Thermal management control method and vehicle

Abstract

The invention provides a thermal management control method and a vehicle, and relates to the technical field of thermal management, and the method comprises the steps: determining the predicted temperature of each heat source based on the actually measured temperature, the temperature measurement noise and the model estimation temperature, and determining the temperature weight coefficient corresponding to each heat source based on the actually measured state parameters; determining a global equivalent temperature based on the actually measured state parameter, the predicted temperature and the temperature weight coefficient of the heat source so as to perform thermal management control based on the global equivalent temperature; wherein the model estimation temperature is determined based on a pre-constructed multi-heat-source temperature coupling state model. According to the method, the multi-heat-source temperature coupling state model is constructed, multiple heat sources of which the temperature is independently controlled originally are brought into a unified thermodynamic framework, a foundation is laid for achieving a global collaborative optimization heat management strategy, the generated global equivalent temperature overall plans the heat management requirement of a multi-heat-source system with a single target, and the heat management efficiency is improved. And the heat redundancy exchange phenomenon caused by control strategy splitting in a traditional scheme is reduced.

Description

Technical Field

[0001] This application relates to the field of thermal management technology, and in particular to a thermal management control method and a vehicle. Background Technology

[0002] With the rapid development of the global new energy vehicle industry, the vehicle thermal management system has become one of the key technologies affecting the driving range, safety performance, and driving experience of electric vehicles. Compared with traditional fuel vehicles, the thermal management system of new energy vehicles is more complex, requiring simultaneous consideration of the temperature control needs of multiple heat sources such as the power battery, drive motor, and motor controller. However, existing thermal management strategies lack unified planning and coordinated scheduling of the temperatures of multiple heat sources, and need to be improved. Summary of the Invention

[0003] In view of this, the purpose of this application is to propose a thermal management control method and vehicle to solve the problem that thermal management strategies in related technologies lack unified planning and coordinated scheduling of temperatures from multiple heat sources.

[0004] To achieve the above objectives, this application provides a thermal management control method, the method comprising:

[0005] The measured state parameters of multiple heat sources in the vehicle are obtained; the measured state parameters include the measured temperature and temperature measurement noise of each heat source. The predicted temperature of each heat source is determined based on the measured temperature, temperature measurement noise, and model estimated temperature, and the temperature weighting coefficient corresponding to each heat source is determined based on the measured state parameters. Based on the measured state parameters, predicted temperature, and temperature weighting coefficient of the heat source, the global equivalent temperature is determined, and thermal management control is performed based on the global equivalent temperature. The model estimates the temperature based on a pre-constructed multi-heat-source temperature coupling state model, which is constructed based on the heat generation power of each heat source, the cooling power acting on each heat source, and the pre-stored heat capacity and thermal conductivity parameters of each heat source.

[0006] Furthermore, determining the global equivalent temperature based on the measured state parameters, predicted temperature, and temperature weighting coefficients of the heat source includes: Based on the temperature weighting coefficient of each heat source and the predicted temperature of each heat source, the first equivalent temperature of each heat source is determined. At least one target heat source is determined among the multiple heat sources, and a target weight coefficient corresponding to each target heat source is determined based on the measured state parameters of each target heat source, so as to determine the second equivalent temperature of each target heat source based on the measured temperature of each target heat source and the corresponding target weight coefficient. Determine a first confidence level corresponding to the first equivalent temperature and a second confidence level corresponding to the second equivalent temperature, and determine the process equivalent temperature based on the first equivalent temperature, the first confidence level, the second equivalent temperature and the second confidence level; The equivalent temperature of the process is verified for safety, and the global equivalent temperature is obtained.

[0007] Furthermore, the target heat source is a motor controller and a battery, and the measured state parameters corresponding to the motor controller also include the motor output torque; The determination of the target weight coefficient corresponding to each target heat source based on the measured state parameters of each target heat source includes: The target heat source temperature difference is determined based on the measured temperature of the motor controller and the measured temperature of the battery. Call the pre-stored weighted sensitivity coefficient corresponding to the temperature difference of the target heat source, and the pre-stored load weight coefficient corresponding to the output torque of the motor; Based on the target heat source temperature difference, weighted sensitivity coefficient, motor output torque, and load weight coefficient, the target weight coefficients corresponding to the motor controller and battery are determined respectively. The target heat source temperature difference and the motor output torque are both positively correlated with the target weight coefficient corresponding to the battery.

[0008] Furthermore, the step of performing a safety verification on the process equivalent temperature and obtaining the global equivalent temperature includes: Based on the measured temperature of the battery, the measured temperature of the motor controller, and the preset safety margin, the corrected measured temperature of the battery and the corrected measured temperature of the motor controller are determined respectively. The maximum value among the process equivalent temperature, the corrected predicted battery temperature, and the corrected measured motor controller temperature is determined as the verification process equivalent temperature. Determine the predicted future temperature of the battery after a preset time period; The equivalent temperature of the verification process is corrected based on the predicted future temperature of the battery, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature.

