Thermal management optimization method and system for frequent start-stop working condition of vehicle

Abstract

The invention discloses a thermal management optimization method and system for a vehicle frequent start-stop working condition, and relates to the technical field of new energy vehicle thermal management, and the method comprises the following steps: collecting vehicle real-time operation data, extracting start-stop characteristics, carrying out the classification and real-time recognition of a current start-stop mode based on the start-stop characteristics, and obtaining the current start-stop mode; predicting the parking duration in combination with the current start-stop mode and the vehicle position information; the power of the cooling system is gradually reduced in a staged mode in the parking process, the power of the cooling system is gradually increased in a staged mode in the starting process, and anti-impact control is executed; and dynamically adjusting the energy distribution priorities of a battery, a motor and a passenger cabin according to the current parking working condition or driving working condition of the vehicle, and performing energy recycling when braking energy recovery or residual cooling capacity exists. Start-stop feature vectors are constructed by collecting vehicle speed, start-stop position features and load states, so that accurate judgment of actual working conditions can be kept in a frequent start-stop environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy vehicle thermal management, and particularly relates to a thermal management optimization method and system for frequent start-stop working conditions of a vehicle. BACKGROUND

[0002] New energy vehicles run frequently in scenarios such as urban logistics distribution, supermarket replenishment and warehouse transportation. The vehicle repeatedly accelerates, decelerates, stops and restarts between short-distance paths, forming a frequent start-stop working condition with obvious characteristics. Under such working conditions, the battery system, electric drive system and air conditioning system of the vehicle continuously experience thermal load changes in a short period of time, resulting in the need for the thermal management control strategy to quickly and accurately respond to changes in the vehicle operating state, so as to ensure that the battery temperature, motor temperature and other components of the vehicle are maintained within a suitable temperature range.

[0003] In the prior art, some vehicle control systems collect vehicle speed, brake signals or part of the operating parameters to count the stop behavior of the vehicle, for judging the operating state of the vehicle. However, the existing scheme generally only makes simple threshold judgments on individual state quantities, and cannot extract complete features from continuous and structured start-stop behaviors, so that the identification of the start-stop mode relies on a single parameter or fixed logic, and it is difficult to adapt to complex start-stop characteristics under different transportation tasks, different routes and different load changes.

[0004] However, the present inventors found at least the following technical problems in the process of implementing the technical solution of the present application: under the frequent start-stop working condition, the prior art fails to accurately identify the start-stop mode of the vehicle based on the structured start-stop characteristics, resulting in the control system being difficult to timely obtain mode information reflecting the actual working condition of the vehicle, thereby affecting the adaptability of subsequent stop duration prediction and thermal management strategy. SUMMARY

[0005] The purpose of the present application is to provide a thermal management optimization method and system for frequent start-stop working conditions of a vehicle, which at least solves the problem that the prior art fails to accurately identify the start-stop mode of the vehicle based on the structured start-stop characteristics, resulting in the control system being difficult to timely obtain mode information reflecting the actual working condition of the vehicle.

[0006] Embodiment One

[0007] The present application provides the following scheme:

[0008] According to one aspect of the present application, a thermal management optimization method for frequent start-stop working conditions of a vehicle is provided, comprising the following steps:

[0009] S1, start-stop mode recognition and classification, collecting real-time vehicle running data, extracting start-stop features, classifying the current start-stop mode based on the start-stop features and recognizing it in real time to obtain the current start-stop mode;

[0010] S2, parking time prediction and thermal inertia utilization, combining the current start-stop mode and vehicle location information to predict the parking time, selecting and executing the corresponding thermal inertia utilization strategy according to the predicted parking time, and using temperature buffer zone for temperature control;

[0011] S3, start-stop transition process optimization, gradually reducing the cooling system power in stages during the parking process, gradually increasing the cooling system power in stages during the starting process, and performing anti-shock control;

[0012] S4, energy optimization distribution, dynamically adjusting the energy distribution priority of the battery, motor and passenger cabin according to whether the vehicle is currently in a parking or driving condition, and performing energy recovery and utilization when there is brake energy recovery or residual cold energy.

[0013] Further, the start-stop mode recognition and classification specifically includes:

[0014] The start-stop frequency, parking time distribution, single driving distance, start-stop location feature and load state in the recent period of time are counted by using a sliding time window as start-stop features;

[0015] Based on the start-stop features, the start-stop mode is classified into one or more of express delivery mode, supermarket delivery mode and warehouse transfer mode by using a Bayesian classifier;

[0016] When the mode change is recognized in real time, a delay timer is started, and only when the new mode condition still meets after the delay timer expires can the mode switching be completed.

[0017] Further, the start-stop mode recognition and classification further includes:

[0018] The start-stop mode is hierarchically classified or online clustered to automatically discover new start-stop modes and dynamically expand the mode library;

[0019] When extracting start-stop features or classification, introduce time period, weather, holiday, driver braking habit, cargo type, GPS trajectory similarity as auxiliary start-stop features.

[0020] Further, the parking time prediction and thermal inertia utilization specifically includes:

[0021] Based on historical parking records, current start-stop mode and vehicle location information, a probability statistical model or its equivalent prediction model is used to obtain the predicted value of the parking time;

[0022] According to the predicted parking time, the thermal inertia utilization strategy is divided into a short parking thermal inertia utilization strategy, a medium parking thermal inertia utilization strategy, and a long parking thermal inertia utilization strategy;

[0023] The short parking thermal inertia utilization strategy is to keep the cooling system at the minimum power;

[0024] The medium parking thermal inertia utilization strategy is to gradually reduce the power of the cooling system and allow the temperature to slowly change within a safe range;

[0025] The long parking thermal inertia utilization strategy is to put the cooling system into a sleep mode and retain a minimum circulating power to prevent local overheating.

