New energy vehicle multi-mode energy collaborative management system and method

Abstract

The invention discloses a new energy automobile endurance mileage increasing system and method based on multi-mode energy collaborative management. The system comprises a multi-mode energy input module which is used for converting solar energy into direct current electric energy through a photovoltaic array and integrating power battery energy and hub motor kinetic energy recovery energy; the predictive management module is used for acquiring front road condition information through a V2X technology and predicting an energy consumption curve based on a deep learning model; the dynamic weight distribution module is used for calculating photovoltaic, battery and kinetic energy recovery weight coefficients through a dynamic weight distribution algorithm according to the real-time vehicle environment information and the prediction result; and the energy distribution control module is used for carrying out weighted distribution on the multi-modal energy based on the weight coefficient and outputting total available energy and a control signal so as to drive electric equipment or charge a battery. According to the invention, the problem of dynamic energy distribution during multi-energy coupling is solved, and the purposes of improving the endurance of urban roads, improving the endurance of expressways, reducing the increment cost of the system and being compatible with a 400V / 800V platform are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy and renewable energy, in particular to a new energy vehicle multi-modal energy collaborative management system and method. BACKGROUND

[0002] At present, the range of new energy vehicles is mainly improved by the following several mainstream technical solutions: Physical superposition battery capacity: increase the battery capacity to improve the endurance, for example, a certain vehicle model increases the lithium iron phosphate battery from 64.8kWh to 82.5kWh, and the NEDC endurance increases from 506km to 700km, but the battery pack weight increases from 378kg to 476kg, and the vehicle weight increases by 98kg. Energy recovery system: about 15% of mechanical energy is recovered through brake energy recovery technology, but the efficiency decreases significantly under low speed working condition. Wind resistance coefficient optimization: the wind resistance coefficient of the vehicle body is optimized to 0.23 or less, but this way often sacrifices the space of the vehicle body.

[0003] However, these existing technical methods have the following main disadvantages: single optimization dimension boundary efficiency is low: each technology operates independently, without considering the synergistic effect of multi-energy coupling, resulting in limited overall energy efficiency improvement. For example, battery expansion increases the power, but the weight increase offsets part of the endurance gain; the efficiency of energy recovery decreases sharply under certain working conditions. The problem of dynamic allocation of multi-energy coupling is not solved: the existing scheme lacks a real-time coordination mechanism for photovoltaic, battery and kinetic energy, and cannot dynamically optimize energy allocation according to road conditions, battery state and other factors, resulting in energy waste or efficiency loss. SUMMARY

[0004] The purpose of the present application is to provide a new energy vehicle multi-modal energy collaborative management system, which can realize dynamic optimization management of multi-energy through three-modal coupling of photovoltaic, kinetic and battery, combined with predictive control and dynamic weight distribution algorithm.

[0005] To achieve this purpose, the present application designs a new energy vehicle multi-modal energy collaborative management system, which comprises: The multi-modal energy input module is used to obtain photovoltaic power generation energy, power battery energy and kinetic energy recovery energy; The energy distribution control module is used to obtain the weight coefficients of photovoltaic power generation energy, power battery energy and kinetic energy recovery energy according to the energy consumption prediction results and the real-time collected vehicle environment information, to perform weighted distribution on the photovoltaic power generation energy, power battery energy and kinetic energy recovery energy according to the weight coefficients, to obtain the distributed energy control signal and the total available energy, and to drive the electric equipment or charge the power battery based on the total available energy and the energy control signal.

[0006] The application integrates photovoltaic power generation, power battery and kinetic energy recovery three energy sources, so that the continuous external energy input of solar energy and the optimized braking energy recovery are realized, the additional energy source is provided for the vehicle under the premise of not significantly increasing the weight of the vehicle, the endurance of the new energy vehicle is effectively improved, the global optimal energy utilization efficiency is obtained according to the predicted performance energy management and dynamic weight distribution algorithm, the intelligentization and adaptive optimization are realized, the high reliability and safety of the system are obtained according to the state machine arbitration mechanism based on priority, the stable operation under various working conditions is ensured, the decision logic when various working modes (such as emergency escape, overtaking, downhill recovery and the like) conflict is determined, the safety and the intention of the driver are always given priority response in the extreme weather, emergency escape and the like, so that the control safety of the vehicle and the reliable execution of the system function are ensured.

