A fast-charging-oriented power battery pre-cooling closed-loop decision method and device

Abstract

The application is suitable for the electric vehicle thermal management field, and provides a power battery pre-cooling closed-loop decision method and device for fast charging, which comprises the following steps: obtaining the state information of the vehicle, candidate fast charging station, power battery, environment, passenger cabin and thermal management system; determining the target fast charging station; predicting the power battery's SOC at the station; solving the target pre-cooling temperature at the station according to the predicted arrival time, SOC at the station, candidate power battery temperature at the station, fast charging time, pre-cooling energy consumption and passenger cabin comfort influence; generating the power battery temperature reference track according to the temperature, and tracking control is performed on the actuators such as compressors, cooling liquid pumps, fans and valves through the hierarchical controller; and rolling updating the target fast charging station, SOC at the station, target pre-cooling temperature at the station and control amount according to the state change, and forming a closed-loop re-planning. The application can shorten the fast charging time, comprehensively optimize the fast charging benefit, pre-cooling energy consumption and passenger cabin comfort, and has good working condition adaptability.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle thermal management technology, and particularly relates to a closed-loop decision-making method and device for pre-cooling power batteries for fast charging. Background Technology

[0002] In electric vehicles, the power battery, as the core energy storage unit, directly impacts the vehicle's range, power response, and charging efficiency. However, during fast charging, the power battery typically withstands a large charging current, which can lead to increased ohmic heat, polarization heat, and side reaction heat within the battery. This not only increases the risk of thermal runaway but may also accelerate battery aging and shorten its lifespan. Furthermore, when the power battery temperature is too high, the battery management system often needs to limit the charging power, thereby extending the charging time. Therefore, pre-cooling the power battery before the vehicle arrives at the target fast charging station, ensuring it is within the suitable temperature range for fast charging upon arrival, has become an important technical approach to improve fast charging efficiency and optimize the charging experience.

[0003] In related technologies, most existing solutions focus on thermal management control based on known navigation destinations or predetermined charging scenarios, or emphasize the dynamic adjustment of cooling execution parameters. While such solutions can improve temperature control performance under specific operating conditions to some extent, they are still mainly limited to local temperature control or execution-level control, lacking global planning for pre-cooling targets oriented towards fast charging energy replenishment needs. Specifically, existing technologies typically initiate thermal management control only after the vehicle destination or charging scenario is determined, lacking a mechanism for joint decision-making between candidate fast charging station selection and pre-cooling strategies. Furthermore, they do not adequately consider dynamic factors such as the available power of candidate fast charging stations, waiting time, detour costs, and changes in station status. In addition, there is a lack of closed-loop decision updates between pre-cooling planning and thermal management execution, making it difficult to simultaneously address the needs of fast charging benefits, pre-cooling energy consumption, and multi-actuator collaborative control.

[0004] Therefore, there is an urgent need to provide a closed-loop decision-making method for pre-cooling power batteries for fast charging, which can integrate candidate fast charging station selection, arrival status prediction, target arrival pre-cooling temperature planning, and thermal management collaborative execution into a unified decision-making process. This allows the pre-cooling strategy to be dynamically adjusted according to the station status, route status, and vehicle status, thereby improving the matching between pre-cooling behavior and fast charging energy replenishment demand. Summary of the Invention

[0005] The purpose of this invention is to provide a closed-loop decision-making method and apparatus for pre-cooling power batteries for fast charging, aiming to solve the problems mentioned in the background art.

[0006] The present invention is implemented as follows: a closed-loop decision-making method for pre-cooling of power batteries for fast charging includes the following steps:

[0007] Step 1: Status information acquisition, including vehicle operation information, candidate fast charging station information, power battery status information, environmental status information, passenger compartment status information, and thermal management system operation status information;

[0008] Step 2: Target fast charging station identification. Multiple candidate fast charging stations are screened and evaluated to identify the target fast charging station.

[0009] Step 3: Arrival Status Prediction. After determining the target fast charging station, the arrival status of the vehicle is predicted online based on the current status of the vehicle and the predicted operating conditions. The arrival status includes at least the SOC (State of Charge) of the power battery.

[0010] Step 4: Solve for the target arrival pre-cooling temperature. Based on the estimated arrival time, arrival SOC, candidate arrival power battery temperature, estimated fast charging time, pre-cooling energy consumption, and impact on passenger cabin comfort, solve for the target arrival pre-cooling temperature.

[0011] Step 5: Pre-cooling target tracking and coordinated control. A temperature reference trajectory of the power battery is generated based on the pre-cooling temperature of the target arrival station. Through a hierarchical controller, multiple thermal management actuators are coordinated and controlled so that the power battery temperature follows the temperature reference trajectory.

[0012] Step 6: Rolling update and closed-loop replanning. During vehicle operation, based on changes in the status of candidate fast charging stations, route status, vehicle status, and thermal management system status, the target fast charging station, arrival SOC, target arrival pre-cooling temperature, and thermal management actuator control quantities are updated to form a closed-loop decision-making process for power battery pre-cooling in fast charging scenarios.

[0013] In a further technical solution, in step 1, the vehicle operation information includes the vehicle's current location, navigation route information, real-time traffic information, and vehicle driving condition information; the candidate fast charging station information includes the candidate fast charging station location, available charging power, service status, and estimated waiting time; the power battery status information includes the current power battery temperature and current SOC; the environmental status information includes the ambient temperature; the passenger compartment status information includes the passenger compartment temperature; and the thermal management system operation status information includes the compressor operation status, coolant pump status, fan status, valve status, and current thermal management circuit temperature information.

