Vehicle temperature adjusting method and training method and device of simulation prediction model

Abstract

The invention provides a vehicle temperature adjustment method and a simulation prediction model training method and device.The method comprises the steps that in response to the situation that a thermal runaway risk exists in a vehicle within a second time period according to first heat pump state data of the vehicle within a first time period, the vehicle body system temperature and the passenger compartment temperature of the vehicle are adjusted, and the thermal runaway risk of the vehicle within the second time period is recognized; wherein the second time period is the next time period of the first time period. The timeliness of thermal runaway risk identification of the vehicle is improved, the precision and accuracy of thermal runaway risk identification are improved, the possibility that the thermal runaway risk occurs in the second time period of prediction of the vehicle is reduced, the temperature control precision of the vehicle is improved, and the cost of vehicle heat pump state data prediction and thermal runaway risk identification is reduced. The safety of vehicle driving and vehicle drivers and passengers is improved, and the service life of part of parts of the vehicle is prolonged.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technology, and in particular to a vehicle temperature adjustment method, a simulation prediction model training method and a device. Background Art

[0002] With the development of technology, vehicles occupy an increasingly important position in people's lives. During the daily driving of the vehicle, the temperature of the vehicle body needs to be maintained within the set temperature range to improve the safety of vehicle driving.

[0003] Therefore, it is very important to accurately control the temperature of the vehicle body. Summary of the Invention

[0004] The present disclosure aims to solve one of the technical problems in the related art at least to a certain extent.

[0005] To this end, a first aspect of the present disclosure proposes a vehicle temperature adjustment method.

[0006] A second aspect of the present disclosure provides a training method for a simulation prediction model.

[0007] A third aspect of the present disclosure provides a vehicle temperature adjustment device.

[0008] A fourth aspect of the present disclosure provides a training device for a simulation prediction model.

[0009] A fifth aspect of the present disclosure provides a vehicle.

[0010] A sixth aspect of the present disclosure provides an electronic device.

[0011] A seventh aspect of the present disclosure provides a computer-readable storage medium.

[0012] An eighth aspect of the present disclosure provides a chip.

[0013] A first aspect of the present disclosure proposes a vehicle temperature adjustment method, comprising: in response to identifying, based on first heat pump status data of the vehicle in a first time period, that the vehicle is at risk of thermal runaway in a second time period, adjusting the body system temperature and passenger compartment temperature of the vehicle, wherein the second time period is a time period subsequent to the first time period.

[0014] The second aspect of the present disclosure proposes a training method for a simulation prediction model, comprising: obtaining a candidate state simulation prediction model, a candidate control simulation prediction model, and a candidate power simulation prediction model to be trained; performing model training on the candidate state simulation prediction model, the candidate control simulation prediction model, and the candidate power simulation prediction model according to the sample heat pump state data of the vehicle in a sample time period, to obtain a trained target state simulation prediction model, a target control simulation prediction model, and a target power simulation prediction model; performing model integration on the target control simulation prediction model, the target state simulation prediction model, and the target power simulation prediction model to obtain an integrated cloud-based vehicle condition simulation prediction model; deploying the cloud-based vehicle condition simulation prediction model on the vehicle to obtain the target vehicle condition simulation prediction model on the vehicle, wherein the target vehicle condition simulation prediction model is used to implement the vehicle temperature adjustment method proposed in the first aspect above.

[0015] A third aspect of the present disclosure proposes a vehicle temperature adjustment device, comprising an adjustment module for adjusting the body system temperature and the passenger compartment temperature of the vehicle in response to identifying, based on first heat pump status data of the vehicle in a first time period, that the vehicle is at risk of thermal runaway in a second time period, wherein the second time period is the next time period of the first time period.

[0016] The fourth aspect of the present disclosure proposes a training device for a simulation prediction model, comprising: an acquisition module for acquiring a candidate state simulation prediction model, a candidate control simulation prediction model and a candidate power simulation prediction model to be trained; a training module for performing model training on the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model respectively according to the sample heat pump state data of the vehicle in the sample time period, to obtain a trained target state simulation prediction model, a target control simulation prediction model and a target power simulation prediction model; an integration module for performing model integration on the target control simulation prediction model, the target state simulation prediction model and the target power simulation prediction model to obtain an integrated cloud-based vehicle condition simulation prediction model; a deployment module for deploying the cloud-based vehicle condition simulation prediction model on the vehicle to obtain the target vehicle condition simulation prediction model on the vehicle, wherein the target vehicle condition simulation prediction model is used to implement the vehicle temperature adjustment device proposed in the third aspect.

[0017] The fifth aspect of the present disclosure proposes a vehicle, which is capable of executing the vehicle temperature adjustment method proposed in the first aspect and / or the simulation prediction model training method proposed in the second aspect.

[0018] The sixth aspect of the present disclosure proposes an electronic device comprising: a processor; a memory for storing executable instructions of the processor; wherein the processor is configured to execute instructions to implement the vehicle temperature adjustment method proposed in the first aspect and / or the training method of the simulation prediction model proposed in the second aspect.

[0019] The seventh aspect of the present disclosure proposes a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by the processor of an electronic device, the electronic device is enabled to execute the vehicle temperature adjustment method proposed in the first aspect and / or the simulation prediction model training method proposed in the second aspect.

[0020] The eighth aspect of the present disclosure proposes a chip comprising one or more interface circuits and one or more processors; the interface circuit is used to receive a signal and send the signal to the processor, the signal comprising a computer instruction stored in a memory, and when the processor executes the computer instruction, the chip executes the vehicle temperature adjustment method proposed in the first aspect and / or the simulation prediction model training method proposed in the second aspect.

[0021] The vehicle temperature adjustment method and device proposed in the present disclosure obtain first heat pump status data of the vehicle within a first time period, wherein, when it is identified that the vehicle has a thermal runaway risk within a second time period based on the first heat pump status data, the body system temperature and the passenger compartment temperature of the vehicle within the second time period are adjusted. In the present disclosure, heat pump status data of a vehicle in a second time period is predicted using first heat pump status data, and a thermal runaway risk assessment is performed on the vehicle in the second time period based on the predicted heat pump status data, thereby improving the timeliness of thermal runaway risk identification for the vehicle. Thermal runaway risk assessment and identification are performed using the heat pump status data of the heat pump system, thereby improving the precision and accuracy of thermal runaway risk identification. After identifying that a thermal runaway risk exists in the vehicle, the vehicle's body system temperature and passenger compartment temperature are adjusted to reduce the possibility of the vehicle experiencing a thermal runaway risk in the predicted second time period. The vehicle temperature is adjusted based on the thermal runaway risk identification result, thereby improving the timeliness of vehicle temperature regulation and thus improving the temperature control accuracy of the vehicle. Compared with the related art in which vehicle data prediction is achieved based on a constructed actual vehicle test bench, the cost of vehicle heat pump status data prediction and thermal runaway risk identification is reduced, and the occurrence of poor prediction and identification accuracy due to insufficient experimental data from the actual vehicle test bench is avoided. The safety of vehicle driving and vehicle drivers and passengers is improved, the service life of some vehicle components is extended, the maintenance cost of the vehicle is reduced, and the driving and riding experience of the vehicle is optimized.

[0022] It should be understood that the contents described in the present disclosure are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and / or additional aspects and advantages of the present disclosure will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0024] Figure 1 Schematic diagram of a vehicle temperature adjustment method according to an embodiment of the present disclosure;

[0025] Figure 2 Schematic diagram of a vehicle temperature adjustment method according to another embodiment of the present disclosure;

[0026] Figure 3 Schematic diagram of a target vehicle condition simulation prediction model according to an embodiment of the present disclosure;

[0027] Figure 4 Schematic diagram of a target state simulation prediction model according to an embodiment of the present disclosure;

[0028] Figure 5 Schematic diagram of a target control simulation prediction model according to an embodiment of the present disclosure;

[0029] Figure 6 Schematic diagram of a target state simulation prediction model according to an embodiment of the present disclosure;

[0030] Figure 7 Schematic diagram of a vehicle temperature adjustment method according to another embodiment of the present disclosure;

[0031] Figure 8 Schematic diagram of a flow chart of a method for training a simulation prediction model according to an embodiment of the present disclosure;

[0032] Figure 9 A flowchart of a training method for a simulation prediction model according to another embodiment of the present disclosure;

[0033] Figure 10 A schematic diagram of a simulation prediction model according to an embodiment of the present disclosure;

[0034] Figure 11 This is a structural diagram of a vehicle temperature adjustment device according to an embodiment of the present disclosure;

[0035] Figure 12 A schematic diagram of the structure of a training device for a simulation prediction model according to an embodiment of the present disclosure;

[0036] Figure 13 A block diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0037] The following describes in detail embodiments of the present disclosure, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0038] The following describes a vehicle temperature adjustment method, a simulation prediction model training method and a device proposed in an embodiment of the present disclosure with reference to the accompanying drawings.

[0039] Figure 1 FIG. 1 is a flow chart of a vehicle temperature adjustment method according to an embodiment of the present disclosure. Figure 1 As shown, the method includes:

[0040] S101 , in response to identifying, based on first heat pump status data of the vehicle in a first time period, that the vehicle has a thermal runaway risk in a second time period, adjusting a body system temperature and a passenger compartment temperature of the vehicle.

[0041] The second time period is the next time period of the first time period.

[0042] During the daily driving of a vehicle, the temperature of the body system and the temperature in the passenger compartment have a certain degree of impact on the vehicle's driving safety and the driving and riding experience. In this scenario, the temperature of the body system and the temperature in the passenger compartment can be adjusted and controlled through the thermal management control strategy set on the vehicle.

[0043] Among them, the body system temperature can be understood as the temperature of components such as batteries, motors, and electronic controls configured in the body, and the passenger compartment temperature can be understood as the spatial temperature of the driving and riding space provided by the vehicle for users.

[0044] Optionally, a corresponding heat pump system can be deployed on the vehicle to achieve thermal management control of the vehicle's body system temperature and passenger compartment temperature through the operation of the heat pump system. In this scenario, relevant status data of the heat pump system can be obtained and marked as heat pump status data of the heat pump system.

[0045] In the disclosed embodiment, relevant information about the thermal management status of the vehicle can be obtained through the heat pump status data, thereby identifying whether the vehicle is at risk of thermal runaway. That is, the heat pump data within a set prediction time period can be predicted based on the heat pump data within a set time period, thereby identifying whether the vehicle is at risk of thermal runaway within the prediction time period.

[0046] Optionally, the set time period can be determined as the first time period, and the heat pump status data of the vehicle in the first time period can be determined as the first heat pump status data, and the predicted time period can be determined as the second time period, wherein the second time period can be the next adjacent time period of the first time period, or it can be the next time period with a set time interval from the first time period, and no specific limitation is made here.

[0047] That is to say, based on the set prediction method of heat pump status data, based on the first heat pump status data in the first time period, the heat pump status data of the vehicle in the second time period that needs to be predicted can be predicted, and based on the predicted heat pump status data, it can be identified whether the vehicle has a thermal runaway risk in the second time period.

[0048] Optionally, when it is identified based on the predicted heat pump status data that the vehicle is at risk of thermal runaway within the second time period, the temperature status of the vehicle within the second time period can be adjusted. When the time reaches the second time period, the body system temperature and the passenger compartment temperature of the vehicle within the second time period can be adjusted, thereby reducing the possibility of the vehicle having a thermal runaway risk within the second time period.

[0049] The vehicle temperature adjustment method proposed in the present disclosure obtains first heat pump status data of the vehicle within a first time period, wherein, when it is identified that the vehicle has a thermal runaway risk within a second time period based on the first heat pump status data, the body system temperature and the passenger compartment temperature of the vehicle within the second time period are adjusted. In the present disclosure, heat pump status data of a vehicle in a second time period is predicted using first heat pump status data, and a thermal runaway risk assessment is performed on the vehicle in the second time period based on the predicted heat pump status data, thereby improving the timeliness of thermal runaway risk identification for the vehicle. Thermal runaway risk assessment and identification are performed using the heat pump status data of the heat pump system, thereby improving the precision and accuracy of thermal runaway risk identification. After identifying that a thermal runaway risk exists in the vehicle, the vehicle's body system temperature and passenger compartment temperature are adjusted to reduce the possibility of the vehicle experiencing a thermal runaway risk in the predicted second time period. The vehicle temperature is adjusted based on the thermal runaway risk identification result, thereby improving the timeliness of vehicle temperature regulation and thus improving the temperature control accuracy of the vehicle. Compared with the related art in which vehicle data prediction is achieved based on a constructed actual vehicle test bench, the cost of vehicle heat pump status data prediction and thermal runaway risk identification is reduced, and the occurrence of poor prediction and identification accuracy due to insufficient experimental data from the actual vehicle test bench is avoided. The safety of vehicle driving and vehicle drivers and passengers is improved, the service life of some vehicle components is extended, the maintenance cost of the vehicle is reduced, and the driving and riding experience of the vehicle is optimized.

