Vehicle Energy Consumption Expert.

[0054] Task: Based on the following scenario, provide a drag correction coefficient for the energy consumption model of a [brand name SUV].

[0055] Scenario description: [Winter, -15 degrees Celsius, light snow, driving at night on a highway with a high risk of icing, wind from the northwest, wind speed 5 m / s].

[0056] Output Requirements: Please return the values ​​of the following key energy demand correction information in JSON format, ranging from [0.5, 1.5]: 1. `battery_performance_factor`: A comprehensive assessment of the negative impact of low temperatures on battery charging and discharging efficiency. 2. `rolling_resistance_factor`: A comprehensive assessment of the increase in tire rolling resistance on icy and slippery roads. 3. `air_resistance_factor`: A comprehensive assessment of the impact of crosswinds or headwinds on air resistance. 4. `hvac_power_demand_kw`: An estimate of the average heating power required to maintain a comfortable interior temperature of 22 degrees Celsius. The energy consumption model includes a preset driving resistance relationship.

[0057] In one embodiment of the present invention, the corresponding reference driving resistance is corrected based on energy demand correction information; the reference energy demand curve is corrected according to the corrected reference driving resistance and the preset motor efficiency operation mapping relationship.

[0058] In one embodiment of the present invention, the large model interface module receives a JSON object returned by the large model and applies these resistance correction coefficients to the corresponding formula terms in the baseline driving resistance, for example, based on the rolling resistance correction coefficient. Reference rolling resistance Make corrections to obtain the actual rolling resistance. ,like, .

[0059] In one embodiment of the present invention, the estimated heating power is added as a continuous additional load to the energy consumption calculation, thereby obtaining a more accurate target energy demand curve that better reflects real-world challenges, to describe the relationship between energy demand and travel distance.

[0060] This invention enhances the baseline energy consumption model by incorporating rich world knowledge and reasoning capabilities from a large model. It transforms unstructured scenario descriptions, such as winter in a northern city or light snow, into quantitative parameter corrections for the baseline energy consumption model. Instead of replacing the baseline model with a complex one, it uses cognitive capabilities to calibrate it. By incorporating the impact of complex real-world scenarios like low temperatures, headwinds, and slippery roads on energy demand, it approximates the actual energy demand in real-world driving scenarios, improving the accuracy of the target energy demand curve. This increased accuracy provides a solid and reliable data benchmark for all subsequent optimization decisions, ensuring the precision and effectiveness of current energy planning strategies from the outset, thereby achieving higher energy utilization efficiency.

[0061] This invention solves the fundamental problem of decoupling the baseline energy consumption model from the vehicle in aftermarket enhancers, and achieves a high degree of adaptability to various vehicle models and complex environments in a lightweight and highly intelligent manner.

[0062] In one embodiment of the present invention, after determining the target energy demand curve based on the preset driving resistance relationship and multimodal travel data, the method further includes: determining at least one target segment based on the target energy demand curve, wherein the target segment is used to characterize a physical key segment; inputting the geographical location information of at least one target segment into a preset language model for event reasoning to obtain corresponding event attribute labels; wherein the event attribute labels are used to obtain the user travel scenario, and the geographical location information is obtained based on multimodal travel data.

[0063] In one embodiment of the present invention, the target section includes at least one of a high-power section and an energy storage section. The high-power section includes a ramp section, and the energy storage section includes a long downhill section.

[0064] In one embodiment of the present invention, the target energy demand curve is scanned, and all target segments are automatically identified and marked by a preset signal processing algorithm and assigned initial labels.

[0065] In one embodiment of the present invention, the initial tag includes the geographic location information and physical characteristics of the target segment.

[0066] In one embodiment of the present invention, for each target segment, the event recognition module extracts its geographical location information and initiates an inference request again through the large model interface module to obtain semantically rich and generalized event attribute tags.

[0067] In one embodiment of the present invention, the event attribute label includes a strategy attribute label.

[0068] In one embodiment of the present invention, an example of a reasoning request corresponding to an event reasoning is as follows, which is only a simple example: Role: Intelligent Driving Strategist.

[0069] Task: Assign strategy attribute labels to the following geographical segments.

[0070] Section A: A high-power uphill road located inside 'XX National Forest Park', a famous 5A-level tourist scenic area.

[0071] Output requirements: From the perspectives of driving experience (NVH, ride comfort), social norms, and potential risks, please select appropriate policy attribute labels for road segment A, choose from the candidate library ['NVH_Sensitive', 'Efficiency_Priority', 'Safety_Priority', 'Scenery_Drive'], and return them in JSON array format.