[0009] Furthermore, the step of correcting the process equivalent temperature based on the predicted future temperature of the battery, and determining the corrected process equivalent temperature as the global equivalent temperature, includes: In response to the determination that the difference between the predicted future temperature of the battery and the upper limit of the battery safe temperature is greater than a preset value, the equivalent temperature of the verification process is corrected based on the verification temperature difference, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature. In response to determining that the temperature difference of the verification process is less than or equal to a preset value, the equivalent temperature of the verification process is corrected based on the preset value, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature.

[0010] Furthermore, determining the predicted future temperature of the battery after a preset time period includes: Based on the predicted temperature of the battery at the previous moment and the actual temperature of the battery at the previous moment, the error covariance at the previous moment is determined. Based on the error covariance and battery temperature measurement noise from the previous moment, determine the gain coefficient for the current moment. Based on the gain coefficient at the current moment, the measured temperature of the battery at the current moment, and the predicted temperature at the previous moment, the predicted temperature of the battery at the current moment is determined. Based on the current battery predicted temperature and the multi-heat source temperature coupling state model, the future predicted battery temperature after a preset time period is determined.

[0011] Furthermore, determining the predicted temperature of each heat source based on the measured temperature, temperature measurement noise, and the pre-constructed multi-heat source temperature coupling state model includes: Construct an observation function to reflect the relationship between the measured temperature and the model-estimated temperature of each heat source; A measurement noise matrix is ​​determined based on the temperature measurement noise of each heat source, and a weighted residual cost function is constructed based on the observation function and the weight matrix using the inverse matrix of the measurement noise matrix as the weight matrix. The predicted temperature of each heat source is obtained by minimizing the weighted residual cost function.

[0012] Furthermore, determining the first confidence level corresponding to the first equivalent temperature includes: The measurement noise matrix at the current moment is determined based on the temperature measurement noise of each heat source at the current moment, and the first error matrix is ​​determined based on the measurement noise matrix at the current moment. Based on the predicted temperature and actual temperature of each heat source at the previous moment, the error covariance matrix at the previous moment is determined. Based on the error covariance matrix at the previous moment and the measurement noise matrix determined by the temperature measurement noise of each heat source, the gain matrix at the current moment is determined. Based on the gain matrix, the error covariance matrix at the previous moment is updated to obtain the error covariance matrix at the current moment, so as to determine the second error matrix through the error covariance matrix at the current moment. The first confidence level is determined based on the first error matrix and the second error matrix; The first confidence level increases with the increase of the first error matrix and decreases with the increase of the second error matrix.

[0013] Furthermore, the measured state parameters also include: water pump speed, pedal opening value, and motor output torque; the determination of the temperature weighting coefficient corresponding to each heat source based on the measured state parameters includes: For each of the aforementioned heat sources, perform the following: The basic weights are determined based on the temperature measurement noise of the heat source; The heat exchange dynamic factor is determined based on the real-time thermal resistance from the heat source to the coolant, the water pump speed, and the pre-stored coolant flow rate influence coefficient. Dynamic operating condition factors are determined based on pedal opening value, motor output torque, pre-stored pedal dynamic influence coefficient, and pre-stored torque influence coefficient. The safety boundary factor is determined based on the measured temperature of the heat source, the pre-stored safety enhancement coefficient, the pre-stored upper limit of the safe temperature, and the pre-stored safety margin. The temperature weight coefficient corresponding to the heat source is determined based on the basic weight, heat exchange dynamic factor, dynamic operating condition factor, and safety boundary factor.

[0014] Based on the same inventive concept, this disclosure also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the program to implement any of the methods described above.

[0015] Based on the same inventive concept, this disclosure also provides a vehicle including the above-mentioned electronic devices.

[0016] As can be seen from the above, the thermal management control method and vehicle provided in this application include: determining the predicted temperature of each heat source based on the measured temperature, temperature measurement noise, and model-estimated temperature, and determining the temperature weight coefficient corresponding to each heat source based on the measured state parameters; determining the global equivalent temperature based on the measured state parameters, predicted temperature, and temperature weight coefficient of the heat source, and performing thermal management control based on the global equivalent temperature; wherein, the model-estimated temperature is determined based on a pre-constructed multi-heat source temperature coupling state model, which is constructed based on the heat generation power of each heat source, the cooling power acting on each heat source, and the pre-stored heat capacity parameters and thermal conductivity parameters of each heat source. This application, by constructing a multi-heat source temperature coupling state model, incorporates multiple heat sources that were originally independently temperature-controlled into a unified thermodynamic framework, laying the foundation for achieving a globally coordinated optimization thermal management strategy. This method uses measured temperature, temperature measurement noise, and model-estimated temperature obtained from a multi-heat-source temperature coupling state model to fit the data, resulting in a predicted temperature that balances model prediction and measured feedback. Simultaneously, it dynamically adjusts the weight coefficients of each heat source based on measured state parameters, ensuring that critical components and high-risk conditions receive priority attention. The resulting global equivalent temperature, with a single objective, coordinates the thermal management needs of the multi-heat-source system. This prevents each actuator from operating independently based on local optima, instead enabling them to collaboratively track this unified objective, fundamentally eliminating the redundant heat exchange phenomenon caused by fragmented control strategies in traditional solutions. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of the thermal management control method according to an embodiment of this application; Figure 2 This is a schematic diagram of a vehicle operation control device according to an embodiment of this application; Figure 3 This is a schematic diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0020] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0021] In related technologies, with the rapid development of the global new energy vehicle industry, the vehicle thermal management system has become one of the key technologies affecting the driving range, safety performance, and driving experience of electric vehicles. Compared with traditional fuel vehicles, the thermal management system of new energy vehicles is more complex, requiring simultaneous temperature control of multiple heat sources such as the power battery, drive motor, motor controller, on-board charger (OBC), DC-DC converter (DCDC), and intelligent driving controller (IDC).

