Vehicle thermal management control method and vehicle

Abstract

The application discloses a kind of vehicle thermal management control method and vehicle.Therein, method includes: first temperature control data is decoded and is handled, obtains current multi-source temperature control data;Current multi-source temperature control data is classified, obtains and multiple attribute characteristic categories corresponding multiclass temperature control data;Multiclass temperature control data is screened, and obtains partial category data;Partial category data is handled with feature fusion, and obtains fusion feature value;Based on fusion feature value and at least one historical acquisition period corresponding historical feature value, determine feature evolution trend;Based on partial category data, feature evolution trend and the preset threshold and data importance of the temperature control data of each attribute characteristic category in partial category data, determine operating condition determination value;According to operating condition determination value, determine vehicle thermal management control strategy.The application solves the technical problems of poor vehicle thermal management control efficiency and low accuracy in related technologies.

Description

Technical Field

[0001] This invention relates to the field of vehicle thermal management control technology, and more specifically, to a vehicle thermal management control method and a vehicle. Background Technology

[0002] With the rapid development of new energy vehicles, the vehicle thermal management system, as a core subsystem for ensuring the safety of power batteries, improving the efficiency of electric drive systems, and optimizing cabin comfort, has evolved from single-component temperature control to a multi-source heterogeneous collaborative control system encompassing batteries, electric drive, electronic control, and cabin air conditioning. Under dynamic conditions such as high-rate fast charging, extreme low / high temperature environments, and rapid acceleration, the system's thermal load exhibits strong coupling, nonlinearity, and time-varying characteristics, posing unprecedented challenges to the real-time performance, accuracy, and response speed of thermal management.

[0003] Existing thermal management technologies mostly employ a centralized data acquisition and fixed-strategy control architecture, failing to systematically optimize for the end-to-end latency from data perception to execution. Current thermal management technologies suffer from long acquisition paths and significant signal attenuation, resulting in substantial time delays for key parameters such as temperature and pressure. Secondly, during the data processing stage, automotive-grade controllers, with limited computing power, struggle to perform real-time fusion and analysis of high-dimensional data, leading to lag in response and poor accuracy in strategy decision-making. Therefore, improving the efficiency and accuracy of vehicle thermal management control is one of the key technical challenges in this field.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This invention provides a vehicle thermal management control method and a vehicle, to at least solve the technical problems of poor efficiency and low accuracy in vehicle thermal management control in related technologies.

[0006] According to one aspect of the present invention, a vehicle thermal management control method is provided, comprising: decoding first temperature control data to obtain current multi-source temperature control data, wherein the first temperature control data is obtained by compressing and encoding the original multi-source temperature control data collected in the current acquisition cycle, and the current multi-source temperature control data is used to describe the thermal state of multiple thermal management components in the vehicle thermal management system; classifying the current multi-source temperature control data to obtain multiple types of temperature control data corresponding to multiple attribute feature categories; filtering the multiple types of temperature control data to obtain partial category data, wherein the data importance of the temperature control data corresponding to each attribute feature category in the partial category data meets the data importance evaluation condition; performing feature fusion processing on the partial category data to obtain fused feature values; determining feature evolution trends based on the fused feature values ​​and historical feature values ​​corresponding to at least one historical acquisition cycle; determining operating condition judgment values ​​based on the partial category data, feature evolution trends, and preset thresholds and data importance of the temperature control data corresponding to each attribute feature category in the partial category data; and determining a vehicle thermal management control strategy according to the operating condition judgment values.

[0007] Optionally, the vehicle thermal management control method further includes: acquiring raw multi-source temperature control data, wherein the raw multi-source temperature control data is acquired by multiple data acquisition devices, and the multiple data acquisition devices are respectively deployed on the surface of multiple thermal management components in the vehicle thermal management system; performing data correction processing and timestamp alignment processing on the raw multi-source temperature control data to obtain preprocessed data; and performing length adaptive encoding on the preprocessed data to obtain first temperature control data.

[0008] Optionally, the preprocessed data is subjected to length adaptive encoding to obtain the first temperature control data, including: classifying the preprocessed data to obtain multi-class preprocessed data corresponding to multiple attribute feature categories; calculating the probability weights corresponding to each of the multi-class preprocessed data using a pre-constructed probability distribution model; determining the encoding length corresponding to each of the multi-class preprocessed data based on the probability weights and the timestamp alignment information corresponding to each of the multi-class preprocessed data; determining the compression gain value corresponding to each of the multi-class preprocessed data based on the encoding length; and compressing and encoding the preprocessed data based on the compression gain value to obtain the first temperature control data.

[0009] Optionally, data filtering is performed on multiple types of temperature control data to obtain partial categories of data, including: determining the data importance of each type of temperature control data based on the corresponding code length; and filtering partial categories of data that meet the data importance assessment conditions from multiple types of temperature control data based on the data importance.

[0010] Optionally, feature fusion processing is performed on some categories of data to obtain fused feature values, including: weighted fusion of some categories of data based on the data importance of temperature control data corresponding to each attribute feature category in the partial category of data to obtain fused feature values.

[0011] Optionally, the operating condition judgment value is determined based on partial category data, feature evolution trends, and preset thresholds and data importance of temperature control data corresponding to each attribute feature category in the partial category data. This includes: determining the initial operating condition judgment value based on partial category data, preset thresholds and data importance of temperature control data corresponding to each attribute feature category in the partial category data; and correcting the initial operating condition judgment value based on feature evolution trends to obtain the operating condition judgment value.

[0012] Optionally, a vehicle thermal management control strategy is determined based on the operating condition judgment value, including: using an operating condition mapping method to determine the operating condition level corresponding to the operating condition judgment value; and determining the vehicle thermal management control strategy based on the operating condition level.

[0013] Optionally, the vehicle thermal management control method further includes: determining the movement trajectory of the valve core based on the operating condition level, the response time of the target valve, and the movement speed of the valve core within the target valve, wherein the target valve is used to characterize the execution unit that executes the vehicle thermal management control strategy; and generating control commands based on the movement trajectory to realize the vehicle thermal management control strategy.

[0014] Optionally, the vehicle thermal management control method further includes: obtaining the target fusion feature value corresponding to the operating condition level; determining the temperature control deviation based on the target fusion feature value and the fusion feature value; and determining the temperature control correction parameter based on the temperature control deviation and the feature evolution trend, wherein the temperature control correction parameter is used to correct the initial temperature control parameter.

[0015] According to another aspect of the present invention, a vehicle is also provided, comprising: a memory storing an executable program; and a processor for running the executable program, wherein the executable program executes the vehicle thermal management control method described in any of the preceding embodiments.

[0016] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is executed, it controls the device where the computer-readable storage medium is located to execute the vehicle thermal management control method described in any of the above embodiments.