[0014] A further technical solution, the specific steps of step 2 are as follows:

[0015] Based on the vehicle's current location and the location of candidate fast charging stations, and combined with road topology, traffic travel time, service capacity of candidate fast charging stations, and waiting time within the station, a comprehensive evaluation index is constructed. The comprehensive evaluation index includes at least one of the following: the driving time required for the vehicle to reach the candidate fast charging station, the detour distance or detour time, the available charging power of the candidate fast charging station, and the expected waiting time within the station.

[0016] For each candidate fast charging station, calculate the remaining driving distance, remaining driving time, and estimated arrival time required for the vehicle to reach the candidate fast charging station, and determine the estimated charging time of the vehicle at the candidate fast charging station based on the available charging power and estimated waiting time of the candidate fast charging station.

[0017] Based on the comprehensive evaluation results of each candidate fast charging station, the candidate fast charging station whose comprehensive evaluation results meet the preset conditions is selected as the target fast charging station.

[0018] When the status or path status of a candidate fast charging station changes, the candidate fast charging station is re-evaluated, and the target fast charging station is updated based on the re-evaluation results.

[0019] A further technical solution, the specific steps of step 3 are as follows:

[0020] An energy consumption prediction model is built based on historical driving data and current operating conditions to predict the SOC of the power battery when the vehicle arrives at the target fast charging station.

[0021] The energy consumption prediction model is constructed using a long short-term memory network. The input of the long short-term memory network includes one or more time series features of the vehicle's current speed, acceleration, road gradient, ambient temperature, occupant thermal comfort requirements, and traction power and thermal management power within a preset historical period. The output is the trend of vehicle energy consumption or SOC changes in the future time domain.

[0022] The prediction and update of arrival SOC is carried out in a rolling manner. That is, at each planning time, the current status and route information are collected again, and the remaining driving energy consumption and thermal management energy consumption are re-estimated to update the arrival SOC prediction value.

[0023] A further technical solution, the specific steps of step 4 are as follows:

[0024] The temperature of the candidate arriving power battery is used as an optimization variable, and a comprehensive cost function is constructed based on the fast charging time, acceptable charging power, pre-cooling energy consumption and passenger cabin comfort impact corresponding to different candidate arriving power battery temperatures.

[0025] Under the premise of meeting the fast charging temperature requirements of the power battery, the comprehensive cost function is solved to obtain the target pre-cooling temperature at the destination under the current operating conditions;

[0026] The comprehensive cost function is solved using a particle swarm optimization algorithm. Each particle is initialized with a different candidate arrival battery temperature. By iteratively updating the velocity and position of the particles, the particle swarm converges to the region with the minimum comprehensive cost in the candidate arrival battery temperature space. The candidate arrival battery temperature corresponding to the globally optimal particle is output as the target arrival pre-cooling temperature.

[0027] In a further technical solution, in step 4, the comprehensive cost function is constructed in the following form:

[0028]

[0029] in, For the comprehensive cost function, This indicates the increase in expected charging time caused by the temperature of the candidate arriving power battery relative to the ideal fast charging condition. This represents the total energy consumed by the thermal management system to achieve the desired temperature of the candidate battery upon arrival, from the current time to the arrival time. This indicates the predicted cabin temperature in the time domain. Indicates the target temperature in the crew cabin. , and These are the weighting coefficients;

[0030] State-dependent adaptive weights are introduced into the comprehensive cost function: when the expected arrival time is short and the current battery temperature is higher than the suitable temperature range for fast charging, the weights are increased. When the expected arrival time is long and the demand for passenger cabin comfort is high, increase... When the expected arrival SOC is low, increase .

[0031] In a further technical solution, in step 5, the hierarchical controller includes an upper-layer tracking coordination controller and a lower-layer execution controller;

[0032] The upper-level tracking and coordination controller is used to receive the target arrival pre-cooling temperature and, in combination with the current power battery temperature, passenger compartment temperature, ambient temperature and thermal management system status, generate a power battery temperature reference trajectory suitable for subsequent long-term time domain.

[0033] The lower-level execution controller is used to perform coordinated optimization control of multiple thermal management actuators based on real-time status feedback within a short prediction time domain, so as to achieve the tracking of the power battery temperature and the passenger compartment temperature to the temperature reference trajectory.

[0034] The thermal management actuator includes a compressor, a coolant pump, a condenser fan, a blower, and valves.

[0035] A further technical solution is that the hierarchical controller adopts a hierarchical model predictive control framework that combines long-time domain control and short-time domain control;

[0036] During long-term control, based on the target arrival pre-cooling temperature, the estimated arrival time, the current temperature of the power battery, and the current temperature of the passenger compartment, the temperature changes of the power battery and the passenger compartment during subsequent driving are predicted, and a power battery temperature reference trajectory is generated in the subsequent control time domain.

[0037] During short-time domain control, the control reference quantities of the compressor, coolant pump, fan and valve are solved based on the temperature reference trajectory, the current temperature of the power battery, the current temperature of the passenger compartment and the current operating status of the thermal management system.

[0038] To achieve hierarchical model predictive control, reduced-order predictive models matching the hierarchical control requirements are established: for the long-term temperature coordination process, a reduced-order thermal model is constructed to characterize the power battery temperature, passenger compartment temperature, and system cooling distribution relationship; for the short-term execution control process, a reduced-order control model is constructed to characterize the dynamic characteristics of the thermal management actuator and the response relationship of the refrigerant circuit and coolant circuit.

[0039] In a further technical solution, step 6 includes the following specific steps:

[0040] During the process of the vehicle driving toward the target fast charging station, the status of the target fast charging station, the vehicle path status, the power battery status, and the execution status of the thermal management system are monitored in real time.