[0022] Current mainstream new energy vehicle thermal management systems primarily employ liquid cooling technology, using independent cooling loops to regulate the temperature of key components such as the battery pack, motor, and electronic control system. Specifically, the power battery is typically maintained within its optimal operating temperature range of 20℃-40℃, while the motor and electronic control system are configured with independent temperature thresholds based on their power characteristics. Each subsystem achieves closed-loop control through dedicated electric water pumps, electronic water valves, and heat exchangers. However, while this distributed control architecture is simple and easy to implement, each heat source operates independently based on local optima, lacking unified planning and coordinated scheduling of the vehicle's overall heat flow.

[0023] In recent years, the industry has gradually developed integrated thermal management systems, which connect battery, motor control, and air conditioning system loops through multi-channel valves or pipelines to form a large circulation loop to improve energy utilization efficiency. However, the integrated thermal management systems in related technologies still mostly adopt threshold-based on-off control at the control strategy level, failing to fully consider the thermodynamic coupling relationship and dynamic thermal inertia characteristics between multiple heat sources. This results in redundant heat exchange during the transfer process, making it difficult to achieve optimal system-level energy consumption.

[0024] For example, in low-temperature winter environments, during the initial startup of a vehicle, the motor and electronic control system rapidly heat up to over 60°C due to the driving load, while the power battery needs to be heated to 25°C before it can discharge normally due to the low ambient temperature. In traditional control strategies, the motor circuit initiates a large-loop cooling system, with heat dissipated directly to the external environment through the front-end radiator; simultaneously, the battery circuit independently activates a PTC heater or heat pump system to extract heat from the environment. The waste heat generated by the motor could ideally be transferred to the battery circuit for preheating via a heat exchanger, but due to the lack of a coordinated control mechanism between the two circuits, the high-temperature coolant and the low-temperature coolant do not establish effective thermal coupling, resulting in redundant energy exchange of "cooling down on one side and heating on the other."

[0025] From the perspective of thermodynamic coupling mechanisms, the multi-heat-source system of a vehicle exhibits significant spatiotemporal correlation characteristics. The heat generated by the power battery during charging and discharging is complementary to the waste heat from the motor and electronic control system. However, in traditional control strategies, temperature sensor sampling points are arranged in isolation, lacking the ability to dynamically identify heat flow paths and accurately predict the thermal inertia delay effect between different heat sources. Furthermore, existing control algorithms mostly employ single-point temperature control modes, failing to establish a unified temperature objective function, resulting in limited information exchange between controllers of various subsystems and making it difficult to achieve optimal global energy allocation.

[0026] Furthermore, dynamic load changes in vehicles pose a challenge to the real-time response capabilities of thermal management systems. Existing control strategies generally employ a passive response mode, activating cooling or heating devices only after a temperature threshold is detected. This lag in control leads to excessive compensation during transient conditions such as rapid acceleration, high-speed driving, or fast charging, causing unnecessary energy waste. Therefore, developing intelligent thermal management strategies that integrate multi-heat-source thermodynamic coupling models and possess predictive collaborative control capabilities has become an important technological direction for improving the energy utilization efficiency of new energy vehicles.

[0027] The following is in conjunction with the appendix Figures 1-3 The present application will be described in conjunction with the embodiments.

[0028] In some embodiments, a thermal management control method is implemented by a vehicle controller or by another controller independent of the vehicle controller. For the sake of convenience in the following description, unless otherwise specified, the method is described using the vehicle controller as an example.

[0029] In some embodiments, the thermal management control method, referencing Figure 1 ,include: S101. Obtain the measured state parameters of multiple heat sources in the vehicle; the measured state parameters include the measured temperature of each heat source. Temperature measurement noise .

[0030] In this step, the status parameters of the multi-heat source system are acquired in real time via the vehicle's CAN bus to construct a global temperature state vector. Specifically, this includes: (1) Heat source temperature acquisition

[0031] Collect power battery temperature Motor temperature DC-DC / OBC temperature Motor controller temperature Battery circuit water temperature Motor circuit water temperature The measured temperature data, along with the actual temperature data from other relevant controllers in each loop, form a temperature observation vector:

[0032] in, The sensor measurement value (including noise) for the temperature of the i-th heat source is given in K or °C.

[0033] It should be noted that each heat source can be equipped with multiple sampling points, and each sampling point can be equipped with a temperature sensor. It can be the average temperature of a heat source at multiple sampling points.

[0034] (2) Temperature measurement noise

[0035] The measurement noise of each heat source was determined by measuring its actual temperature. Forming a measurement noise matrix .