[0017] This invention provides a vehicle thermal management control method, comprising: decoding first temperature control data to obtain current multi-source temperature control data, wherein the first temperature control data is obtained by compressing and encoding the original multi-source temperature control data collected in the current acquisition cycle, and the current multi-source temperature control data is used to describe the thermal state of multiple thermal management components in the vehicle thermal management system; classifying the current multi-source temperature control data to obtain multiple types of temperature control data corresponding to multiple attribute feature categories; filtering the multiple types of temperature control data to obtain partial category data, wherein the data importance of the temperature control data corresponding to each attribute feature category in the partial category data meets the data importance evaluation condition; performing feature fusion processing on the partial category data to obtain fused feature values; determining feature evolution trends based on the fused feature values ​​and historical feature values ​​corresponding to at least one historical acquisition cycle; determining operating condition judgment values ​​based on the partial category data, feature evolution trends, and preset thresholds and data importance of the temperature control data corresponding to each attribute feature category in the partial category data; and determining a vehicle thermal management control strategy based on the operating condition judgment values. This invention first decodes the compressed and encoded first temperature control data to restore the highly timely multi-source thermal state data, avoiding information distortion or accumulated delays caused by transmission compression of the original data and improving the reliability of the input data. Second, the decoded current multi-source temperature control data is classified according to attribute feature categories, and precise screening is performed based on data importance assessment conditions to eliminate redundant and low-value data, significantly reducing the data dimension and computational burden of subsequent processing and improving the system's operating efficiency under the limited computing power of automotive-grade controllers. Furthermore, feature fusion is performed on some of the filtered category data to generate compact fused feature values, and feature evolution trends are calculated in combination with historical feature values ​​to enhance the ability to perceive dynamic changes in thermal state and avoid misjudgments caused by relying solely on instantaneous values. Finally, the partial category data, feature evolution trends, preset thresholds, and data importance are combined to comprehensively calculate the operating condition judgment value, achieving highly sensitive and low-latency quantitative identification of complex operating conditions, thereby accurately matching the corresponding control strategy and avoiding the response lag and control inaccuracy caused by traditional fixed rules or single thresholds. In summary, this invention achieves the technical effect of improving the efficiency and accuracy of vehicle thermal management control, thereby solving the technical problems of poor efficiency and low accuracy in vehicle thermal management control in related technologies. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0019] Figure 1 This is a flowchart of a vehicle thermal management control method according to one embodiment of the present invention;

[0020] Figure 2This is a flowchart illustrating a data acquisition process according to one embodiment of the present invention;

[0021] Figure 3 This is a flowchart illustrating an adaptive Huffman coding compression process according to one embodiment of the present invention.

[0022] Figure 4 This is a flowchart illustrating multi-source data fusion and feature extraction according to one embodiment of the present invention;

[0023] Figure 5 This is a flowchart illustrating the working condition identification process according to one embodiment of the present invention;

[0024] Figure 6 This is a flowchart illustrating the actuator optimization and incremental feedback according to one embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] According to an embodiment of the present invention, an embodiment of a vehicle thermal management control method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0028] This invention provides a vehicle thermal management control method. Figure 1This is a flowchart of a vehicle thermal management control method according to one embodiment of the present invention, such as... Figure 1 As shown, the vehicle thermal management control method includes the following steps:

[0029] Step S101: Decode the first temperature control data to obtain the current multi-source temperature control data. The first temperature control data is obtained by compressing and encoding the original multi-source temperature control data collected in the current acquisition cycle. The current multi-source temperature control data is used to describe the thermal state of multiple thermal management components in the vehicle thermal management system.

[0030] Step S102: Classify the current multi-source temperature control data to obtain multiple types of temperature control data corresponding to various attribute feature categories;

[0031] Step S103: Data filtering is performed on multiple types of temperature control data to obtain partial category data. Among them, the data importance of the temperature control data corresponding to each attribute feature category in the partial category data meets the data importance evaluation conditions.

[0032] Step S104: Perform feature fusion processing on some category data to obtain fused feature values;

[0033] Step S105: Determine the feature evolution trend based on the fused feature value and the historical feature value corresponding to at least one historical acquisition cycle;

[0034] Step S106: Determine the working condition judgment value based on partial category data, feature evolution trend, and preset threshold and data importance of temperature control data corresponding to each attribute feature category in partial category data;

[0035] Step S107: Determine the vehicle thermal management control strategy based on the operating condition judgment value.

[0036] The first temperature control data mentioned above refers to the compressed and encoded temperature control data stream.

[0037] In one optional embodiment, raw temperature, pressure, rotation speed and other signals are collected by a micro-sensor module at the acquisition end, and then compressed to generate the first temperature control data through an adaptive Huffman coding algorithm.

[0038] Multi-source temperature control data refers to real-time thermal status data from multiple key subsystems in the vehicle thermal management system, including battery cell temperature, electric drive winding temperature, electronic control module temperature, cabin evaporator outlet temperature, ambient temperature, battery current, drive torque, etc. Each type of data represents the operating status of different heat sources or thermal control points.

[0039] Optionally, in the vehicle thermal management system, to reduce Ethernet communication load and shorten data transmission latency, adaptive Huffman coding is used at the acquisition end to compress the raw multi-source temperature control data such as temperature, pressure, and speed, forming a compact "first temperature control data," which is a binary data stream after encoding and compression. At the control unit end, this data needs to be decoded in real time to restore the original physical quantity sequence. The decoding process relies on a dynamic probability dictionary established during the encoding stage. This dictionary is automatically generated by the system in each acquisition cycle based on the sliding window statistics of the data distribution, ensuring that consistent mapping rules are used at both the encoding and decoding ends.

[0040] For example, the encoded data frame received from the acquisition module is checked by verifying the frame header and synchronization identifier to determine the data integrity. Then, based on the Huffman tree structure pre-stored in the current period, the binary stream is parsed bit by bit to restore the variable-length code to the original value, such as the battery cell temperature value (unit: °C), the electric drive winding temperature (unit: °C), etc., and finally a structured and time-aligned "current multi-source temperature control data" is formed.

[0041] In one optional embodiment, the decoded current multi-source temperature control data is automatically classified using a preset feature classification mapping table to obtain multiple types of temperature control data.

[0042] For example, based on the thermal management function logic, the current multi-source temperature control data is functionally categorized into four categories, including: heat source category (such as battery temperature, electric drive winding temperature, and electronic control module temperature), heat sink category (such as cabin evaporator outlet temperature and coolant inlet temperature), operating condition drive category (battery charging and discharging current, drive motor torque, and vehicle speed), and environmental disturbance category (ambient temperature, humidity, and solar radiation intensity).

[0043] In addition, it can be divided according to thermal management components, as well as other division methods (such as division according to preset attribute characteristics).

[0044] In addition, the data can also be divided into battery temperature data, electric drive winding temperature data, electronic control module temperature data, drive motor torque data, vehicle speed data, etc., according to the description of the data itself.

[0045] Data importance is a quantitative indicator used to assess the weight of a certain type of temperature control data in thermal management decisions.

[0046] In one alternative embodiment, data importance is calculated from two sub-dimensions: change sensitivity (the absolute value of the difference between the current value and the previous period value, reflecting the priority of dynamic response) and coding efficiency ratio (the ratio of the number of bits occupied by the data in the compression stage to its information entropy, reflecting the density of regulatory information carried per unit of data).

[0047] By using data importance assessment criteria (such as threshold filtering or importance ranking filtering), only data whose importance meets the assessment criteria are retained, while data with weak changes or redundant coding are removed.

[0048] The aforementioned feature fusion processing refers to mapping multiple types of highly important temperature control data into a low-dimensional decision space, eliminating dimensional differences, and forming a single feature value that comprehensively represents the thermal state of the vehicle.

[0049] The aforementioned feature evolution trends reflect the direction and rate of change of the fused feature values ​​over time, which can be used to predict the development trend of thermal state and identify dangerous operating conditions that are "about to be triggered" in advance.

[0050] The above-mentioned operating condition judgment value is a comprehensive scoring index. It generates a continuous value by integrating current data, trends, thresholds, and importance to characterize the urgency of the current thermal state, corresponding to different operating condition levels.

[0051] The aforementioned thermal management control strategy refers to a set of predefined execution instructions triggered by the operating condition level, including valve opening, water pump speed, compressor frequency, cooling circuit switching, etc.