[0041] When the detected changes in the service capacity of the target fast charging station, changes in the expected waiting time, changes in the path traffic status, deviations in the power battery status, or deviations in the execution of the thermal management system meet the preset conditions, the rolling replanning mechanism is triggered, and steps 2 to 5 are re-executed.

[0042] The vehicle's current location, power battery temperature, passenger compartment temperature, current SOC, ambient temperature, and thermal management system operating status are reacquired according to a preset cycle. Based on the latest road information and candidate fast charging station status, the remaining path length, estimated arrival time, arrival SOC, and target arrival pre-cooling temperature are recalculated, and the power battery temperature reference trajectory for subsequent stages is regenerated.

[0043] Another objective of this invention is to provide a closed-loop decision-making device for pre-cooling power batteries for fast charging, based on the above method, comprising:

[0044] The data acquisition module and data storage module are used to collect and store information on candidate fast charging stations, power battery status, environmental status, passenger compartment status, and vehicle operating conditions.

[0045] The candidate fast charging station selection module is used to construct a comprehensive evaluation index based on the vehicle's current location, candidate fast charging station locations, navigation route information, road topology, traffic travel time, available charging power of candidate fast charging stations, service status, and estimated waiting time, and select the candidate fast charging station with the best comprehensive benefit or the lowest comprehensive cost as the target fast charging station.

[0046] The arrival state of charge prediction module is used to predict the SOC of a vehicle when it arrives at the target fast charging station online.

[0047] The charging revenue calculation module and the pre-cooling target solution module are used to calculate the fast charging revenue and pre-cooling cost corresponding to different candidate arrival temperatures, and solve the target arrival pre-cooling temperature online.

[0048] The hierarchical control module is used to perform upper-level temperature planning and lower-level execution control solving. The upper-level temperature planning generates a long-term time-domain temperature reference trajectory based on the target arrival pre-cooling temperature. The lower-level execution control solving solves the thermal management actuator control reference quantity based on the temperature reference trajectory and real-time status.

[0049] The actuator control module is used to coordinate the thermal management actuator to perform temperature control actions, so that the power battery temperature changes according to the temperature reference trajectory.

[0050] The rolling update module is used to trigger replanning when the status of candidate fast charging stations, path status, or vehicle status changes, and to re-call the candidate fast charging station selection module to the hierarchical control module for updating.

[0051] The present invention provides a closed-loop decision-making method and apparatus for pre-cooling power batteries for fast charging, which has the following advantages:

[0052] (1) Significantly shorten fast charging time: By pre-solving the optimal target pre-cooling temperature during driving, the power battery is in a suitable fast charging temperature range when it arrives at the fast charging station, avoiding the limitation of charging power caused by high temperature, and effectively shortening the actual fast charging time.

[0053] (2) Comprehensive optimization of fast charging benefits, pre-cooling energy consumption and passenger cabin comfort: Construct a comprehensive cost function that includes the expected increase in charging time, thermal management system energy consumption and passenger cabin temperature deviation, and introduce state-related adaptive weights to achieve adaptive comprehensive optimal decision under different operating conditions.

[0054] (3) Achieve precise temperature tracking and coordinated control: A hierarchical controller including an upper-level tracking and coordination controller and a lower-level execution controller is adopted to generate a long-term temperature reference trajectory and coordinate actuators such as compressors, coolant pumps, fans, and valves in the short-term domain, thereby improving control accuracy and system energy efficiency.

[0055] (4) Possesses closed-loop rolling replanning capability: Real-time monitoring of fast charging station status, path status, power battery status and thermal management system execution status, triggering rolling updates and closed-loop replanning, so that the pre-cooling strategy always matches the latest operating conditions, enhancing robustness and adaptability.

[0056] (5) Reduce the energy consumption of the thermal management system and extend the driving range: avoid excessive pre-cooling and introduce energy consumption costs in the lower control layer so that the pre-cooling process can operate in a more energy-efficient way, reducing the consumption of the vehicle's electricity.

[0057] (6) Improve fast charging safety and battery life: Ensure that the temperature of the power battery does not exceed the upper limit of fast charging before arrival at the station, reduce the safety risks caused by high temperature fast charging, and extend the cycle life of the power battery.

[0058] (7) Strong engineering applicability: It can be deployed based on existing vehicle sensors, bus communication and cloud data. The algorithms involved (long short-term memory network, particle swarm optimization, model predictive control) have good real-time performance and portability value. Attached Figure Description

[0059] Figure 1 A flowchart of a closed-loop decision-making method for pre-cooling power batteries for fast charging, provided as an embodiment of the present invention;

[0060] Figure 2 A schematic diagram illustrating the process of selecting a target fast charging station;

[0061] Figure 3 The flowchart shows the workflow of the arrival state of charge prediction module.

[0062] Figure 4 A schematic diagram for assessing the benefits of pre-cooling;

[0063] Figure 5 This is a schematic diagram of a precooling coordinated tracking control structure based on hierarchical model predictive control.

[0064] Figure 6 This is a schematic diagram of a closed-loop decision-making device for pre-cooling power batteries for fast charging, provided in an embodiment of the present invention.

[0065] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0066] In the attached diagram: Data acquisition module 101; Data storage module 102; Candidate fast charging station selection module 103; Arrival state of charge prediction module 104; Charging revenue calculation module 105; Pre-cooling target solution module 106; Hierarchical control module 107; Actuator control module 108; Rolling update module 109; Computation unit 201; Read-only memory 202; Random access memory 203; Bus 204; I / O interface 205; Vehicle controller 206; Input 207; Output 208; Data acquisition 209; Vehicle sensor 210. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0068] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0069] like Figures 1-5 As shown, this invention provides a closed-loop decision-making method for pre-cooling power batteries for fast charging. This method aims to dynamically plan the optimal pre-cooling strategy for the power battery as the vehicle travels to the target fast charging station, and ensures its precise execution through closed-loop control. Specifically, it includes the following steps:

[0070] Step 1: Obtain status information;

[0071] First, obtain vehicle operation information, candidate fast charging station information, power battery status information, environmental status information, passenger compartment status information, and thermal management system operation status information.