[0036]

[0037]

[0038] in, :and Column vectors of the same dimension represent the deviation of the corresponding input values, that is, they reflect the uncertainty of the sensor's measurement values ​​or the sensor's accuracy. : Normal binomial distribution formula, 0: expected value and variance are 0; Or R: Measurement noise matrix; The measurement noise of the battery, motor, DC-DC / OBC, motor controller, battery circuit water temperature, and motor circuit water temperature sensor are represented in sequence. : A diagonal matrix, with only the diagonal lines being non-zero; representing the independent noise of each sensor (without cross-correlation).

[0039] For example, the temperature measurement noise To determine the variance of the measured temperature based on the i-th heat source, for example, data from the current moment and several adjacent moments can be used for variance calculation. In specific implementation, a sliding window method is used to estimate the temperature measurement noise variance of each heat source in real time: the window length is set. (Example 5-10) Collect sensor measurements at various times within the acquisition window. Calculate the sample variance as an estimate of the measurement noise variance at the current time:

[0040] in, This represents the sample mean of the measurements within the window. This dynamic estimation method can adapt to measurement accuracy drift caused by sensor aging, electromagnetic interference, and changes in operating conditions.

[0041] (3) Extended state parameter acquisition In addition to the measured temperature, the real-time thermal resistance from the i-th heat source to the coolant can also be obtained simultaneously. Fan speed Pump speed Accelerator pedal opening value Brake pedal opening value Motor output torque The temperature of the motor and battery, etc., are used to prepare for the calculation of the temperature weighting coefficients of each heat source in subsequent embodiments.

[0042] S102, Based on the measured temperature and temperature measurement noise The model estimates the temperature to determine the predicted temperature of each heat source, and the measured state parameters are used to determine the temperature weighting coefficient corresponding to each heat source. .

[0043] The model estimates the temperature based on a pre-constructed multi-heat-source temperature coupling state model, which is constructed based on the heat generation power of each heat source, the cooling power acting on each heat source, and the pre-stored heat capacity and thermal conductivity parameters of each heat source.

[0044] For example, the process of determining the predicted temperature mainly utilizes a multi-heat-source temperature coupled state model for state recursion prediction. This model is built based on the law of energy conservation and describes the dynamic law of the temperature evolution of each heat source over time. At the current sampling moment, based on the optimal temperature estimate from the previous moment and the input variables at the current moment (including the heat generation power of each heat source, the cooling power acting on each heat source, etc.), the estimated temperature of each heat source is obtained through state transition calculation. Simultaneously, the measured temperature of each heat source is obtained through the vehicle's CAN bus. Although this measured temperature includes sensor measurement noise, it directly reflects the true physical state of the thermal management system. An observation function is constructed to reflect the relationship between the measured temperature and the model-estimated temperature. A measurement noise matrix is ​​determined based on the temperature measurement noise of each heat source. Using the inverse of this matrix as the weight matrix, a weighted residual cost function is constructed. The smaller the measurement noise, the higher the sensor accuracy, and the greater the weight of the residual of the corresponding heat source in the cost function. By minimizing this weighted residual cost function, the optimal predicted temperature that minimizes the overall deviation between the model output and the measured data is obtained, achieving optimal fitting between the thermodynamic model and the actual vehicle measurement data.

[0045] For example, the determination of temperature weighting coefficients can be based on a dynamic adaptive mechanism constructed from multi-dimensional measured state parameters, comprehensively evaluating the importance of each heat source from four levels: basic weight, heat exchange dynamics, dynamic operating conditions, and safety boundary. The basic weight can be directly determined by temperature measurement noise; the lower the noise, the higher the weight, reflecting trust in high-precision sensor data. The heat exchange dynamics level considers the real-time thermal resistance between the heat source and the coolant, as well as the coolant flow rate; when heat exchange efficiency decreases, the weight of the heat source increases accordingly. The dynamic operating conditions level monitors the rate of change of the accelerator pedal and brake pedal, as well as the motor output torque; heat sources with rapidly changing heat loads receive higher weights. The safety boundary level increases the weight sharply when the temperature approaches the safety limit. After multiplying the weighting factors of the above four levels and normalizing them, the final temperature weighting coefficient for each heat source is obtained, used for subsequent weighted fusion of the global equivalent temperature, ensuring that critical heat sources, high-risk conditions, and heat exchange-limited scenarios receive priority in vehicle thermal management decisions.

[0046] The multi-heat-source temperature coupling state model is constructed based on the heat generation power of each heat source, the cooling power acting on each heat source, and pre-stored heat capacity and thermal conductivity parameters of each heat source. For example, the multi-heat-source temperature coupling state model is as follows:

[0047] in, ; ; .

[0048] in, Process noise, and The column vectors of the same dimension represent the deviations that the corresponding input values ​​produce during the process, such as process deviations caused by cooling coefficient efficiency, sensor aging, battery degradation and aging, etc. : Normal binomial distribution formula, 0: expected variance is 0; Q: process noise matrix; I is the identity matrix, ensuring the initial parameter values.

[0049] The derivation of this model is as follows: (1) Define the temperature vector of each heat source Heat capacity matrix Thermal conductivity matrix Input vector The details are as follows:

[0050] in, Battery temperature (°C / K); Motor temperature (°C / K); : DCDC / OBC temperature (°C / K); Motor controller temperature (°C / K); Battery circuit water temperature (°C / K); Motor circuit water temperature (°C / K).