[0052] Optionally, the current operating condition level is determined based on the magnitude of the operating condition judgment value, and then a control strategy corresponding to the current operating condition level is determined. The control strategy includes: opening command of the integrated valve body, cooling water pump speed, electronic expansion valve opening, heat pump system mode switching, etc.

[0053] This invention provides a vehicle thermal management control method, comprising: decoding first temperature control data to obtain current multi-source temperature control data, wherein the first temperature control data is obtained by compressing and encoding the original multi-source temperature control data collected in the current acquisition cycle, and the current multi-source temperature control data is used to describe the thermal state of multiple thermal management components in the vehicle thermal management system; classifying the current multi-source temperature control data to obtain multiple types of temperature control data corresponding to multiple attribute feature categories; filtering the multiple types of temperature control data to obtain partial category data, wherein the data importance of the temperature control data corresponding to each attribute feature category in the partial category data meets the data importance evaluation conditions; performing feature fusion processing on the partial category data to obtain fused feature values; determining feature evolution trends based on the fused feature values ​​and historical feature values ​​corresponding to at least one historical acquisition cycle; determining operating condition judgment values ​​based on the partial category data, feature evolution trends, and preset thresholds and data importance of the temperature control data corresponding to each attribute feature category in the partial category data, wherein the operating condition judgment values ​​are used to quantitatively characterize the operating condition scenario of the vehicle; and determining a vehicle thermal management control strategy based on the operating condition judgment values. This invention first decodes the compressed and encoded first temperature control data to restore the highly timely multi-source thermal state data, avoiding information distortion or accumulated delays caused by transmission compression of the original data and improving the reliability of the input data. Second, the decoded current multi-source temperature control data is classified according to attribute feature categories, and precise screening is performed based on data importance assessment conditions to eliminate redundant and low-value data, significantly reducing the data dimension and computational burden of subsequent processing and improving the system's operating efficiency under the limited computing power of automotive-grade controllers. Furthermore, feature fusion is performed on some of the filtered category data to generate compact fused feature values, and feature evolution trends are calculated in combination with historical feature values ​​to enhance the ability to perceive dynamic changes in thermal state and avoid misjudgments caused by relying solely on instantaneous values. Finally, the partial category data, feature evolution trends, preset thresholds, and data importance are combined to comprehensively calculate the operating condition judgment value, achieving highly sensitive and low-latency quantitative identification of complex operating conditions, thereby accurately matching the corresponding control strategy and avoiding the response lag and control inaccuracy caused by traditional fixed rules or single thresholds. In summary, this invention achieves the technical effect of improving the efficiency and accuracy of vehicle thermal management control, thereby solving the technical problems of poor efficiency and low accuracy in vehicle thermal management control in related technologies.

[0054] The vehicle thermal management control method in the embodiments of this application will be further described below.

[0055] Optionally, the vehicle thermal management control method further includes:

[0056] Step S1081: Obtain raw multi-source temperature control data, wherein the raw multi-source temperature control data is acquired by multiple data acquisition devices, and the multiple data acquisition devices are respectively deployed on the surface of multiple thermal management components in the vehicle thermal management system.

[0057] Step S1082: Perform data correction and timestamp alignment on the original multi-source temperature control data to obtain preprocessed data;

[0058] Step S1083: Perform length adaptive encoding on the preprocessed data to obtain the first temperature control data.

[0059] The aforementioned data acquisition device includes a low-power embedded sensing module that integrates a miniature temperature sensor, a pressure sensor, and a signal conditioning circuit.

[0060] Multiple data acquisition devices are directly attached to the metal / ceramic surface of the thermal management components (such as the battery cell casing, electric drive winding insulation layer, electronic control IGBT substrate, and outer wall of the cabin evaporator copper tube) using thermally conductive adhesive or metal clamping structures, rather than being placed in pipes or cavities far away from the thermal management components, to achieve near-source sensing.

[0061] The aforementioned data correction process refers to noise suppression, zero-point drift compensation, and dynamic response compensation of the original multi-source temperature control data to improve the data's authenticity and stability, rather than simply performing linear calibration.

[0062] The aforementioned timestamp alignment process refers to mapping data from different acquisition devices with independent clock sources to the system reference time of the vehicle control unit in a unified manner, thereby eliminating timing misalignments caused by crystal oscillator deviations or network transmission jitter.

[0063] The aforementioned length adaptive coding refers to a compression algorithm that adjusts the coding length in real time according to the dynamic changes in data. The core idea is to use long codes for high-frequency changing data and short codes for low-frequency stable data, so as to maximize compression efficiency while ensuring information integrity.

[0064] For example, miniature data acquisition modules are set near the battery cells, electric drive windings, electronic control modules, and cabin evaporators. The acquisition modules are directly attached to the surfaces of the heat-generating and temperature-controlling components to minimize the distance between the data acquisition points and the components, avoid signal attenuation and path loss during data transmission, reduce the initial delay of data acquisition from the source, and ensure that the core data such as temperature and pressure can be quickly output to the processing unit.

[0065] By directly attaching the micro data acquisition module to the surfaces of key heat-generating and temperature-controlled components such as battery cells, electric drive windings, electronic control modules, and cabin evaporators, the acquisition path length is minimized, physically compressing the signal propagation link. Simultaneously, a local preprocessing mechanism enhances the effectiveness of the raw data, forming a seamless low-latency data link with subsequent data compression and fusion steps. In this process, the acquisition latency is defined as a coupling function of the acquisition path length, signal attenuation coefficient, and sampling response time. A latency expression model is constructed to quantitatively describe the effect of the attached arrangement. The total acquisition latency of a single acquisition module can be expressed as:

[0066] (1)

[0067] in, This indicates the total acquisition latency of a single acquisition module. This represents the equivalent path length from the data acquisition point to the processing unit. This indicates the speed at which a signal propagates in a wire or sensing medium. This represents the signal attenuation coefficient per unit length. This indicates the sampling response time of the sensor body, achieved by attaching the acquisition module to the sensor body. By approaching the minimum value, the propagation delay and attenuation accumulation are reduced simultaneously, ensuring that temperature and pressure data can enter a highly available state immediately upon generation. The purpose of this expression is to characterize the direct impact of physical layout on acquisition efficiency and to provide a quantitative basis for subsequent system optimization.

[0068] Based on the contactless acquisition method, to avoid local measurement noise interfering with the subsequent fusion accuracy, a lightweight dynamic correction mechanism is introduced within the acquisition module to correct the original measurement values ​​in real time. By constructing a data acquisition correction model, sensing errors and transient disturbances are suppressed. The corrected effective acquisition data can be expressed as:

[0069] (2)

[0070] in, This indicates the valid data collected after correction. This represents the data directly output by the sensor. Indicates the noise suppression coefficient. This represents the rate of change of the original data over time. By suppressing high-frequency change components, abrupt noise is weakened while the true thermal change trend is preserved. The purpose of constructing a data acquisition correction model is to improve data quality without increasing the computational burden and reduce the complexity of subsequent fusion calculations from the source.

[0071] Furthermore, to achieve consistency of multi-source acquired data in the time dimension, a unified timestamp alignment mechanism is introduced to map the data output by each acquisition module to a unified time base. A time deviation compensation model is used to synchronize and correct the multi-source data, which can be specifically represented as follows:

[0072] (3)

[0073] in, This indicates the deviation between the local time of the acquisition module and the system reference time. This indicates the internal timing value of the data acquisition module. This indicates a unified time base for the entire vehicle control unit. This refers to synchronized data after time alignment. By forward compensation for time deviations, data from different subsystems are made comparable at the same time, thereby avoiding misjudgments caused by time misalignment during subsequent fusion. The purpose of this mechanism is to provide a unified time sequence basis for multi-source data fusion.