[0072] NVH is an abbreviation for Noise, Vibration, and Harshness. NVH_Sensitive represents the NVH-sensitive strategy, Efficiency_Priority represents the efficiency-priority strategy, Safety_Priority represents the safety-priority strategy, and Scenery_Drive represents the scenery-driving strategy.

[0073] In one embodiment of the present invention, the large model interface module receives the return results of the large model (e.g., ["NVH_Sensitive", "Scenery_Drive"]). The event recognition module writes these semantic tags back into the path planning data corresponding to the target segment, thereby enriching a simple physical climb into a unique generalized climb event with multi-dimensional strategy guidance. For example, it can be understood whether the climb is in a scenic area requiring quietness or on a highway where full power can be delivered, thus assigning it different strategy attributes. This is a prerequisite for realizing refined and humanized strategies and an important source of information input for dynamically determining the target energy cost relationship.

[0074] In one embodiment of the present invention, an example of an initial label is as follows: Physical events { Physical characteristics: High power consumption Location: "XX National Forest Park" } Examples of generalized event attribute tags are as follows: Physical events { Physical characteristics: High power consumption Location: "XX National Forest Park" "Policy Attributes": [NVH_Sensitive, Scenery_Drive] } Step S230: Based on multimodal travel data, determine at least one user travel scenario, and perform parameter adjustment reasoning on the preset energy cost relationship according to the preset language model and at least one user travel scenario to obtain at least one target energy cost relationship.

[0075] Among them, parameter item adjustment includes at least one of cost item adjustment and cost weight adjustment, and cost item adjustment includes at least one of cost item increase / decrease and cost item internal parameter adjustment.

[0076] In one embodiment of the present invention, the preset energy cost relationship includes pre-setting at least one performance cost that affects energy planning, thereby optimizing energy planning by minimizing the value of the preset energy cost relationship. The performance cost includes, but is not limited to, the cost value corresponding to at least one of the following performance metrics: economic performance, NVH (noise, vibration, and harshness), driving efficiency, safety performance, and battery performance.

[0077] In one embodiment of the present invention, a reasoning request corresponding to parameter adjustment is constructed based on multimodal travel data and input into a large language model for reasoning, thereby adjusting the parameter items of the preset energy cost relationship.

[0078] In one embodiment of the present invention, the dynamic target configuration module in the cloud is activated. Instead of executing a fixed optimization algorithm, it allows the large model to integrate all input information, such as path planning data labeled with event attributes, to form a travel profile of the current trip. Based on the travel profile, it dynamically and programmatically customizes the target energy cost relationship, i.e., the cost function, for the underlying optimization algorithm.

[0079] In one embodiment of the present invention, the travel profile includes at least one user travel scenario.

[0080] In one embodiment of the present invention, the user's travel scenario is obtained based on the entire current travel route or segmented based on waypoint data.

[0081] In one embodiment of the present invention, the cost weight is adjusted, for example, by automatically increasing the cost weight of the NVH penalty (NVH_Penalty) item in scenic road sections to forcibly suppress the start of the range extender.

[0082] In one embodiment of the present invention, for the increase or decrease of cost items, for example, a battery health penalty (Battery_Health_Penalty) item is introduced and weighted under extreme low temperature conditions to avoid the occurrence of conditions that damage the battery.

[0083] In one embodiment of the invention, modifications are made to the internal parameters of the cost item, for example, by modifying the temperature-related penalty coefficient in the battery degradation model.

[0084] In one embodiment of the invention, all outputs of the large model in the cloud must be verified by a rule and physical constraint verification engine before application. Any output exceeding a preset safety boundary will be rejected, and subsequent planning will be carried out through a default safety policy pre-calibrated by experts, thereby preventing the illusion of a large model. For example, limiting the maximum charging SOC of the battery in low-temperature environments.

[0085] In one embodiment of the invention, the large model returns specific configuration JSON. The dynamic target configuration module parses this configuration JSON and converts it into a mathematical expression that the optimization solver module can recognize.

[0086] This invention uses a large model to adjust and reason about preset energy cost relationships based on user travel scenarios. While ensuring high economic efficiency, it can also take into account multiple dimensions of user needs such as comfort, noise and vibration (NVH), long-term battery health and safety, and provide a more humanized and intelligent driving experience.

[0087] This invention elevates vehicle energy planning from solving a fixed mathematical optimization problem to a system where a large model dynamically defines the mathematical optimization problem itself based on travel profiles (such as family trips and business trips) and event attribute labels. This solves the problem of fixed and singular optimization objectives in related technologies, such as fuel economy.