[0050] In the above embodiment, the vehicle temperature adjustment method can be combined with Figure 2 Further understanding, Figure 2 FIG. 1 is a flow chart of a vehicle temperature adjustment method according to another embodiment of the present disclosure. Figure 2 As shown, the method includes:

[0051] S201, obtaining a target vehicle condition simulation prediction model deployed on a vehicle.

[0052] In an embodiment of the present disclosure, a model for simulating and predicting vehicle condition data such as heat pump data of the vehicle may be deployed on the vehicle. The model may be determined as a target vehicle condition simulation prediction model deployed on the vehicle.

[0053] Among them, the target vehicle simulation prediction model can be trained in the cloud, and the model trained in the cloud can be deployed on the vehicle through a preset deployment strategy, thereby realizing the deployment of the target vehicle control simulation prediction model on the vehicle.

[0054] Optionally, obtain the cloud-based vehicle condition simulation prediction model sent by the server and the vehicle-side model loading environment of the vehicle, wherein the vehicle-side model loading environment is constructed based on the signal access layer, data preprocessing layer, inference layer, feedback layer and fault tolerance layer.

[0055] In the embodiment of the present disclosure, the model that needs to be deployed and has been trained in the cloud and sent to the vehicle side via the server can be determined as a cloud-based vehicle condition simulation prediction model. After the vehicle receives the cloud-based simulation prediction model, it can build an environment for the cloud-based vehicle condition simulation prediction model on the vehicle side based on the method of building the environment required for model deployment in related technologies, and determine the built environment that can deploy the cloud-based vehicle condition simulation prediction model as the vehicle-side model loading environment.

[0056] Optionally, relevant signal access layer, data preprocessing layer, inference layer, feedback layer and fault tolerance layer can be deployed in the initial construction environment of the vehicle, wherein the signal access layer can be used to obtain the vehicle's multi-source sensor data through the vehicle-side service (CarService), and complete signal analysis and scheduling tasks through the central processing unit (CPU) resident process; the data preprocessing layer can be used to control the CPU to preprocess the input data based on the input requirements of the neural network processor (NeuralProcessing Unit, NPU), complete tensor conversion and batch processing; the inference layer is used to support the NPU to run the quantization model, support dynamic batch processing and priority scheduling; the feedback layer is used to control the NPU to notify the CPU through the interrupt mechanism to obtain the inference results and the CPU to post-process and output them to the decision layer of thermal control management after execution; and the fault tolerance layer is used to monitor the NPU status, switch the CPU lightweight model or historical data backup when timeout or abnormality occurs.

[0057] Furthermore, based on the deployment of the signal access layer, data preprocessing layer, inference layer, feedback layer and fault tolerance layer in the initial construction environment, a completed vehicle-side model loading environment is obtained.

[0058] Optionally, the cloud-based vehicle condition simulation prediction model is deployed in the vehicle-side model loading environment to obtain a target vehicle condition simulation prediction model deployed in the vehicle-side model loading environment.

[0059] In the embodiment of the present disclosure, based on the model deployment method in the relevant technology, the cloud-based vehicle condition simulation prediction model can be deployed to the established vehicle-side model loading environment, and the model deployed in the lane model loading environment can be determined as the target vehicle condition simulation prediction model deployed on the vehicle.

[0060] S202 : Based on the target vehicle condition simulation prediction model and the first heat pump state data, thermal runaway risk identification is performed on the vehicle in the second time period.

[0061] Optionally, a target state simulation prediction model, a target control simulation prediction model, and a target power simulation prediction model in the target vehicle condition simulation prediction model are obtained.

[0062] In the disclosed embodiment, the target vehicle condition simulation prediction model includes a plurality of trained simulation prediction models.

[0063] Among them, the model for state dimension data prediction included in the target vehicle condition simulation prediction model can be determined as the target state simulation prediction model in the target vehicle condition simulation prediction model, the model for control dimension data prediction can be determined as the target control simulation prediction model in the target vehicle condition simulation prediction model, and the model for power dimension data prediction can be determined as the target power simulation prediction model in the target vehicle condition simulation prediction model.

[0064] Optionally, the target state simulation prediction model, the target control simulation prediction model and the target power simulation prediction model included in the target vehicle condition simulation prediction model can be used to predict the second heat pump state data in the second time period based on the first heat pump state data in the first time period.

[0065] For example, Figure 3 As shown, set Figure 3 The control model shown is the target control simulation prediction model, the state model is the target state simulation prediction model, and the power model is the target power simulation prediction model. Figure 3 In the illustrated scenario, the control model 1 , the state model 1 , and the power model 1 can be used to predict data for the time period 2 based on the data for the time period 1 , thereby obtaining predicted data for the time period 2 .

[0066] In addition, the control model 2, state model 2 and power model 2 predict the data of time period 3 (not shown in the figure) based on the data of time period 2, and obtain the predicted data of time period 3 (not shown in the figure), and so on, until the predicted data of time period n is obtained.

[0067] In this scenario, based on the first heat pump state data, the state dimension data of the vehicle in the second time period can be predicted through the target state simulation prediction model, the control dimension data of the vehicle in the second time period can be predicted through the target control simulation prediction model, and the power dimension data of the vehicle in the second time period can be predicted through the target power simulation prediction model, thereby obtaining the second heat pump state data predicted for the vehicle in the second time period.

[0068] Optionally, predicted state data of the vehicle in the second time period is obtained according to the target state simulation prediction model and the first heat pump state data.

[0069] In the embodiment of the present disclosure, the data required for the target state simulation prediction model to perform simulation prediction can be obtained from the first heat pump state data as input data of the target state simulation prediction model, and the data can be input into the target state simulation prediction model to obtain the predicted data of the vehicle state dimension in the second time period output by the target state simulation prediction model as the predicted state data of the vehicle in the second time period.

[0070] The prediction of the compressor and passenger compartment status data can be understood in conjunction with the following:

[0071] Optionally, a compressor state parameter set and a compressor state control parameter set of the vehicle in the first time period, as well as a passenger compartment state parameter set and a passenger compartment state control parameter set of the vehicle in the first time period are obtained from the first heat pump state data.

[0072] In the disclosed embodiment, the first heat pump data may be screened to obtain relevant parameters that can characterize the compressor state of the vehicle within the first time period, and form a required state parameter set as the compressor state parameter set in the first heat pump data.

[0073] Furthermore, a set of related control parameters required to achieve a corresponding compressor state is obtained to form a compressor state control parameter set for the vehicle in a first time period.

[0074] Accordingly, the first heat pump data may be screened to obtain relevant parameters of the vehicle's passenger compartment state within the first time period, thereby forming a passenger compartment state parameter set of the vehicle within the first time period.

[0075] Furthermore, relevant control parameters required to achieve the corresponding passenger compartment state are obtained to form a passenger compartment state control parameter set of the vehicle in the first time period.

[0076] Optionally, the compressor status parameter set includes at least the compressor high-pressure parameter set, the compressor low-pressure parameter set, the compressor intake temperature parameter set, the compressor exhaust temperature parameter set, the compressor internal cooling temperature parameter set, the compressor high-pressure coolant heater inlet water temperature parameter set, the compressor high-pressure coolant heater outlet water temperature parameter set and the passenger compartment temperature parameter set of the vehicle in the first time period.

[0077] Optionally, the compressor state control parameter set includes at least the compressor active air intake grille (AGS) opening parameter set, the compressor cooling fan speed parameter set, the compressor blower speed parameter set, the compressor expansion valve opening parameter set, the compressor speed parameter set, the compressor positive temperature coefficient thermistor (PTC) power parameter set, the compressor mechanism hot water pump pulse width modulation (PWM) parameter set, the compressor ambient temperature parameter set, the passenger compartment temperature parameter set and the compressor internal and external circulation ratio parameter set of the vehicle in the first time period.

[0078] In the embodiment of the present disclosure, parameters of each component can be collected based on the data collection equipment deployed on the components to which the above-mentioned parameters belong, so as to obtain the required compressor state parameter set and compressor state control parameter set. The heat pump data of the first time period can also be predicted based on the compressor state dimension data of the previous time period corresponding to the first time period, so as to obtain the compressor state parameter set and compressor state control parameter set of the first time period. No specific limitation is made here.

[0079] Optionally, the passenger compartment status parameter set includes at least a passenger compartment temperature parameter set and a passenger compartment cold temperature parameter set of the vehicle in a first time period.

[0080] Optionally, the passenger compartment state control parameter set includes at least a blower air volume parameter set and a passenger compartment environment factor parameter set of the vehicle in a first time period.

[0081] In the embodiment of the present disclosure, parameters of each component can be collected based on the data collection equipment deployed on the components to which the above-mentioned parameters belong, so as to obtain the required passenger compartment state parameter set and passenger compartment state control parameter set. Alternatively, the heat pump data of the first time period can be predicted based on the passenger compartment state dimension data of the previous time period corresponding to the first time period, so as to obtain the passenger compartment state parameter set and passenger compartment state control parameter set of the first time period. No specific limitation is made here.

[0082] like Figure 4 As shown, the target state simulation prediction model can include Figure 4The target compressor state simulation prediction model and the target passenger compartment state simulation prediction model are shown.

[0083] Optionally, a predicted compressor state parameter set predicted for the vehicle in the second time period is obtained based on the compressor state parameter set and the compressor state control parameter set, and a target compressor state simulation prediction model in the target state simulation prediction model.

[0084] In the embodiment of the present disclosure, the model used for predicting the compressor state quantity in the target state simulation prediction model may be determined as the target compressor state simulation prediction model.

[0085] Among them, the compressor state parameter set and the compressor state control parameter set can be input into the target compressor state simulation prediction model, and the input parameter set can be subjected to feature extraction and data prediction by the target compressor state simulation prediction model, and then the corresponding prediction data can be output, wherein the output data can be determined as the predicted compressor state parameter set obtained by the target compressor state simulation prediction model based on the compressor state parameter set and the compressor state control parameter set.

[0086] Optionally, the predicted compressor status parameter set includes at least a predicted compressor high-pressure parameter set, a predicted compressor low-pressure parameter set, a predicted compressor intake temperature parameter set, a predicted compressor exhaust temperature parameter set, a predicted compressor internal cooling temperature parameter set, a predicted compressor high-pressure coolant heater (High Voltage Coolant Heater, HVCH) inlet water temperature parameter set and a predicted compressor high-pressure coolant heater outlet water temperature parameter set predicted by the vehicle within the second time period.

[0087] As shown in Table 1:

[0088] Table 1

[0089]

[0090]

[0091] As shown in the above table, a compressor state parameter set can be formed based on the various parameters included in the state parameters in the variable column, and a compressor state control parameter set can be formed based on the various parameters included in the control parameters in the variable column. Then, based on the target compressor state simulation prediction model, the various output parameters included in the output column shown in the above table are obtained to form a predicted compressor state parameter set.

[0092] It should be noted that any parameter set in the parameter sets shown in the above table may be composed of parameters obtained at each of multiple moments in the first time period.

[0093] Optionally, a predicted passenger compartment state parameter set predicted for the vehicle in the second time period is obtained based on the passenger compartment state parameter set and the passenger compartment state control parameter set, as well as the target passenger compartment state simulation prediction model in the target state simulation prediction model.

[0094] In the embodiment of the present disclosure, the model used for predicting the state quantity of the passenger compartment in the target state simulation prediction model may be determined as the target passenger compartment state simulation prediction model.

[0095] Among them, the crew compartment state parameter set and the crew compartment state control parameter set can be input into the target crew compartment state simulation prediction model, and the target crew compartment state simulation prediction model is used to perform feature extraction and data prediction on the input parameter set, and then output corresponding prediction data, wherein the output data can be determined as the predicted crew compartment state parameter set obtained by the target crew compartment state simulation prediction model based on the crew compartment state parameter set and the crew compartment state control parameter set.

[0096] Table 2

[0097]

[0098] Among them, the passenger compartment state parameter set composed of various parameters included in the state parameters in the above table, and the passenger compartment state control parameter set composed of various parameters included in the above control parameters and environmental factors can be input into the target passenger compartment state simulation prediction model to obtain the predicted passenger compartment temperature output by the model.

[0099] Optionally, the predicted passenger compartment state parameter set includes at least a predicted passenger compartment temperature of the vehicle predicted within the second time period.

[0100] In this scenario, a parameter set including at least the predicted passenger compartment temperature of the vehicle predicted in the second time period may be determined as the predicted passenger compartment state parameter set of the vehicle in the second time period.