[0074] It is easy to understand that through the synergistic effect of the above-mentioned close-fitting arrangement, dynamic correction and time alignment, the data output by the acquisition end is optimized in terms of latency, accuracy and consistency. This provides high-quality, low-redundancy data input for the data compression processing based on adaptive Huffman coding in the subsequent steps, thereby further reducing the transmission load and shortening the overall link latency.

[0075] Figure 2 This is a flowchart illustrating a data acquisition process according to one embodiment of the present invention, such as... Figure 2 The original temperature control data of battery cells, electric drive windings, electronic control modules and cabin evaporators are collected by the closely fitted acquisition modules. The original temperature control data is dynamically corrected, noise is suppressed and time is aligned to unify the timestamp, so as to obtain low-latency and high-precision synchronous data.

[0076] Optionally, the preprocessed data is subjected to length adaptive encoding to obtain the first temperature control data, including:

[0077] Step S10831: Classify the preprocessed data to obtain multi-class preprocessed data corresponding to multiple attribute feature categories;

[0078] Step S10832: Calculate the probability weights corresponding to the various preprocessed data using a pre-built probability distribution model.

[0079] Step S10833: Based on the probability weights and the timestamp alignment information corresponding to the various types of preprocessed data, determine the encoding length corresponding to each type of preprocessed data.

[0080] Step S10834: Based on the encoding length, determine the compression benefit values ​​corresponding to the various types of preprocessed data;

[0081] Step S10835: Based on the compression gain value, the preprocessed data is compressed and encoded to obtain the first temperature control data.

[0082] Optionally, the data can be divided into battery temperature data, electric drive winding temperature data, electronic control module temperature data, drive motor torque data, vehicle speed data, etc., according to the description of the data itself.

[0083] The aforementioned probability distribution model refers to a statistical model used to quantify the frequency and variation of each data category under the current thermal state. Its core is the symbol frequency statistics within the sliding time window, and a dynamic rate of change correction factor is introduced to enhance the response sensitivity to sudden operating conditions.

[0084] The aforementioned probability weights represent the "information contribution" of a certain type of data to the overall thermal state description within the current period, and are the core basis for determining its encoding length.

[0085] The aforementioned coding length refers to the number of Huffman coded bits allocated to each type of data, which determines the space it occupies in the transmission packet.

[0086] The timestamp alignment information mentioned above refers to the actual acquisition time after correction for each type of data, used to evaluate its timing consistency and response priority on the system timeline.

[0087] The compression benefit value mentioned above is a quantitative indicator for measuring the "information retention efficiency after compression" of a certain type of data.

[0088] The aforementioned compression coding refers to assigning a unique binary code sequence to each type of data using an improved Huffman tree based on a determined code length.

[0089] Optionally, upon obtaining Based on this, data compression processing is carried out. By introducing an adaptive Huffman coding mechanism, multi-source synchronous data is structurally reorganized and probabilistically modeled. This significantly reduces the data volume while maintaining the physical meaning and control precision, thereby reducing transmission overhead in the Ethernet link and compressing overall latency. In this process, a data symbol probability distribution model is first constructed to reflect the frequency changes of different collected data under the current thermal state, as shown below:

[0090] (4)

[0091] in, Indicates the first Class data symbols at time Adaptive probability weights, This indicates the frequency of the symbol's occurrence within the sliding time window. This indicates the total number of symbol categories. Indicates the dynamic change inhibition coefficient. This model represents the data value of the corresponding category. By introducing a time change rate term, it suppresses high-fluctuation data with weights, enabling stable data to obtain shorter code lengths in encoding, thereby improving the overall compression efficiency. It also dynamically matches the operating status of the thermal management system, allowing the encoding strategy to adaptively adjust with changes in operating conditions.

[0092] After obtaining the probability distribution, an improved code length generation model is constructed, transforming the traditional static Huffman coding into a dynamic coding mechanism oriented towards real-time hot data streams, as shown below:

[0093] (5)

[0094] in, Indicates the first Encoding length of class data, This represents the corresponding probability weight. This represents the time-sensitive adjustment coefficient. The time alignment deviation obtained from the above steps is represented by the time alignment information. By incorporating the time alignment information into the coding length calculation, data with larger time deviations are encoded with higher priority, thereby prioritizing the recovery of key timing data during transmission and avoiding the impact of delay superposition on subsequent control judgments. The purpose of this coding length generation model is to simultaneously consider data importance and time consistency during the compression process.

[0095] During the encoding process, to further reduce the proportion of redundant data, a data validity screening mechanism is introduced. A compression benefit determination function is constructed to make a retention or compression decision for each type of data, specifically expressed as follows:

[0096] (6)

[0097] in, Indicates the first The compression benefit value of class data, and These represent the synchronization data values ​​at the current time and the previous time, respectively. This indicates the corresponding encoding length. When the data variation is small but the encoding length is large, ... The reduction triggers compression or merging processes, allowing redundant data to be actively reduced during the encoding stage. The purpose of the compression benefit judgment function is to achieve a balance between data volume and information volume, ensuring that every bit transmitted has effective control value.

[0098] It should be noted that through the synergistic effect of the above-mentioned probabilistic modeling, dynamic coding, and benefit screening, the data compression rate is consistently maintained above 50%, while also maintaining... The thermal state information it carries is not distorted, which significantly reduces the risk of congestion in the Ethernet link and shortens the transmission time from the acquisition end to the control end, providing an efficient and structurally optimized data input foundation for subsequent multi-source data fusion and feature extraction.

[0099] Figure 3 This is a flowchart illustrating an adaptive Huffman coding compression process according to one embodiment of the present invention, as shown below. Figure 3 As shown, through the synergistic effect of probabilistic modeling, dynamic coding, and benefit screening, the data compression rate is consistently above 50%.

[0100] Optionally, data filtering is performed on multiple types of temperature control data to obtain partial categories of data, including:

[0101] Step S1031: Determine the data importance of each type of temperature control data based on the encoding length of each type of temperature control data.

[0102] Step S1032: Based on data importance, select a subset of data categories from multiple types of temperature control data that meet the data importance assessment criteria.

[0103] The aforementioned data categories refer to the high-value data retained after importance screening. These data are the sole input source for subsequent feature fusion and working condition identification, and their quantity is far less than the original data.

[0104] After the compressed encoded data is decoded and restored to Next, multi-source data fusion and core feature extraction are performed. By constructing a feature selection mechanism for thermal management control, battery temperature, electric drive speed, cabin set temperature, and ambient temperature are mapped to a unified feature set. Simultaneously, data components that do not contribute to temperature control decisions are removed, thereby reducing computational dimensionality and shortening processing latency. In this process, a feature importance assessment model is first established to quantitatively filter various types of synchronous data, specifically as follows:

[0105] (7)

[0106] in, Indicates the first Feature weights (i.e., data importance) of class data. and This represents the current synchronization data with the previous time step. Indicates the total number of data categories. The encoding length is represented by a feature importance assessment model that evaluates data importance by coupling the magnitude of change with encoding cost. This prioritizes the retention of data that changes significantly and is efficiently encoded, thus proactively filtering out redundant data. The above calculation process further transforms information from the data compression stage into a basis for feature selection, reducing unnecessary computational burden.

[0107] In one optional embodiment, the data importance of multiple types of temperature control data is sorted from high to low, and a preset number of temperature control data categories with the highest data importance are selected as partial category data.