[0088] Because aftermarket range extenders follow the principle of minimizing intrusion, they cannot be deeply integrated into the complex electronic and electrical architecture of a vehicle like original equipment manufacturer (OEM) range extenders. This limits the dimensions of multimodal travel data they can acquire and the fine-grained control permissions they can execute. Consequently, their optimization algorithms can only pursue a single, simple optimization objective and cannot execute intelligent strategies that require multidimensional information input and complex collaborative control. This invention solves the technical problem of a single optimization objective by adjusting and reasoning about the preset energy cost relationship through parameter adjustments.

[0089] Step S240: Perform global energy planning based on the target energy demand curve and at least one target energy cost relationship to obtain the current energy planning strategy.

[0090] The current energy planning strategy includes at least one of an energy management strategy and a thermal management strategy for controlling the operation of the thermal management system. The energy management strategy includes at least one of a range extender operation strategy for controlling the generation of the range extender and an external charging strategy for controlling the charging of the vehicle battery on the external power grid.

[0091] In one embodiment of the present invention, the current energy planning strategy is obtained by using optimization algorithms such as Dynamic Programming (DP), Model Predictive Control (MPC), Genetic Algorithm (GA), or Reinforcement Learning (RL) with the goal of minimizing the energy cost relationship of all objectives.

[0092] In one embodiment of the present invention, the present invention uses a dynamic programming algorithm for decision optimization.

[0093] In one embodiment of the present invention, the current energy planning strategy is obtained based on the discretization of the current travel route.

[0094] In one embodiment of the present invention, the range extender operating strategy includes the initial power curve of the range extender, or the range extender operating strategy includes the initial power curve of the range extender and the battery state of charge curve.

[0095] In one embodiment of the present invention, the external charging strategy includes charging scenario parameters and charging configuration parameters. The charging configuration parameters include parameters such as charging current, charging power, charging upper limit, and charging duration; the charging scenario parameters include parameters such as charging location and charging time period.

[0096] In one embodiment of the present invention, the thermal management strategy includes at least one of a temperature threshold control strategy and a predictive management strategy for the thermal management system. The temperature threshold control strategy is used to control the start / stop of cooling or heating based on a temperature threshold; the predictive management strategy is used to intervene in thermal management in advance before the temperature threshold control strategy is triggered.

[0097] In one embodiment of the present invention, global energy planning is performed based on the target energy demand curve and at least one target energy cost relationship to obtain the current energy planning strategy. This includes: determining the stage state and decision action for each travel stage, and calculating the stage cost after executing each decision action in different stage states for the same travel stage; using dynamic programming to perform reverse optimization on the stage state, decision action, and stage cost corresponding to each travel stage, with minimizing the total travel cost as the optimization objective; selecting one decision action from multiple decision actions corresponding to each travel stage to obtain the current energy planning strategy from the starting point to the end point of the current travel route; wherein, the stage state of each travel stage is based on the current travel route of the discretized vehicle, and / or the state transition obtained by executing the corresponding decision action under the energy demand constraint of the corresponding stage; the stage state includes at least one of battery state of charge and battery temperature; each decision action includes at least one of energy management decision and thermal management decision; the energy management decision includes at least one of range extender operation decision and external charging decision; the stage energy demand is obtained based on the target energy demand curve and the travel distance corresponding to the travel stage; the stage cost is determined based on the target energy cost relationship corresponding to the travel stage; and the total travel cost is obtained based on the sum of the stage costs corresponding to the selected decision actions executed in each travel stage.

[0098] In one embodiment of the present invention, the current travel distance is discretized into multiple travel stages according to a preset second distance interval, which can be set to 100 meters. The state space of each travel stage consists of discretized battery SOC and battery temperature. For example, the battery SOC is discretized with an accuracy of 1%, and the battery temperature is discretized with an accuracy of 5°C.

[0099] In one embodiment of the invention, the range extender operation decision includes turning off the range extender or turning it on at different discrete range extender power levels, such as 1kW, 2kW, ..., 10kW, etc. In one embodiment of the present invention, for each stage state of each travel phase, after making a decision action, the new battery SOC and new battery temperature for transitioning to the next travel phase are calculated based on the stage energy demand of that travel phase. The stage cost incurred in executing this decision action is calculated based on the target energy cost relationship configured in the large model.