[0007] Preferably, the formula for calculating the total available energy is: ; Wherein, is the total available energy, is the power battery energy, is the photovoltaic power generation energy, is the kinetic energy recovery energy, , and are weight coefficients of the power battery energy, the photovoltaic power generation energy and the kinetic energy recovery energy respectively.

[0008] Preferably, when the preset energy distribution mode trigger condition is met, the weight coefficients are distributed according to the coefficient distribution rules corresponding to different energy distribution modes.

[0009] Preferably, when any preset energy distribution mode trigger condition is not met, the weight coefficients are calculated by a dynamic weight distribution algorithm according to the energy consumption prediction results and the real-time collected vehicle environment information.

[0010] Preferably, the method for obtaining the photovoltaic power generation energy, the power battery energy and the kinetic energy recovery energy comprises: The absorbed solar energy is converted into direct current energy through a DCDC converter to obtain the photovoltaic power generation energy; the state of the power battery is monitored and the power battery energy is output; the kinetic energy is recovered through the wheel hub motor when the vehicle brakes or goes downhill, and is converted into direct current energy through an inverter rectifier to obtain the kinetic energy recovery energy.

[0011] Preferably, when any preset energy distribution mode trigger condition is not met, the method for obtaining the weight coefficients comprises: The baseline weights for photovoltaic power generation, power battery energy, and kinetic energy recovery are calculated. The battery baseline weight is obtained based on the current state of charge (SOC) of the battery, and the higher the SOC, the greater the weight. The photovoltaic baseline weight is obtained based on the ratio of the current photovoltaic power generation to the total power demand of the vehicle. The kinetic energy recovery baseline weight is obtained based on the ratio of the predicted kinetic energy recovery power to the total power demand. An adjustment factor based on the optimization objective is introduced, including a battery life factor and a system efficiency factor; The weighted value is obtained by multiplying the baseline weight of each energy source by the corresponding adjustment factor, and then normalized to obtain the weight coefficient.

[0012] The beneficial effects of this invention are as follows: This invention proposes a multimodal energy collaborative management system for new energy vehicles. It integrates solar energy, power battery energy, and kinetic energy recovery energy through a multimodal energy input module. A predictive management module, based on vehicle-to-everything (V2X) communication technology and deep learning models, predicts future energy consumption curves. A dynamic weight allocation algorithm calculates weight coefficients in real time, effectively solving the dynamic allocation problem in existing technologies involving multi-energy coupling, and significantly improving the driving range of new energy vehicles. Through the three-modal coupling of photovoltaic, kinetic energy, and battery energy, combined with the efficient conversion of perovskite-silicon tandem photovoltaic films and kinetic energy recovery from wheel hub motors, it intelligently manages energy consumption under different lighting and road conditions. This invention can switch operating modes, such as prioritizing photovoltaic mode or kinetic energy recovery mode, thereby increasing range in urban areas. Compared to simply expanding battery capacity, this invention optimizes energy utilization efficiency and reduces incremental system costs, resulting in higher economic efficiency. The central arbitrator and multi-level priority rules in the dynamic weight allocation module ensure real-time and secure mode switching, improving system reliability and response speed. The DC-DC adaptive conversion module, compatible with 400V / 800V platforms, enhances system adaptability and applicability, making it widely applicable to new energy vehicles with different voltage platforms. Through multi-modal energy collaborative management, predictive control, and intelligent weight allocation, this invention effectively overcomes the low boundary efficiency problem of single-dimensional optimization, improving overall energy efficiency and robustness, and has broad application prospects. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the structure of the present invention; Figure 2 This is a schematic diagram of the DQN model operation; Figure 3 This is a block diagram illustrating the principle of vehicle energy management. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to represent selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0015] The system integrates perovskite-silicon tandem photovoltaic films on the roof and hood, kinetic energy recovery from wheel hub motors, and a power battery. It uses an energy distribution controller to calculate weighting coefficients in real time and predicts energy consumption curves based on V2X road condition information and a deep learning model, dynamically switching operating modes. This approach effectively improves energy utilization efficiency, overcomes the limitations of single-dimensional optimization, and solves the dynamic allocation problem in multi-energy coupling scenarios. The invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1 A multimodal energy collaborative management system for new energy vehicles, such as Figure 1 As shown, it includes: The multimodal energy input module is used to acquire photovoltaic power generation energy, power battery energy, and kinetic energy recovery energy; The energy distribution control module is used to obtain the weighting coefficients of photovoltaic power generation energy, power battery energy and kinetic energy recovery energy based on the energy consumption prediction results and real-time vehicle environmental information. Based on the weighting coefficients, the photovoltaic power generation energy, power battery energy and kinetic energy recovery energy are weighted and distributed to obtain the distributed energy control signal and total available energy. Based on the total available energy and energy control signal, the module drives the electrical equipment or charges the power battery.