[0108] Optionally, feature fusion processing is performed on some categories of data to obtain fused feature values, including: weighted fusion of some categories of data based on the data importance of temperature control data corresponding to each attribute feature category in the partial category of data to obtain fused feature values.

[0109] After initial screening, a multi-source fusion representation model is constructed for the retained data (i.e., data from some categories), mapping different physical quantities to a unified thermal management control space to form a low-dimensional feature vector, specifically represented as follows:

[0110] (8)

[0111] in, This represents the combined eigenvalues ​​after fusion. This indicates the number of data categories retained after filtering. Indicates the first Historical mean of the data type This represents the corresponding standardized scale parameter. The multi-source fusion expression model, which represents the feature weights, eliminates the dimensional differences between different physical quantities through standardization and achieves a unified expression of multi-source information through weight superposition. This enables battery temperature, electric drive speed, cabin set temperature, and ambient temperature to form a compact representation within the same decision space. The purpose of calculating the comprehensive feature value is to compress high-dimensional data into low-dimensional features while maintaining sensitivity to changes in thermal state.

[0112] Furthermore, to enhance the responsiveness of features to dynamic changes in operating conditions, a feature evolution gradient model is introduced to perform trend enhancement processing on the fused features, specifically expressed as follows:

[0113] (9)

[0114] in, Represents the feature evolution gradient (i.e., used to represent the feature evolution trend). and These represent the fused feature values ​​of the current time step and the previous time step, respectively. This indicates time alignment deviation. By introducing a time compensation term, the gradient calculation is made more stable, avoiding the impact of time error amplification on trend judgment. The purpose of this feature evolution gradient model is to highlight the rate of change of thermal state and provide a more sensitive criterion for subsequent working condition identification.

[0115] During the multi-source data fusion process, key feature parameters such as battery temperature, electric drive speed, cabin set temperature and ambient temperature are extracted. Redundant data that is not related to thermal management and control is discarded, simplifying the data fusion calculation process, reducing the amount of computation in the model calculation process, adapting to the computing power limitations of automotive-grade controllers, and avoiding control lag problems caused by excessive calculation time.

[0116] It should be noted that through the synergistic effect of the above-mentioned feature weight evaluation, multi-source fusion and evolutionary gradient extraction, a core feature set with low dimension, high information density and time consistency is formed, which effectively reduces the computational burden of automotive-grade controllers and avoids control lag caused by redundant data participating in the calculation, providing direct input for rapid identification of subsequent dynamic operating conditions.

[0117] Figure 4 This is a flowchart illustrating multi-source data fusion and feature extraction according to one embodiment of the present invention, such as... Figure 4 As shown, core features are obtained through the synergistic effect of data decoding, feature weight evaluation, multi-source fusion and evolutionary gradient extraction, including but not limited to battery temperature, electric drive speed, cabin temperature and ambient temperature.

[0118] Optionally, based on partial category data, feature evolution trends, and preset thresholds and data importance of temperature control data corresponding to each attribute feature category in the partial category data, the operating condition judgment value is determined, including:

[0119] Step S1061: Determine the initial working condition judgment value based on the preset threshold and data importance of the temperature control data corresponding to each attribute feature category in the partial category data;

[0120] Step S1062: Based on the feature evolution trend, the initial working condition judgment value is corrected to obtain the working condition judgment value.

[0121] The above-mentioned data categories refer to high-value temperature control data (such as battery temperature, electric drive temperature, charging current, drive torque, etc.) that have been retained after importance screening, and redundant and low-value data have been removed.

[0122] The aforementioned preset thresholds refer to the critical values ​​that are pre-set in the thermal management control strategy for each type of key temperature control data and trigger specific operating condition responses, such as: battery temperature threshold of 70°C, charging current threshold of 400A, and driving torque threshold of 300N·m.

[0123] The aforementioned evolutionary trends reflect the dynamic development trend of the thermal state, rather than the static values ​​at a single moment.

[0124] In the obtained and Based on this, a dynamic operating condition rapid identification mechanism is constructed. By pre-mapping key thresholds such as high-rate fast charging current, ambient temperature, and drive torque to a unified judgment space, the control unit can complete real-time operating condition identification without performing complex model calculations. In this process, a multi-threshold unified judgment function is first constructed, which normalizes and fuses the thresholds of different physical quantities, specifically expressed as:

[0125] (10)

[0126] in, This represents the current operating condition judgment value (i.e., the initial operating condition judgment value). Indicates the first Temperature control data, including but not limited to battery current, ambient temperature, and drive torque, Indicates the first The preset threshold corresponding to the temperature control data. Indicates the first The feature weights (i.e., data importance) corresponding to temperature control data. This indicates the number of data categories involved in the judgment (i.e., the number of categories of partial data). By scaling up real-time data with thresholds and adding weights, different physical quantities can be compared uniformly on the same scale. The purpose of calculating the initial working condition judgment value is to achieve rapid fusion judgment of multi-dimensional threshold conditions and avoid the time overhead caused by judging one by one.

[0127] After completing the unified judgment, a dynamic enhancement factor is introduced to incorporate the feature evolution gradient into the working condition identification process, thereby improving the response speed to sudden working conditions. The calculation of the working condition judgment value is as follows:

[0128] (11)

[0129] in, This represents the enhanced operating condition judgment value (i.e., the operating condition judgment value). This represents the gradient amplification factor. This represents the feature evolution gradient. By amplifying the trend of change, rapid temperature rise or drastic load changes receive higher weight in the judgment process, thereby enabling early identification of extreme operating conditions. The purpose of calculating the operating condition judgment value is to transform static threshold judgment into a dynamic response mechanism, improving the system's sensitivity to transient changes.

[0130] Optionally, based on the operating condition judgment value, a vehicle thermal management control strategy is determined, including:

[0131] Step S1071: Use the working condition mapping method to determine the working condition level corresponding to the working condition judgment value;

[0132] Step S1072: Determine the vehicle thermal management control strategy based on the operating condition level.

[0133] The aforementioned operating condition levels refer to discretizing continuous operating condition judgment values ​​into several predefined level labels with clear control, such as "Level 1" and "Level 2", which are used to drive control plans at different levels.

[0134] The aforementioned vehicle thermal management control strategy refers to a set of predefined execution instructions triggered by the operating condition level, covering specific actions such as cooling circuit switching, valve opening, water pump / compressor speed, and charging power limitation.

[0135] Furthermore, to achieve rapid classification and triggering of operating conditions, a hierarchical judgment model is introduced, mapping the enhanced judgment value (i.e., the operating condition judgment value) to different operating condition level ranges. The hierarchical judgment model is specifically represented as follows:

[0136] (12)

[0137] in, Indicates the operating condition level number. Indicates the grading scale parameter, This represents the enhanced operating condition judgment value. Through discretization, continuous judgment values ​​are quickly mapped to specific operating condition categories. Different levels correspond to specific scenarios such as high-rate fast charging, extreme environments, or rapid acceleration, thereby directly triggering the preset temperature control emergency plan. The purpose of using the hierarchical judgment model is to transform the complex continuous judgment process into a low-computation integer mapping operation, significantly improving the execution efficiency of the control unit.

[0138] It should be noted that dynamic operating condition judgment standards such as 4C and above high-rate fast charging current threshold, extreme ambient temperature threshold, and rapid acceleration torque threshold are preset in advance. The preset thresholds are entered into the control unit in advance, and the collected real-time data is compared with the preset thresholds in real time. Various dynamic operating conditions can be quickly identified without complicated operating condition analysis and calculation, and the corresponding temperature control emergency plan can be triggered in a timely manner.