[0072] The vehicle operation information includes at least the vehicle's current location, navigation route information, real-time traffic information, and vehicle driving condition information; the candidate fast charging station information includes at least the candidate fast charging station location, available charging power, service status, and estimated waiting time; the power battery status information includes at least the current power battery temperature and current SOC; the environmental status information includes at least the ambient temperature; the passenger compartment status information includes at least the passenger compartment temperature; and the thermal management system operation status information includes at least the compressor operation status, coolant pump status, fan status, valve status, and current thermal management circuit temperature information.

[0073] In device-based implementations (such as...) Figure 6The data acquisition module 101 and data storage module 102 are used to acquire and store state variables and environmental variables related to fast charging pre-cooling in real time. The state variables include at least: vehicle current location, vehicle speed, acceleration, power battery temperature, power battery SOC, passenger compartment temperature, current operating status of the thermal management system, and air conditioning load status. The environmental variables include at least: ambient temperature, location and fast charging capability of candidate fast charging stations, navigation path information, remaining path length, and road congestion status.

[0074] The power battery temperature is directly collected or estimated by the battery management system (BMS), the power battery state of charge (SOC) is provided by the BMS, and the passenger compartment temperature is obtained by in-cabin temperature sensors. The current operating status of the thermal management system includes compressor speed, coolant pump speed, fan speed, valve opening, refrigerant circuit pressure, and coolant circuit temperature. Candidate fast charging station information is provided by the vehicle navigation system, cloud service platform, or station-side information platform.

[0075] Step 2: Identify the target fast charging station;

[0076] Multiple candidate fast charging stations are screened and evaluated to determine the target fast charging station. Specifically, based on the vehicle's current location and the locations of candidate fast charging stations, combined with road topology, traffic travel time, service capacity of candidate fast charging stations, and waiting time within the stations, multiple candidate fast charging stations are comprehensively evaluated to determine the target fast charging station.

[0077] When determining target fast charging stations, the pre-cooling factor of the power battery is incorporated into the station selection process to construct a comprehensive evaluation index. This comprehensive evaluation index includes at least one of the following: the driving time required for the vehicle to reach the candidate fast charging station, the detour distance or time, the available charging power of the candidate fast charging station, and the estimated waiting time within the station. For each candidate fast charging station, the remaining driving distance, remaining driving time, and estimated arrival time required for the vehicle to reach that station are calculated. Based on the available charging power and estimated waiting time of the candidate fast charging station, the estimated charging time for the vehicle at that station is determined. Based on the comprehensive evaluation results corresponding to each candidate fast charging station, the candidate fast charging station whose comprehensive evaluation results meet preset conditions (e.g., optimal comprehensive benefit or minimum comprehensive cost) is selected as the target fast charging station.

[0078] For each candidate fast-charging station, the remaining path length and estimated arrival time of the vehicle from its current location to the station are first calculated (based on navigation route length, speed limit information, real-time traffic flow data, and the vehicle's current driving status). Then, evaluation indicators for the candidate fast-charging station are constructed by combining the station's rated charging power, current availability, estimated waiting time within the station, and the potential thermal state of the battery upon arrival. These evaluation indicators include at least the total arrival time, additional detour costs, the feasibility of the battery temperature reaching a suitable fast-charging range upon arrival, and the available charging capacity of the candidate fast-charging station. For different candidate fast-charging stations, the pre-cooling implementation difficulty and fast-charging benefits under current operating conditions are estimated, and based on these estimates, the stations are ranked and selected.

[0079] When the status of a candidate fast charging station or the route changes (e.g., a candidate fast charging station stops operating, the expected waiting time increases, or the congestion on the route to the target fast charging station increases), the candidate fast charging stations are re-evaluated, and the target fast charging station is updated based on the re-evaluation results.

[0080] Step 3: Arrival status prediction;

[0081] After identifying the target fast charging station, the vehicle's arrival status is predicted online based on its current state and predicted operating conditions. The arrival status includes at least the state of charge (SOC) of the power battery.

[0082] Specifically, an energy consumption prediction model is constructed based on historical driving data and current operating conditions to predict the SOC of the power battery when the vehicle arrives at the target fast charging station.

[0083] In one specific implementation, the energy consumption prediction model is constructed using a Long Short-Term Memory (LSTM) network. The inputs to the LSTM network include one or more time-series features selected from the vehicle's current speed, acceleration, road gradient, ambient temperature, occupant thermal comfort requirements, and traction power and thermal management power over a preset historical period. Optionally, it also includes current State of Charge (SOC), passenger compartment load information, and remaining path information provided by navigation. The output is the future time-domain trend of vehicle energy consumption or SOC change, and the predicted SOC of the power battery when the vehicle arrives at the target fast-charging station is determined based on the output.

[0084] The prediction and update of arrival SOC is performed in a rolling manner. That is, at each planning time of the vehicle, the system re-collects the current state and route information, and re-estimates the remaining driving energy consumption and thermal management energy consumption to update the arrival SOC prediction value. By incorporating the impact of thermal management power into the arrival state estimation process, the arrival SOC prediction is made to better reflect the actual changes in driving conditions.

[0085] Step 4: Determine the target pre-cooling temperature upon arrival;

[0086] Based on the estimated arrival time, arrival SOC, candidate arrival power battery temperature, estimated fast charging time, pre-cooling energy consumption, and impact on passenger cabin comfort, the target arrival pre-cooling temperature is calculated.