[0101] Optionally, based on the predicted compressor state parameter set and the predicted passenger compartment state parameter set, predicted state data of the vehicle in the second time period is obtained.

[0102] In the embodiment of the present disclosure, the predicted compressor state parameter set and the predicted passenger compartment state parameter set can be integrated based on the data format set for the predicted state data, and the integrated data can be determined as the predicted state data of the vehicle in the second time period.

[0103] In the embodiment of the present disclosure, the prediction of the control dimension data of the compressor and the passenger compartment can be understood in combination with the following:

[0104] Optionally, prediction control data of the vehicle in the second time period is obtained according to the target control simulation prediction model and the first heat pump state data.

[0105] Optionally, a compressor air-conditioning side control parameter set of the vehicle in the first time period and a passenger compartment control parameter set of the vehicle in the first time period are obtained from the first heat pump state data.

[0106] In the embodiment of the present disclosure, the first heat pump status data can be screened to obtain the relevant control parameters of the vehicle's compressor on the air-conditioning side in the first time period, thereby forming a compressor air-conditioning side control parameter set of the vehicle's compressor in the first time period.

[0107] Correspondingly, relevant control parameters of the vehicle for the passenger compartment during the first time period can also be obtained from the first heat pump state data, thereby forming a passenger compartment control parameter set of the vehicle during the first time period.

[0108] Optionally, the compressor air-conditioning side control parameter set includes at least the compressor desired wind temperature (DWT) request parameter set of the vehicle in the first time period, the compressor ambient temperature parameter set, the compressor speed parameter set, the compressor target superheat parameter set, the compressor target subcooling parameter set, the compressor actual superheat parameter set, the compressor actual subcooling parameter set, the compressor intake temperature parameter set, the compressor internal cooling temperature parameter set, the compressor saturated high pressure parameter set, the compressor target saturated high parameter set, the compressor saturated low pressure parameter set, the compressor target saturated low pressure parameter set, the vehicle interior temperature parameter set, the compressor exhaust temperature parameter set, the compressor high-pressure coolant heater inlet water temperature parameter set, the compressor high-pressure coolant heater outlet water temperature parameter set, the compressor desired wind temperature request parameter set and the compressor water pump speed pulse width modulation parameter set.

[0109] Optionally, the passenger cabin control parameter set includes at least the passenger cabin ambient temperature parameter set of the vehicle in the first time period, the passenger cabin interior temperature parameter set, the passenger cabin main driver's seat set temperature, the passenger cabin co-driver's seat set temperature, the passenger cabin rear seat set temperature, the passenger cabin automatic level, the passenger cabin main driver's seat initial interior temperature, the co-driver's seat initial interior temperature, the passenger cabin rear seat initial interior temperature, vehicle speed, passenger cabin left side sunlight parameters and passenger cabin right side sunlight parameters.

[0110] like Figure 5 As shown, the target control simulation prediction model includes Figure 5 The target compressor control simulation prediction model and the target passenger compartment control simulation prediction model are shown.

[0111] Optionally, a predicted compressor control parameter set predicted for the vehicle in the second time period is obtained based on the compressor air-conditioning side control parameter set and the target compressor control simulation prediction model in the target control simulation prediction model.

[0112] In the embodiment of the present disclosure, the compressor air-conditioning side control parameter set can be input into the target compressor control simulation prediction model, and the input data can be feature extracted and data predicted by the target compressor control simulation prediction model to obtain the output data of the model as the predicted compressor control parameter set predicted for the vehicle in the second time period.

[0113] Among them, the prediction parameters output by the model can include Figure 5 The expansion valve opening, compressor speed, positive temperature coefficient thermistor power and hot water pump pulse width adjustment parameters shown may also include other control parameters on the air conditioning side of the compressor, which are not specifically limited here.

[0114] Optionally, the predicted compressor control parameter set includes at least a predicted expansion valve opening, a predicted compressor speed, a predicted positive temperature coefficient thermistor power, and a predicted hot water pump pulse width modulation parameter set of the vehicle in the second time period.

[0115] As shown in Table 3:

[0116] Table 3:

[0117]

[0118]

[0119] Among them, the compressor air-conditioning side control parameter set composed of each air-conditioning side control parameter in the above table can be input into the target compressor control simulation prediction model, and the input data can be feature extracted and data predicted by the target compressor control simulation prediction model, and the output parameters represented on the output are composed of the predicted compressor control parameter set output by the target compressor control simulation prediction model based on the output parameters.

[0120] Optionally, a predicted passenger compartment control parameter set predicted for the vehicle in the second time period is obtained according to the passenger compartment control parameter set and a target passenger compartment control simulation prediction model in the target control simulation prediction model.

[0121] In the embodiment of the present disclosure, the target control simulation prediction model also includes a target passenger compartment control simulation prediction model, which can use the passenger compartment control parameter set obtained from the first heat pump state data as the input data of the model. The input data is subjected to data feature extraction and data prediction through the target passenger compartment control simulation prediction model, thereby obtaining the control parameters of the vehicle's passenger compartment in the second time period output by the model, and then determining the set of various control parameters as the predicted passenger compartment control parameter set.

[0122] like Figure 5 As shown, the prediction data output by the model can include Figure 5 The blower speed and desired air temperature request shown may also include other passenger compartment control parameters, which are not specifically limited here.

[0123] Optionally, the predicted passenger cabin control parameter set includes at least a predicted blower speed and a predicted expected air temperature request predicted for the vehicle within the second time period.

[0124] As shown in Table 4:

[0125] Table 4:

[0126]

[0127] It can be understood that through the parameter set composed of various passenger compartment control parameters shown in the above table, and the target passenger compartment simulation prediction model, the blower speed and expected wind temperature request shown in the above table can be obtained as the control parameters of the passenger compartment predicted by the vehicle in the second time period, and then constitute the predicted passenger compartment control parameter set.

[0128] Optionally, predicted control data of the vehicle in the second time period is obtained based on the predicted compressor control parameter set and the predicted passenger compartment control parameter set.

[0129] In the embodiment of the present disclosure, data integration may be performed on the predicted compressor control parameter set and the predicted passenger compartment control parameter set based on a preset integration strategy for the control data, thereby obtaining integrated predicted control data.

[0130] Optionally, predicted power data of the vehicle in the second time period is obtained according to the target power simulation prediction model and the first heat pump state data.

[0131] like Figure 6 As shown, the target power simulation prediction model can include Figure 6 The target compressor power simulation prediction model, the target windage energy consumption simulation prediction model and the target fan power simulation prediction model are shown.

[0132] In an embodiment of the present disclosure, the target power simulation prediction model can be used to predict the power dimension data of the vehicle in the second time period based on the first heat pump state data, and the predicted data output by the model can be determined as the predicted power data of the vehicle in the second time period.

[0133] Optionally, a compressor power parameter set, a windage energy consumption parameter set, and a fan power parameter set of the vehicle in a first time period are obtained from the first heat pump state data.

[0134] In the embodiment of the present disclosure, the first heat pump status data can be screened to obtain the power parameters of the vehicle's compressor in the first time period, forming a compressor power parameter set for the first time period, and the windage energy consumption parameters of the vehicle in the first time period are obtained to form a windage energy consumption parameter set for the first time period, and the power parameters of the vehicle's cooling fan and other related fan components in the first time period are obtained to form a fan power parameter set for the first time period.

[0135] Optionally, the compressor power parameter set includes at least the compressor speed, compressor displacement, compressor saturated high pressure, compressor exhaust temperature, compressor intake temperature and compressor saturated low pressure of the vehicle in the first time period.

[0136] Optionally, the wind resistance energy consumption parameter set includes at least the vehicle speed, the active air intake grille opening and the cooling fan speed of the vehicle in the first time period.

[0137] Optionally, the fan power parameter set includes at least a cooling fan speed of the vehicle in a first time period.

[0138] It should be noted that the acquisition process of the above parameters can complete the understanding of the acquisition method of the compressor state parameter set and the compressor state control parameter set proposed in the above embodiments, which will not be repeated here.

[0139] Optionally, a predicted compressor power of the vehicle predicted in the second time period is obtained according to the compressor power parameter set and the target compressor power simulation prediction model in the target power simulation prediction model.

[0140] In an embodiment of the present disclosure, the compressor power parameter set can be input into the target compressor power simulation prediction model, and the compressor power of the vehicle in the second time period can be predicted based on the compressor power parameter set through the model capability of the model, and the prediction result output by the model is determined as the predicted compressor power of the vehicle in the second time period.

[0141] As shown in Table 5:

[0142] Table 5:

[0143]

[0144] It can be seen from the above table that the parameters included in the input column are the compressor power parameters of the input target compressor power simulation prediction model. Based on the set of compressor power parameters and the model capability of the target compressor power simulation model, the predicted compressor power of the vehicle's compressor in the second time period shown in the output column of the above table can be obtained as the predicted compressor power.

[0145] As an example, the predicted compressor power can be obtained based on a set algorithm. The algorithm formula is as follows:

[0146]

[0147] In the above formula, v represents the power and speed of the air-conditioning compressor, w represents the compressor displacement, and p represents the compressor speed. high Indicates saturated high pressure, T2 indicates compressor exhaust temperature, T1 indicates compressor inlet temperature, p low Indicates the saturation low pressure correlation, where k1 and i represent the coefficients to be fitted.

[0148] Optionally, the predicted wind resistance energy consumption of the vehicle in the second time period is obtained according to the wind resistance energy consumption parameter set and the target wind resistance power simulation prediction model in the target power simulation prediction model.

[0149] In the embodiment of the present disclosure, the wind resistance energy consumption parameter set can be input into the target wind resistance energy consumption simulation prediction model, and the wind resistance energy consumption of the vehicle in the second time period can be predicted based on the wind resistance energy consumption parameter set through the model capability of the model, and the prediction result output by the model is determined as the predicted wind resistance energy consumption of the vehicle in the second time period.

[0150] As shown in Table 6:

[0151] Table 6:

[0152]

[0153] It can be seen from the above table that the various parameters included in the input column are the various wind resistance energy consumption parameters input into the target wind resistance energy consumption simulation prediction model. Based on the set of various wind resistance energy consumption parameters and the model capability of the target wind resistance energy consumption simulation model, the predicted wind resistance energy consumption of the vehicle in the second time period shown in the output column of the above table can be obtained as the predicted wind resistance energy consumption.

[0154] As an example, the predicted wind resistance energy consumption can be obtained based on a set algorithm. The algorithm formula is as follows:

[0155] C a =k2*ags 2 +k3*ags+k4

[0156] In the above formula, Cd Indicates the drag coefficient, F a The wind resistance energy consumption is expressed as, ρ represents air density, v represents vehicle speed, A represents frontal area, and k2, k3, and k4 represent fitting coefficients.

[0157] Optionally, a predicted fan power of the vehicle predicted in the second time period is obtained according to the fan power parameter set and a target fan power simulation prediction model in the target power simulation prediction model.

[0158] In an embodiment of the present disclosure, a fan power parameter set can be input into a target fan power simulation prediction model, and the fan power of the vehicle in the second time period can be predicted based on the fan power parameter set through the model capability of the model, and the prediction result output by the model is determined as the predicted fan power of the vehicle in the second time period.

[0159] As shown in Table 7:

[0160] Table 7:

[0161]

[0162] As can be seen from the above table, the parameters included in the input column are the fan power parameters of the input target fan power simulation prediction model. Based on the set of fan power parameters and the model capability of the target fan power simulation model, the fan power predicted for the vehicle in the second time period shown in the output column of the above table can be obtained as the predicted fan power.

[0163] As an example, the predicted fan power can be obtained based on a set algorithm. The algorithm formula is as follows:

[0164] F f =k5*fan 2 +k6fan+b

[0165] In the above formula, fan represents the fan speed, k5 and k6 represent fitting coefficients, and b represents a constant.

[0166] Optionally, based on the predicted compressor power, the predicted windage energy consumption and the predicted fan power, predicted power data of the vehicle in the second time period is obtained.

[0167] In an embodiment of the present disclosure, data integration is performed on the predicted compressor power, predicted windage energy consumption, and predicted fan power based on a preset data integration strategy, and the integrated data is determined as the predicted power data of the vehicle in the second time period.

[0168] Optionally, second heat pump state data predicted for the vehicle in the second time period is obtained based on the predicted state data, predicted control data and predicted power data, and thermal runaway risk identification is performed on the vehicle in the second time period based on the second heat pump state data.

[0169] Optionally, the predicted state data, the predicted control data and the predicted power data are integrated to obtain second heat pump state data.

[0170] In the disclosed embodiment, the predicted state data, the predicted control data, and the predicted power data may be further integrated, and the integrated data may be determined as the second heat pump state data predicted for the vehicle in the second time period.