[0139] Furthermore, the preset strategy instruction table is queried to determine the vehicle thermal management control strategy corresponding to the current operating condition level number.

[0140] It is easy to understand that through the synergistic effect of the above-mentioned unified judgment, dynamic enhancement and hierarchical mapping, the millisecond-level identification of key dynamic operating conditions can be achieved, avoiding the delay problem caused by complex model reasoning, enabling the temperature control system to enter the response state in advance, and providing clear and timely control command input for the rapid action of the actuator.

[0141] Figure 5 This is a flowchart illustrating the operation condition identification process according to one embodiment of the present invention, such as... Figure 5 As shown, the process first identifies core features and determines evolutionary gradients, then performs unified threshold judgment and normalization comparison, then uses dynamic enhancement factors to amplify trends, further hierarchically maps discrete operating condition levels, and finally triggers emergency plans corresponding to different operating conditions (such as high-rate fast charging, extreme environments, or rapid acceleration).

[0142] Optionally, the vehicle thermal management control method further includes:

[0143] Step S1091: Based on the operating condition level, the response time of the target valve, and the movement speed of the valve core inside the target valve, determine the movement trajectory of the valve core, wherein the target valve is used to characterize the execution unit that executes the vehicle thermal management control strategy.

[0144] Step S1092: Based on the motion trajectory, generate control commands to implement the vehicle thermal management control strategy.

[0145] The aforementioned target valve refers to a multi-channel intelligent regulating valve integrated into the vehicle's thermal management system. It is typically a four-way or six-way valve body that integrates the switching functions of coolant, refrigerant, and bypass flow paths, and is the core actuator for achieving directional heat distribution.

[0146] The aforementioned valve core refers to a precision component inside the target valve that can move axially or rotatably. It is responsible for controlling the opening and closing of the flow channel and the distribution of flow. Its material, quality, friction characteristics, and movement path directly affect the response speed.

[0147] The aforementioned motion trajectory refers to the time-displacement curve of the valve core as it moves from its current position to the target position. It is the physical basis for achieving "fast, smooth, and low-impact" control.

[0148] The above response time refers to the maximum allowable time required for the target valve to complete the target opening degree from receiving the control command, which is preset by the thermal management strategy level.

[0149] The aforementioned control commands refer to discrete pulse sequences, PWM signals, or CAN frame commands sent to valve drivers (such as stepper motors or DC servo motors) to precisely drive the valve core to move along a planned trajectory.

[0150] In output Building upon this foundation, rapid response optimization at the actuator level is achieved through lightweighting and motion trajectory co-design of the integrated valve body and valve core structure. This enables control commands to be executed with lower resistance and shorter paths, thereby eliminating the impact of actuator delay on the overall thermal management response. In this process, a valve core motion resistance model is first constructed to quantitatively describe the dynamic characteristics after ceramic valve core replacement.

[0151] (13)

[0152] in, This represents the total resistance experienced by the valve core during its movement. This represents the coefficient of friction between the valve core and the valve body contact surface. Indicates the quality of the valve core. Represents gravitational acceleration. Indicates the fluid damping coefficient. This indicates the valve core movement speed, achieved by using lightweight ceramic materials. Significantly reduced, while ceramic surface properties make The decrease in resistance leads to a reduction in overall resistance. The above valve core motion resistance model explains the direct impact of valve core material replacement on response speed from a dynamic perspective.

[0153] Based on reducing resistance, the valve core's stroke trajectory is optimized. By constructing an optimal motion path function, the valve core reduces ineffective displacement while completing the opening and closing action, thereby shortening the execution time. The optimal motion path function is expressed as follows:

[0154] (14)

[0155] in, This indicates the optimal stroke of the valve core. Indicates valve response time. Indicates the valve core at time The speed of movement, Indicates the operating condition adjustment coefficient. Indicates the operating condition level number. This indicates the upper limit of the operating condition level. By introducing an operating condition level adjustment term, the valve core automatically compresses its motion path under high-level operating conditions, achieving a faster opening or closing response. The purpose of using the optimal motion path function is to directly map the control decision results to the actuator's motion strategy, reducing unnecessary mechanical travel.

[0156] Furthermore, to achieve overall optimization of valve response time, an execution response time model is introduced, coupling resistance and stroke into a unified evaluation index:

[0157] (15)

[0158] in, Indicates the valve's response time. This indicates the optimized itinerary. This indicates the average speed of the valve core. Indicates the drag effect coefficient. Indicates total resistance. Indicates valve core quality, achieved by simultaneously reducing and ,make This significantly shortens the valve start-stop response time, reducing it by more than 30%. The purpose of using the execution response time model is to establish a unified evaluation system for the performance of the execution end, enabling the structural optimization effect to be quantified.

[0159] It is easy to understand that, through the synergistic effect of the aforementioned material optimization, path compression, and response modeling, the actuator is able to respond quickly to [various factors]. The control commands are significantly reduced in execution latency, providing a high-response execution foundation for subsequent feedback-based incremental parameter correction, thereby achieving closed-loop rapid regulation of the vehicle's thermal management system. The valve core structure of the existing integrated valve body is optimized by replacing the traditional metal valve core with a lightweight ceramic valve core, reducing frictional resistance during valve core movement. Simultaneously, the valve core stroke trajectory is optimized, shortening the valve core's start-stop movement distance, reducing valve start-stop response time by more than 30%, accelerating the execution speed of control commands, and alleviating latency issues at the execution end.

[0160] Optionally, the vehicle thermal management control method further includes:

[0161] Step S10101: Obtain the target fusion feature value corresponding to the working condition level;

[0162] Step S10102: Determine the temperature control deviation based on the target fusion feature value and the fusion feature value;

[0163] Step S10103: Based on the temperature control deviation and the characteristic evolution trend, determine the temperature control correction parameters, wherein the temperature control correction parameters are used to correct the initial temperature control parameters.

[0164] In the obtained fast execution response Based on this, a closed-loop feedback control mechanism is constructed. By introducing an incremental parameter correction method, the control unit only makes local adjustments to the current temperature control deviation, avoiding the computational burden of recalculating all parameters, thereby further improving control efficiency and response speed. In this process, a temperature control deviation expression model is first established to quantify the difference between the target thermal state and the actual thermal state, specifically expressed as:

[0165] (16)

[0166] in, This indicates the temperature control deviation at the current moment. This represents the fusion feature value obtained from the above steps. Indicates according to The set target thermal state characteristic value is used to map the multi-source thermal state deviations to the same space through the model. The purpose of this formula is to provide a direct basis for subsequent parameter correction and to make the correction process have a clear direction.

[0167] After obtaining the deviation, an incremental parameter update model is constructed to limit the changes in control parameters within the deviation-driven range, thereby reducing unnecessary redundant calculations. The incremental parameter update model is specifically represented as follows:

[0168] (17)

[0169] in, This represents the incremental correction value of the control parameter. This represents the deviation adjustment coefficient. Indicates the trend compensation coefficient. Indicates temperature control deviation. The gradient representing the feature evolution obtained from the above steps is used to introduce a trend term, enabling the system to perform feedforward correction before the deviation expands, thereby reducing control lag. The purpose of using incremental parameter update models is to achieve dual-driven regulation based on deviation and trend, improving control accuracy and response speed.