[0100] In one embodiment of the present invention, please refer to Figure 3 , Figure 3 This is a schematic diagram of a decision optimization based on dynamic programming provided in one embodiment of the present invention. Figure 3As shown, the left vertical line represents the starting point of the current journey, and the right vertical line represents the ending point. The hollow circle represents the decision-making actions and stage costs for each stage of the journey. The battery SOC decreases sequentially from top to bottom in each stage state. The dynamic programming (DP) algorithm recursively calculates the optimal future cost to reach the destination for each stage state of each journey stage based on the stage cost, and stores the optimal decision-making action that leads to this optimal future cost. Figure 3 As shown by the dashed line; Figure 3 As shown by the solid line, starting from the starting point of the current journey, by tracing forward along the stored optimal decision action, a theoretically optimal SOC curve and initial power curve covering the entire current journey can be obtained.

[0101] This invention enables truly strategically significant current energy planning strategies. For example, it can accurately calculate the need to pre-charge the battery to a precise SOC value on flat roads with optimal power to cope with extremely long uphill climbs of tens of kilometers; or, to maximize energy recovery from long downhill slopes, it needs to maintain the battery SOC at an optimal low level in advance to reserve energy storage space. This end-to-end planning capability of the invention breaks through the limitations of the step-by-step approach in related technologies, ensuring the lowest total energy consumption of the vehicle throughout the entire current journey, achieving true global optimization.

[0102] In one embodiment of the present invention, please refer to Figure 4 , Figure 4 This is a flowchart illustrating global energy planning based on dynamically configured target energy cost relationships, provided in one embodiment of the present invention. Figure 4 As shown, the system inputs multimodal travel data with event attribute labels, such as external weather data and route planning data, to form a travel profile. This travel profile is then used to construct inference requests, which are used to derive the target energy cost relationship (i.e., the cost function) through large-scale model inference, and output to the dynamic target configuration module. The optimization solution module is then activated to discretize the space, obtaining discretized decision actions and stage states for multiple travel phases. The optimal decision action is obtained by recursively tracing back from the endpoint to the starting point, and the battery state-of-charge curve and the range extender's initial power curve can be output through forward tracing.

[0103] In one embodiment of the present invention, after obtaining the current energy planning strategy, the method further includes: smoothing the initial power curve in the range extender's operating strategy to obtain a target power curve; dividing the vehicle's current travel distance into multiple task segments; sampling the target power curve according to a preset first distance interval to obtain a power command mapping relationship corresponding to each task segment, wherein the power command mapping relationship is used to characterize the mapping relationship between distance and target power within a task segment; and determining the energy consumption reference strategy information for each task segment based on each power command mapping relationship and the geofence of each task segment; wherein the geofence is obtained based on multimodal travel data.

[0104] In one embodiment of the present invention, the target energy planning strategy includes energy consumption reference strategy information.

[0105] In one embodiment of the present invention, in order to completely reduce the computational burden on the vehicle, the strategy generation and translation module in the cloud is activated and the output initial power curve is smoothed. The smoothed initial power curve is sampled at a fixed preset first distance interval to generate an extremely detailed, discrete power instruction lookup table, that is, a power instruction mapping relationship. In this way, the initial power curve is compiled into a simple instruction format that can be quickly queried by the vehicle and executed without complex calculations, thus realizing the function of heavy computation in the cloud and light execution on the vehicle.

[0106] In one embodiment of the present invention, the preset first distance interval can be set to 100 meters. The power command lookup table is shown in the following table: Table 1 Power Command Mapping Relationship

[0107] In one embodiment of the present invention, in each task segment, the corresponding geofence and power command mapping relationship are combined to obtain the energy consumption reference strategy information of each task segment, and then sent to the vehicle.

[0108] In one embodiment of the present invention, the strategy generation and translation module divides the entire current travel route into several macroscopic task segments. Each task segment serves as a strategy unit, and its corresponding energy consumption reference strategy information includes: the geofence of the task segment, the task objective, a user-friendly explanatory text generated by the large model, and a core power command lookup table. Finally, these strategy units are combined into a complete strategy JSON file and returned to the vehicle. An example of a strategy JSON file is shown below: { "strategy_id":"TRIP_XXXXXX_ABC", "segments":[ { "segnent_id":1, "geo_fence":"[Polygon GeoJSON…]", "Task goal": "Efficiently navigate congested urban roads" "human_readable_explanation":"The road ahead is urban; a pure electric vehicle priority strategy will be adopted to reduce noise and fuel consumption." "action_lookup_table":[ {"distance:0,"power":0}, {"distance:100,"power":0}, ... ] }, ... ] } Among them, strategy_id is used to represent the strategy identity, such as identifying it by the date "XXXXXX" to the target location "ABC"; segments are used to represent task segments; sequential_id is used to represent the segment identity of a task segment, such as the first task segment; geo_fence is used to represent geofences, such as polygon coordinate information; taskgoal is used to represent the task objective; human_readable_explanation is used to represent human-readable explanation text; action_lookup_table is used to represent the power instruction lookup table, such as the target power being 0 when the distance within the task segment is 0 and 100.