[0171] As an example, the target vehicle condition simulation prediction model is deployed in a vehicle-side model loading environment based on the signal access layer, data preprocessing layer, inference layer, feedback layer, and fault tolerance layer. In this example, the prediction process of the target vehicle condition simulation prediction model can be understood in combination with the following:

[0172] For any data prediction round, the data prediction link can be that the CPU receives the first heat pump status data as an input signal from the vehicle service end, processes the signal and sends it to the NPU, and the NPU predicts the relevant data of the vehicle in the second time period based on the first heat pump status data. After the NPU prediction is completed, it returns the inference structure to the CPU, and finally the CPU outputs the complete prediction data consisting of the predicted state data obtained based on the target state simulation prediction model, the predicted control data obtained based on the target control simulation prediction model, and the predicted power data obtained based on the target power simulation prediction model, that is, the second heat pump status data predicted for the vehicle in the second time period.

[0173] Optionally, in response to the second heat pump data matching a preset thermal runaway risk determination condition, it is determined that the vehicle has a thermal runaway risk within the second time period.

[0174] In the disclosed embodiment, respective thermal runaway risk determination conditions can be set based on the various parameters included in the second heat pump status data. For any parameter, when the parameter matches its corresponding thermal runaway risk determination condition, it can be determined that the parameter may be a thermal runaway risk parameter when the vehicle is at risk of thermal runaway.

[0175] In this scenario, when a set proportion of parameters in the second heat pump status data are thermal runaway risk parameters, it can be determined that the second heat pump status data matches the corresponding thermal runaway risk judgment condition. In this scenario, it can be determined that based on the current vehicle's thermal control management strategy, the vehicle is at risk of thermal runaway when the second time period is reached.

[0176] S203 , in response to identifying that the vehicle has a thermal runaway risk in the second time period, updating the thermal control strategy in the second time period to adjust the vehicle body system temperature and the passenger compartment temperature in the second time period.

[0177] Optionally, in response to identifying that the vehicle is at risk of thermal runaway within the second time period, an updated thermal control strategy for the vehicle is generated based on the second heat pump status data, wherein the updated thermal control strategy is obtained based on a thermal control strategy acquisition model deployed in the cloud.

[0178] In the embodiment of the present disclosure, a thermal control strategy model is deployed on the cloud server corresponding to the vehicle side. When it is identified that the vehicle is at risk of thermal runaway in the second time period, a new thermal control strategy for the vehicle to use in the second time period can be generated by the thermal control strategy model deployed on the cloud server. This thermal control strategy can be determined as the updated thermal control strategy of the vehicle.

[0179] As an example, Figure 7 As shown, in Figure 7 In the scenario shown, the cloud server is deployed with Figure 7 The big data platform and vehicle thermal management simulation system shown in the figure, wherein the vehicle thermal management simulation system is deployed with Figure 7 The simulation prediction model and thermal control strategy model are shown.

[0180] In this example, when the output result of the target vehicle condition simulation prediction model deployed on the vehicle identifies that the vehicle has a risk of thermal runaway in the second time period, the identification result can be uploaded to the interactive link between the vehicle and the cloud. Figure 7 The big data platform shown in the figure receives the corresponding data and transmits it to Figure 7 The vehicle thermal management simulation system is shown.

[0181] like Figure 7 As shown, the vehicle thermal management simulation system is deployed with a thermal control strategy model. In this scenario, the thermal control strategy model can obtain the vehicle's current thermal control strategy data through the information interaction link between the vehicle and the cloud, and generate a new thermal control strategy that needs to be updated for the vehicle based on the relevant data transmitted from the big data platform and each vehicle, as the vehicle's updated thermal control strategy.

[0182] like Figure 7 As shown in Figure 2, there is an interactive link between the vehicle and the thermal control strategy model that can transmit information in both directions. Figure 7 In the illustrated scenario, after the thermal control strategy model generates an updated thermal control strategy, the updated thermal control strategy can be sent to the vehicle based on the interactive link, so that the vehicle can update its currently used thermal control strategy.

[0183] It should be noted that Figure 7The simulation prediction model shown is used to obtain an undeployed cloud vehicle condition simulation prediction model and send the cloud vehicle condition simulation prediction model to the vehicle so that the vehicle can be deployed with the target vehicle condition simulation prediction model.

[0184] Optionally, based on the updated thermal control strategy, the thermal control strategy of the vehicle corresponding to the second time period is updated to adjust the body system temperature and passenger compartment temperature of the vehicle in the second time period, wherein the updated thermal control strategy is transmitted to the vehicle based on the cloud.

[0185] In the disclosed embodiment, the thermal control strategy of the vehicle reaching the second time period can be updated based on the thermal control strategy update method in the related art, wherein the updated thermal control strategy can overwrite the original thermal control strategy, or other preset update methods can be used, which are not specifically limited here.

[0186] Furthermore, after reaching the second time period and completing the update of the thermal control strategy, the vehicle's body system temperature and passenger compartment temperature in the second time period will be adjusted along with the implementation of the new thermal control strategy, thereby reducing the possibility of thermal runaway risk in the vehicle during the second time period.

[0187] It should be noted that the thermal control strategy update of the vehicle based on the updated thermal control strategy proposed in the embodiment of the present disclosure can be performed in a scenario where the vehicle is offline, or in other scenarios where the thermal control strategy update can be achieved and will not affect the vehicle's driving. No specific limitation is made here.

[0188] Optionally, in response to identifying that there is no thermal runaway risk for the vehicle in the second time period, thermal runaway risk identification is continued for the vehicle in a third time period based on the target wheel simulation prediction model and the second heat pump status data, wherein the third time period is the next predicted time period of the second time period.

[0189] In the embodiment of the present disclosure, the vehicle may not be at risk of thermal runaway in the second time period. It can be understood that when it is identified based on the second heat pump status data that the vehicle is not at risk of thermal runaway in the second time period, there is no need to update the thermal control strategy used by the vehicle in the second time period.

[0190] Optionally, based on the second heat pump status data predicted in the second time period, the heat pump status data of the vehicle in the next third time period that needs to be predicted can be predicted to continue to identify the thermal runaway risk of the vehicle in the third time period.

[0191] It can be understood that, in the embodiment of the present disclosure, a single-step iteration method can be used to predict the heat pump status data in each time period. For the first time period of the current time step, the heat pump status data of the second time period corresponding to the next time step can be predicted based on the first heat pump status data in the first time period, and the data of the third time period corresponding to the next time step of the second time period can be predicted based on the predicted second heat pump status data of the second time period corresponding to the next time step, and so on.

[0192] It should be noted that each simulation prediction model in the target vehicle condition simulation prediction model deployed on the vehicle proposed in the embodiment of the present disclosure can be obtained by training with relevant data with the vehicle identification of the vehicle, so that the prediction state of the target vehicle condition simulation prediction model deployed on the vehicle is more adapted to the actual prediction requirements of the vehicle.

[0193] The vehicle temperature adjustment method proposed in the present disclosure predicts the heat pump status data in each time period based on a single-step iteration method through the prediction of each simulation prediction model included in the target vehicle condition simulation prediction model. Compared with the data prediction based on the sub-module model in the related art, the method avoids the error in model prediction accuracy caused by the series connection of sub-modules, improves the prediction accuracy of the target vehicle condition simulation prediction model, determines the thermal runaway risk of the vehicle in the second time period based on the predicted heat pump status data, improves the timeliness of the vehicle's thermal runaway risk identification, determines and identifies the thermal runaway risk based on the heat pump status data of the heat pump system, improves the precision and accuracy of the thermal runaway risk identification, adjusts the vehicle's body system temperature and passenger compartment temperature after identifying the thermal runaway risk of the vehicle, reduces the possibility of the vehicle having a thermal runaway risk in the predicted second time period, updates the thermal control strategy to adjust the vehicle's body system temperature and passenger compartment temperature when it reaches the second time period, and improves the vehicle's temperature control timeliness and temperature control accuracy.

[0194] Regarding the target vehicle condition simulation prediction model proposed in the above embodiment, the present disclosure also proposes a training method for the simulation prediction model, which can be combined with Figure 8 understand, Figure 8 FIG. 1 is a flow chart of a training method for a simulation prediction model according to an embodiment of the present disclosure. Figure 8 As shown, the method includes:

[0195] S801, obtaining candidate state simulation prediction models, candidate control simulation prediction models, and candidate power simulation prediction models to be trained.

[0196] In the disclosed embodiment, the model that needs to be deployed on the vehicle can be trained in the cloud, and the trained model can be sent to the vehicle side, wherein the model for predicting the state dimension data of the vehicle can be determined as the candidate state simulation prediction model to be trained, the model for predicting the control dimension data of the vehicle can be determined as the candidate control simulation prediction model to be trained, and the model for predicting the power dimension data of the vehicle can be determined as the candidate power simulation prediction model to be trained.

[0197] As an example, the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model can be constructed based on the long short-term memory network (LSTM) in the relevant technology, that is, the model structure of the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model can be a timing network structure obtained based on LSTM.

[0198] Can be combined Figure 9 The schematic diagram of the LSTM model is shown as Figure 9 As shown:

[0199] y t The compressor high pressure, compressor low pressure, intake temperature, exhaust temperature, saturated high pressure, etc. at time t

[0200] x t are the factors that affect the change of state quantity at time t, including state quantity and control quantity.

[0201] c t Represents the cell state of the LSTM network, which determines which information can be retained and input to the next state and which will be forgotten and discarded.

[0202] The LSTM network consists of three gated networks: forget gate, input gate, and output gate, and is used to learn long- and short-cycle temporal relationships.

[0203] Forget gate: Forget gate f t Determine which information is forgotten and discarded. An activation function (sigmoid) is used to transform the hidden state h of the previous moment into t-1 And input x t Mapped to a value between 0 and 1, by comparing it with the cell state c at the previous moment t-1 The multiplication operation of determines the forgetting and discarding ratio.

[0204] f t =σ(W f [h t-1 ,x t ]+b f )

[0205] Input gate: Input gate i t Determine the new input information x t The degree of influence on the current cell state. Determine which information needs to be updated. Cell state c is updated by multiplication operation. t Updates.

[0206] i t =σ(W i [h t-1 ,x t ]+b i )

[0207]

[0208] Output gate: output gate o t Controls the current input x t , the hidden state h at the previous moment t-1 The effect on the output is achieved by multiplying the cell state c t Get the final output h t .

[0209] o t =σ(W o [h t-1 ,x t ]+b o )

[0210] h t =o t *tanh(c t )

[0211] The final output y t =W t h t . The multidimensional vector h t Mapped to a multidimensional scalar y t .

[0212] S802, based on the sample heat pump state data of the vehicle in the sample time period, the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model are trained respectively to obtain the trained target state simulation prediction model, the target control simulation prediction model and the target power simulation prediction model.

[0213] In an embodiment of the present disclosure, for a vehicle that currently needs to be deployed with a model, data of the heat pump system within a set historical time period can be extracted from the relevant data of the vehicle, thereby obtaining the heat pump status data of the vehicle's heat pump system within the historical time period, wherein the historical time period for data extraction can be determined as a sample time period, and the heat pump status data extracted within the sample time period can be determined as the sample heat pump status data of the vehicle.

[0214] Optionally, based on the sample generation method in the related art, samples required for model training of the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model can be generated based on the sample heat pump state data.

[0215] Furthermore, based on the model training method in the relevant technology, the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model are respectively trained until the training is completed, and the trained models are respectively determined as the trained target state simulation prediction model, the target control simulation prediction model and the target power simulation prediction model.

[0216] S803 , integrating the target control simulation prediction model, the target state simulation prediction model, and the target power simulation prediction model to obtain an integrated cloud-based vehicle condition simulation prediction model.

[0217] In the embodiment of the present disclosure, based on the model integration algorithm in the relevant technology, the trained target control simulation prediction model, target state simulation prediction model and target power simulation prediction model can be integrated algorithmically, and the model formed by integrating these models can be determined as the cloud vehicle condition simulation prediction model of the vehicle on the cloud server.

[0218] S804: deploying the cloud vehicle condition simulation prediction model on the vehicle to obtain a target vehicle condition simulation prediction model on the vehicle.

[0219] Among them, the target vehicle condition simulation prediction model is used to achieve the above Figures 1 to 7 The vehicle temperature adjustment method proposed in the embodiment.

[0220] In the disclosed embodiment, after obtaining the cloud-based vehicle condition simulation prediction model, the cloud server can send it to the vehicle, and based on the model deployment method configured on the vehicle, deploy the cloud server to the environment built for it by the vehicle, thereby realizing the deployment of the cloud-based vehicle condition simulation prediction model on the vehicle.

[0221] Among them, the model after the deployment operation on the vehicle is completed can be determined as the target vehicle condition simulation prediction model on the vehicle.