[0087] Specifically, the temperature of the candidate arriving power batteries is used as an optimization variable. A comprehensive cost function is constructed based on the fast charging time, acceptable charging power, pre-cooling energy consumption, and impact on passenger cabin comfort corresponding to different candidate arriving power battery temperatures. Under the premise of meeting the fast charging temperature requirements of the power batteries, the comprehensive cost function is solved to obtain the target arriving pre-cooling temperature under the current operating conditions.

[0088] In one specific implementation, multiple candidate battery temperatures for arrival at the target charging station are first selected as a search space within a preset temperature range. These candidate battery temperatures are located near a suitable fast-charging temperature range, such as discrete or continuous temperature values ​​between 10°C and 35°C. For each candidate battery temperature, the average acceptable charging power and estimated charging time after the battery enters the fast-charging phase at that candidate battery temperature are estimated by combining the expected arrival SOC, the rated power of the target fast-charging station, the battery charging power boundary, and the preset charging cutoff SOC. Simultaneously, based on the pre-cooling process during the vehicle's journey from the current moment to the target fast-charging station, the energy consumption of the thermal management system required to bring the battery to the candidate battery temperature is calculated, and the impact of the pre-cooling process on passenger cabin comfort is estimated. Then, the expected charging time, thermal management system energy consumption, and passenger cabin temperature deviation are incorporated into a comprehensive cost function.

[0089] The comprehensive cost function can be constructed in the following form:

[0090]

[0091] in, For the comprehensive cost function, This indicates the increase in expected charging time caused by the temperature of the candidate arriving power battery relative to the ideal fast charging condition. This represents the total energy consumed by the thermal management system to achieve the desired temperature of the candidate battery upon arrival, from the current time to the arrival time. This indicates the predicted cabin temperature in the time domain. Indicates the target temperature in the crew cabin. , and represents the weighting coefficients. This comprehensive cost function allows for the simultaneous consideration of fast charging performance, pre-cooling energy consumption, and passenger cabin comfort during the same solution process.

[0092] The particle swarm optimization algorithm is used to solve the comprehensive cost function. Specifically: each particle is initialized with a different candidate arriving battery temperature, and each particle is assigned a corresponding velocity and position; for each particle's corresponding candidate arriving battery temperature, the arrival SOC prediction result, charging revenue calculation result, and pre-cooling energy consumption are sequentially called to obtain the corresponding comprehensive cost value; the particle's velocity and position are updated according to the current particle's historical best position and global best position, so that the particle swarm gradually converges to the region with the minimum comprehensive cost in the candidate arriving battery temperature space; when the preset number of iterations or convergence conditions are reached, the candidate arriving battery temperature corresponding to the global best particle is output as the target arriving pre-cooling temperature at the current moment (i.e., the optimal pre-cooling target temperature).

[0093] In some implementations, to further enhance adaptability to different operating conditions, state-dependent adaptive weights are introduced into the comprehensive cost function. Specifically: when the expected arrival time is short and the current battery temperature is higher than the suitable temperature range for fast charging, the weight of the battery temperature deviation term upon arrival is increased (i.e., When the expected arrival time is long and the demand for passenger cabin comfort is high, the weight of the passenger cabin comfort item should be increased (i.e., When the expected SOC at arrival is low, increase the weight of the thermal management energy consumption item (i.e., By using the above weight adjustment method, the online solution process for the precooling target can adaptively adjust the optimization focus according to different working conditions.

[0094] Step 5: Pre-cooling target tracking and collaborative control;

[0095] Based on the target arrival pre-cooling temperature, a temperature reference trajectory for the power battery is generated. Through a hierarchical controller, thermal management actuators such as compressors, coolant pumps, fans, and valves are coordinated and controlled to ensure that the power battery temperature follows the temperature reference trajectory.

[0096] The hierarchical controller includes an upper-level tracking and coordination controller and a lower-level execution controller.

[0097] Upper-level tracking and coordination controller: used to receive the target arrival pre-cooling temperature and, in combination with the current power battery temperature, passenger compartment temperature, ambient temperature and thermal management system status, generate a temperature reference trajectory suitable for subsequent long-term time domain.

[0098] Lower-level execution controller: Within a short prediction time domain, it performs coordinated optimization control on multiple thermal management actuators based on real-time status feedback to achieve tracking of the power battery temperature and passenger compartment temperature along a temperature reference trajectory. The thermal management actuators include compressors, coolant pumps, condenser fans, blowers, and valves. Based on the temperature reference trajectory generated by the upper-level tracking and coordination controller, the lower-level execution controller coordinates the control of compressor speed, coolant pump speed, fan speed, and valve opening, ensuring the refrigerant circuit, power battery coolant circuit, and passenger compartment air conditioning circuit operate in harmony. This allows the power battery temperature to change according to the reference trajectory and improves system energy efficiency.

[0099] In one implementation, the hierarchical control employs a hierarchical model predictive control framework that combines long-term and short-term control. During long-term control, based on the target arrival pre-cooling temperature, estimated arrival time, current battery temperature, and current passenger compartment temperature, the temperature changes of the battery and passenger compartment during subsequent travel are predicted, and a battery temperature reference trajectory is generated for the subsequent control time domain. During short-term control, based on the temperature reference trajectory, current battery temperature, current passenger compartment temperature, and the current operating state of the thermal management system, control reference quantities for thermal management actuators such as the compressor, coolant pump, fan, and valves are determined.