[0109] In one embodiment of the present invention, the humanized explanatory text is derived based on the user's travel scenario.

[0110] In one embodiment of the present invention, different task segments are divided based on different types of geographical segments and / or different user travel scenarios in the current trip. These different types of geographical segments include, but are not limited to, uphill segments, downhill segments, scenic segments, urban segments, and highway segments. User travel scenarios include, but are not limited to, tourism, daily commuting, and business travel.

[0111] In one embodiment of the present invention, the determination of the user's travel scenario corresponds to the type of geographical segment in the current travel route.

[0112] In one embodiment of the present invention, the target power curve corresponding to each task segment is sampled according to a preset first distance interval. The preset first distance interval may be the same or different for different task segments.

[0113] In one embodiment of the present invention, after determining the energy consumption reference strategy information for each task segment, the method further includes: obtaining the vehicle's current location information and the distance traveled; determining the task segment to which the current location information belongs as the target segment based on the geofence, and determining the distance traveled by the vehicle on the target segment based on the distance traveled and the current location information; determining the target data pairs on both sides of the distance traveled on the target segment based on the power command mapping relationship of the target segment, wherein the target data pairs include the target power and the distance within the task segment; interpolating the target data pairs on both sides based on the distance traveled on the segment to obtain the vehicle's power to be controlled, and sending it to the range extender controller to control the range extender to generate electricity.

[0114] In one embodiment of the invention, the distance traveled is used to represent the total distance traveled within the current travel route. The distance traveled within a road segment is used to represent the total distance traveled within the target road segment.

[0115] In one embodiment of the present invention, the vehicle receives and parses the policy JSON file, i.e., the target energy planning policy, sent from the cloud, and the policy caching and management module loads it into memory. The task scheduling and execution module starts a high-frequency timer, for example, 100 milliseconds. See also... Figure 5 , Figure 5 This is a flowchart illustrating the vehicle-side policy execution phase according to an embodiment of the present invention. Figure 5 As shown, within each preset scheduling cycle, the following steps are performed: 1. Obtain the vehicle's real-time location and mileage: Obtain the vehicle's real-time travel distance and current location information, such as information from the Global Positioning System (GPS); 2. Determine the assigned task segment: Determine the task segment to which the current location information belongs and obtain the power command mapping relationship for that task segment; 3. Calculate the distance traveled within the segment: Calculate the distance the vehicle has traveled on that task segment; 4. Look up tables and interpolate: In the power command mapping relationship, use efficient algorithms such as binary search to find the two closest target data pairs on both sides of the travel distance of the segment, and calculate the current power to be controlled through linear interpolation; 5. Generate hardware commands: Encapsulate the calculated power to be controlled into hardware commands; 6. Send commands via the bus: Send hardware commands to the range extender controller via the vehicle bus interface.

[0116] In one embodiment of the invention, the target data pair includes target power and distance within the task segment.

[0117] In one embodiment of the present invention, the power to be controlled is determined as follows: Equation (2) in, For the power to be controlled, The target power is the first target data pair before the distance already traveled on the road segment. The target power is the first target data pair after the target has traveled a certain distance. This represents the distance already traveled on the road segment. This represents the distance within the task segment for the first target data pair before the distance already traveled on that segment. The distance within the task segment is the first target data pair after the distance already traveled on the road segment.

[0118] In one embodiment of the present invention, if the vehicle has traveled a distance of 150m on the target road segment, and the power command mapping relationship is shown in Table 1, then the target data pair is the data points corresponding to 100m and 200m, and the power to be controlled is 5.5 + (6.2 - 5.5). (150-100) / (200-100)=5.85KW.