[0016] In some preferred embodiments, the method for obtaining photovoltaic power generation energy, power battery energy, and kinetic energy recovery energy includes: The absorbed solar energy is converted into DC power by a DC-DC converter to obtain photovoltaic power generation energy; the status of the power battery is monitored and the power battery energy is output; the kinetic energy is recovered by the hub motor when the vehicle brakes or goes downhill, and is converted into DC power by the inverter to obtain kinetic energy recovery energy.

[0017] In some preferred embodiments of this invention, when determining solar energy, solar energy is absorbed by a photovoltaic array, converted by a DC-DC converter to an energy distribution controller, and then output to the power equipment and the power battery by the control core.

[0018] In some preferred embodiments of the present invention, when determining electrical energy, the power battery outputs the energy to the energy distribution controller, and then the control core outputs the energy to the electrical equipment.

[0019] In determining mechanical energy, in some preferred embodiments of the present invention, kinetic energy is recovered through vehicle braking or downhill driving, and the energy is converted into energy distribution controller via inverter rectification, and then output to electrical equipment or to charge the power battery by the control core.

[0020] For multimodal energy collaborative management, in some preferred embodiments, the vehicle energy management principle is as follows: Figure 3 As shown, the three-mode coupling of photovoltaics, kinetic energy, and batteries is achieved through perovskite-silicon tandem photovoltaic films covering the roof and hood, and kinetic energy recovery via wheel hub motors. In optional embodiments, the thickness of this material is ≤0.5mm, and the weight is ≤1.2Kg / ㎡. The total coverage area is approximately 5.2㎡, including 2.5㎡ for the roof, 2.0㎡ for the hood, and 0.7㎡ for the trunk lid. The photoelectric conversion efficiency is expected to reach 26% under automotive conditions. A flexible polyimide substrate with a bending radius of up to 25mm is used. The substrate shape is consistent with the body's bonding surface. A modified acrylic adhesive with high light transmittance (>95%) is used to achieve seamless vacuum bonding with the curved steel or aluminum surfaces of the vehicle body or the roof glass, without affecting the original aerodynamics of the vehicle body. The estimated vehicle characteristics under different lighting conditions based on a total area of ​​5.2㎡ are shown in Table 1. Table 1 Vehicle Characteristics Table In some preferred embodiments, the method for obtaining energy consumption prediction results includes: acquiring road condition information ahead through vehicle-to-everything (V2X) communication technology, and combining the road condition information with a deep learning-based prediction model to predict the future energy consumption curve, thereby obtaining the energy consumption prediction result. Of course, predicting forward driving energy consumption during vehicle operation essentially involves predicting forward energy consumption based on the vehicle's current driving conditions, including but not limited to energy consumption parameters such as vehicle speed and torque, combined with forward road condition information, including but not limited to inclines, declines, straight sections, and curves.

[0021] For predictive control based on deep learning, in some preferred embodiments, road condition information 3km ahead is obtained through V2X, and the energy consumption curve is predicted 1km in advance and the working mode is pre-adjusted. Among them, the road condition information is a multi-dimensional data set, mainly including geographic elevation information (a series of discrete points with an accuracy of ±0.1 meters), traffic event information (including congestion level: smooth traffic, slow traffic, congested, severe congestion; accident location; construction area; traffic light phase and timing), dynamic traffic flow information (including average vehicle speed v, traffic density ρ), and environmental information (real-time light intensity I, wind speed w, wind direction, and ambient temperature T).

[0022] In some preferred embodiments, photovoltaic power generation energy, power battery energy, and kinetic energy recovery energy are stored in an energy distribution controller, and then delivered to electrical equipment or used to charge the power battery through a central arbitrator in the energy distribution controller.

[0023] In some preferred embodiments, by emphasizing the role of the energy distribution controller, energy is required to be stored in the controller before being distributed. A central arbitrator is introduced to centrally manage the energy flow, improving the accuracy and reliability of energy distribution and avoiding energy conflicts or waste. For example, the energy distribution controller can query the status of photovoltaic, battery, and recycling systems in real time and perform smooth distribution according to weighting coefficients, ensuring the stability of energy transmission. Especially in the case of multiple source inputs, it can achieve seamless switching and improve the overall system efficiency.