[0100] In a further implementation, to achieve the hierarchical model predictive control, reduced-order predictive models matching the hierarchical control requirements are established: for the temperature coordination process on a longer timescale, a reduced-order thermal model is constructed to characterize the power battery temperature, passenger compartment temperature, and system cooling distribution; for the execution control process on a shorter timescale, a reduced-order control model is constructed to characterize the dynamic characteristics of the thermal management actuator and the response relationships of the refrigerant circuit and coolant circuit. Based on this, the temperature tracking error, system energy consumption, and control quantity changes are collectively constructed into a hierarchical model predictive control optimization problem, which is then transformed into a comprehensive cost function minimization problem for solution.

[0101] In some implementations, a robustness enhancement mechanism is introduced when performing pre-cooling target temperature tracking: the weights of the battery temperature tracking item and the passenger compartment comfort item are adjusted based on changes in the battery temperature, passenger compartment temperature, and thermal management system operating status. When the battery temperature deviates significantly from the reference trajectory, the weight of the battery temperature tracking item is increased; when the passenger compartment temperature deviates from the comfortable temperature range, the weight of the passenger compartment comfort item is increased. This allows the hierarchical controller to adjust its control priorities promptly when driving conditions, vehicle status, and thermal management system status change.

[0102] Step 6: Rolling updates and closed-loop replanning;

[0103] During vehicle operation, the target fast charging station, arrival SOC, target arrival pre-cooling temperature, and thermal management system control quantities are updated based on changes in the status of candidate fast charging stations, route status, vehicle status, and thermal management system status, thereby forming a closed-loop decision-making process for power battery pre-cooling in fast charging scenarios.

[0104] Specifically, during the vehicle's journey to the target fast charging station, the status of the target fast charging station, the vehicle's route status, the power battery status, and the execution status of the thermal management system are monitored in real time. When changes in the target fast charging station's service capacity, the estimated waiting time, the route's traffic conditions, the power battery's status deviation, or the thermal management system's execution deviation are detected and meet preset conditions, a rolling replanning mechanism is triggered, and steps 2 to 5 above are re-executed.

[0105] In one implementation, the system employs a rolling update method: it re-acquires the vehicle's current location, current battery temperature, current passenger compartment temperature, current state of charge (SOC), ambient temperature, and the thermal management system's operating status according to a preset cycle. Based on the latest road information and candidate fast-charging station status, it recalculates the remaining path length, estimated arrival time, arrival SOC, evaluation results for each candidate temperature, and the target arrival pre-cooling temperature. When the target fast-charging station, road congestion level, vehicle energy consumption status, or thermal management system's execution capability changes, it regenerates the battery temperature reference trajectory for subsequent stages based on the updated status information and controls the thermal management system's execution through a hierarchical controller.

[0106] like Figure 3 and Figure 6 As shown, another embodiment of the present invention provides a closed-loop decision-making device for pre-cooling power batteries for fast charging. Based on the above method, it includes: a data acquisition module 101, a data storage module 102, a candidate fast charging station selection module 103, an arrival state of charge prediction module 104, a charging revenue calculation module 105, a pre-cooling target solution module 106, a hierarchical control module 107, an actuator control module 108, and a rolling update module 109.

[0107] The data acquisition module 101 and data storage module 102 are used to collect and store candidate fast charging station information, battery status (temperature, SOC), environmental status (temperature), passenger compartment status (temperature) and vehicle operating information (vehicle speed, acceleration, navigation path, etc.).

[0108] The candidate fast charging station selection module 103 is used to construct a comprehensive evaluation index (including driving time, detour cost, available charging power, and estimated waiting time) based on the vehicle's current location, candidate fast charging station location, navigation route information, road topology, traffic travel time, available charging power of candidate fast charging stations, service status, and estimated waiting time, and select the candidate fast charging station with the best comprehensive benefit or the lowest comprehensive cost as the target fast charging station.

[0109] The arrival state of charge prediction module 104 is used to predict the SOC of the vehicle when it arrives at the target fast charging station online. Preferably, an LSTM time series prediction model is used, with inputs including current vehicle speed, acceleration, road gradient, ambient temperature, historical traction power, and thermal management power, and outputting the predicted SOC value upon arrival.

[0110] The charging revenue calculation module 105 and the pre-cooling target solution module 106 are used to calculate the fast charging revenue (estimated fast charging time, acceptable charging power) and pre-cooling cost (thermal management energy consumption, impact on passenger cabin comfort) corresponding to different candidate arrival temperatures, and to solve for the target arrival pre-cooling temperature (optimal pre-cooling target temperature) online. The comprehensive cost function can be solved using a particle swarm optimization algorithm.

[0111] The hierarchical control module 107 is used to perform upper-level temperature planning and lower-level execution control solving. The upper-level temperature planning generates a long-term time-domain temperature reference trajectory based on the target arrival pre-cooling temperature, and the lower-level execution control solving solves the thermal management actuator control reference quantity based on the temperature reference trajectory and real-time status.

[0112] The actuator control module 108 is used to coordinate the thermal management actuators such as the compressor, coolant pump, condenser fan, blower and valve to perform temperature control actions, so that the power battery temperature changes according to the temperature reference trajectory.

[0113] The rolling update module 109 is used to trigger replanning when the status of the candidate fast charging station, the path status, or the vehicle status changes, and to call the candidate fast charging station selection module 103 to the hierarchical control module 107 for updating again.

[0114] In a preferred embodiment of the present invention, data transmission between modules is achieved using wired or wireless communication.

[0115] The wired communication methods include CAN bus, LIN bus, FlexRay bus, and vehicle Ethernet, etc.

[0116] The wireless communication methods include cellular communication, Wi-Fi, Bluetooth, or vehicle-to-everything (V2X) communication. Information on candidate fast-charging stations, traffic information, and navigation information from cloud platforms or external service platforms is transmitted to the vehicle controller via a wireless communication link.