[0222] The simulation prediction model training method proposed in the present invention trains the candidate state simulation prediction model, candidate control simulation prediction model and candidate power simulation prediction model of the vehicle in the cloud, thereby improving the training efficiency of each simulation prediction model and optimizing the training effect of each simulation prediction model. The cloud-based vehicle condition simulation prediction model obtained by integrating the trained models is deployed on the vehicle side, eliminating the need for model training on the vehicle side, saving training algorithm resources on the vehicle side, reducing the model training cost and deployment cost, optimizing the training method and training effect of the target vehicle condition simulation prediction model, and improving the prediction accuracy and prediction efficiency of the heat pump state data in the scenario of heat pump state data prediction and thermal runaway risk identification based on the target vehicle condition simulation prediction model, thereby improving the recognition accuracy and recognition efficiency of thermal runaway risk identification, improving the driving and riding safety of vehicles deployed with the target vehicle condition simulation prediction model, and optimizing the driving and riding experience of the vehicle.

[0223] In the above embodiment, the model training of the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model can be combined with Figure 10 Further understanding, Figure 10 FIG. 1 is a flow chart of a training method for a simulation prediction model according to another embodiment of the present disclosure. Figure 10 As shown, the method includes:

[0224] S1001 , performing model training on a candidate state simulation prediction model based on sample heat pump state data of a vehicle in a sample time period to obtain a trained target state simulation prediction model.

[0225] Optionally, a candidate compressor state simulation prediction model and a candidate passenger compartment state simulation prediction model are obtained from the candidate state simulation prediction models.

[0226] In the embodiment of the present disclosure, the candidate state simulation prediction model includes a model for predicting state dimension data of the compressor, namely, the candidate compressor state simulation prediction model, and a model for predicting state dimension data of the passenger compartment, namely, the candidate passenger compartment state simulation prediction model.

[0227] Optionally, a first training sample set is generated based on the sample heat pump state data, the sample compressor state parameter set and the sample compressor state control parameter set of the vehicle within the sample time period, to train the candidate compressor state simulation prediction model to obtain a trained target compressor state simulation prediction model.

[0228] In the embodiment of the present disclosure, relevant parameters of the vehicle's compressor state in a sample time period can be obtained from the sample heat pump state data to form a sample compressor state parameter set, and relevant parameters of the state control required to achieve each sample compressor state parameter can be obtained to form a sample compressor state control parameter set.

[0229] Furthermore, based on the sample generation method in the related art, sample generation processing is performed on the sample compressor state parameter set and the sample compressor state control parameter set, and the generated samples are determined as the first training sample set used for model training of the candidate compressor state simulation prediction model.

[0230] In this scenario, based on the model training method in the relevant technology, the candidate compressor state simulation prediction model can be trained through each first training sample in the first training sample set until the training is completed, thereby obtaining a trained target compressor state simulation prediction model.

[0231] Among them, the corresponding model training end condition can be set based on the output result of the model, and the corresponding model training end condition can also be set based on the training round, which is not specifically limited here.

[0232] For the candidate compressor state simulation prediction model in the training process, when it is identified that it meets the preset model training end conditions, the model training can be ended, and the model obtained at the end of the last round of training is determined as the trained target compressor state simulation prediction model.

[0233] Optionally, a second training sample set is generated based on the sample heat pump state data, the sample passenger compartment state parameter set and the sample passenger compartment control parameter set of the vehicle within the sample time period, so as to train the candidate passenger compartment simulation prediction model and obtain a trained target passenger compartment state simulation prediction model.

[0234] In the disclosed embodiment, relevant parameters of the passenger compartment state of the vehicle within the sample time period can be obtained from the sample heat pump state data to form a sample passenger compartment state parameter set, and the state control parameters required to achieve each sample passenger compartment state parameter can be obtained to form a sample passenger compartment state control parameter set.

[0235] Furthermore, based on the sample generation method in the related art, sample generation processing is performed on the sample passenger compartment state parameter set and the sample passenger compartment control parameter set, and the generated samples are determined as the second training sample set used for model training of the candidate passenger compartment state simulation prediction model.

[0236] It should be noted that the training of the candidate passenger compartment state simulation prediction model can be understood in conjunction with the training process of the above-mentioned candidate compressor state simulation prediction model, and will not be repeated here.

[0237] Optionally, a target state simulation prediction model is obtained based on a target compressor state simulation prediction model and a target passenger compartment state simulation prediction model.

[0238] In an embodiment of the present disclosure, based on a preset model integration method, the trained target compressor state simulation prediction model and the trained target passenger compartment state simulation prediction model can be integrated, and the integrated model can be determined as the target state simulation prediction model.

[0239] S1002 , performing model training on a candidate control simulation prediction model based on sample heat pump state data of the vehicle in a sample time period to obtain a trained target control simulation prediction model.

[0240] Optionally, a candidate compressor control simulation prediction model and a candidate passenger compartment control simulation prediction model are obtained from the candidate control simulation prediction models.

[0241] In the embodiment of the present disclosure, the candidate control simulation prediction model includes a model for predicting control dimension data for the compressor, namely, the candidate compressor control simulation prediction model, and a model for predicting control dimension data for the passenger compartment, namely, the candidate passenger compartment control simulation prediction model.

[0242] Optionally, a third training sample set is generated based on the sample compressor air-conditioning side control parameter set of the vehicle within the sample time period in the sample heat pump state data to train the candidate compressor control simulation prediction model to obtain a trained target compressor control simulation prediction model.

[0243] In the disclosed embodiment, relevant control parameters of the vehicle's compressor on the air conditioning side during a sample time period can be obtained from the sample heat pump status data to form a sample compressor air conditioning side control parameter set.

[0244] Furthermore, based on the sample generation method in the related art, sample generation processing is performed on the sample compressor air-conditioning side control parameter set, and the generated samples are determined as the third training sample set used for model training of the candidate compressor control simulation prediction model.

[0245] It should be noted that the training of the candidate compressor control simulation prediction model can be understood in conjunction with the training process of the above-mentioned candidate compressor state simulation prediction model, and will not be described in detail here.

[0246] Optionally, a fourth training sample set is generated based on a sample passenger compartment control parameter set of the vehicle within a sample time period in the sample heat pump state data to train a candidate passenger compartment control simulation prediction model to obtain a trained target passenger compartment control simulation prediction model.

[0247] In the disclosed embodiment, relevant control parameters of the vehicle's passenger compartment in a sample time period may be obtained from the sample heat pump status data to form a sample passenger compartment control parameter set.

[0248] Furthermore, based on the sample generation method in the related art, sample generation processing is performed on the sample passenger compartment control parameter set, and the generated samples are determined as the fourth training sample set used in model training of the candidate passenger compartment control simulation prediction model.

[0249] It should be noted that the training of the candidate passenger compartment control simulation prediction model can be understood in conjunction with the training process of the above-mentioned candidate compressor state simulation prediction model, and will not be repeated here.

[0250] Optionally, a target control simulation prediction model is obtained based on the target compressor control simulation prediction model and the target passenger compartment control simulation prediction model.

[0251] In the embodiment of the present disclosure, based on a preset model integration method, the trained target compressor control simulation prediction model and the trained target passenger compartment control simulation prediction model can be integrated, and the integrated model can be determined as the target control simulation prediction model.

[0252] S1003 , performing model training on a candidate power simulation prediction model based on sample heat pump status data of the vehicle in a sample time period to obtain a trained target power simulation prediction model.

[0253] Optionally, a candidate compressor power simulation prediction model, a candidate windage energy consumption simulation prediction model, and a candidate fan power simulation prediction model are obtained from the candidate power simulation prediction models.

[0254] In the embodiment of the present disclosure, the candidate power simulation prediction model includes a model for performing data prediction on compressor power, namely, a candidate compressor power simulation prediction model, a model for performing data prediction on vehicle windage energy consumption, namely, a candidate windage energy consumption simulation prediction model, and a model for performing data prediction on fan power of equipment such as the vehicle's cooling fan, namely, a candidate fan power simulation prediction model.

[0255] Optionally, a fifth training sample set is generated based on a sample compressor power parameter set of the vehicle within a sample time period in the sample heat pump state data to train the candidate compressor power simulation prediction model to obtain a trained target compressor power simulation prediction model.

[0256] In the disclosed embodiment, the power parameters of the vehicle's compressor within a sample time period may be obtained from the sample heat pump status data to form a sample compressor power parameter set.

[0257] Furthermore, based on the sample generation method in the related art, sample generation processing is performed on the sample compressor power parameter set, and the generated samples are determined as the fifth training sample set used for model training of the candidate compressor power simulation prediction model.

[0258] It should be noted that the training of the candidate compressor power simulation prediction model can be understood in conjunction with the training process of the above-mentioned candidate compressor state simulation prediction model, and will not be repeated here.

[0259] Optionally, a sixth training sample set is generated based on a sample wind resistance energy consumption parameter set of the vehicle within a sample time period in the sample heat pump state data to train the candidate wind resistance energy consumption simulation prediction model to obtain a trained target wind resistance energy consumption simulation prediction model.

[0260] In the embodiment of the present disclosure, the wind resistance energy consumption parameters of the vehicle within the sample time period can be obtained from the sample heat pump status data to form a sample wind resistance energy consumption parameter set.

[0261] Furthermore, based on the sample generation method in the related art, sample generation processing is performed on the sample wind resistance energy consumption parameter set, and the generated samples are determined as the sixth training sample set used in model training of the candidate wind resistance energy consumption simulation prediction model.

[0262] It should be noted that the training of the candidate wind resistance energy consumption simulation prediction model can be understood in conjunction with the training process of the above-mentioned candidate compressor state simulation prediction model, and will not be repeated here.

[0263] Optionally, a seventh training sample set is generated based on a sample fan power parameter set of the vehicle within a sample time period in the sample heat pump state data to train the candidate fan power simulation prediction model to obtain a trained target fan power simulation prediction model.

[0264] In the disclosed embodiment, power parameters of the vehicle's cooling fan and other equipment within a sample time period may be obtained from the sample heat pump status data to form a sample fan power parameter set.

[0265] Furthermore, based on the sample generation method in the related art, sample generation processing is performed on the sample fan power parameter set, and the generated samples are determined as the seventh training sample set used in model training of the candidate fan power simulation prediction model.

[0266] It should be noted that the training of the candidate fan power simulation prediction model can be understood in conjunction with the training process of the above-mentioned candidate compressor state simulation prediction model, and will not be repeated here.

[0267] Optionally, a target power simulation prediction model is obtained based on a target compressor power simulation prediction model, a target windage energy consumption simulation prediction model, and a target fan power simulation prediction model.

[0268] In the embodiment of the present disclosure, based on a preset model integration method, the trained standard compressor power simulation prediction model, the target wind resistance energy consumption simulation prediction model and the target fan power simulation prediction model can be integrated, and then the integrated model can be determined as the target power simulation prediction model.

[0269] The simulation prediction model training method proposed in the present invention trains the candidate state simulation prediction model, candidate control simulation prediction model and candidate power simulation prediction model of the vehicle in the cloud, thereby improving the training efficiency of each simulation prediction model and optimizing the training effect of each simulation prediction model. The cloud-based vehicle condition simulation prediction model obtained by integrating the trained models is deployed on the vehicle side, eliminating the need for model training on the vehicle side, saving training algorithm resources on the vehicle side, reducing the model training cost and deployment cost, optimizing the training method and training effect of the target vehicle condition simulation prediction model, and improving the prediction accuracy and prediction efficiency of the heat pump state data in the scenario of heat pump state data prediction and thermal runaway risk identification based on the target vehicle condition simulation prediction model, thereby improving the recognition accuracy and recognition efficiency of thermal runaway risk identification, improving the driving and riding safety of vehicles deployed with the target vehicle condition simulation prediction model, and optimizing the driving and riding experience of the vehicle.

[0270] Corresponding to the vehicle temperature adjustment methods proposed in the above-mentioned embodiments, an embodiment of the present disclosure further proposes a vehicle temperature adjustment device. Since the vehicle temperature adjustment device proposed in the embodiment of the present disclosure corresponds to the vehicle temperature adjustment methods proposed in the above-mentioned embodiments, the implementation method of the above-mentioned vehicle temperature adjustment method is also applicable to the vehicle temperature adjustment device proposed in the embodiment of the present disclosure, and will not be described in detail in the following embodiments.

[0271] Figure 11 FIG. 1 is a schematic structural diagram of a vehicle temperature adjustment device according to an embodiment of the present disclosure. Figure 11 As shown, the vehicle temperature adjustment device 1100 includes an adjustment module 1101, wherein:

[0272] The adjustment module 1101 is configured to adjust the vehicle's body system temperature and passenger compartment temperature in response to identifying a risk of thermal runaway of the vehicle in a second time period based on first heat pump status data of the vehicle in a first time period, wherein the second time period is a time period subsequent to the first time period.