[0051]

[0052] in, Heat capacity (J / K) is related to material quality and specific heat capacity. ; Mass (kg) Specific heat capacity (J / (kg·K)); Battery temperature heat capacity (J / (kg·K)) Motor temperature heat capacity (J / (kg·K)); : Temperature heat capacity of DCDC / OBC (J / (kg·K)); Motor controller temperature heat capacity (J / (kg·K)); : Battery circuit water temperature heat capacity J / (kg·K); Heat capacity of water in motor circuit (J / (kg·K)).

[0053]

[0054] in, Thermal conductivity (W / K) is the thermal conductivity between heat sources i and j.

[0055]

[0056] in, Heat generation power (such as the heat generation power of batteries, motors, motor controllers, DC-DC / OBC / IDC, etc.); : Cooling power applied to each heat source, i.e., the heat power removed by the cooling system (negative values ​​indicate cooling); , The formula for dynamic influence is given, where, ; ; in, The parameters affecting heat generation in the formula, in order, are: vehicle speed / electric power / torque / ambient temperature / accelerator pedal opening / brake pedal opening; : Cooling thermal power of heat source i, negative values ​​indicate cooling, positive values ​​indicate heating (in practice, it is usually negative). The convective heat transfer coefficient (W / (m²·K)) between the coolant and the heat source i depends on the coolant flow rate, material, surface roughness, etc. : Heat exchange area (m²) of coolant in contact with heat source i, surface area of ​​heat source i in contact with coolant; : The temperature of the coolant at heat source i (°C / K), usually the coolant inlet or average temperature; Temperature of heat source i (°C / K).

[0057] In addition, for ease of calculation, we can also... Simply put, it is the difference between input power and output power. Taking a motor as an example, the difference between the AC power input to the motor stator and the mechanical power output from the motor shaft is the heat generation power of the motor body.

[0058] (2) Based on the defined temperature vectors of each heat source Heat capacity matrix Thermal conductivity matrix Input vector The energy conservation equation is constructed as follows: C·

[0059] This energy conservation equation is a differential equation in the continuous time domain, describing the continuous change of the heat source temperature over time.

[0060] (3) Discretize the above energy conservation equation to obtain the multi-heat source temperature coupling state model, as follows: ; in, ; .

[0061] Therefore, the multi-heat-source temperature coupling state model is determined as follows: .

[0062] The discretization process essentially divides the time axis into equally spaced sampling periods Δt (100ms-1s in the example), where k represents the k-th sampling time. This discretization process is implemented using the forward Euler method, assuming that input quantities such as heat flux density and cooling power exist in each sampling period. By maintaining constancy, the continuous thermodynamic dynamic process is transformed into an algebraic recursive relationship at each sampling point. Therefore, the state vector T(k) in the discrete equation represents the instantaneous temperature estimate of each heat source at the k-th sampling time, rather than the average temperature or continuous curve over the time period. This provides a computational basis in a discrete-time frame for subsequent estimation calculations based on the sampling point sequence.

[0063] S103. Based on the measured state parameters, predicted temperature and temperature weighting coefficient of the heat source, determine the global equivalent temperature, and perform thermal management control based on the global equivalent temperature. For example, the determination of the global equivalent temperature can employ a weighted fusion mechanism, combining the predicted temperatures of each heat source and their corresponding temperature weighting coefficients for calculation. First, the predicted temperatures of each heat source are multiplied by their respective temperature weighting coefficients and summed, then divided by the sum of the weighting coefficients to obtain a preliminary weighted average equivalent temperature. Subsequently, a safety boundary constraint is introduced, comparing this weighted average equivalent temperature with the actual temperatures of each critical heat source minus their safety margins. The maximum value is taken as the equivalent temperature after safety constraints, ensuring it does not fall below the safety lower limit of any critical component. Finally, combined with predictive safety correction, if it is predicted that the temperature of a critical heat source will approach its safety upper limit at some future moment, the current equivalent temperature is proactively compensated and increased to form the final global equivalent temperature. This equivalent temperature is not a simple arithmetic average but dynamically reflects the overall thermal state and potential risks of the multi-heat-source system.

[0064] When performing thermal management control based on the global equivalent temperature, this equivalent temperature is output as a unified control target to each heat source controller, replacing the independently set temperature thresholds for each subsystem in traditional solutions. The actuators, such as the battery thermal management module, motor circuit thermal management module, and air conditioning circuit thermal management module, no longer make independent decisions based on local sensors. Instead, they collaboratively track the global equivalent temperature and adjust execution variables such as electric water pump speed, electronic water valve opening, fan speed, and heat pump operating power to converge the temperatures of each heat source towards this unified target. When the global equivalent temperature rises, the system prioritizes enhancing heat dissipation; when the equivalent temperature decreases, the system comprehensively utilizes waste heat from the motor or activates heating devices, achieving tiered utilization and dynamic allocation of heat between heat sources, significantly reducing the energy consumption of the vehicle's thermal management system.