[0117] The control status, temperature reference trajectory, and thermal management actuator commands inside the vehicle are transmitted in real time via the vehicle bus or vehicle Ethernet.

[0118] like Figure 7 As shown, an electronic device provided in another embodiment of the present invention, based on the above method, includes a computing unit 201, a read-only memory 202, a random access memory 203, and an I / O interface 205 connected via a bus 204;

[0119] The computing unit 201 is used to execute method steps, and the read-only memory 202 and random access memory 203 are used to store program instructions, model parameters, and running data. The I / O interface 205 is connected to the input 207, output 208, data acquisition 209, and vehicle sensors 210, and is used to receive information related to vehicle status, environmental status, and candidate fast charging stations, and to output corresponding control commands. In addition, the electronic device is also connected to the vehicle controller 206 to control the relevant execution components of the thermal management system.

[0120] Another embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above-described method. The storage medium can be any form of non-transitory storage medium, such as read-only memory (ROM), random access memory (RAM), flash memory, solid-state drive (SSD), or hard disk drive.

[0121] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A closed-loop decision-making method for pre-cooling power batteries for fast charging, characterized in that, Includes the following steps: Step 1: Status information acquisition, including vehicle operation information, candidate fast charging station information, power battery status information, environmental status information, passenger compartment status information, and thermal management system operation status information; Step 2: Target fast charging station identification. Multiple candidate fast charging stations are screened and evaluated to identify the target fast charging station. Step 3: Arrival Status Prediction. After determining the target fast charging station, the arrival status of the vehicle is predicted online based on the current status of the vehicle and the predicted operating conditions. The arrival status includes at least the SOC of the power battery upon arrival. Step 4: Solve for the target arrival pre-cooling temperature. Based on the estimated arrival time, arrival SOC, candidate arrival power battery temperature, estimated fast charging time, pre-cooling energy consumption, and impact on passenger cabin comfort, solve for the target arrival pre-cooling temperature. Step 5: Pre-cooling target tracking and coordinated control. A temperature reference trajectory of the power battery is generated based on the pre-cooling temperature of the target arrival station. Through a hierarchical controller, multiple thermal management actuators are coordinated and controlled so that the power battery temperature follows the temperature reference trajectory. Step 6: Rolling update and closed-loop replanning. During vehicle operation, based on changes in the status of candidate fast charging stations, path status, vehicle status, and thermal management system status, the target fast charging station, arrival SOC, target arrival pre-cooling temperature, and thermal management actuator control quantities are updated to form a closed-loop decision-making process for power battery pre-cooling in fast charging scenarios. The specific steps of step 4 are as follows: The temperature of the candidate arriving power battery is used as an optimization variable, and a comprehensive cost function is constructed based on the fast charging time, acceptable charging power, pre-cooling energy consumption and passenger cabin comfort impact corresponding to different candidate arriving power battery temperatures. Under the premise of meeting the fast charging temperature requirements of the power battery, the comprehensive cost function is solved to obtain the target pre-cooling temperature at the destination under the current operating conditions; Among them, the particle swarm optimization algorithm is used to solve the comprehensive cost function. Each particle is initialized with a different candidate arrival power battery temperature. By iteratively updating the velocity and position of the particles, the particle swarm converges to the region with the minimum comprehensive cost in the candidate arrival power battery temperature space. The candidate arrival power battery temperature corresponding to the globally optimal particle is output as the target arrival precooling temperature. In step 4, the comprehensive cost function is constructed in the following form: in, For the comprehensive cost function, This indicates the increase in expected charging time caused by the temperature of the candidate arriving power battery relative to the ideal fast charging condition. This represents the total energy consumed by the thermal management system to achieve the desired temperature of the candidate battery upon arrival, from the current time to the arrival time. This indicates the predicted cabin temperature in the time domain. Indicates the target temperature in the crew cabin. , and These are the weighting coefficients; State-dependent adaptive weights are introduced into the comprehensive cost function: when the expected arrival time is short and the current battery temperature is higher than the suitable temperature range for fast charging, the weights are increased. When the expected arrival time is long and the demand for passenger cabin comfort is high, increase... When the expected arrival SOC is low, increase .

2. The closed-loop decision-making method for pre-cooling power batteries for fast charging as described in claim 1, characterized in that, In step 1, the vehicle operation information includes the vehicle's current location, navigation route information, real-time traffic information, and vehicle driving condition information; the candidate fast charging station information includes the candidate fast charging station location, available charging power, service status, and estimated waiting time; the power battery status information includes the current power battery temperature and current SOC; the environmental status information includes the ambient temperature; the passenger compartment status information includes the passenger compartment temperature; and the thermal management system operation status information includes the compressor operation status, coolant pump status, fan status, valve status, and current thermal management circuit temperature information.

3. The closed-loop decision-making method for pre-cooling power batteries for fast charging according to claim 1, characterized in that, The specific steps of step 2 are as follows: Based on the vehicle's current location and the location of candidate fast charging stations, and combined with road topology, traffic travel time, service capacity of candidate fast charging stations, and waiting time within the station, a comprehensive evaluation index is constructed. The comprehensive evaluation index includes at least one of the following: the driving time required for the vehicle to reach the candidate fast charging station, the detour distance or detour time, the available charging power of the candidate fast charging station, and the expected waiting time within the station. For each candidate fast charging station, calculate the remaining driving distance, remaining driving time, and estimated arrival time required for the vehicle to reach the candidate fast charging station, and determine the estimated charging time of the vehicle at the candidate fast charging station based on the available charging power and estimated waiting time of the candidate fast charging station. Based on the comprehensive evaluation results of each candidate fast charging station, the candidate fast charging station whose comprehensive evaluation results meet the preset conditions is selected as the target fast charging station. When the status or path status of a candidate fast charging station changes, the candidate fast charging station is re-evaluated, and the target fast charging station is updated based on the re-evaluation results.