[0273] In the disclosed embodiment, the adjustment module 1101 is further used to: obtain a target vehicle condition simulation prediction model deployed on the vehicle; obtain second heat pump status data of the vehicle in a second time period based on the target vehicle condition simulation prediction model and the first heat pump status data, so as to identify the thermal runaway risk of the vehicle in the second time period; in response to identifying the thermal runaway risk of the vehicle in the second time period, update the thermal control strategy in the second time period, so as to adjust the body system temperature and the passenger compartment temperature of the vehicle in the second time period.

[0274] In the embodiment of the present disclosure, the adjustment module 1101 is also used to: in response to identifying that there is no thermal runaway risk for the vehicle in the second time period, continue to identify the thermal runaway risk of the vehicle in the third time period based on the target wheel simulation prediction model and the second heat pump status data, wherein the third time period is the next predicted time period of the second time period.

[0275] In the embodiment of the present disclosure, the adjustment module 1101 is also used to: obtain the target state simulation prediction model, the target control simulation prediction model and the target power simulation prediction model in the target vehicle condition simulation prediction model; obtain the predicted state data of the vehicle in the second time period based on the target state simulation prediction model and the first heat pump state data; obtain the predicted control data of the vehicle in the second time period based on the target control simulation prediction model and the first heat pump state data; obtain the predicted power data of the vehicle in the second time period based on the target power simulation prediction model and the first heat pump state data; obtain the second heat pump state data predicted for the vehicle in the second time period based on the predicted state data, the predicted control data and the predicted power data, and identify the thermal runaway risk of the vehicle in the second time period based on the second heat pump state data.

[0276] In the embodiment of the present disclosure, the adjustment module 1101 is also used to: obtain the compressor state parameter set and the compressor state control parameter set of the vehicle in the first time period, as well as the passenger compartment state parameter set and the passenger compartment state control parameter set of the vehicle in the first time period from the first heat pump state data; obtain the predicted compressor state parameter set predicted for the vehicle in the second time period based on the compressor state parameter set and the compressor state control parameter set, and the target compressor state simulation prediction model in the target state simulation prediction model; obtain the predicted passenger compartment state parameter set predicted for the vehicle in the second time period based on the passenger compartment state parameter set and the passenger compartment state control parameter set, and the target passenger compartment state simulation prediction model in the target state simulation prediction model; and obtain the predicted state data of the vehicle in the second time period based on the predicted compressor state parameter set and the predicted passenger compartment state parameter set.

[0277] In the embodiment of the present disclosure, the compressor state parameter set includes at least the compressor high-pressure parameter set, the compressor low-pressure parameter set, the compressor intake temperature parameter set, the compressor exhaust temperature parameter set, the compressor internal cooling temperature parameter set, the compressor high-pressure coolant heater inlet water temperature parameter set, the compressor high-pressure coolant heater outlet water temperature parameter set and the passenger compartment temperature parameter set in the first time period of the vehicle; the compressor state control parameter set includes at least the compressor active air intake grille opening parameter set, the compressor cooling fan speed parameter set, the compressor blower speed parameter set, the compressor expansion valve opening parameter set, the compressor speed parameter set, the compressor positive temperature coefficient thermistor power parameter set, the compressor mechanism hot water pump pulse width modulation parameter set, the compressor ambient temperature parameter set, and the passenger compartment temperature parameter set in the first time period of the vehicle. and the compressor internal and external circulation ratio parameter set; the predicted compressor state parameter set, at least including the predicted compressor high-pressure parameter set, the predicted compressor low-pressure parameter set, the predicted compressor intake temperature parameter set, the predicted compressor exhaust temperature parameter set, the predicted compressor internal cooling temperature parameter set, the predicted compressor high-pressure coolant heater inlet water temperature parameter set and the predicted compressor high-pressure coolant heater outlet water temperature parameter set predicted by the vehicle in the second time period; the passenger compartment state parameter set, at least including the passenger compartment temperature parameter set and the passenger compartment internal cooling temperature parameter set of the vehicle in the first time period; the passenger compartment state control parameter set, at least including the blower air volume parameter set and the passenger compartment environmental factor parameter set of the vehicle in the first time period; the predicted passenger compartment state parameter set, at least including the predicted passenger compartment temperature predicted by the vehicle in the second time period.

[0278] In the embodiment of the present disclosure, the adjustment module 1101 is also used to: obtain the compressor air-conditioning side control parameter set of the vehicle in the first time period, and the passenger compartment control parameter set of the vehicle in the first time period from the first heat pump status data; obtain the predicted compressor control parameter set predicted for the vehicle in the second time period based on the compressor air-conditioning side control parameter set and the target compressor control simulation prediction model in the target control simulation prediction model; obtain the predicted passenger compartment control parameter set predicted for the vehicle in the second time period based on the passenger compartment control parameter set and the target passenger compartment control simulation prediction model in the target control simulation prediction model; obtain the predicted control data of the vehicle in the second time period based on the predicted compressor control parameter set and the predicted passenger compartment control parameter set.

[0279] In the embodiment of the present disclosure, the compressor air conditioning side control parameter set includes at least a compressor desired air temperature request parameter set of the vehicle in the first time period, a compressor ambient temperature parameter set, a compressor speed parameter set, a compressor target superheat parameter set, a compressor target subcooling parameter set, a compressor actual superheat parameter set, a compressor actual subcooling parameter set, a compressor intake temperature parameter set, a compressor internal cooling temperature parameter set, a compressor saturated high pressure parameter set, a compressor target saturated high parameter set, a compressor saturated low pressure parameter set, a compressor target saturated low pressure parameter set, an in-vehicle temperature parameter set, a compressor exhaust temperature parameter set, a compressor high-pressure coolant heater inlet water temperature parameter set, a compressor high-pressure coolant heater outlet water temperature parameter set, a compressor desired air temperature request parameter set, and a compressor water pump Speed ​​pulse width modulation parameter set; predicted compressor control parameter set, including at least the predicted expansion valve opening, predicted compressor speed, predicted positive temperature coefficient thermistor power and predicted hot water pump pulse width modulation parameter set of the vehicle in the second time period; passenger cabin control parameter set, including at least the passenger cabin ambient temperature parameter set of the vehicle in the first time period, the passenger cabin interior temperature parameter set, the passenger cabin main driver's seat set temperature, the passenger cabin co-driver's seat set temperature, the passenger cabin rear seat set temperature, the passenger cabin automatic level, the passenger cabin main driver's seat initial interior temperature, the passenger cabin co-driver's seat initial interior temperature, the passenger cabin rear seat initial interior temperature, vehicle speed, passenger cabin left side sunlight parameter and passenger cabin right side sunlight parameter; predicted passenger cabin control parameter set, including at least the predicted blower speed and predicted expected wind temperature request predicted by the vehicle in the second time period.

[0280] In the embodiment of the present disclosure, the adjustment module 1101 is also used to: obtain the compressor power parameter set, windage energy consumption parameter set and fan power parameter set of the vehicle in the first time period from the first heat pump status data; obtain the predicted compressor power predicted for the vehicle in the second time period based on the compressor power parameter set and the target compressor power simulation prediction model in the target power simulation prediction model; obtain the predicted windage energy consumption predicted for the vehicle in the second time period based on the windage energy consumption parameter set and the target windage power simulation prediction model in the target power simulation prediction model; obtain the predicted fan power predicted for the vehicle in the second time period based on the fan power parameter set and the target fan power simulation prediction model in the target power simulation prediction model; and obtain the predicted power data of the vehicle in the second time period based on the predicted compressor power, predicted windage energy consumption and predicted fan power.

[0281] In the disclosed embodiment, the compressor power parameter set includes at least the compressor speed, compressor displacement, compressor saturated high pressure, compressor exhaust temperature, compressor intake temperature and compressor saturated low pressure of the vehicle in the first time period; the wind resistance energy consumption parameter set includes at least the vehicle speed, active air intake grille opening and cooling fan speed of the vehicle in the first time period; the fan power parameter set includes at least the cooling fan speed of the vehicle in the first time period.

[0282] In the disclosed embodiment, the adjustment module 1101 is further used to: integrate the predicted state data, the predicted control data, and the predicted power data to obtain second heat pump state data; and determine that the vehicle is at risk of thermal runaway within the second time period in response to the second heat pump data matching a preset thermal runaway risk determination condition.

[0283] In the disclosed embodiment, the adjustment module 1101 is also used to: obtain the cloud-based vehicle condition simulation prediction model sent by the server, and the vehicle-side model loading environment of the vehicle, wherein the vehicle-side model loading environment is constructed based on the signal access layer, data preprocessing layer, inference layer, feedback layer and fault tolerance layer; deploy the cloud-based vehicle condition simulation prediction model in the vehicle-side model loading environment to obtain the target vehicle condition simulation prediction model deployed in the vehicle-side model loading environment.

[0284] In the disclosed embodiment, the adjustment module 1101 is further used to: in response to identifying that the vehicle is at risk of thermal runaway within the second time period, generate an updated thermal control strategy for the vehicle based on the second heat pump status data, wherein the updated thermal control strategy is obtained based on a thermal control strategy acquisition model deployed on the cloud; based on the updated thermal control strategy, update the thermal control strategy of the vehicle corresponding to the second time period to adjust the body system temperature and the passenger compartment temperature of the vehicle within the second time period, wherein the updated thermal control strategy is transmitted to the vehicle based on the cloud.

[0285] The vehicle temperature adjustment method proposed in the present disclosure obtains first heat pump status data of the vehicle within a first time period, wherein, when it is identified that the vehicle has a thermal runaway risk within a second time period based on the first heat pump status data, the body system temperature and the passenger compartment temperature of the vehicle within the second time period are adjusted. In the present disclosure, heat pump status data of a vehicle in a second time period is predicted using first heat pump status data, and a thermal runaway risk assessment is performed on the vehicle in the second time period based on the predicted heat pump status data, thereby improving the timeliness of thermal runaway risk identification for the vehicle. Thermal runaway risk assessment and identification are performed using the heat pump status data of the heat pump system, thereby improving the precision and accuracy of thermal runaway risk identification. After identifying that a thermal runaway risk exists in the vehicle, the vehicle's body system temperature and passenger compartment temperature are adjusted to reduce the possibility of the vehicle experiencing a thermal runaway risk in the predicted second time period. The vehicle temperature is adjusted based on the thermal runaway risk identification result, thereby improving the timeliness of vehicle temperature regulation and thus improving the temperature control accuracy of the vehicle. Compared with the related art in which vehicle data prediction is achieved based on a constructed actual vehicle test bench, the cost of vehicle heat pump status data prediction and thermal runaway risk identification is reduced, and the occurrence of poor prediction and identification accuracy due to insufficient experimental data from the actual vehicle test bench is avoided. The safety of vehicle driving and vehicle drivers and passengers is improved, the service life of some vehicle components is extended, the maintenance cost of the vehicle is reduced, and the driving and riding experience of the vehicle is optimized.

[0286] Corresponding to the training methods of simulation prediction models proposed in the above-mentioned embodiments, an embodiment of the present disclosure further proposes a training device for a simulation prediction model. Since the training device for simulation prediction models proposed in the embodiment of the present disclosure corresponds to the training methods for simulation prediction models proposed in the above-mentioned embodiments, the implementation method of the above-mentioned training method for simulation prediction models is also applicable to the training device for simulation prediction models proposed in the embodiment of the present disclosure, and will not be described in detail in the following embodiments.

[0287] Figure 12 FIG. 1 is a structural diagram of a training device for a simulation prediction model according to an embodiment of the present disclosure. Figure 12 As shown, the training device 1200 of the simulation prediction model includes an acquisition module 121, a training module 122, an integration module 123 and a deployment module 124, wherein:

[0288] An acquisition module 121 is used to acquire candidate state simulation prediction models, candidate control simulation prediction models, and candidate power simulation prediction models to be trained;

[0289] A training module 122 is configured to perform model training on a candidate state simulation prediction model, a candidate control simulation prediction model, and a candidate power simulation prediction model based on sample heat pump state data of the vehicle in a sample time period, to obtain a trained target state simulation prediction model, a target control simulation prediction model, and a target power simulation prediction model;

[0290] An integration module 123 is configured to integrate the target control simulation prediction model, the target state simulation prediction model, and the target power simulation prediction model to obtain an integrated cloud-based vehicle condition simulation prediction model;

[0291] The deployment module 124 is used to deploy the cloud vehicle condition simulation prediction model on the vehicle to obtain the target vehicle condition simulation prediction model on the vehicle, wherein the target vehicle condition simulation prediction model is used to implement the above claims. Figure 11 A vehicle temperature adjustment device is provided in an embodiment.