[0065] The vehicle thermal management control method in this embodiment includes: determining the predicted temperature of each heat source based on the measured temperature, temperature measurement noise, and model-estimated temperature; determining the temperature weight coefficient corresponding to each heat source based on the measured state parameters; determining the global equivalent temperature based on the measured state parameters, predicted temperature, and temperature weight coefficient of the heat source, and performing thermal management control based on the global equivalent temperature; wherein, the model-estimated temperature is determined based on a pre-constructed multi-heat source temperature coupling state model, which is constructed based on the heat generation power of each heat source, the cooling power acting on each heat source, and the pre-stored heat capacity and thermal conductivity parameters of each heat source. This embodiment incorporates multiple heat sources that were originally independently temperature-controlled into a unified thermodynamic framework by constructing a multi-heat source temperature coupling state model, laying the foundation for a globally coordinated optimization thermal management strategy. This method uses the measured temperature, temperature measurement noise, and model-estimated temperature obtained from the multi-heat source temperature coupling state model for data fitting to obtain a predicted temperature that takes into account both model prediction and measured feedback. At the same time, it dynamically adjusts the weight coefficient of each heat source according to the measured state parameters to ensure that key components and high-risk conditions receive priority attention. The global equivalent temperature generated on this basis coordinates the thermal management needs of a multi-heat source system with a single objective. This enables each actuator to no longer operate independently according to local optima, but to collaboratively track the unified objective, fundamentally eliminating the redundant heat exchange phenomenon caused by the fragmented control strategies in traditional solutions.

[0066] In some embodiments, the step S102 based on the measured temperature and temperature measurement noise... or And model-estimated temperatures to determine the predicted temperatures of each heat source, including: S201, Construct a system to reflect the measured temperatures of each heat source. Temperature estimated by model The observation function of the relationship between them:

[0067] in, This refers to the temperature observation vector formed by the measured temperatures of each heat source, as described in S101 above. This refers to the measurement noise matrix formed by the measurement noise of each heat source, as described in S101. The model estimate of the temperature vector for each heat source obtained from the multi-heat-source temperature coupling state model in the aforementioned embodiments can also be called the current state estimate; H is the temperature observation matrix, as follows:

[0068] Where H is the identity matrix. The positions of the non-zero elements of this observation matrix intuitively reflect the physical layout strategy of the sensors in the vehicle's thermal management system. If the temperature of all heat sources can be directly measured, it is a full observation matrix; if in practical applications only the temperature of the battery and motor controller is measured, it is a sparse matrix.

[0069] In this step, an observation function is used to establish a mapping relationship from the theoretical state space to the sensor measurement space. This observation function clarifies the correspondence between the theoretical predicted temperature of each heat source and the actual sensor arrangement, i.e., which heat sources' predicted temperatures can be directly measured and which need to be indirectly calculated through the model. Through this mapping, abstract thermodynamic state variables are associated with specific physical measurements, providing a unified dimensional benchmark for subsequent residual calculations. This construction process fully considers the flexibility and scalability of the vehicle's sensor layout. Whether it is a full-observation configuration or an under-observation configuration, it can adapt to different hardware solutions by adjusting the mapping relationship, ensuring the reusability and portability of the algorithm platform across different vehicle models, and reducing the development cost and cycle of the thermal management control system.

[0070] The following S202 and S203 are implemented using the weighted least squares (WLS) method.

[0071] S202. Determine the measurement noise matrix R based on the temperature measurement noise of each heat source, and use the inverse matrix of the measurement noise matrix... As a weight matrix Construct a weighted residual cost function based on the observation function and the weight matrix. .

[0072] The weighted residual cost function is as follows:

[0073] in, Weight matrix, using The mathematical derivation of maximum likelihood estimation also guarantees statistical optimality. The weights reflect the reliability of the observations: that is, the smaller the noise, the larger the weight. In this step, a measurement noise matrix is ​​constructed based on the measurement noise of each heat source temperature sensor. Its inverse matrix is ​​then used as the weight matrix to construct a weighted residual cost function. This weight matrix essentially quantifies the reliability of data from different sensors: sensors with high accuracy and low noise are assigned larger weights, and their measurement residuals play a more significant role in the cost function; sensors with low accuracy and high noise are assigned smaller weights, and their measurement deviations are effectively suppressed from affecting the overall optimization objective. This differentiated weighting mechanism based on measurement uncertainty allows the weighted residual cost function to objectively reflect the engineering intuition that "high-precision data should be prioritized for fitting, while low-precision data should be adopted with caution," avoiding the contamination of the optimal solution by noise caused by simple averaging or equal weighting, and significantly improving the accuracy and reliability of temperature estimation.

[0074] S203. By minimizing the weighted residual cost function, the predicted temperature of each heat source is obtained. .

[0075] Specifically, the gradient of the weighted residual cost function is calculated:

[0076] Set the gradient to zero:

[0077] WLS estimated solution:

[0078] The purpose of calculating the gradient is to determine the optimization direction of the current state estimate T(k) (i.e., the model estimated temperature) (i.e., how to adjust T(k) to optimize the cost function). Minimize); the purpose of setting the gradient to zero is to find the minimum point of the cost function, thereby solving for the optimal state estimate of each heat source. (i.e., predicting temperature).