[0026] Further, the parking time prediction and thermal inertia utilization further include temperature buffer zone control:

[0027] The temperature control target of the battery or motor or passenger cabin is set to a normal temperature range and a buffer temperature range that is larger in extension;

[0028] Only when the actual temperature exceeds the buffer temperature range, the actuator adjustment action is triggered.

[0029] Further, the start-stop transition process optimization specifically includes:

[0030] The parking process is divided into a deceleration prediction stage and a parking stable stage, and the power of the cooling system is gradually reduced according to the vehicle deceleration signal in the deceleration prediction stage, and the corresponding thermal inertia utilization strategy is executed in the parking stable stage;

[0031] The starting process is divided into a starting preparation stage, an acceleration stage, and a cruising stage, the cooling system power is activated or boosted in advance in the starting preparation stage, the actuator power is quickly boosted in the acceleration stage, and the control power is smoothly transitioned to normal in the cruising stage.

[0032] Further, the start-stop transition process optimization further includes the following anti-shock control:

[0033] The rate of change of the total power of the cooling system is limited so that it does not exceed a preset threshold;

[0034] The multiple actuators are scheduled in time to avoid simultaneous large actions of multiple actuators.

[0035] Further, the energy optimization allocation specifically includes:

[0036] In the parking working condition, the available refrigeration or heating energy is allocated in the order of battery temperature maintenance priority, motor natural cooling second, and passenger cabin comfort third;

[0037] In the driving working condition, the available refrigeration or heating energy is allocated in the order of motor cooling and battery temperature maintenance priority, and passenger cabin comfort second.

[0038] The distribution weight corresponding to each priority is adjusted in real time according to the current working condition.

[0039] Further, the energy optimization distribution further comprises:

[0040] During brake energy recovery, the recovered energy is preferentially used for pre-cooling of the battery or the electric drive system.

[0041] When parking, the remaining cold energy of the passenger cabin or other circuits is transferred to the battery circuit or the motor circuit through valve or pipeline switching.

[0042] A thermal management optimization system for frequent start-stop working conditions of a vehicle comprises a vehicle thermal management controller or a thermal management domain controller.

[0043] The controller obtains vehicle operating state signals through a vehicle CAN bus or Ethernet, and controls an air conditioner compressor, a cooling fan, a water pump and a valve actuator based on the operating state signals, so as to achieve thermal management optimization for frequent start-stop working conditions of the vehicle.

[0044] Through the above scheme, the following beneficial technical effects are obtained:

[0045] The present application constructs a start-stop feature vector by collecting vehicle speed, acceleration, parking duration distribution, driving mileage, start-stop position characteristics and load state, and uses a Bayesian classifier to identify the start-stop mode in real time. The feature vector is structured, the data source is clear, and after being combined with the classification model, the controller can obtain the start-stop mode to which the vehicle currently belongs in time, so that the actual working condition can be accurately determined in the frequent start-stop environment, thereby providing a reliable basis for subsequent parking duration prediction and thermal management strategy selection.

[0046] The present application predicts the parking duration based on historical parking records, the current start-stop mode and position information, and according to the prediction result, the parking behavior is divided into three categories of short parking, medium parking and long parking, and corresponding thermal inertia utilization strategies such as minimum power maintenance, gradual power reduction and hibernation mode are selected. At the same time, a double-interval control mode of normal temperature interval and buffer temperature interval is introduced, and the actuator is triggered to act only when the temperature exceeds the interval, so that the mechanical load caused by frequent start-stop of the actuator during parking can be reduced, and the temperature regulation stability of the cooling system under different parking categories can be improved.

[0047] The application sets multiple temperature control stages such as deceleration prediction, parking stability, start preparation, acceleration and cruising during parking and starting processes, and adjusts the power changes of the compressor, fan and water pump according to the preset logic in each stage. Meanwhile, in the energy optimization distribution, the energy priorities of the battery, motor and passenger cabin are adjusted according to whether the vehicle is in a parking condition or a driving condition, and the energy is reused during the brake energy recovery period or when there is remaining cold energy, so that the distribution of cooling resources in the start-stop process is more in line with the actual heat load demand, and the continuity and coordination of the overall thermal management system operation are improved. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 is a flow chart of a thermal management optimization method for frequent start-stop conditions of a vehicle;

[0049] Figure 2 is a control strategy flow chart of a thermal management optimization method and system for frequent start-stop conditions of a vehicle. DETAILED DESCRIPTION

[0050] The technical solutions of the application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0051] Please refer to Figure 1 and Figure 2 , a thermal management optimization method for frequent start-stop conditions of a vehicle, comprising the following steps:

[0052] S1, start-stop mode recognition and classification, collecting real-time running data of the vehicle, extracting start-stop characteristics, classifying and recognizing the current start-stop mode in real time based on the start-stop characteristics, and obtaining the current start-stop mode;

[0053] S2, parking time prediction and thermal inertia utilization, combining the current start-stop mode and vehicle location information to predict the parking time of this time, selecting and executing the corresponding thermal inertia utilization strategy according to the predicted parking time, and using the temperature buffer zone method for temperature control;

[0054] S3, start-stop transition process optimization, gradually reducing the cooling system power in stages during the parking process, gradually increasing the cooling system power in stages during the starting process, and performing anti-shock control;

[0055] S4, energy optimization distribution, dynamically adjusting the energy distribution priorities of the battery, motor and passenger cabin according to whether the vehicle is in a parking condition or a driving condition, and performing energy recovery and utilization when there is brake energy recovery or remaining cold energy.