[0292] In the embodiment of the present disclosure, the training module 122 is also used to: obtain a candidate compressor state simulation prediction model and a candidate passenger compartment state simulation prediction model in the candidate state simulation prediction model; generate a first training sample set based on the sample compressor state parameter set and the sample compressor state control parameter set of the vehicle in the sample time period in the sample heat pump state data, so as to train the candidate compressor state simulation prediction model and obtain a trained target compressor state simulation prediction model; generate a second training sample set based on the sample passenger compartment state parameter set and the sample passenger compartment control parameter set of the vehicle in the sample time period in the sample heat pump state data, so as to train the candidate passenger compartment simulation prediction model and obtain a trained target passenger compartment state simulation prediction model; obtain a target state simulation prediction model based on the target compressor state simulation prediction model and the target passenger compartment state simulation prediction model.

[0293] In the embodiment of the present disclosure, the training module 122 is also used to: obtain the candidate compressor control simulation prediction model and the candidate passenger compartment control simulation prediction model in the candidate control simulation prediction model; generate a third training sample set based on the sample compressor air-conditioning side control parameter set of the vehicle in the sample time period in the sample heat pump status data, so as to train the candidate compressor control simulation prediction model and obtain a trained target compressor control simulation prediction model; generate a fourth training sample set based on the sample passenger compartment control parameter set of the vehicle in the sample time period in the sample heat pump status data, so as to train the candidate passenger compartment control simulation prediction model and obtain a trained target passenger compartment control simulation prediction model; obtain a target control simulation prediction model based on the target compressor control simulation prediction model and the target passenger compartment control simulation prediction model.

[0294] In the embodiment of the present disclosure, the training module 122 is also used to: obtain a candidate compressor power simulation prediction model, a candidate wind resistance energy consumption simulation prediction model and a candidate fan power simulation prediction model from the candidate power simulation prediction model; generate a fifth training sample set based on the sample compressor power parameter set of the vehicle within the sample time period in the sample heat pump status data to train the candidate compressor power simulation prediction model and obtain a trained target compressor power simulation prediction model; generate a sixth training sample set based on the sample wind resistance energy consumption parameter set of the vehicle within the sample time period in the sample heat pump status data to train the candidate wind resistance energy consumption simulation prediction model and obtain a trained target wind resistance energy consumption simulation prediction model; generate a seventh training sample set based on the sample fan power parameter set of the vehicle within the sample time period in the sample heat pump status data to train the candidate fan power simulation prediction model and obtain a trained target fan power simulation prediction model; obtain a target power simulation prediction model based on the target compressor power simulation prediction model, the target wind resistance energy consumption simulation prediction model and the target fan power simulation prediction model.

[0295] The training device of the simulation prediction model proposed in the present invention trains the candidate state simulation prediction model, candidate control simulation prediction model and candidate power simulation prediction model of the vehicle in the cloud, thereby improving the training efficiency of each simulation prediction model and optimizing the training effect of each simulation prediction model. The cloud-based vehicle condition simulation prediction model obtained by integrating the trained models is deployed on the vehicle side, and there is no need to train the model on the vehicle side, which saves the training algorithm resources on the vehicle side, reduces the training cost and deployment cost of the model, optimizes the training method and training effect of the target vehicle condition simulation prediction model, and improves the prediction accuracy and prediction efficiency of the heat pump state data in the scenario of predicting the heat pump state data and identifying the thermal runaway risk based on the target vehicle condition simulation prediction model, thereby improving the recognition accuracy and recognition efficiency of the thermal runaway risk recognition, improving the driving and riding safety of the vehicle deployed with the target vehicle condition simulation prediction model, and optimizing the driving and riding experience of the vehicle.

[0296] To achieve the above embodiments, the present disclosure also provides a vehicle capable of executing the vehicle temperature adjustment method provided in the above embodiments.

[0297] To achieve the above embodiments, the present disclosure also provides an electronic device, a computer-readable storage medium, and a computer program product.

[0298] Figure 13 FIG. 1 is a block diagram of an electronic device 1300 according to an embodiment of the present disclosure. Figure 13 As shown, the electronic device 1300 includes a memory 1301, a processor 1302, and a computer program stored in the memory 1301 and executable on the processor 1302. When the processor 1302 executes the program instructions, the vehicle temperature adjustment method and / or the simulation prediction model training method provided in the above embodiments are implemented.

[0299] In order to implement the above embodiments, the present disclosure also proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it implements the vehicle temperature adjustment method and / or simulation prediction model training method provided in the above embodiments.

[0300] In order to implement the above embodiments, the present disclosure also proposes a computer program product on which a computer program is stored. When the computer program is executed by a processor, it implements the vehicle temperature adjustment method and / or the simulation prediction model training method provided in the above embodiments.

[0301] In order to implement the above embodiments, the present disclosure also proposes a chip, including one or more interface circuits and one or more processors; the interface circuit is used to receive signals and send the signals to the processors, and the signals include computer instructions stored in a memory. When the processor executes the computer instructions, the chip executes the vehicle temperature adjustment method and / or simulation prediction model training method provided in the above embodiments.

[0302] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0303] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0304] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0305] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0306] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement the present invention: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0307] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0308] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0309] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

[0310] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0311] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A vehicle temperature adjustment method, characterized in that: The method comprises: In response to identifying a risk of thermal runaway of the vehicle in a second time period based on first heat pump status data of the vehicle in a first time period, adjusting a body system temperature and a passenger compartment temperature of the vehicle, wherein the second time period is a time period subsequent to the first time period.

2. The method according to claim 1, characterized in that In response to identifying that the vehicle has a thermal runaway risk in a second time period based on first heat pump status data of the vehicle in a first time period, adjusting a body system temperature and a passenger compartment temperature of the vehicle includes: Obtaining a target vehicle condition simulation prediction model deployed on the vehicle; obtaining second heat pump state data of the vehicle in a second time period based on the target vehicle condition simulation prediction model and the first heat pump state data, so as to identify a thermal runaway risk of the vehicle in the second time period; In response to identifying that the vehicle has a thermal runaway risk during the second time period, a thermal control strategy during the second time period is updated to adjust the vehicle body system temperature and the passenger compartment temperature during the second time period.

3. The method according to claim 2, characterized in that The method further comprises: In response to identifying that there is no thermal runaway risk for the vehicle in the second time period, thermal runaway risk identification is continued for the vehicle in a third time period based on the target wheel simulation prediction model and the second heat pump status data, wherein the third time period is the next predicted time period of the second time period.

4. The method according to claim 2, characterized in that The obtaining, based on the target vehicle condition simulation prediction model and the first heat pump state data, second heat pump state data of the vehicle in the second time period to identify a thermal runaway risk of the vehicle in the second time period includes: Obtaining a target state simulation prediction model, a target control simulation prediction model, and a target power simulation prediction model in the target vehicle condition simulation prediction model; Obtaining predicted state data of the vehicle in the second time period according to the target state simulation prediction model and the first heat pump state data; Obtaining prediction control data for the vehicle in the second time period according to the target control simulation prediction model and the first heat pump state data; Obtaining predicted power data of the vehicle in the second time period according to the target power simulation prediction model and the first heat pump state data; The second heat pump state data predicted for the vehicle in the second time period is obtained based on the predicted state data, the predicted control data and the predicted power data, and the thermal runaway risk of the vehicle in the second time period is identified based on the second heat pump state data.

5. The method according to claim 4, characterized in that The step of obtaining predicted state data of the vehicle in the second time period based on the target state simulation prediction model and the first heat pump state data includes: acquiring, from the first heat pump state data, a compressor state parameter set and a compressor state control parameter set of the vehicle in the first time period, and a passenger compartment state parameter set and a passenger compartment state control parameter set of the vehicle in the first time period; Obtaining a predicted compressor state parameter set for the vehicle during the second time period based on the compressor state parameter set and the compressor state control parameter set, and a target compressor state simulation prediction model in the target state simulation prediction model; Obtaining a predicted passenger compartment state parameter set for the vehicle within the second time period based on the passenger compartment state parameter set, the passenger compartment state control parameter set, and a target passenger compartment state simulation prediction model in the target state simulation prediction model; The predicted state data of the vehicle in the second time period is obtained based on the predicted compressor state parameter set and the predicted passenger compartment state parameter set.

6. The method according to claim 5, characterized in that The compressor status parameter set includes at least a compressor high-pressure parameter set, a compressor low-pressure parameter set, a compressor intake air temperature parameter set, a compressor exhaust temperature parameter set, a compressor internal cooling temperature parameter set, a compressor high-pressure coolant heater inlet water temperature parameter set, a compressor high-pressure coolant heater outlet water temperature parameter set, and a passenger compartment temperature parameter set of the vehicle within the first time period; The compressor state control parameter set includes at least a compressor active air intake grille opening parameter set, a compressor cooling fan speed parameter set, a compressor blower speed parameter set, a compressor expansion valve opening parameter set, a compressor speed parameter set, a compressor positive temperature coefficient thermistor power parameter set, a compressor hot water pump pulse width modulation parameter set, a compressor ambient temperature parameter set, a passenger compartment temperature parameter set, and a compressor internal and external circulation ratio parameter set for the vehicle within the first time period; The predicted compressor state parameter set includes at least a predicted compressor high-pressure parameter set, a predicted compressor low-pressure parameter set, a predicted compressor intake temperature parameter set, a predicted compressor exhaust temperature parameter set, a predicted compressor internal cooling temperature parameter set, a predicted compressor high-pressure coolant heater inlet water temperature parameter set, and a predicted compressor high-pressure coolant heater outlet water temperature parameter set, all of which are predicted for the vehicle during the second time period; The passenger compartment state parameter set includes at least a passenger compartment temperature parameter set and a passenger compartment cold temperature parameter set of the vehicle within the first time period; The passenger compartment state control parameter set includes at least a blower air volume parameter set and a passenger compartment environment factor parameter set of the vehicle in the first time period; The predicted passenger compartment state parameter set includes at least a predicted passenger compartment temperature of the vehicle predicted within the second time period.

7. The method according to claim 4, characterized in that The obtaining, according to the target control simulation prediction model and the first heat pump state data, prediction control data of the vehicle in the second time period includes: acquiring, from the first heat pump state data, a compressor air-conditioning side control parameter set of the vehicle in the first time period and a passenger compartment control parameter set of the vehicle in the first time period; Obtaining a predicted compressor control parameter set for the vehicle during the second time period based on the compressor air-conditioning side control parameter set and a target compressor control simulation prediction model in the target control simulation prediction model; obtaining a predicted passenger compartment control parameter set for the vehicle during the second time period based on the passenger compartment control parameter set and a target passenger compartment control simulation prediction model in the target control simulation prediction model; The predicted control data of the vehicle in the second time period is obtained based on the predicted compressor control parameter set and the predicted passenger compartment control parameter set.

8. The method according to claim 7, characterized in that The compressor air-conditioning side control parameter set includes at least a compressor desired air temperature request parameter set, a compressor ambient temperature parameter set, a compressor speed parameter set, a compressor target superheat parameter set, a compressor target subcooling parameter set, a compressor actual superheat parameter set, a compressor actual subcooling parameter set, a compressor intake air temperature parameter set, a compressor internal cooling temperature parameter set, a compressor saturated high pressure parameter set, a compressor target saturated high pressure parameter set, a compressor saturated low pressure parameter set, a compressor target saturated low pressure parameter set, an in-vehicle temperature parameter set, a compressor exhaust temperature parameter set, a compressor high-pressure coolant heater inlet water temperature parameter set, a compressor high-pressure coolant heater outlet water temperature parameter set, a compressor desired air temperature request parameter set, and a compressor water pump speed pulse width modulation parameter set; The predicted compressor control parameter set includes at least a predicted expansion valve opening, a predicted compressor speed, a predicted positive temperature coefficient thermistor power, and a predicted hot water pump pulse width modulation parameter set of the vehicle in the second time period; The passenger cabin control parameter set includes at least a passenger cabin ambient temperature parameter set of the vehicle during the first time period, a passenger cabin interior temperature parameter set, a passenger cabin driver's seat set temperature, a passenger cabin passenger seat set temperature, a passenger cabin rear seat set temperature, a passenger cabin automatic level, a passenger cabin driver's seat initial interior temperature, a passenger cabin passenger seat initial interior temperature, a passenger cabin rear seat initial interior temperature, a vehicle speed, a passenger cabin left side sunlight parameter, and a passenger cabin right side sunlight parameter; The predicted passenger compartment control parameter set includes at least a predicted blower speed and a predicted expected air temperature request of the vehicle predicted during the second time period.