[0119] This invention transforms complex cloud-based strategies into discrete power command mapping relationships, thereby completing all complex and computationally intensive planning tasks in the cloud at once. The vehicle-side target controller no longer needs to perform any complex real-time optimization calculations; its core tasks are simplified to extremely lightweight table lookups and linear interpolation, significantly reducing the computational burden on the target controller and improving the reliability, real-time performance, and response speed of strategy execution. This invention not only makes it applicable to vehicles with aftermarket range extenders that have limited computing power, but also avoids the risk of control instability caused by real-time calculation delays or failures on the vehicle side. It greatly improves the reliability, accuracy, and real-time response speed of current energy planning strategies in actual driving. It solves the problem in related technologies that require complex controllers, such as proportional-integral-derivative (PID) and model predictive control (MPC), to track the battery SOC curve sent from the cloud in real time, resulting in high computing power consumption and high control robustness requirements on the vehicle side. It also resolves the contradiction between high computing power on the cloud and low latency and high reliability on the vehicle side, ensuring that the results of the optimization algorithm can be reproduced on the vehicle side in a lightweight, accurate, and stable manner. This is especially critical for vehicles with aftermarket range extenders that have limited computing power and is a technical guarantee for the implementation of the solution.

[0120] In one embodiment of the present invention, after obtaining the current energy planning strategy, the method further includes: acquiring the vehicle's network status and new multimodal travel data, wherein the new multimodal travel data is obtained based on the new travel route; determining the cloud or vehicle as the energy planning end based on the network status; if the energy planning end is the vehicle end, and the lowest route matching degree between the new travel route and at least one historical travel route is greater than or equal to a preset matching threshold, then determining the historical energy planning strategy corresponding to the highest route matching degree as the travel energy planning strategy for the new travel route; if the energy planning end is the vehicle end, and the lowest route matching degree between the new travel route and at least one historical travel route is less than the preset matching threshold, then performing global energy planning based on the local language model and the new multimodal travel data to obtain the travel energy planning strategy for the new travel route; wherein the preset language model includes the vehicle's built-in local language model, which is used to infer the target energy demand curve and at least one target energy cost relationship corresponding to the new travel route.

[0121] In one embodiment of the present invention, the historical energy planning strategy is obtained based on the current energy planning strategy or the target energy planning strategy stored locally in the vehicle.

[0122] In one embodiment of the present invention, the target energy planning strategy includes energy consumption reference strategy information for multiple task segments.

[0123] In one embodiment of the present invention, the historical energy planning strategy includes at least one of the stored energy planning strategies for frequently used user routes and the energy planning strategy for the most recent travel route.

[0124] In one embodiment of the present invention, when the vehicle is unable to connect to the network at the start of a journey, the vehicle will automatically downgrade and invoke a built-in lightweight local energy planning model based on offline maps and preset fixed rules to execute a safety net strategy. If the new journey closely matches a cached historical journey, the corresponding historical energy planning strategy can be reused. The robustness of strategy generation is improved through strategy caching and offline global energy planning.

[0125] In one embodiment of the present invention, the lightweight local energy planning model includes a local language model. The preset fixed rules include parameter adjustment rules corresponding to preset energy cost relationships.

[0126] In one embodiment of the present invention, the present invention can use real vehicle energy consumption data to correct the preset driving resistance relationship; it can also use at least one of real vehicle energy consumption data and user feedback data to fine-tune the large model through reinforcement learning, such as reinforcement learning from human feedback (RLHF), so that its world knowledge and reasoning energy can continue to evolve.

[0127] Please see Figure 6 , Figure 6 This is a block diagram of a vehicle energy planning device provided in one embodiment of the present invention. This device can be applied to… Figure 1 The implementation environment shown is specifically configured in the cloud 1100 or vehicle 1200. This device can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applicable.

[0128] like Figure 6 As shown, a vehicle energy planning device 600 according to an embodiment of the present invention includes: a data acquisition module 610, an energy demand determination module 620, a cost determination module 630, and an energy planning module 640.

[0129] Among them, the data acquisition module 610 is used to acquire multimodal travel data of the vehicle while it is in motion; The energy demand determination module 620 is used to determine the target energy demand curve based on the preset driving resistance relationship and multimodal travel data; The cost determination module 630 is used to determine at least one user travel scenario based on multimodal travel data, and to perform parameter adjustment reasoning on the preset energy cost relationship according to the preset language model and at least one user travel scenario to obtain at least one target energy cost relationship. The parameter adjustment includes at least one of cost item adjustment and cost weight adjustment. The cost item adjustment includes at least one of cost item increase / decrease and internal parameter adjustment of the cost item. The energy planning module 640 is used to perform global energy planning based on the target energy demand curve and at least one target energy cost relationship to obtain a current energy planning strategy. The current energy planning strategy includes at least one of an energy management strategy and a thermal management strategy for controlling the operation of the thermal management system. The energy management strategy includes at least one of a range extender operation strategy for controlling the generation of the range extender and an external charging strategy for controlling the charging of the vehicle battery on the external power grid.