[0056] Specifically, in order to facilitate unified data processing and decision-making for the above steps, a quantifiable start-stop feature vector needs to be constructed for the vehicle start-stop behavior, which is described in the following form:

[0057] (1)

[0058] wherein: represents the number of starts and stops within a preset sliding time window;

[0059] represents the vehicle travel distance within the time window;

[0060] represents the cumulative duration of short-time parking periods;

[0061] represents the cumulative duration of medium-time parking periods;

[0062] represents the cumulative duration of long-time parking periods;

[0063] represents the quantified value of the vehicle load state.

[0064] The above feature vector is used to perform start-stop mode recognition. According to the statistical characteristics of the feature vector , its posterior probability belonging to each start-stop mode is calculated, and the current start-stop mode is obtained through the maximum posterior principle. Its posterior probability can be expressed as:

[0065] (2)

[0066] wherein:

[0067] represents the target start-stop mode;

[0068] represents a preset set of start-stop modes:

[0069] represents the prior probability of mode ,

[0070] represents the probability density of the feature vector under mode .

[0071] The parking duration prediction adopts a statistical model or an equivalent prediction model to output the predicted duration of the current parking after inputting the current start-stop mode, geographic location information, task state information and historical parking data. The parking prediction result is used to select different thermal inertia utilization strategies. According to the parking prediction result, different thermal management modes are selected for short-time parking, medium-time parking and long-time parking, and normal temperature intervals and buffer temperature intervals are set when processing temperature, so as to reduce the repeated action of the actuator.

[0072] The control of the start-stop transition process is segmented into vehicle deceleration phase, parking stabilization phase, start preparation phase, acceleration phase and cruising phase. The power output of the cooling system is gradually reduced or increased according to the stage requirement. In this process, the total power change rate is limited, and the timing staggered scheduling is adopted for multiple actuators, so that the actuator action meets the control stability requirement.

[0073] The energy optimization distribution selects different energy distribution priorities according to whether the vehicle is in a parking working condition or a driving working condition, and distributes the available thermal management energy to the battery, motor and cabin system respectively. During brake energy recovery, part of the recovered energy is used for pre-cooling of the battery or electric drive circuit, and in the parking state, whether the remaining cold energy is switched to other systems via the cooling circuit is selected according to the current cold energy distribution.

[0074] Each step is calculated and executed by the whole vehicle thermal management controller. The controller collects vehicle speed, acceleration, brake signal, drive parameter, battery information, motor state, location information and environmental parameter through the vehicle bus, and issues control instructions to the compressor, fan, water pump and valve actuator.

[0075] In this embodiment, the start-stop mode recognition and classification specifically includes:

[0076] The start-stop frequency, parking duration distribution, single driving mileage, start-stop location feature and load state in the recent period of time are counted by using a sliding time window as the start-stop feature;

[0077] Based on the start-stop feature, the start-stop mode is classified into one or more of the express delivery mode, the supermarket delivery mode and the warehouse transfer mode by using a Bayesian classifier;

[0078] When the mode change is recognized in real time, a delay timer is started. Only when the new mode condition still meets after the delay timer expires, the mode switching can be completed.

[0079] Specifically, in the embodiment, the start-stop mode recognition and classification includes statistics on real-time vehicle running data, feature extraction on start-stop behavior, classification calculation on features, and decision control on mode switching process. To realize the above functions, the embodiment sets a sliding time window, continuously collects vehicle running signals and forms a start-stop feature vector. The collected running signals include vehicle speed, acceleration, driving torque, brake state, gear state, driving motor output power, battery temperature, motor temperature, vehicle load estimation value, and vehicle position information. The length of the sliding time window is a fixed value, and the time interval of the window sliding back each time is a fixed value, to ensure continuous recording of the vehicle start-stop behavior.

[0080] The start-stop frequency, parking duration distribution, single driving mileage, start-stop location feature, and load state are counted in the sliding time window. The embodiment combines the above statistical indicators to form a start-stop feature vector X, which is used to represent the start-stop behavior mode of the vehicle in the current time period. The form of the feature vector X is formula (1).

[0081] The above feature vector is input into a Bayesian classifier. The Bayesian classifier calculates the posterior probability of each mode according to the statistical relationship between the feature vector and the preset start-stop mode set, and outputs the one with the maximum probability as the current start-stop mode. The posterior probability is specifically formula (2).

[0082] After the classifier obtains the real-time determination result, the embodiment sets a delay timer for processing mode switching. If the mode type output by the classifier changes, the delay timer is started. During the counting period of the delay timer, it is continuously detected whether the new mode condition remains consistent. Only when the counting is completed and the new mode condition does not change, the controller can complete the mode switching. If the new mode condition is not met during the counting stage, the mode switching is cancelled. The embodiment ensures stable recognition of the start-stop mode through the above control, avoiding frequent changes of the mode due to short-term data fluctuations.

[0083] In the embodiment, the start-stop transition process optimization specifically includes:

[0084] The parking process is divided into a deceleration prediction stage and a parking stable stage. In the deceleration prediction stage, the cooling system power is gradually reduced according to the vehicle deceleration signal, and in the parking stable stage, the corresponding thermal inertia utilization strategy is executed;

[0085] The starting process is divided into a starting preparation stage, an acceleration stage, and a cruising stage. In the starting preparation stage, the cooling system power is activated or improved in advance, in the acceleration stage, the actuator power is quickly improved, and in the cruising stage, it is smoothly transitioned to the normal control power.

[0086] Specifically, in the present embodiment, the start-stop transition process optimization includes thermal management adjustment in the parking phase and the start-up phase. The parking process is divided into a deceleration prediction phase and a parking stabilization phase. The deceleration prediction phase is triggered by a vehicle speed reduction trend, and when the vehicle speed decreases to a predetermined interval, the controller gradually reduces the output power of the cooling system according to the deceleration signal. The steps of reducing the output power include: reducing the compressor speed, reducing the cooling fan duty cycle, and reducing the cooling water pump flow. The above operations are processed in stages according to the vehicle deceleration, so that the cooling system can complete the power attenuation in a continuous manner.