[0024] Regarding the total available energy, in some preferred embodiments, the formula for calculating the total available energy is as follows: ; in, The total available energy is calculated through real-time integration: , For power battery energy, For photovoltaic power generation energy, To recover energy from kinetic energy , and These are the weighting coefficients for power battery energy, photovoltaic power generation energy, and kinetic energy recovery energy, respectively.

[0025] In some preferred embodiments, by clarifying the calculation formula for total available energy and quantifying the contribution of each energy source, a mathematical basis for dynamic optimization is provided, enabling the system to calculate available energy based on real-time data. Combined with weighting coefficients, it can effectively address changes in road conditions. For example, by adjusting α, β, and γ, energy combination optimization can be achieved, ensuring the scientific nature of energy calculations and providing a basis for predictive control, thereby improving the accuracy of range prediction.

[0026] Regarding the judgment criteria, in some preferred embodiments, when the vehicle speed and sunlight intensity in the real-time collected vehicle environmental information are both greater than a set threshold, the basic photovoltaic priority mode one is preferentially adopted to determine the outcome. This is a preset value; If the downhill slope gradient in the real-time vehicle environment information is greater than a set threshold, the kinetic energy recovery mode will be prioritized. This is a preset value.

[0027] In some preferred embodiments, the preset energy distribution mode triggering conditions include vehicle speed, light intensity, battery temperature, brake pedal, road congestion, gradient, throttle opening, rainstorm and sandstorm detection, and whether the airbag and manual emergency switch are triggered.

[0028] In some preferred embodiments, photovoltaic or kinetic energy recovery modes are given priority under specific conditions. For example, photovoltaic mode is given priority when the vehicle speed is high and the sunlight is strong. A rule-based triggering mechanism is introduced, which can quickly respond to high-yield scenarios and maximize energy utilization.

[0029] In some preferred embodiments, when the preset energy allocation mode triggering conditions are met, the weighting coefficients are allocated according to the coefficient allocation rules corresponding to different energy allocation modes.

[0030] Regarding energy distribution modes, some preferred embodiments specifically include: The pure battery drive mode is triggered when the vehicle speed exceeds a set value or within a set time period. In an optional embodiment, the time period is from 17:00 to 7:00 in winter and from 18:00 to 6:00 in summer.

[0031] The second basic photovoltaic priority mode is triggered when the irradiance is greater than a set value and the SOC is greater than a set percentage. The supercharging photovoltaic priority mode is triggered when the battery temperature is below a set value and the sunlight intensity is above a set threshold. In an optional embodiment, the sunlight intensity is greater than 800W / m². 2 This is considered strong light. The standard kinetic energy recovery mode is triggered when the brake pedal travel meets the set percentage range.

[0032] The strong energy recovery mode is triggered when the road conditions ahead are congested or downhill. The downhill-specific recovery mode is triggered when the slope is greater than a set value and continues for a set distance. The congestion optimization mode is triggered when the vehicle speed is less than the set value and the air conditioning is turned on. Low-temperature combined heating mode is triggered when the battery temperature is lower than a set value and the SOC is lower than a set percentage. The quick overtaking mode is triggered when the throttle opening is greater than a set percentage and continues for a set duration. Extreme weather safety mode is triggered when heavy rain or sandstorms are detected. The vehicle-to-everything (V2X) communication technology predicts cruise mode, triggered by the condition that the 5G-V2X signal is valid. Emergency escape mode is triggered by either airbag activation or manual emergency switch activation.

[0033] The specifics are shown in Table 2 below: Table 2 Energy Distribution Pattern Table In some preferred embodiments, 12 energy distribution modes are listed, such as pure battery drive, photovoltaic priority, and kinetic energy recovery. The flexible mode library covering all operating conditions provides diverse energy strategies to adapt to complex driving environments. Mode switching can effectively balance energy consumption and comfort, and the preset modes simplify the decision-making process, ensuring that the system can operate safely and efficiently in various scenarios (such as overtaking, low temperature, and extreme weather).

[0034] In some preferred embodiments, the present invention designs a state machine-based central arbiter, and the energy allocation mode is divided into three priorities: P0 Level: Safety and Regulations Priority: The highest level. When the P0 level mode is selected, it will be executed immediately without conditions. It includes extreme weather safety mode and emergency escape mode.