[0130] It should be noted that the vehicle energy planning device and the vehicle energy planning method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the vehicle energy planning device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0131] Embodiments of the present invention also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by one or more processors, cause the electronic device to implement the vehicle energy planning method provided in the above embodiments.

[0132] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a computer system for an electronic device provided in one embodiment of the present invention. Figure 7 The computer system 700 of the illustrated electronic device is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0133] like Figure 7 As shown, the computer system 700 includes a central processing unit 701, which can perform various appropriate actions and processes based on a program stored in a read-only memory 702 or a program loaded from a storage section 708 into a random access memory 703, such as executing the methods described in the above embodiments. The random access memory 703 also stores various programs and data required for system operation. The central processing unit 701, the read-only memory 702, and the random access memory 703 are interconnected via a bus 704. An input / output interface 705 is also connected to the bus 704.

[0134] The following components are connected to the input / output interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 710 as needed so that computer programs read from it can be installed into the storage section 708 as needed.

[0135] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 709, and / or installed from removable medium 711. When the computer program is executed by central processing unit 701, it performs various functions defined in the system of the present invention.

[0136] The computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM), flash memory, optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. Computer programs contained on computer-readable media can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0137] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0138] The units described in the embodiments of the present invention can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself. Therefore, the technical solutions according to the embodiments of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, portable hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of the present invention.

[0139] Another aspect of the present invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a computer's processor, causes the computer to perform the vehicle energy planning methods provided in the above embodiments. This computer-readable storage medium may be included in the electronic devices described in the above embodiments, or it may exist independently and not incorporated into the electronic devices.

[0140] In the above embodiments, unless otherwise specified, the use of ordinal numbers such as "first" and "second" to describe common objects only indicates that they refer to different instances of the same object, rather than indicating that the objects being described must be in a given order, whether temporally, spatially, sequentially, or in any other way.

[0141] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A vehicle energy planning method, characterized in that, The method includes: Acquire multimodal travel data of vehicles while in motion; The target energy demand curve is determined based on the preset driving resistance relationship and the multimodal travel data; Based on the multimodal travel data, at least one user travel scenario is determined, and the preset energy cost relationship is inferred by parameter adjustment according to the preset language model and at least one user travel scenario to obtain at least one target energy cost relationship. The parameter adjustment includes at least one of cost item adjustment and cost weight adjustment. The cost adjustment includes at least one of cost item increase / decrease and internal parameter adjustment of cost item. Global energy planning is performed based on the target energy demand curve and at least one of the target energy cost relationships to obtain a current energy planning strategy. The current energy planning strategy includes at least one of an energy management strategy and a thermal management strategy for controlling the operation of the thermal management system. The energy management strategy includes at least one of a range extender operation strategy for controlling the generation of the range extender and an external charging strategy for controlling the charging of the vehicle battery on the external power grid.

2. The vehicle energy planning method according to claim 1, characterized in that, The target energy demand curve is determined based on the preset driving resistance relationship and the multimodal travel data, including: The reference driving resistance of multiple driving path segments is determined based on the path planning data and the preset driving resistance relationship, and the reference energy demand curve is determined based on the preset motor efficiency operation mapping relationship and each of the reference driving resistances. External environmental data is input into a large language model to perform energy demand calibration inference, and energy demand correction information is obtained. The energy demand correction information includes at least one of additional load and resistance correction coefficients. The reference energy demand curve is corrected based on the energy demand correction information and the corresponding reference driving resistance to obtain the target energy demand curve. The multimodal travel data includes the route planning data and the external environment data, and the preset language model includes the large language model running in the cloud.

3. The vehicle energy planning method according to claim 1, characterized in that, After determining the target energy demand curve based on the preset driving resistance relationship and the multimodal travel data, the process also includes: At least one target segment is determined based on the target energy demand curve, and the target segment is used to characterize the physical key segment; The geographical location information of at least one of the target segments is input into the preset language model for event reasoning to obtain the corresponding event attribute labels; The event attribute tags are used to obtain the user's travel scenario, and the geographical location information is obtained based on the multimodal travel data.