[0087] When the vehicle is completely stopped, it enters the parking stabilization phase. In the parking stabilization phase, the controller selects the corresponding thermal inertia utilization strategy according to the parking time prediction result in step S2. If it is predicted to be short-time parking, the minimum power output of the cooling system is maintained; if it is predicted to be medium-time parking, the output power of the cooling system is gradually reduced and the temperature is allowed to change naturally within the allowed range; if it is predicted to be long-time parking, the compressor is stopped and the minimum cooling water pump flow for temperature circulation is maintained.

[0088] The start-up process is divided into a start-up preparation phase, an acceleration phase and a cruising phase. In the start-up preparation phase, the controller activates the cooling system or increases the power output of the cooling system according to the acceleration demand of the vehicle to provide temperature control capability for subsequent load increase. The processing of the start-up preparation phase includes: increasing the target output of the compressor in advance, increasing the fan duty cycle, and increasing the water pump flow. When the vehicle enters the acceleration phase from static, the thermal load increases, and the controller quickly increases the power of the actuator according to the torque instruction of the acceleration phase. When the vehicle enters the cruising phase, the controller gradually adjusts the power of each actuator to the stable control state corresponding to the current driving condition.

[0089] In the process of implementing the start-stop transition optimization, the present embodiment sets up an anti-shock control to limit the power change of multiple actuators. The anti-shock control includes: limiting the total power change rate, and arranging the adjustment actions of the compressor, the fan and the water pump in time sequence so that they do not change greatly at the same time. Through the above technical means, the present embodiment keeps the temperature change of the vehicle within the required range during the start-stop process, and ensures the continuity and stability of the actuator action.

[0090] In the present embodiment, the start-stop mode recognition and classification also includes:

[0091] Hierarchical classification or online clustering of start-stop modes is performed to automatically discover new start-stop modes and dynamically expand the mode library;

[0092] When extracting start-stop features or classification, time period, weather, holiday, driver braking habit, cargo type, GPS trajectory similarity are introduced as auxiliary start-stop features.

[0093] The start-stop transition process optimization further includes the following anti-shock control:

[0094] The rate of change of the total power of the cooling system is limited so as not to exceed a preset threshold;

[0095] The plurality of actuators are time-sequentially staggered to avoid simultaneous large movements of the plurality of actuators;

[0096] The energy optimization allocation specifically includes:

[0097] In the parking condition, the available cooling or heating energy is allocated according to the order of battery temperature maintenance priority, motor natural cooling second, and cabin comfort third;

[0098] In the driving condition, the available cooling or heating energy is allocated according to the order of motor cooling and battery temperature maintenance priority, and cabin comfort second;

[0099] The allocation weight corresponding to each priority is adjusted in real time according to the current condition.

[0100] Specifically, the start-stop mode recognition and classification further includes extended processing of the start-stop mode to adapt to diversified start-stop behaviors in the urban logistics vehicle operating environment. After completing the basic classification, the start-stop mode is further processed by hierarchical classification or online clustering. The hierarchical classification adopts a mode division method from coarse to fine, first divides the start-stop behavior into basic categories according to the overall distribution of the start-stop feature vector, and then performs secondary division based on different parking segment proportions, driving path features, and load change trends, etc. information, and forms a mode structure with clear hierarchical relationship through iterative method.

[0101] Online clustering continuously records the feature vectors within the sliding time window, updates the cluster center based on the distance measurement between the feature vectors, and forms new cluster categories for the feature sets not covered by the existing modes. The controller adds the category to the mode library when the number of new cluster categories detected reaches a set threshold, so that the mode library can be gradually supplemented and updated during long-term vehicle operation, enhancing the applicability of start-stop mode recognition.

[0102] Multiple auxiliary features are introduced in the extraction of start-stop features or classification. Auxiliary features include time period information, weather information, holiday information, driver braking habit, cargo type and GPS trajectory similarity. Time period information contains the corresponding relationship between the current time belonging to the hour interval and the historical start-stop behavior. Weather information includes temperature, humidity and precipitation conditions, which are obtained through environmental sensors or external services. Holiday information is obtained according to the date correspondence. Driver braking habit is obtained by long-term statistical braking signal change, including braking frequency and brake pedal change rate. Cargo type is provided by vehicle operation task information, which is used to supplement the load state change. GPS trajectory similarity is obtained based on the similarity measurement of historical trajectory sequence and current trajectory sequence, so as to identify repeated route scenarios. The above auxiliary features form an expanded feature vector, and are used as weight parameters in the classification process to participate in probability calculation, thereby improving the matching degree of pattern determination result and actual working condition.

[0103] The start-stop transition process optimization also includes anti-shock control. The content of anti-shock control includes limiting the change rate of the total power of the cooling system so that it does not exceed the preset threshold. The controller records the current power output value , according to the set maximum change rate limit formula does not exceed the pre-set rising limit value, and ensures that the descending change amount when the power decreases does not exceed the descending limit value. The change rate limit is realized by step adjustment, which does not directly change the target power of the actuator, but splits the target power into multiple transition values, and outputs in multiple control cycles.

[0104] Anti-shock control also includes time sequence staggered scheduling of multiple actuators to avoid simultaneous large action of multiple actuators. Specifically, the controller sets different priority adjustment sequences for the compressor, fan, water pump and valve actuators. Only one actuator is allowed to make large adjustment in the same control cycle, and the remaining actuators remain small adjustment or remain unchanged. If the compressor speed needs to be adjusted, the compressor is preferentially adjusted in power. In the next control cycle, the fan duty cycle is adjusted, and in the next cycle, the water pump flow is adjusted. Through the time-sharing adjustment method, the instantaneous load fluctuation caused by the superposition of the actions of each actuator is reduced, so that the heat management control process remains continuous.