[0035] P1 Level with Driver Intent Priority: The second level, when the P1 level mode is selected, responds to the driver's active operation in real time, and is only triggered when there is no P0 level command, including the fast overtaking mode. P2 level, which prioritizes system energy efficiency: The third level, when there are no P0 or P1 level instructions, is designed for deterministic high-yield scenarios and aims to maximize energy efficiency. It is decided autonomously by the system and includes a strong kinetic energy recovery mode and a downhill-specific recovery mode. Learning-oriented optimization-priority P3 level: The third level, in the absence of P0 and P1 level instructions, dynamically calculates the optimal choice through a set model and makes decisions autonomously. It includes pure battery drive mode, basic photovoltaic priority mode, supercharging photovoltaic priority mode, standard kinetic energy recovery mode, congestion optimization mode, low temperature combined heating mode, and vehicle-to-everything (V2X) communication technology predictive cruise mode.

[0036] In some preferred embodiments, such as Figure 2 As shown, the model is a DQN model. DQN refers to a decision-making process in which a deep reinforcement learning agent trained offline interacts with the environment in real time when running online.

[0037] Step 1: The state perception and coding system encodes the current vehicle state into a state vector St, which serves as the input to the DQN model and accurately describes the vehicle's state at time t.

[0038] Step 2: Q-value calculation and action-related state vector St are input into the DQN neural network. The output layer of the network has multiple nodes, each node corresponds to a specific action (working mode), and its output value is the Q-value of that action.

[0039] Step 3: The optimal action selection system executes a simple decision-making logic: selecting the action with the highest Q value. Here, a* is the selected optimal working mode, which interacts with the environment to obtain a new state S′, stores it, and uses it for model updates. The dynamically updated weights are then fed back to the DQN neural network, and the process continues in a loop.

[0040] Regarding mode switching delay, in some preferred embodiments, the switching delay is defined according to different priorities: P0 or P1 level switching (hard real-time): Delay: ≤ 50 ms; Technical safeguards: Implemented via hardware interrupts or the highest priority real-time task. The entire process, from detecting a condition being met (such as a change in the airbag sensor signal) to the controller issuing a mode-switching command, must be completed within 50ms to ensure safety and immediate response.

[0041] P2 / P3 level handover (soft real-time): Delay: 100 ms - 200 ms Technical safeguards: As a background optimization task, this delay includes the entire process of sensor data fusion (0~20ms), model inference calculation (0~30ms), and power device pre-synchronization and soft start (0~50-150ms). The aim is to achieve seamless and smooth switching, avoiding power interruption or jerking.

[0042] In some preferred embodiments, the energy allocation mode is divided into P0, P1, P2, and P3 priority levels, establishing a hierarchical decision-making logic to ensure safety and real-time performance and avoid mode conflicts. Level P0 is executed unconditionally with priority, level P1 responds to driver intent, and level P2 or P3 is optimized by the system. Intelligent decision-making is achieved through a priority arbitrator, and the mode is switched immediately upon the occurrence of a safety event, thereby improving system reliability and response speed.

[0043] In some preferred embodiments, a weighting coefficient is calculated using a dynamic weighting algorithm based on the energy consumption prediction results and real-time collected vehicle environmental information. The weighting coefficient is adjusted in real time based on the energy distribution controller. The energy distribution controller includes a basic safety filter: based on whether the real-time collected vehicle environmental information reaches a set threshold, it determines whether to enter a certain fixed charging mode; Weighted Allocation Algorithm: If a fixed charging mode is not entered, the weighted allocation algorithm is entered. The weight coefficient is calculated by combining the energy consumption prediction results with the real-time collected vehicle parameters through a dynamic weight allocation algorithm.

[0044] In some preferred embodiments, by refining the dynamic weight allocation algorithm, including a basic safety filter and a weighted allocation algorithm, multi-layer optimization logic is introduced to adjust the weight coefficients in real time and avoid energy abuse. The algorithm updates the weights through a long short-term memory network and combines them with a battery protection threshold to optimize energy allocation while ensuring safety. For example, by dynamically adjusting α, β, and γ through efficiency factors, the system's adaptability is improved.

[0045] In some preferred embodiments, when no preset energy allocation mode triggering conditions are met, the weight coefficient is calculated by a dynamic weight allocation algorithm based on the energy consumption prediction results and the real-time collected vehicle environment information.