9. The method according to claim 4, characterized in that Obtaining predicted power data of the vehicle in the second time period according to the target power simulation prediction model and the first heat pump state data includes: acquiring, from the first heat pump state data, a compressor power parameter set, a windage energy consumption parameter set, and a fan power parameter set of the vehicle in the first time period; Obtaining a predicted compressor power of the vehicle during the second time period according to the compressor power parameter set and a target compressor power simulation prediction model in the target power simulation prediction model; Obtaining predicted wind resistance energy consumption of the vehicle during the second time period according to the wind resistance energy consumption parameter set and the target wind resistance power simulation prediction model in the target power simulation prediction model; Obtaining a predicted fan power of the vehicle during the second time period according to the fan power parameter set and a target fan power simulation prediction model in the target power simulation prediction model; The predicted power data of the vehicle in the second time period is obtained based on the predicted compressor power, the predicted windage energy consumption, and the predicted fan power.

10. The method according to claim 9, characterized in that The compressor power parameter set includes at least a compressor speed, a compressor displacement, a compressor saturated high pressure, a compressor exhaust temperature, a compressor intake temperature, and a compressor saturated low pressure of the vehicle during the first time period; The wind resistance energy consumption parameter set includes at least the vehicle speed, active air intake grille opening, and cooling fan speed of the vehicle during the first time period; The fan power parameter set includes at least a cooling fan speed of the vehicle during the first time period.

11. The method according to claim 4, characterized in that The step of obtaining the second heat pump state data predicted for the vehicle in the second time period based on the predicted state data, the predicted control data, and the predicted power data, and identifying a thermal runaway risk for the vehicle in the second time period based on the second heat pump state data includes: performing data integration on the predicted state data, the predicted control data, and the predicted power data to obtain the second heat pump state data; In response to the second heat pump data matching a preset thermal runaway risk determination condition, it is determined that the vehicle has a thermal runaway risk within the second time period.

12. The method according to claim 2, characterized in that The obtaining of the target vehicle condition simulation prediction model deployed on the vehicle includes: Obtaining a cloud-based vehicle condition simulation prediction model and a vehicle-side model loading environment for the vehicle, where the vehicle-side model loading environment is constructed based on a signal access layer, a data preprocessing layer, an inference layer, a feedback layer, and a fault-tolerance layer; The cloud-based vehicle condition simulation prediction model is deployed in the vehicle-side model loading environment to obtain the target vehicle condition simulation prediction model deployed in the vehicle-side model loading environment.

13. The method according to claim 12, characterized in that In response to identifying that the vehicle has a thermal runaway risk within the second time period, updating a thermal control strategy within the second time period to adjust the vehicle body system temperature and the passenger compartment temperature within the second time period includes: In response to identifying that the vehicle is at risk of thermal runaway within the second time period, generating an updated thermal control strategy for the vehicle based on the second heat pump status data, wherein the updated thermal control strategy is obtained based on a thermal control strategy acquisition model deployed in the cloud; Based on the updated thermal control strategy, the thermal control strategy corresponding to the second time period is updated for the vehicle to adjust the body system temperature and the passenger compartment temperature of the vehicle within the second time period, wherein the updated thermal control strategy is transmitted to the vehicle based on the cloud.

14. A training method for a simulation prediction model, characterized in that: The method comprises: Obtaining candidate state simulation prediction models, candidate control simulation prediction models, and candidate power simulation prediction models to be trained; According to the sample heat pump state data of the vehicle in the sample time period, the candidate state simulation prediction model, the candidate control simulation prediction model and the candidate power simulation prediction model are respectively trained to obtain the trained target state simulation prediction model, the target control simulation prediction model and the target power simulation prediction model; Integrate the target control simulation prediction model, the target state simulation prediction model, and the target power simulation prediction model to obtain an integrated cloud-based vehicle condition simulation prediction model; The cloud-based vehicle condition simulation prediction model is deployed on the vehicle to obtain the target vehicle condition simulation prediction model on the vehicle, wherein the target vehicle condition simulation prediction model is used to implement the vehicle temperature adjustment method according to any one of claims 1 to 13.

15. The method according to claim 14, characterized in that The performing model training on the candidate state simulation prediction model according to the sample heat pump state data of the vehicle in the sample time period to obtain a trained target state simulation prediction model includes: Obtaining a candidate compressor state simulation prediction model and a candidate passenger compartment state simulation prediction model from the candidate state simulation prediction models; generating a first training sample set based on the sample heat pump state data, a sample compressor state parameter set and a sample compressor state control parameter set of the vehicle within the sample time period, to train the candidate compressor state simulation prediction model to obtain a trained target compressor state simulation prediction model; generating a second training sample set based on the sample heat pump state data, a sample passenger compartment state parameter set and a sample passenger compartment control parameter set of the vehicle within the sample time period, to train the candidate passenger compartment simulation prediction model to obtain a trained target passenger compartment state simulation prediction model; The target state simulation prediction model is obtained based on the target compressor state simulation prediction model and the target passenger compartment state simulation prediction model.

16. The method according to claim 14, characterized in that The method of training the candidate control simulation prediction model based on the sample heat pump state data of the vehicle in the sample time period to obtain a trained target control simulation prediction model includes: acquiring a candidate compressor control simulation prediction model and a candidate passenger compartment control simulation prediction model from the candidate control simulation prediction models; generating a third training sample set based on the sample heat pump state data and the sample compressor air-conditioning side control parameter set of the vehicle within the sample time period to train the candidate compressor control simulation prediction model to obtain a trained target compressor control simulation prediction model; generating a fourth training sample set based on a sample passenger compartment control parameter set of the vehicle within the sample time period in the sample heat pump state data to train the candidate passenger compartment control simulation prediction model to obtain a trained target passenger compartment control simulation prediction model; The target control simulation prediction model is obtained based on the target compressor control simulation prediction model and the target passenger compartment control simulation prediction model.

17. The method according to claim 14, characterized in that The method of training the candidate power simulation prediction model based on the sample heat pump state data of the vehicle in the sample time period to obtain a trained target power simulation prediction model includes: Obtaining a candidate compressor power simulation prediction model, a candidate windage energy consumption simulation prediction model, and a candidate fan power simulation prediction model from the candidate power simulation prediction models; generating a fifth training sample set based on the sample compressor power parameter set of the vehicle within the sample time period in the sample heat pump state data, so as to train the candidate compressor power simulation prediction model to obtain a trained target compressor power simulation prediction model; generating a sixth training sample set based on the sample wind resistance energy consumption parameter set of the vehicle in the sample time period in the sample heat pump state data, so as to train the candidate wind resistance energy consumption simulation prediction model to obtain a trained target wind resistance energy consumption simulation prediction model; generating a seventh training sample set based on the sample fan power parameter set of the vehicle in the sample time period in the sample heat pump state data, so as to train the candidate fan power simulation prediction model to obtain a trained target fan power simulation prediction model; The target power simulation prediction model is obtained based on the target compressor power simulation prediction model, the target windage energy consumption simulation prediction model and the target fan power simulation prediction model.

18. A vehicle temperature adjustment device, characterized in that: The device comprises: An adjustment module is configured to adjust a body system temperature and a passenger compartment temperature of the vehicle in response to identifying a risk of thermal runaway of the vehicle in a second time period based on first heat pump status data of the vehicle in a first time period, wherein the second time period is a time period subsequent to the first time period.

19. The device according to claim 18, characterized in that The adjustment module is further used to: Obtaining a target vehicle condition simulation prediction model deployed on the vehicle; obtaining second heat pump state data of the vehicle in a second time period based on the target vehicle condition simulation prediction model and the first heat pump state data, so as to identify a thermal runaway risk of the vehicle in the second time period; In response to identifying that the vehicle has a thermal runaway risk during the second time period, a thermal control strategy during the second time period is updated to adjust the vehicle body system temperature and the passenger compartment temperature during the second time period.

20. The device according to claim 19, characterized in that The adjustment module is further used to: In response to identifying that there is no thermal runaway risk for the vehicle in the second time period, thermal runaway risk identification is continued for the vehicle in a third time period based on the target wheel simulation prediction model and the second heat pump status data, wherein the third time period is the next predicted time period of the second time period.

21. The device according to claim 19, characterized in that The adjustment module is further used to: Obtaining a target state simulation prediction model, a target control simulation prediction model, and a target power simulation prediction model in the target vehicle condition simulation prediction model; Obtaining predicted state data of the vehicle in the second time period according to the target state simulation prediction model and the first heat pump state data; Obtaining prediction control data for the vehicle in the second time period according to the target control simulation prediction model and the first heat pump state data; Obtaining predicted power data of the vehicle in the second time period according to the target power simulation prediction model and the first heat pump state data; The second heat pump state data predicted for the vehicle in the second time period is obtained based on the predicted state data, the predicted control data and the predicted power data, and the thermal runaway risk of the vehicle in the second time period is identified based on the second heat pump state data.

22. A training device for a simulation prediction model, characterized in that: The device comprises: An acquisition module, used to acquire candidate state simulation prediction models, candidate control simulation prediction models, and candidate power simulation prediction models to be trained; a training module for performing model training on the candidate state simulation prediction model, the candidate control simulation prediction model, and the candidate power simulation prediction model, respectively, based on sample heat pump state data of the vehicle in a sample time period, to obtain trained target state simulation prediction models, target control simulation prediction models, and target power simulation prediction models; An integration module is used to integrate the target control simulation prediction model, the target state simulation prediction model, and the target power simulation prediction model to obtain an integrated cloud-based vehicle condition simulation prediction model; A deployment module is used to deploy the cloud-based vehicle condition simulation prediction model on the vehicle to obtain the target vehicle condition simulation prediction model on the vehicle, wherein the target vehicle condition simulation prediction model is used to implement the vehicle temperature adjustment device described in any one of claims 15-18 above.

23. The device according to claim 22, characterized in that The training module is further used to: Obtaining a candidate compressor state simulation prediction model and a candidate passenger compartment state simulation prediction model from the candidate state simulation prediction models; generating a first training sample set based on the sample heat pump state data, a sample compressor state parameter set and a sample compressor state control parameter set of the vehicle within the sample time period, to train the candidate compressor state simulation prediction model to obtain a trained target compressor state simulation prediction model; generating a second training sample set based on the sample heat pump state data, a sample passenger compartment state parameter set and a sample passenger compartment control parameter set of the vehicle within the sample time period, to train the candidate passenger compartment simulation prediction model to obtain a trained target passenger compartment state simulation prediction model; The target state simulation prediction model is obtained based on the target compressor state simulation prediction model and the target passenger compartment state simulation prediction model.

24. The device according to claim 22, characterized in that The training module is further used to: acquiring a candidate compressor control simulation prediction model and a candidate passenger compartment control simulation prediction model from the candidate control simulation prediction models; generating a third training sample set based on the sample heat pump state data and the sample compressor air-conditioning side control parameter set of the vehicle within the sample time period to train the candidate compressor control simulation prediction model to obtain a trained target compressor control simulation prediction model; generating a fourth training sample set based on a sample passenger compartment control parameter set of the vehicle within the sample time period in the sample heat pump state data to train the candidate passenger compartment control simulation prediction model to obtain a trained target passenger compartment control simulation prediction model; The target control simulation prediction model is obtained based on the target compressor control simulation prediction model and the target passenger compartment control simulation prediction model.

25. The device according to claim 22, characterized in that The training module is further used to: Obtaining a candidate compressor power simulation prediction model, a candidate windage energy consumption simulation prediction model, and a candidate fan power simulation prediction model from the candidate power simulation prediction models; generating a fifth training sample set based on the sample compressor power parameter set of the vehicle within the sample time period in the sample heat pump state data, so as to train the candidate compressor power simulation prediction model to obtain a trained target compressor power simulation prediction model; generating a sixth training sample set based on the sample wind resistance energy consumption parameter set of the vehicle in the sample time period in the sample heat pump state data, so as to train the candidate wind resistance energy consumption simulation prediction model to obtain a trained target wind resistance energy consumption simulation prediction model; generating a seventh training sample set based on the sample fan power parameter set of the vehicle in the sample time period in the sample heat pump state data, so as to train the candidate fan power simulation prediction model to obtain a trained target fan power simulation prediction model; The target power simulation prediction model is obtained based on the target compressor power simulation prediction model, the target windage energy consumption simulation prediction model and the target fan power simulation prediction model.

26. A vehicle, characterized in that: The vehicle is capable of performing the method according to any one of claims 1 to 13 and / or 14.

27. An electronic device, characterized in that: include: processor; a memory for storing executable instructions for the processor; The processor is configured to execute instructions to implement the method according to any one of claims 1-13 and / or 14.

28. A computer-readable storage medium, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the method according to any one of claims 1-13 and / or 14.

29. A chip, characterized in that: The chip comprises one or more interface circuits and one or more processors; the interface circuit is used to receive a signal and send the signal to the processor, the signal including a computer instruction stored in a memory, and when the processor executes the computer instruction, the chip executes the steps of the method described in any one of claims 1-13 and / or 14.

Images