[0105] Energy optimization distribution includes setting the energy priority of the battery, motor and passenger cabin according to whether the vehicle is currently in a parking working condition or a driving working condition. In the parking working condition, the available refrigeration or heating energy is preferentially distributed for battery temperature maintenance, so that the battery temperature is maintained within the target interval. The motor is in a non-driving state and only needs to maintain natural cooling, so the required energy is low, and therefore the motor is lower than the battery in the energy distribution sequence. The temperature regulation of the passenger cabin is not a control task that must be performed in the parking state, so it is located at the last in the priority sequence.

[0106] In driving conditions, the motor is in driving state and generates a large amount of heat, and the priorities of motor cooling and battery temperature regulation are the same. The controller calculates the energy requirements of the two according to the motor output torque, motor coil temperature, battery current and its trend, and ensures their thermal management requirements at the same time according to the priorities. The priority of cabin temperature control in the driving process is lower than that of the motor and the battery, and the controller allocates the remaining energy to it according to the available energy.

[0107] The distribution weight corresponding to the above energy priority is adjusted in real time according to the current working condition. The controller estimates the energy demand according to the vehicle speed, driving torque, ambient temperature and temperature estimates of the battery and motor. Based on the energy demand, a weight coefficient is formed to determine the distribution ratio of the cooling system output energy among different thermal management objects. The update period of the weight coefficient is consistent with the refresh period of the vehicle bus signal, so that the energy distribution can be dynamically adjusted with the change of working condition.

[0108] The above-mentioned mode expansion, auxiliary feature introduction, impact prevention control and energy priority setting make the data flow and control logic consistent among the start-stop mode recognition and classification, start-stop transition control and energy distribution strategy. The controller performs the above-mentioned judgment and adjustment based on the real-time collected vehicle signals, and realizes the comprehensive thermal management in the frequent start-stop working condition.

[0109] In the embodiment, the parking duration prediction and thermal inertia utilization specifically include:

[0110] Based on the historical parking record, the current start-stop mode and the vehicle location information, a probability statistical model or its equivalent prediction model is used to obtain the predicted value of the parking duration of this time;

[0111] According to the predicted parking duration, the thermal inertia utilization strategy is divided into a short parking thermal inertia utilization strategy, a medium parking thermal inertia utilization strategy and a long parking thermal inertia utilization strategy;

[0112] The short parking thermal inertia utilization strategy is to keep the cooling system running at the minimum power;

[0113] The medium parking thermal inertia utilization strategy is to gradually reduce the power of the cooling system and allow the temperature to change slowly within a safe range;

[0114] The long parking thermal inertia utilization strategy is to make the cooling system enter a sleep mode and retain a minimum circulating power to prevent local overheating;

[0115] Specifically, the parking duration prediction model is trained based on historical operation data in the vehicle development stage. The training data includes parking start time, parking end time, geographical position sequence, load state sequence and corresponding start-stop mode label under multiple working conditions. After the controller performs sample cleaning and feature extraction on the above-mentioned data, a probability statistical method is used to generate a mapping relationship The mapping relationship is stored in the controller in a parameterized form, and does not need to be retrained during online operation, only an inference process is performed.

[0116] In the parking duration prediction process, the embodiment extracts sample data related to parking behavior from vehicle historical operation data, the sample data including parking start time, parking end time, parking location, start-stop mode at that time, and task type information. For the current parking process, a prediction feature vector is constructed , the feature vector is:

[0117] ;

[0118] , wherein,

[0119] represents the start-stop mode identifier identified at present:

[0120] represents the location information identifier of the current vehicle;

[0121] represents the task state information identifier at present;

[0122] represents the statistical quantity identifier associated with the historical parking behavior.

[0123] Based on the above feature vector , the embodiment uses a probability statistical model or an equivalent prediction model to output a prediction value of the parking duration at present. The prediction process can be represented as:

[0124] ;

[0125] , wherein, represents the prediction value of the parking duration at present,

[0126] represents the prediction mapping relationship obtained by training the historical sample.

[0127] The prediction mapping relationship is determined by the controller according to historical data in the development stage, and is fixed in the control strategy, and is directly used during online operation.

[0128] According to the predicted parking duration , the embodiment sets the short parking, medium parking, and long parking threshold values. The upper threshold value of the short parking is set as , the upper threshold value of the medium parking is set as , and the parking category is determined as:

[0129] ;

[0130] short stop;

[0131] medium stop;

[0132] wherein,

[0133] represents the upper threshold of short stop parking duration.

[0134] represents the upper threshold of medium stop parking duration.

[0135] In the short stop thermal inertia utilization strategy, the controller sets the output power of the cooling system to the minimum maintenance power , keeping the compressor, water pump and fan running in a low power state for a long time, relying on the heat capacity of the system to maintain the battery and motor temperature within the set temperature range. At this time, the output power of the cooling system can be represented as:

[0136] ;

[0137] wherein, represents the total output power of the cooling system at time t during the short stop.

[0138] In the medium stop thermal inertia utilization strategy, the controller gradually reduces the cooling system power during the parking period according to the power value at the parking start time and the predicted parking duration. Let the cooling system power at the parking start time be , the target minimum power in the medium stop stage be , and the parking prediction duration be , then the power change in the medium stop stage can be represented as:

[0139] ;

[0140] wherein,

[0141] represents the target minimum power in the medium stop thermal inertia utilization stage:

[0142] represents the cumulative time from the parking start time, satisfying .

[0143] In the above process, the controller monitors the battery temperature motor temperature and cabin temperature , and sets the target temperature range to the normal temperature range and the buffer temperature range. When the temperature is within the normal temperature range, maintain the current power adjustment plan, when the temperature exceeds the buffer temperature range, adjust the power reduction rate or stop further reducing the power, to keep the temperature within the safe range of variation.