[0079] In this step, an optimization algorithm is used to find the temperature estimate that minimizes the weighted residual cost function, which serves as the predicted temperature output for each heat source. This solution process essentially seeks a set of optimal temperature estimates that minimizes the overall deviation between the model's predicted output and the weighted measured data, thereby achieving optimal fusion between thermodynamic model deduction and real-vehicle measurement feedback. The minimization operation fully utilizes the convexity of the weighted residual cost function, avoiding local minima traps. The final predicted temperature conforms to the physical laws of energy conservation and heat conduction, and closely matches the measured data after precision weighting by sensors. This provides a high-precision, low-noise state estimation basis for the subsequent generation of the global equivalent temperature, effectively supporting the refined control requirements of the vehicle's thermal management system.

[0080] Compared to the decentralized approach of traditional independent control strategies where each heat source makes independent decisions based on its local sensors, this embodiment first establishes a mapping relationship from the theoretical state space to the sensor measurement space using an observation function. Then, it introduces a quantitative assessment of sensor accuracy differences through the measurement noise covariance matrix, achieving differentiated processing of data reliability. Finally, by minimizing the weighted residuals, the predicted temperature achieves an optimal balance between physical plausibility and data fit. This architecture significantly improves the accuracy and robustness of temperature estimation: the advantages of high-precision sensors are fully utilized, the influence of low-precision sensors is effectively suppressed, and the weights of model prediction and measured feedback are adaptively adjusted according to operating conditions. The resulting predicted temperature lays a reliable foundation for subsequent calculations of the global equivalent temperature, enabling the vehicle's thermal management system to achieve precise allocation and tiered utilization of heat between heat sources under complex operating conditions, ultimately achieving the dual optimization goals of energy consumption reduction and safety improvement.

[0081] In some embodiments, the measured state parameters further include: water pump speed, pedal opening value, and motor output torque; the step of determining the temperature weighting coefficient corresponding to each heat source based on the measured state parameters in S102 includes: For each of the aforementioned heat sources, perform the following: S301. Determine the basic weights based on the temperature measurement noise of the heat source. .

[0082] The process for determining the temperature measurement noise of the temperature sensors for each heat source has been mentioned in S102 above, and can be used... or This means that the base weights in this step can be... .

[0083] In this step, the basic weight of each heat source is determined by the reciprocal of the square of the temperature measurement noise. The lower the temperature measurement noise, the higher the accuracy of the sensor, and the greater the basic weight of the corresponding heat source. Thus, the measured data of the high-precision sensor is preferentially adopted in the subsequent weighted calculation.

[0084] S302. Determine the heat exchange dynamic factor based on the real-time thermal resistance from the heat source to the coolant, the water pump speed, and the pre-stored coolant flow rate influence coefficient. .

[0085]

[0086] in, : Real-time thermal resistance from the i-th heat source to the coolant; : Flow impact coefficient; Pump speed (%)

[0087] For example, the real-time thermal resistance of the battery to the coolant is: ;in, : Initial thermal resistance of the battery (can be a pre-calibrated value); Aging factor, in example, 0.001-0.005 (years). - ¹As vehicles age, coolant may deposit sediment, and scale may form on cooling pipes, leading to increased thermal resistance. Vehicle service life; For example, the real-time thermal resistance of the motor controller to the coolant is: ;in, : Initial thermal resistance of the motor controller (can be a pre-calibrated value); Dust accumulation coefficient (example: 0.0001-0.001 (h·%)) -1 ); The cumulative workload of the fan is the integral of the fan speed over time, reflecting the degree of fan usage, ranging from 0 to 10000 h·%. Fan speed (%) Thermal resistance influence coefficient (example is 0.005-0.01).

[0088] In this step, the real-time heat exchange efficiency between the heat source and the coolant is comprehensively evaluated. When the thermal resistance increases or the coolant flow rate decreases, resulting in a decrease in heat exchange efficiency, the heat exchange dynamic factor increases accordingly, giving the heat source with limited heat exchange a higher weight of attention.

[0089] S303. Determine dynamic operating condition factors based on pedal opening value, motor output torque, pre-stored pedal dynamic influence coefficient, and pre-stored torque influence coefficient. .

[0090]

[0091] in, Accelerator pedal dynamic influence coefficient (example is 0.3-0.5); , These are the accelerator pedal opening value (0-100%) and the brake pedal opening value (0-100%), respectively. Brake pedal dynamic influence coefficient (example is 0.1-0.3); Torque influence coefficient (typical value: 0.1-0.3); Torque Motor output torque (Nm).

[0092] In this step, the rate of change of the accelerator pedal and brake pedal, as well as the output torque level of the motor, are monitored. When the vehicle is in a state of rapid acceleration, rapid deceleration, or high load, the dynamic operating condition factor is increased, which increases the weight of the heat source with rapidly changing heat load to match the transient response requirements.

[0093] S304. Determine the safety boundary factor based on the measured temperature of the heat source, the pre-stored safety enhancement factor, the pre-stored upper limit of the safe temperature, and the pre-stored safety margin. .

[0094] ; in, Safety enhancement factor (example is 3.0-5.0); The safe temperature limits for each heat source are as follows: Battery: 55℃; Motor: 130℃; Motor controller: 105℃; DC-DC converter: 60℃; Intelligent driving controller (IDC): 80℃. Safety margin (example: 3-5℃).