[0046] In some preferred embodiments, when no preset energy allocation mode triggering conditions are met, the method for obtaining the weighting coefficient includes: The baseline weights for photovoltaic power generation, power battery energy, and kinetic energy recovery are calculated. The battery baseline weight is obtained based on the current state of charge (SOC) of the battery, and the higher the SOC, the greater the weight. The photovoltaic baseline weight is obtained based on the ratio of the current photovoltaic power generation to the total power demand of the vehicle. The kinetic energy recovery baseline weight is obtained based on the ratio of the predicted kinetic energy recovery power to the total power demand. An adjustment factor based on the optimization objective is introduced, including a battery life factor and a system efficiency factor; The weighted value is obtained by multiplying the baseline weight of each energy source by the corresponding adjustment factor, and then normalized to obtain the weight coefficient.

[0047] For the weighting algorithms of α, β, and γ for power batteries, photovoltaics, and kinetic energy recovery, in some preferred embodiments, the system also sets rule-based emergency thresholds and boundary conditions to ensure safety and real-time performance. These thresholds are embedded as hard constraints into the optimization problem described above. 1. Battery protection threshold (highest priority hard constraint) Discharge protection trigger condition: SOC≤SOC min +δ soc ; Charging protection trigger condition: SOC ≥ SOC max -δ soc ; High-temperature derating trigger condition: T batt ≥T max -δ temp ; δsoc represents the battery's safety margin, typically a calibrated value; δtemp represents the temperature safety margin, also typically a calibrated value; and SOC represents the remaining charge of the power battery. min The minimum remaining charge of the power battery, SOC max T represents the maximum remaining capacity of the power battery. battFor the power battery temperature, T max Maximum temperature of the power battery.

[0048] The vehicle control system requests power demand from the energy distribution controller. The energy distribution controller then requests available power and current status from the photovoltaic system, energy recovery system, and battery management system (BMS). After receiving the response feedback, it calculates the corresponding α, β, and γ values ​​and initiates the final power distribution to the motor controller.

[0049] In some preferred embodiments, an energy distribution controller includes: First layer: Basic rule filter (mode enablement and security priority). This layer determines whether to enter a certain fixed mode based on the highest priority rule. The rules are as follows: Rule R1 (Battery Protection): If SOC <SOC low_threshold If the minimum threshold is set, then (α, β, γ) = (1.0, 0, 0) is set to force battery-driven operation and disable photovoltaic charging to protect battery life.

[0050] Rule R2 (Photovoltaic Invalid): If P solar <P solar_min (Minimum illumination threshold), then set β = 0.

[0051] Rule R3 (Forced Regeneration): If the brake pedal depth > calibrated threshold, then set γ = γ_max (maximum threshold).

[0052] Second layer: Weighted allocation algorithm. If the fixed pattern of the first layer is not triggered, the weighted allocation calculation is entered at this layer. The rules are as follows: (1) Calculate the baseline weight of each energy source Battery benchmark weight W batt_base This is a function based on SOC (State of Charge). The higher the SOC, the more likely the battery will be used. The specific formula is as follows: ; in, Set a maximum threshold for the battery; Set a threshold for the lowest possible battery level. These are the Changshu parameters that have been pre-set and adjusted experimentally.

[0053] Photovoltaic benchmark weight The specific formula for the ratio of current photovoltaic power generation to total power demand is as follows: ; in, Photovoltaic power generation capacity, This represents the total power required by the entire vehicle.

[0054] Recovering benchmark weights The specific formula for the ratio of predicted recovered energy to total power demand is as follows: ; in, To recover total power.

[0055] (2) Calculate the adjustment factor based on the optimization objective. Battery life factor To ensure the battery current is as smooth as possible, the specific formula is as follows: ; Where k2 is a pre-set constant parameter adjusted experimentally, and ΔP demand This represents the change in the total power demand of the entire vehicle. The system efficiency factor is used to prioritize the use of energy sources with higher instantaneous efficiency, and its specific formula is as follows: Photovoltaic factor: (Photovoltaic system efficiency factor); Recovery factor: (Recovery system efficiency factor); (3) Calculate the final weights and normalize them. Calculate the weighted sum: ; ; ; This refers to the total energy of the battery. Total photovoltaic energy, To recover total energy; Normalization yields the final assigned weights: S total =W batt_final + W solar_final + W regen_final ; α=W batt_final / S total ; β=W solar_final / S total ; γ=W regen_final / S total Among them, S total This represents the total weight.