[0170] Furthermore, to ensure the stability and convergence of the parameter update process, an incremental constraint model is introduced to limit the magnitude of each correction, preventing overshoot or oscillation in the system. The incremental constraint model is specifically expressed as follows:

[0171] (18)

[0172] in, This indicates the control parameters at the current moment. This represents the control parameters at the previous moment. Indicates the incremental correction value. This represents the deviation suppression coefficient. This represents the temperature control deviation. By adaptively scaling the correction magnitude, control updates become smoother when the deviation is large and more agile when the deviation is small. The purpose of using the incremental constraint model is to maintain system stability while ensuring fast response.

[0173] By adopting an incremental parameter correction method instead of the traditional full parameter recalculation mode, incremental correction is only performed on the deviation between the previous control parameter and the actual temperature control effect during the temperature control feedback regulation process. There is no need to recalculate the full parameters, which greatly reduces the calculation time of feedback regulation, realizes the rapid correction of control parameters, and further improves the response efficiency of the overall temperature control system.

[0174] Through the synergistic effect of the above-mentioned deviation modeling, incremental update and constraint adjustment, the control parameters can be quickly optimized locally in each cycle, avoiding the calculation delay caused by full recalculation, and forming an efficient closed loop with the rapid execution of the execution unit. This enables the vehicle thermal management system to maintain a high response and high stability under dynamic conditions, thereby completing a low-latency optimization closed loop from data acquisition to execution feedback.

[0175] It should be noted that in the above embodiments, miniature wireless sensors can be used to replace near-end acquisition, LZ77 algorithms can replace Huffman coding, PCA algorithms can be used to complete feature dimensionality reduction, lightweight machine learning can be used to achieve working condition recognition, electromagnetic fast valves can replace ceramic valve cores, and fuzzy incremental control can be used to complete parameter correction, all of which can achieve similar technical objectives.

[0176] Figure 6 This is a flowchart illustrating the actuator optimization and incremental feedback according to one embodiment of the present invention, as shown below. Figure 6 As shown, by optimizing the valve core (using ceramic materials and stroke compression), the valve start-stop response time is reduced by more than 30%. During temperature control feedback regulation, only the deviation between the previous regulation parameters and the actual temperature control effect is incrementally corrected, eliminating the need to recalculate all parameters. This significantly reduces the calculation time for feedback regulation and enables rapid correction of the regulation parameters.

[0177] According to another aspect of the present invention, a vehicle is also provided, comprising: a memory storing an executable program; and a processor for running the executable program, wherein the executable program executes the vehicle thermal management control method described in any of the preceding embodiments.

[0178] Optionally, in this embodiment, the executable program performs the following steps when it runs:

[0179] Step S101: Decode the first temperature control data to obtain the current multi-source temperature control data. The first temperature control data is obtained by compressing and encoding the original multi-source temperature control data collected in the current acquisition cycle. The current multi-source temperature control data is used to describe the thermal state of multiple thermal management components in the vehicle thermal management system.

[0180] Step S102: Classify the current multi-source temperature control data to obtain multiple types of temperature control data corresponding to various attribute feature categories;

[0181] Step S103: Data filtering is performed on multiple types of temperature control data to obtain partial category data. Among them, the data importance of the temperature control data corresponding to each attribute feature category in the partial category data meets the data importance evaluation conditions.

[0182] Step S104: Perform feature fusion processing on some category data to obtain fused feature values;

[0183] Step S105: Determine the feature evolution trend based on the fused feature value and the historical feature value corresponding to at least one historical acquisition cycle;

[0184] Step S106: Determine the working condition judgment value based on partial category data, feature evolution trend, and preset threshold and data importance of temperature control data corresponding to each attribute feature category in partial category data;

[0185] Step S107: Determine the vehicle thermal management control strategy based on the operating condition judgment value.

[0186] Optionally, the executable program executes the following steps during runtime: acquiring raw multi-source temperature control data, wherein the raw multi-source temperature control data is acquired by multiple data acquisition devices, which are respectively deployed on the surface of multiple thermal management components in the vehicle thermal management system; performing data correction processing and timestamp alignment processing on the raw multi-source temperature control data to obtain preprocessed data; and performing length adaptive encoding on the preprocessed data to obtain first temperature control data.

[0187] Optionally, the executable program executes the following steps during runtime: classifying the preprocessed data to obtain multiple types of preprocessed data corresponding to various attribute feature categories; calculating the probability weights corresponding to each type of preprocessed data using a pre-built probability distribution model; determining the encoding lengths corresponding to each type of preprocessed data based on the probability weights and the timestamp alignment information corresponding to each type of preprocessed data; determining the compression gain values ​​corresponding to each type of preprocessed data based on the encoding lengths; and compressing and encoding the preprocessed data based on the compression gain values ​​to obtain the first temperature control data.

[0188] Optionally, the executable program performs the following steps when it runs: determining the data importance of each type of temperature control data based on the encoding length of each type of temperature control data; and selecting a subset of data that meet the data importance assessment criteria from the multiple types of temperature control data based on the data importance.

[0189] Optionally, the executable program performs the following steps when it runs: weighted fusion of the partial category data according to the data importance of the temperature control data corresponding to each attribute feature category in the partial category data to obtain fused feature values.

[0190] Optionally, the executable program executes the following steps during runtime: determining an initial operating condition judgment value based on the preset threshold and data importance of the temperature control data corresponding to each attribute feature category in the partial category data; and correcting the initial operating condition judgment value based on the feature evolution trend to obtain the operating condition judgment value.

[0191] Optionally, the executable program performs the following steps when it runs: using a working condition mapping method to determine the working condition level corresponding to the working condition judgment value; and determining the vehicle thermal management control strategy based on the working condition level.

[0192] Optionally, the executable program executes the following steps during runtime: determining the movement trajectory of the valve core based on the operating condition level, the response time of the target valve, and the movement speed of the valve core within the target valve, wherein the target valve is used to characterize the execution unit that implements the vehicle thermal management control strategy; and generating control commands based on the movement trajectory to realize the vehicle thermal management control strategy.

[0193] Optionally, the executable program executes the following steps during runtime: obtaining the target fusion feature value corresponding to the operating condition level; determining the temperature control deviation based on the target fusion feature value and the fusion feature value; and determining the temperature control correction parameter based on the temperature control deviation and the feature evolution trend, wherein the temperature control correction parameter is used to correct the initial temperature control parameter.

[0194] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is executed, it controls the device where the computer-readable storage medium is located to execute the vehicle thermal management control method described in any of the above embodiments.

[0195] Optionally, in this embodiment, the executable program can be configured to store an executable program for performing the following steps:

[0196] Step S101: Decode the first temperature control data to obtain the current multi-source temperature control data. The first temperature control data is obtained by compressing and encoding the original multi-source temperature control data collected in the current acquisition cycle. The current multi-source temperature control data is used to describe the thermal state of multiple thermal management components in the vehicle thermal management system.

[0197] Step S102: Classify the current multi-source temperature control data to obtain multiple types of temperature control data corresponding to various attribute feature categories;

[0198] Step S103: Data filtering is performed on multiple types of temperature control data to obtain partial category data. Among them, the data importance of the temperature control data corresponding to each attribute feature category in the partial category data meets the data importance evaluation conditions.

[0199] Step S104: Perform feature fusion processing on some category data to obtain fused feature values;

[0200] Step S105: Determine the feature evolution trend based on the fused feature value and the historical feature value corresponding to at least one historical acquisition cycle;

[0201] Step S106: Determine the working condition judgment value based on partial category data, feature evolution trend, and preset threshold and data importance of temperature control data corresponding to each attribute feature category in partial category data;

[0202] Step S107: Determine the vehicle thermal management control strategy based on the operating condition judgment value.