[0144] In the long thermal inertia utilization strategy, if the predicted parking duration is determined to be long, the controller will put the cooling system into hibernation after completing the necessary temperature safety check, turn off the compressor and fan, and set the water pump to the minimum circulating power , ensuring that the cooling medium circulates between key components at a low flow rate to prevent local areas from generating excessive temperatures. At this time, the total power of the cooling system can be approximately expressed as:

[0145] ;

[0146] wherein,

[0147] represents the minimum circulating power for preventing local overheating.

[0148] In this embodiment, the energy optimization distribution further includes:

[0149] During brake energy recovery, the recovered energy is preferentially used for pre-cooling of the battery or electric drive system;

[0150] When parking, the remaining cold of the passenger cabin or other circuits is transferred to the battery circuit or motor circuit through valve or pipeline switching.

[0151] Specifically, the energy optimization distribution further includes redistributing the cold during brake energy recovery and when parking. During brake energy recovery, the feedback power under braking conditions is obtained through the motor controller , the recovery energy conversion efficiency is set as , and the braking time interval is The available recovery energy during this braking process is expressed as:

[0152] ;

[0153] wherein,

[0154] represents the recovery energy available for thermal management:

[0155] represents the conversion efficiency from brake energy to available electric energy;

[0156] represents the brake feedback power at time .

[0157] The controller distributes between the battery circuit and the electric drive circuit according to the battery pre-cooling demand and the electric drive system pre-cooling demand. Let the energy required for battery pre-cooling be , and the energy required for electric drive system pre-cooling be When greater than or equal to , the precooling demand of both is satisfied according to a preset ratio: when less than , the is sequentially allocated to the battery circuit and the electric drive circuit for reducing the initial temperature in the subsequent working condition according to a preset priority.

[0158] In the parking process, the embodiment also redistributes the remaining cold energy in the passenger cabin circuit or other cooling circuits. The controller estimates the remaining cold energy of the passenger cabin circuit by measuring the temperature and flow rate of the refrigerant entering and leaving the heat exchanger in the passenger cabin refrigerant circuit When it is detected that the passenger cabin temperature has reached or is close to the set lower limit and the vehicle is in a parked state, the controller adjusts the valves or pipe switching devices in the cooling system to switch the part of the refrigerant flow connected to the passenger cabin circuit to the battery circuit or the motor circuit. The switching process includes closing the part of the valve connected to the passenger cabin evaporator, and opening the bypass valve connected to the battery cooling plate or the motor cooling jacket, so that the refrigerant originally circulating in the passenger cabin circuit flows into the battery circuit or the motor circuit.

[0159] Through the above parking duration prediction, thermal inertia utilization strategy division, cooling system staged power control, and redistribution of recovered energy and remaining cold energy, the thermal management resources in the parking and braking stages are reasonably scheduled, so that the battery circuit and the electric drive circuit are in a suitable temperature state when the vehicle is started subsequently, thereby forming a complete control link between the steps of the present application.

[0160] To realize the switching between the short parking, medium parking and long parking strategies, the embodiment constructs a parking strategy state machine in the controller. The inputs of the state machine include a parking start signal, a parking prediction duration temperature interval state and actuator feedback signal. The state machine enters the corresponding state according to the time period, records the switching time stamp when the state is switched, and reconfigures the power adjustment rate, actuator action sequence and temperature update period after switching.

[0161] In the embodiment, the parking duration prediction and thermal inertia utilization also include temperature buffer zone control:

[0162] The temperature control target of the battery or motor or passenger cabin is set to the normal temperature interval and the buffer temperature interval which is larger in extension;

[0163] Only when the actual temperature exceeds the buffer temperature interval, the actuator adjustment action is triggered.

[0164] Specifically, in the temperature buffer zone control process, the controller first determines the normal temperature interval according to the working characteristics of different thermal management objects. The normal temperature interval is composed of the lower limit of the target temperature and the upper limit of the target temperature, and is used to represent the direct demand of the thermal management object under the current working condition. Taking the battery as an example, its normal temperature interval can be recorded as The interval is determined by structured parameters such as battery cell model, charging and discharging current, and environmental temperature.

[0165] On the basis of the normal temperature interval, the embodiment further sets an extended buffer temperature interval. The buffer temperature interval is used for the controller to judge the hysteresis interval when the actuator is executed, so that the control strategy is not frequently adjusted by the actuator due to slight temperature fluctuations. The buffer temperature interval of the battery can be expressed as:

[0166]

[0167] Wherein,

[0168] represents the real-time measured battery temperature;

[0169] represents the temperature expansion of the buffer interval, which is determined by the cooling capacity of the thermal management system, the heat load change rate of the typical working condition of the vehicle, and the heat capacity parameter

[0170] .

[0171] Similarly, the normal temperature interval and the buffer temperature interval are also set for the motor temperature and the cabin temperature , and the forms are as follows:

[0172]

[0173]

[0174] Wherein, is the normal temperature interval boundary of the motor;

[0175] is the normal temperature interval boundary of the cabin;

[0176] are the buffer interval expansion of the motor and the cabin, respectively.

[0177] The controller in operation is and Periodic sampling is performed to compare the temperature sampling value with the corresponding buffer interval. When the temperature is between the normal temperature interval and the buffer temperature interval, the controller maintains the current actuator state and does not perform any adjustment operation on the compressor, fan, water pump or valve actuator. This control mode reduces the number of actuator actions, so that the actuator avoids unnecessary starting actions in the parked state.