[0056] In some preferred embodiments, energy recovery is triggered at the optimal time and with optimal power based on vehicle status, driver intent, and predicted road conditions. The triggering conditions are managed by a multi-level decision-making system, as follows: 1. Driver-triggered operation, this is the highest priority and responds in real time: (1) Brake pedal signal trigger: Condition: Brake pedal travel S brake > S brake_threshold_1 ; Where S brake_threshold_1 The first threshold for brake pedal travel is 5%-10% of the travel in an optional embodiment. Once the brake pedal is detected to be depressed, the system intervenes immediately. Initially, the braking force is primarily provided by the kinetic energy recovery system to maximize energy recovery. As the pedal travel increases, the system gradually increases the proportion of mechanical braking according to a preset braking force distribution curve to ensure braking efficiency and safety.

[0057] Output: The weight increases rapidly, reaching The maximum value is 1.0.

[0058] (2) Triggered by fully releasing the accelerator pedal ("one-pedal mode"): Condition: Accelerator pedal opening A accel = 0, and vehicle speed v>v min ; Among them, v min The minimum regenerative braking speed is set at 5 km / h in an optional embodiment to avoid excessive jerking during low-speed crawling. The system interprets this state as the driver's intention to decelerate, immediately triggering moderate-intensity regenerative braking to generate a noticeable drag sensation for deceleration and energy recovery.

[0059] Output: The weights are dynamically adjusted based on vehicle speed and a preset deceleration curve. In an optional embodiment, the weights are maintained. It is in the range of 0.4-0.6.

[0060] 2. Predictive strategy triggering, this is intelligent system triggering: This condition does not depend on driver operation, but is based on V2X and high-precision map information to activate the recycling system in advance, achieving "predictive" energy saving.

[0061] (1) Predictive triggering based on road conditions ahead: Condition: A scenario requiring deceleration is detected within 3 kilometers ahead via V2X.

[0062] When there is congestion ahead: if the system predicts that the average vehicle speed will decrease significantly, it will increase the weight of γ in advance and smoothly to achieve smooth deceleration and reduce the use of mechanical brakes; When there is a red light or stop sign ahead: the system calculates the distance between the current vehicle speed and the target stopping point, predicts an optimal deceleration curve, and controls the vehicle to coast and decelerate along this curve by adjusting the γ weight; When the speed limit changes or there is a curve: if the system anticipates that the vehicle needs to reduce its speed ahead, it will trigger recovery in advance to help the vehicle smoothly reduce to a safe speed.

[0063] (2) Predictive triggering based on geographic information: Conditions: High-precision map or V2I information showing the road slope ahead. ; in, The downhill slope threshold is -3% in an optional embodiment. The system predicts that the vehicle will accelerate under the action of gravity when going downhill. In order to avoid unnecessary acceleration and subsequent mechanical braking, the system will enter the high recovery mode in advance, use the recovery resistance to limit the vehicle speed, and continuously convert the gravitational potential energy into electrical energy.

[0064] Output: The weight can be set to In an optional embodiment, The value is above 0.7 and continues throughout the downhill section.

[0065] Example 2 A multimodal energy collaborative management method for new energy vehicles, comprising: The absorbed solar energy is converted into DC power by a DC-DC converter to obtain photovoltaic power generation energy; the status of the power battery is monitored and the power battery energy is output; the kinetic energy is recovered by the hub motor when the vehicle brakes or goes downhill, and is converted into DC power by the inverter to obtain kinetic energy recovery energy. By acquiring road condition information ahead through vehicle-to-everything (V2X) communication technology, and combining this information with a deep learning-based prediction model, the future energy consumption curve is predicted, thus obtaining the energy consumption prediction result. Based on the energy consumption prediction results and the real-time collected vehicle environmental information, the weight coefficient is calculated through a dynamic weight allocation algorithm. When the triggering conditions of the energy allocation mode are met, the weight coefficient is allocated according to the coefficient allocation rules corresponding to different energy allocation modes. Based on the weighting coefficients, the photovoltaic power generation energy, the power battery energy, and the kinetic energy recovery energy are weighted and allocated to obtain the allocated energy control signal and the total available energy. Based on the total available energy and the energy control signal, the electrical equipment is driven or the power battery is charged.

[0066] Example 3 A computer program product includes a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the method described in Embodiment 2.

[0067] Example 4 A new energy vehicle equipped with the system described in claim 1.