[0203] Optionally, the executable program can be configured to store an executable program for performing the following steps: acquiring raw multi-source temperature control data, wherein the raw multi-source temperature control data is acquired by multiple data acquisition devices, which are respectively deployed on the surface of multiple thermal management components in the vehicle thermal management system; performing data correction processing and timestamp alignment processing on the raw multi-source temperature control data to obtain preprocessed data; and performing length adaptive encoding on the preprocessed data to obtain first temperature control data.

[0204] Optionally, the executable program executes the following steps during runtime: classifying the preprocessed data to obtain multiple types of preprocessed data corresponding to various attribute feature categories; calculating the probability weights corresponding to each type of preprocessed data using a pre-built probability distribution model; determining the encoding lengths corresponding to each type of preprocessed data based on the probability weights and the timestamp alignment information corresponding to each type of preprocessed data; determining the compression gain values ​​corresponding to each type of preprocessed data based on the encoding lengths; and compressing and encoding the preprocessed data based on the compression gain values ​​to obtain the first temperature control data.

[0205] Optionally, the executable program can be configured to store an executable program for performing the following steps: determining the data importance of each type of temperature control data based on the encoding length of each type of temperature control data; and selecting a subset of data that meet the data importance assessment criteria from the multiple types of temperature control data based on the data importance.

[0206] Optionally, the executable program can be configured to store an executable program for performing the following steps: weighting and fusing the partial category data according to the data importance of the temperature control data corresponding to each attribute feature category in the partial category data to obtain a fused feature value.

[0207] Optionally, the executable program can be configured to store an executable program for performing the following steps: determining an initial operating condition judgment value based on a subset of data, a preset threshold for temperature control data corresponding to each attribute feature category in the subset of data, and the importance of the data; and correcting the initial operating condition judgment value based on the feature evolution trend to obtain the operating condition judgment value.

[0208] Optionally, the executable program can be configured to store an executable program for performing the following steps: using a working condition mapping method to determine the working condition level corresponding to the working condition judgment value; and based on the working condition level, determining the vehicle thermal management control strategy.

[0209] Optionally, the executable program can be configured to store an executable program for performing the following steps: determining the movement trajectory of the valve core based on the operating condition level, the response time of the target valve, and the movement speed of the valve core within the target valve, wherein the target valve is used to characterize the execution unit that executes the vehicle thermal management control strategy; and generating control commands based on the movement trajectory to implement the vehicle thermal management control strategy.

[0210] Optionally, the executable program can be configured to store an executable program for performing the following steps: obtaining the target fusion feature value corresponding to the operating condition level; determining the temperature control deviation based on the target fusion feature value and the fusion feature value; and determining the temperature control correction parameter based on the temperature control deviation and the feature evolution trend, wherein the temperature control correction parameter is used to correct the initial temperature control parameter.

[0211] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0212] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0213] In the embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The system embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of units or modules may be electrical or other forms.

[0214] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0215] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0216] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0217] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A vehicle thermal management control method, characterized in that, include: The first temperature control data is decoded to obtain the current multi-source temperature control data. The first temperature control data is obtained by compressing and encoding the original multi-source temperature control data collected in the current acquisition cycle. The current multi-source temperature control data is used to describe the thermal state of multiple thermal management components in the vehicle thermal management system. The current multi-source temperature control data is classified to obtain multiple types of temperature control data corresponding to various attribute feature categories; Data filtering is performed on the multiple types of temperature control data to obtain partial category data, wherein the data importance of the temperature control data corresponding to each attribute feature category in the partial category data meets the data importance evaluation conditions; The aforementioned category data are subjected to feature fusion processing to obtain fused feature values; Based on the fused feature values ​​and historical feature values ​​corresponding to at least one historical acquisition cycle, the feature evolution trend is determined; Based on the partial category data, the feature evolution trend, the preset threshold of the temperature control data corresponding to each attribute feature category in the partial category data, and the data importance, the working condition judgment value is determined; Based on the aforementioned operating condition judgment values, a vehicle thermal management control strategy is determined.

2. The vehicle thermal management control method according to claim 1, characterized in that, The vehicle thermal management control method further includes: The original multi-source temperature control data is acquired by multiple data acquisition devices, which are respectively deployed on the surface of multiple thermal management components in the vehicle thermal management system. The original multi-source temperature control data is subjected to data correction and timestamp alignment processing to obtain preprocessed data; The preprocessed data is subjected to length adaptive encoding to obtain the first temperature control data.

3. The vehicle thermal management control method according to claim 2, characterized in that, The preprocessed data is subjected to length adaptive encoding to obtain the first temperature control data, including: The preprocessed data is classified to obtain multiple types of preprocessed data corresponding to the various attribute feature categories; Using a pre-constructed probability distribution model, calculate the probability weights corresponding to the various types of preprocessed data. Based on the probability weights and the timestamp alignment information corresponding to the various types of preprocessed data, the encoding lengths corresponding to the various types of preprocessed data are determined. Based on the encoding length, determine the compression benefit values ​​corresponding to the various types of preprocessed data respectively; Based on the compression gain value, the preprocessed data is compressed and encoded to obtain the first temperature control data.

4. The vehicle thermal management control method according to claim 3, characterized in that, Data filtering is performed on the various types of temperature control data to obtain the partial categories of data, including: Based on the encoding lengths corresponding to the various types of temperature control data, the importance of the data corresponding to the various types of temperature control data is determined. Based on the data importance, select the partial categories of data that meet the data importance assessment conditions from the multiple types of temperature control data.

5. The vehicle thermal management control method according to claim 4, characterized in that, The fused feature values ​​are obtained by performing feature fusion processing on the aforementioned partial category data, including: Based on the data importance of the temperature control data corresponding to each attribute feature category in the partial category data, the partial category data is weighted and fused to obtain the fused feature value.

6. The vehicle thermal management control method according to claim 4, characterized in that, Based on the partial category data, the feature evolution trend, and the preset threshold and data importance of the temperature control data corresponding to each attribute feature category in the partial category data, the operating condition judgment value is determined, including: Based on the partial category data, the preset threshold of the temperature control data corresponding to each attribute feature category in the partial category data, and the data importance, the initial working condition judgment value is determined; Based on the aforementioned feature evolution trend, the initial working condition judgment value is corrected to obtain the working condition judgment value.

7. The vehicle thermal management control method according to claim 6, characterized in that, Based on the operating condition determination value, the vehicle thermal management control strategy is determined, including: The working condition level corresponding to the working condition judgment value is determined by using a working condition mapping method. Based on the operating condition level, the vehicle thermal management control strategy is determined.

8. The vehicle thermal management control method according to claim 7, characterized in that, The vehicle thermal management control method further includes: Based on the operating condition level, the response time of the target valve, and the movement speed of the valve core within the target valve, the movement trajectory of the valve core is determined, wherein the target valve is used to characterize the execution unit that executes the vehicle thermal management control strategy; Based on the motion trajectory, control commands are generated to implement the vehicle thermal management control strategy.

9. The vehicle thermal management control method according to claim 7, characterized in that, The vehicle thermal management control method further includes: Obtain the target fusion feature value corresponding to the operating condition level; Based on the target fusion feature value and the fusion feature value, the temperature control deviation is determined; Based on the temperature control deviation and the feature evolution trend, temperature control correction parameters are determined, wherein the temperature control correction parameters are used to correct the initial temperature control parameters.

10. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the executable program, wherein the executable program executes the vehicle thermal management control method according to any one of claims 1 to 9 when it runs.

Images