4. The closed-loop decision-making method for pre-cooling power batteries for fast charging according to claim 1, characterized in that, The specific steps of step 3 are as follows: An energy consumption prediction model is built based on historical driving data and current operating conditions to predict the SOC of the power battery when the vehicle arrives at the target fast charging station. The energy consumption prediction model is constructed using a long short-term memory network. The input of the long short-term memory network includes one or more time series features of the vehicle's current speed, acceleration, road gradient, ambient temperature, occupant thermal comfort requirements, and traction power and thermal management power within a preset historical period. The output is the trend of vehicle energy consumption or SOC changes in the future time domain. The prediction and update of arrival SOC is carried out in a rolling manner. That is, at each planning time, the current status and route information are collected again, and the remaining driving energy consumption and thermal management energy consumption are re-estimated to update the arrival SOC prediction value.

5. The closed-loop decision-making method for pre-cooling power batteries for fast charging according to claim 1, characterized in that, In step 5, the hierarchical controller includes an upper-layer tracking coordination controller and a lower-layer execution controller; The upper-level tracking and coordination controller is used to receive the target arrival pre-cooling temperature and, in combination with the current power battery temperature, passenger compartment temperature, ambient temperature and thermal management system status, generate a power battery temperature reference trajectory suitable for subsequent long-term time domain. The lower-level execution controller is used to perform coordinated optimization control of multiple thermal management actuators based on real-time status feedback within a short prediction time domain, so as to achieve the tracking of the power battery temperature and the passenger compartment temperature to the temperature reference trajectory. The thermal management actuator includes a compressor, a coolant pump, a condenser fan, a blower, and valves.

6. The closed-loop decision-making method for pre-cooling power batteries for fast charging according to claim 5, characterized in that, The hierarchical controller adopts a hierarchical model predictive control framework that combines long-time domain control and short-time domain control; During long-term control, based on the target arrival pre-cooling temperature, the estimated arrival time, the current temperature of the power battery, and the current temperature of the passenger compartment, the temperature changes of the power battery and the passenger compartment during subsequent driving are predicted, and a power battery temperature reference trajectory is generated in the subsequent control time domain. During short-time domain control, the control reference quantities of the compressor, coolant pump, fan and valve are solved based on the temperature reference trajectory, the current temperature of the power battery, the current temperature of the passenger compartment and the current operating status of the thermal management system. To achieve hierarchical model predictive control, reduced-order predictive models matching the hierarchical control requirements are established: for the long-term temperature coordination process, a reduced-order thermal model is constructed to characterize the power battery temperature, passenger compartment temperature, and system cooling distribution relationship; for the short-term execution control process, a reduced-order control model is constructed to characterize the dynamic characteristics of the thermal management actuator and the response relationship of the refrigerant circuit and coolant circuit.

7. The closed-loop decision-making method for pre-cooling power batteries for fast charging according to claim 1, characterized in that, Step 6 includes the following specific steps: During the process of the vehicle driving toward the target fast charging station, the status of the target fast charging station, the vehicle path status, the power battery status, and the execution status of the thermal management system are monitored in real time. When the detected changes in the service capacity of the target fast charging station, changes in the expected waiting time, changes in the path traffic status, deviations in the power battery status, or deviations in the execution of the thermal management system meet the preset conditions, the rolling replanning mechanism is triggered, and steps 2 to 5 are re-executed. The vehicle's current location, power battery temperature, passenger compartment temperature, current SOC, ambient temperature, and thermal management system operating status are reacquired according to a preset cycle. Based on the latest road information and candidate fast charging station status, the remaining path length, estimated arrival time, arrival SOC, and target arrival pre-cooling temperature are recalculated, and the power battery temperature reference trajectory for subsequent stages is regenerated.

8. A closed-loop decision-making device for pre-cooling power batteries for fast charging, based on the closed-loop decision-making method for pre-cooling power batteries for fast charging as described in any one of claims 1-7, characterized in that, include: The data acquisition module and data storage module are used to collect and store information on candidate fast charging stations, power battery status, environmental status, passenger compartment status, and vehicle operating conditions. The candidate fast charging station selection module is used to construct a comprehensive evaluation index based on the vehicle's current location, candidate fast charging station locations, navigation route information, road topology, traffic travel time, available charging power of candidate fast charging stations, service status, and estimated waiting time, and select the candidate fast charging station with the best comprehensive benefit or the lowest comprehensive cost as the target fast charging station. The arrival state of charge prediction module is used to predict the SOC of the power battery when the vehicle arrives at the target fast charging station online. The charging revenue calculation module and the pre-cooling target solution module are used to calculate the fast charging revenue and pre-cooling cost corresponding to different candidate arrival temperatures, and solve the target arrival pre-cooling temperature online. The hierarchical control module is used to perform upper-level temperature planning and lower-level execution control solving. The upper-level temperature planning generates a long-term time-domain temperature reference trajectory based on the target arrival pre-cooling temperature. The lower-level execution control solving solves the thermal management actuator control reference quantity based on the temperature reference trajectory and real-time status. The actuator control module is used to coordinate the thermal management actuator to perform temperature control actions, so that the power battery temperature changes according to the temperature reference trajectory. The rolling update module is used to trigger replanning when the status of candidate fast charging stations, path status, or vehicle status changes, and to re-call the candidate fast charging station selection module to the hierarchical control module for updating.

Images