[0068] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A new energy vehicle multi-modal energy collaborative management system, characterized in that, Comprising: A multi-modal energy input module is used to obtain photovoltaic power generation energy, power battery energy and kinetic energy recovery energy; An energy distribution control module is used to obtain weight coefficients of photovoltaic power generation energy, power battery energy and kinetic energy recovery energy according to energy consumption prediction results and real-time collected vehicle environment information, to perform weighted distribution on photovoltaic power generation energy, power battery energy and kinetic energy recovery energy according to the weight coefficients, and to obtain an energy control signal after distribution and total available energy.

2. The multi-modal energy collaborative management system of a new energy vehicle according to claim 1, characterized in that: The formula for calculating the total available energy is: ; wherein, is the total available energy, is the power battery energy, is the photovoltaic power generation energy, is the kinetic energy recovery energy, , and are the weight coefficients of the power battery energy, the photovoltaic power generation energy and the kinetic energy recovery energy, respectively.

3. The multi-modal energy collaborative management system of a new energy vehicle according to claim 1, characterized in that: When the preset energy distribution mode trigger condition is met, the weight coefficients are distributed according to the coefficient distribution rules corresponding to different energy distribution modes.

4. The multi-modal energy collaborative management system of a new energy vehicle according to claim 1, characterized in that: When none of the preset energy distribution mode trigger conditions is met, the weight coefficients are obtained by a dynamic weight distribution algorithm according to the energy consumption prediction results and the real-time collected vehicle environment information.

5. The multi-modal energy collaborative management system of a new energy vehicle according to claim 1, characterized in that: The method for obtaining photovoltaic power generation energy, power battery energy and kinetic energy recovery energy comprises: The absorbed solar energy is converted into direct current electric energy through a DCDC converter to obtain photovoltaic power generation energy; the state of the power battery is monitored and the power battery energy is output; the kinetic energy is recovered through the wheel hub motor when the vehicle brakes or goes downhill, and is converted into direct current electric energy through an inverter rectifier to obtain kinetic energy recovery energy.

6. The multi-modal energy collaborative management system of a new energy vehicle according to claim 1, characterized in that: The preset energy distribution mode trigger condition factors include vehicle speed, illumination, battery temperature, brake pedal, road congestion, slope, throttle opening, rainstorm and sandstorm detection, airbag and manual emergency switch triggering.

7. The multi-modal energy collaborative management system of a new energy vehicle according to claim 4, characterized in that: When none of the preset energy distribution mode trigger conditions is met, the method for obtaining the weight coefficients comprises: A reference weight of photovoltaic power generation energy, power battery energy and kinetic energy recovery energy is calculated, the battery reference weight is obtained based on the current battery state of charge, the higher the SOC, the greater the weight tendency; the photovoltaic reference weight is obtained based on the ratio of the current photovoltaic power generation power to the total demand power of the vehicle; the recovery reference weight is obtained based on the ratio of the predicted kinetic energy recovery power to the total demand power; An adjustment factor based on an optimization target is introduced, including a battery life factor and a system efficiency factor; The reference weight of each energy is multiplied by the corresponding adjustment factor to obtain a weighted value, which is normalized to obtain the weight coefficients.

8. A new energy vehicle multi-modal energy collaborative management method, characterized in that, Comprising: The absorbed solar energy is converted into direct current electric energy through a DCDC converter to obtain photovoltaic power generation energy; The state of the power battery is monitored, and power battery energy is output; kinetic energy is recovered by the wheel hub motor when the vehicle brakes or goes downhill, and is converted into direct current energy via an inverter rectifier to obtain kinetic energy recovery energy; Road condition information in front is obtained through vehicle-to-everything communication technology, and a future energy consumption curve is predicted based on a deep learning prediction model combined with the road condition information in front to obtain an energy consumption prediction result; According to the energy consumption prediction result and the real-time collected vehicle environment information, a weight coefficient is calculated through a dynamic weight distribution algorithm, and when a trigger condition of an energy distribution mode is met, the weight coefficient is distributed according to a coefficient distribution rule corresponding to different energy distribution modes; According to the weight coefficient, the photovoltaic power generation energy, the power battery energy and the kinetic energy recovery energy are weighted and distributed to obtain an energy control signal after distribution and total available energy.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by a processor to implement the steps of the method in claim 8.

10. A new energy vehicle, characterized in that, The system as claimed in any one of claims 1 to 7 is mounted.

Images