[0178] When the real-time temperature exceeds the boundary of the buffer temperature interval, the controller selects a corresponding adjustment strategy according to the temperature deviation direction and the deviation degree. If the temperature exceeds the upper limit of the buffer interval, the controller performs a power increase process on the compressor, and adjusts the fan duty ratio and the water pump flow according to the demand, so that the cooling capacity meets the temperature drop demand. If the temperature is lower than the lower limit of the buffer interval, the controller performs a heating or reduced refrigeration output strategy, depending on the system architecture and the heat pump, PTC or waste heat recovery device configured for the vehicle.

[0179] In the parked working condition, the temperature buffer interval control is used in combination with the thermal inertia utilization strategy. When the predicted parking time length belongs to short parking, the cooling system remains to run at the minimum power, but due to the existence of the buffer temperature interval, the controller can determine whether the temperature change needs to adjust the actuator. When the predicted parking time length belongs to medium parking or long parking, the buffer interval judgment logic is also applicable, so that the adjustment action of the cooling system is related to the parking time length classification, and the temperature stability under different parking categories is realized.

[0180] In the starting process stage, the temperature buffer interval still participates in the judgment of actuator action. In the preparation stage before starting, the controller determines whether the power needs to be raised in advance according to the temperature conditions of the battery and the motor. When the temperature is within the normal interval or the buffer interval, the pre-activation strategy is executed in a standard step-by-step manner; when the temperature exceeds the buffer interval, the pre-activation power is adjusted according to the temperature deviation direction, so that the starting stage can meet the subsequent heat load change demand.

[0181] In the whole temperature buffer interval control process, the controller updates the temperature signal according to a fixed sampling period, and uses it as the input data of the temperature adjustment decision. The design of the buffer temperature interval is integrated with the actuator action logic, so that the control strategy can remain consistent in the parked, starting and driving working conditions, avoid the wear and tear of the actuator caused by frequent start-stop, and at the same time maintain the accuracy of temperature control.

[0182] Extension of the buffer temperature interval The extension of the buffer temperature interval is determined by the system heat capacity, the cooling medium flow, the maximum power change rate of the actuator and the temperature change speed in the typical working condition. The temperature change rate model is established by thermal balance experiment or simulation in the vehicle development stage, so as to provide parameter basis for the buffer interval.

[0183] Embodiment two

[0184] In this embodiment, a thermal management optimization system for vehicle frequent start-stop working conditions includes a vehicle thermal management controller or a thermal management domain controller.

[0185] The controller obtains vehicle operating state signals through the vehicle CAN bus or Ethernet, and controls the air conditioning compressor, cooling fan, water pump, and valve actuator based on the operating state signals to achieve thermal management optimization for vehicle frequent start-stop working conditions.

[0186] Specifically, the software structure of the controller includes a start-stop mode recognition module, a parking time prediction module, a thermal inertia utilization module, a temperature buffer control module, a start-stop transition control module, an energy distribution module, and an actuator driving module, which are connected by a data sharing structure and executed by the central processor in a unified scheduling cycle according to a predetermined order.

[0187] The vehicle thermal management controller internally includes a processor, a memory, and a vehicle communication interface. The processor is used for computing and logically judging the collected data, the memory is used for storing the thermal management control strategy, the start-stop mode recognition model, the parking time prediction model, and the actuator scheduling parameters. The vehicle communication interface is used for data interaction with the vehicle information bus to realize real-time collection and instruction issuance.

[0188] The controller receives vehicle operating state signals through the CAN bus or Ethernet, and the received signals include vehicle speed signals, acceleration signals, brake signals, gear signals, battery temperature, battery current, battery voltage, battery cell temperature difference provided by the battery management system, stator temperature, speed, and output power provided by the motor controller, and air conditioning circuit temperature, ambient temperature, and valve state provided by the body control system. The controller constructs a start-stop feature vector, a parking prediction feature vector, and temperature control parameters according to the above signals.

[0189] When performing start-stop mode recognition, the controller inputs the real-time collected signals into the mode recognition logic, and the processor completes feature statistics and model operation, and generates a current start-stop mode identification according to the mode recognition result. The mode identification is stored in the system state register area for subsequent input of parking time prediction and energy distribution strategy.

[0190] When performing parking time prediction, the controller inputs the current start-stop mode, vehicle location information, and historical data index results into the prediction model module. The prediction model module outputs a parking time prediction value according to the model calculation method, and submits the prediction result to the thermal inertia utilization strategy module, which selects the power regulation strategy of the cooling system according to the parking time category.

[0191] In the process of executing the start-stop transition control, the controller uses the real-time speed, acceleration and driving torque data of the vehicle to determine whether the vehicle is in deceleration, stable stop, start or acceleration state. The processor calculates the cooling system power adjustment according to the state determination result, and generates a phased target power value according to the power change rate limit condition. The controller decomposes the target power value into control instructions corresponding to the actuators, including compressor speed target value, fan duty cycle target value, water pump flow target value and valve opening target value.

[0192] In the process of executing energy optimization distribution, the controller inputs the current working condition identification, battery temperature, motor temperature and cabin temperature into the energy distribution module. The energy distribution module generates different energy weights according to the working condition priority setting, and calculates the target energy distribution proportion available for battery cooling, motor cooling and cabin adjustment. Then the energy distribution proportion is converted into actual actuator control instructions by the actuator scheduling module, and submitted to the output interface for issuance.

[0193] During brake energy recovery, the controller receives the brake feedback power signal through the motor controller interface, calculates the available pre-cooling energy according to the feedback power and energy conversion efficiency. The pre-cooling energy calculation result is input into the battery pre-cooling or motor pre-cooling logic, which generates pre-cooling instructions for corresponding actuators to keep the pre-cooling process consistent with the vehicle state.

[0194] For the redistribution of the remaining cold energy, the controller estimates the remaining cold energy of the circuit according to the inlet and outlet temperatures and flow information of the cabin cooling system, and determines whether to perform circuit switching according to the current parking state and battery or motor temperature demand. If the switching condition is met, the controller issues opening degree adjustment instructions to the bypass valve or switching valve through the valve drive interface, realizing the switching of part of the refrigerant flow path from the cabin circuit to the battery or motor cooling circuit.