4. The vehicle energy planning method according to claim 3, characterized in that, Global energy planning is performed based on the target energy demand curve and at least one of the target energy cost relationships to obtain the current energy planning strategy, including: Determine the stage state and decision action for each stage of the journey, and calculate the stage cost after executing each decision action in different stage states for the same stage of the journey. The dynamic programming method is used to perform reverse optimization on the stage state, decision action and stage cost corresponding to each of the aforementioned travel stages. With minimizing the total travel cost as the optimization objective, one decision action is selected from multiple decision actions corresponding to each travel stage to obtain the current energy planning strategy from the starting point to the destination of the current travel route. The stage state of each travel stage is obtained based on the discretized current travel distance of the vehicle and / or the state transition under the energy demand constraint of the corresponding stage after executing the corresponding decision action. The stage state includes at least one of battery state of charge and battery temperature. Each decision action includes at least one of energy management decision and thermal management decision. The energy management decision includes at least one of range extender operation decision and external charging decision. The stage energy demand is obtained based on the target energy demand curve and the travel distance corresponding to the travel stage. The stage cost is determined based on the target energy cost relationship corresponding to the travel stage. The total travel cost is obtained based on the sum of the stage costs corresponding to the selected decision action in each travel stage.

5. The vehicle energy planning method according to claim 1, characterized in that, After obtaining the current energy planning strategy, it also includes: The network status of the vehicle and new multimodal travel data are obtained, which are based on the new travel route; Based on the network status, the cloud or vehicle terminal will be identified as the energy planning terminal; If the energy planning end is the vehicle end, and the lowest route matching degree between the new travel route and at least one historical travel route is greater than or equal to a preset matching threshold, then the historical energy planning strategy corresponding to the highest route matching degree is determined as the travel energy planning strategy for the new travel route. If the energy planning end is the vehicle end, and the minimum route matching degree between the new travel route and at least one historical travel route is less than a preset matching threshold, then global energy planning is performed based on the local language model and the new multimodal travel data to obtain the travel energy planning strategy for the new travel route. The preset language model includes the local language model built into the vehicle, which is used to infer the target energy demand curve and at least one target energy cost relationship corresponding to the new travel route.

6. The vehicle energy planning method according to any one of claims 1-5, characterized in that, After obtaining the current energy planning strategy, it also includes: The initial power curve in the range extender's operating strategy is smoothed to obtain the target power curve; The vehicle's current travel route is divided into multiple task segments; The target power curve is sampled according to a preset first distance interval to obtain the power command mapping relationship corresponding to each task segment. The power command mapping relationship is used to characterize the mapping relationship between distance and target power within a task segment. Based on the power command mapping relationship and the geofence of each task segment, determine the energy consumption reference strategy information of each task segment; The geofence is obtained based on the multimodal travel data.

7. The vehicle energy planning method according to claim 6, characterized in that, After determining the energy consumption reference strategy information for each of the aforementioned task segments, the process also includes: Obtain the vehicle's current location information and the distance traveled; Based on the geofence, the task segment to which the current location information belongs is determined as the target segment, and based on the distance already traveled and the current location information, the distance already traveled by the vehicle on the target segment is determined; Based on the power command mapping relationship of the target road segment, target data pairs are determined on both sides of the distance already traveled on the road segment. The target data pairs include the target power and the distance within the task road segment. Based on the distance traveled on the road segment, the target data pairs on both sides are interpolated to obtain the power to be controlled for the vehicle, which is then sent to the range extender controller to control the range extender to generate electricity.

8. A vehicle energy planning device, characterized in that, The device includes: The data acquisition module is used to acquire multimodal travel data of vehicles while they are in motion; An energy demand determination module is used to determine a target energy demand curve based on a preset driving resistance relationship and the multimodal travel data; The cost determination module is used to determine at least one user travel scenario based on the multimodal travel data, and to perform parameter adjustment reasoning on the preset energy cost relationship according to the preset language model and at least one user travel scenario to obtain at least one target energy cost relationship. The parameter adjustment includes at least one of cost item adjustment and cost weight adjustment. The cost adjustment includes at least one of cost item increase / decrease and internal parameter adjustment of the cost item. An energy planning module is used to perform global energy planning based on the target energy demand curve and at least one of the target energy cost relationships to obtain a current energy planning strategy. The current energy planning strategy includes at least one of an energy management strategy and a thermal management strategy for controlling the operation of a thermal management system. The energy management strategy includes at least one of a range extender operation strategy for controlling the generation of the range extender and an external charging strategy for controlling the charging of the vehicle battery on an external power grid.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the vehicle energy planning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by the computer's processor, causes the computer to perform the vehicle energy planning method according to any one of claims 1 to 7.

Images