[0195] During the entire execution process, the controller executes data sampling, model operation, control quantity calculation and instruction issuance according to a fixed cycle period. The processor updates the system state variables in each control period, including start-stop mode identification, stop prediction result, temperature measurement result and actuator feedback state. The controller continuously monitors the actuator feedback signal, checks the instruction execution, and adjusts the control instruction again when a control deviation is detected, to ensure that the cooling system output meets the target power change curve.

[0196] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions recorded in the above embodiments can be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A thermal management optimization method for a vehicle frequent start-stop working condition, characterized in that, The method comprises the following steps: S1, start-stop mode recognition and classification, collecting real-time vehicle running data, extracting start-stop characteristics, classifying the current start-stop mode based on the start-stop characteristics and recognizing it in real time to obtain the current start-stop mode; S2, parking time prediction and thermal inertia utilization, combining the current start-stop mode and vehicle location information to predict the parking time, selecting and executing the corresponding thermal inertia utilization strategy according to the predicted parking time, and using the temperature buffer zone method for temperature control; S3, start-stop transition process optimization, gradually reducing the cooling system power in stages during the parking process, gradually increasing the cooling system power in stages during the starting process, and performing anti-shock control; S4, energy optimization distribution, dynamically adjusting the energy distribution priority of the battery, motor and passenger cabin according to whether the vehicle is currently in a parking or driving condition, and performing energy recovery and utilization when there is brake energy recovery or residual cold energy.

2. The method of claim 1, wherein, The start-stop mode recognition and classification specifically comprises: Using a sliding time window to count the start-stop frequency, parking time distribution, single driving mileage, start-stop location characteristics and load state in the recent period as start-stop characteristics; Based on the start-stop characteristics, using a Bayesian classifier to classify the start-stop mode into one or more of the express delivery mode, the supermarket delivery mode and the warehouse transfer mode; When the mode change is recognized in real time, a delay timer is started, and only when the new mode condition still meets after the delay timer expires can the mode switching be completed.

3. The method of claim 1, wherein, The start-stop mode recognition and classification further comprises: Hierarchical classification or online clustering of the start-stop mode to automatically discover new start-stop modes and dynamically expand the mode library; Introducing time period, weather, holiday, driver braking habit, cargo type, GPS trajectory similarity as auxiliary start-stop characteristics when extracting start-stop characteristics or classification.

4. The method of claim 1, wherein, The parking time prediction and thermal inertia utilization specifically comprises: Based on historical parking records, current start-stop mode and vehicle location information, using a probability statistical model or an equivalent prediction model to obtain the predicted value of the parking time; According to the predicted parking time, the thermal inertia utilization strategy is divided into a short parking thermal inertia utilization strategy, a medium parking thermal inertia utilization strategy and a long parking thermal inertia utilization strategy; The short parking thermal inertia utilization strategy is to keep the cooling system running at the minimum power; The medium parking thermal inertia utilization strategy is to gradually reduce the cooling system power and allow the temperature to change slowly within the safe range; The long parking thermal inertia utilization strategy is to make the cooling system enter the sleep mode and retain the minimum circulating power to prevent local overheating.

5. The method of claim 1, wherein, The parking time prediction and thermal inertia utilization further comprises temperature buffer zone control: The temperature control target of the battery or motor or passenger cabin is set to a normal temperature interval and a buffer temperature interval larger than the normal temperature interval; Only when the actual temperature exceeds the buffer temperature interval, the actuator adjustment action is triggered.

6. The method of claim 1, wherein, The start-stop transition process optimization specifically comprises: The parking process is divided into a deceleration prediction stage and a parking stable stage, the cooling system power is gradually reduced according to the vehicle deceleration signal in the deceleration prediction stage, and the corresponding thermal inertia utilization strategy is executed in the parking stable stage; The starting process is divided into a starting preparation stage, an acceleration stage and a cruising stage. In the starting preparation stage, the power of the cooling system is activated or boosted in advance. In the acceleration stage, the power of the actuators is rapidly boosted. In the cruising stage, the power is smoothly transitioned to the normal control power.

7. The method of claim 1, wherein, The start-stop transition process optimization further includes the following anti-shock control: The rate of change of the total power of the cooling system is limited so as not to exceed a preset threshold; The multiple actuators are time-sequentially staggered and scheduled to avoid simultaneous large movements of multiple actuators.

8. The method of claim 1, wherein, The energy optimization distribution specifically includes: In the parking working condition, the available refrigeration or heating energy is distributed according to the order of battery temperature maintenance priority, motor natural cooling second, and cabin comfort third; In the driving working condition, the available refrigeration or heating energy is distributed according to the order of motor cooling and battery temperature maintenance highest priority, and cabin comfort second; The distribution weight corresponding to each priority is adjusted in real time according to the current working condition.

9. The method of claim 1, wherein, The energy optimization distribution further includes: During the brake energy recovery, the recovered energy is preferentially used for pre-cooling of the battery or the electric drive system; When parking, the remaining cold of the cabin or other circuits is transferred to the battery circuit or the motor circuit through valve or pipeline switching.

10. A thermal management optimization system for a vehicle frequently starting and stopping operating condition, characterized in that, A vehicle frequent start-stop working condition-oriented thermal management optimization method according to any one of claims 1-9, comprising a vehicle thermal management controller or a thermal management domain controller; The controller obtains vehicle operating state signals through a vehicle CAN bus or Ethernet, and controls the air conditioner compressor, cooling fan, water pump and valve-type actuators based on the operating state signals to achieve thermal management optimization for the vehicle frequent start-stop working condition.

Images