[0095] In this step, the degree to which the temperature of each heat source approaches the safety upper limit is assessed. When the measured temperature enters the safety margin range, the safety boundary factor increases sharply, which greatly increases the weight of heat sources at risk of overheating to trigger protective control.

[0096] S305, Based on the aforementioned basic weights Heat exchange dynamic factor Dynamic operating condition factors and safety boundary factor Determine the temperature weighting coefficient corresponding to the heat source. .

[0097] Right now, ; In this step, the basic weight, heat exchange dynamic factor, dynamic operating condition factor and safety boundary factor are multiplied and normalized to obtain the final temperature weight coefficient of each heat source, thereby achieving a comprehensive quantitative assessment of sensor accuracy, heat exchange status, operating condition severity and thermal safety risk.

[0098] This embodiment constructs a multi-dimensional, adaptive heat source importance assessment system through a four-layer progressive weight calculation architecture consisting of basic weights, dynamic heat exchange factors, dynamic operating condition factors, and safety boundary factors. This architecture first establishes a baseline of confidence in sensor accuracy using basic weights, ensuring that high-precision measurement data is prioritized. Then, the dynamic heat exchange factor senses changes in the heat transfer efficiency of the thermal management system in real time, enabling heat sources with increased thermal resistance or insufficient flow to receive timely weight compensation. Next, the dynamic operating condition factor captures transient characteristics of driving behavior, allowing critical heat sources to respond quickly under severe conditions such as rapid acceleration and high load. Finally, the safety boundary factor solidifies the thermal safety baseline, triggering a weight surge to achieve proactive defense when the temperature approaches its limit. The product fusion and normalization of the four layers of factors allow the temperature weight coefficients of each heat source to evolve in real time with the vehicle's operating status. This avoids the one-sidedness of single-dimensional assessment and overcomes the rigidity of fixed weight allocation, providing a scientific and reasonable priority ranking basis for subsequent weighted fusion of global equivalent temperatures. This significantly improves the adaptive capability and thermal safety reliability of the vehicle's thermal management system under complex operating conditions.

[0099] In some embodiments, determining the global equivalent temperature based on the measured state parameters, predicted temperature, and temperature weighting coefficient of the heat source in step S103 includes: S401, Based on the temperature weighting coefficient of each heat source and the predicted temperature of each of the heat sources. Determine the first equivalent temperature of each heat source. .

[0100] ; in, The predicted temperatures of each heat source obtained from S203 are... The weighting coefficients can be preset or determined in S305.

[0101] This step fuses the predicted temperatures of each heat source to generate a first equivalent temperature. This first equivalent temperature is essentially the theoretical highest priority value under least squares estimation, providing a physical constraint benchmark for subsequent fusion with the second equivalent temperature driven by measured data, effectively avoiding the physical inconsistencies that may arise from purely data-driven methods.

[0102] S402. Among the multiple heat sources, at least one target heat source (such as a battery or motor controller) is determined, and based on the measured state parameters of each target heat source, a target weight coefficient corresponding to each target heat source is determined, so as to determine the second equivalent temperature of each target heat source based on the measured temperature of each target heat source and the corresponding target weight coefficient. .

[0103] The target heat source is a motor controller and a battery, and the measured state parameters corresponding to the motor controller also include the motor output torque.

[0104] Step S402, which involves determining the target weight coefficient for each target heat source based on its measured state parameters, includes: determining the temperature difference between the target heat sources based on the measured temperature of the motor controller and the measured temperature of the battery. ; Invoke the pre-stored weighted sensitivity coefficient corresponding to the temperature difference of the target heat source, and the pre-stored load weight coefficient corresponding to the output torque of the motor; Based on the temperature difference of the target heat source Weighted sensitivity coefficient Motor output torque and load weighting coefficient Determine the target weight coefficients corresponding to the motor controller and the battery respectively; wherein, the target weight coefficient corresponding to the battery is: The target weight coefficient corresponding to the motor controller is The target heat source temperature difference and the motor output torque are both positively correlated with the target weight coefficient corresponding to the battery.

[0105] in, = ; ; in, This refers to the actual measured temperature of the battery. The actual temperature measured for the motor controller; , : Design parameters (TBD) (Weighted sensitivity coefficient); : Load weighting coefficient; Motor output torque; 。

[0106] In this step, the most critical target heat sources (such as the power battery and motor controller) affecting the overall vehicle safety and performance are selected from the multi-heat source system. Based on their measured state parameters, target weight coefficients are dynamically determined to generate a second equivalent temperature. This second equivalent temperature is an estimate calculated directly from sensor measurement data, reflecting the real physical state of the key heat sources in real time, unaffected by least squares estimation or parameter drift. By introducing target weight coefficients, the proportions of the battery and motor controller in the equivalent temperature can be dynamically adjusted according to their real-time temperature differences, load requirements, and safety margins: increasing the battery weight when the battery temperature is low and preheating is required, and increasing the motor controller weight when the motor controller experiences a sudden temperature rise under high load. This data-focused mechanism targeting key heat sources ensures the second equivalent temperature's rapid response to extreme conditions and sudden risks, compensating for the lag in least squares estimation under transient conditions.

[0107] S403, Determine the equivalent temperature with respect to the first temperature. The corresponding first confidence level and the second equivalent temperature The corresponding second confidence level The equivalent temperature is determined based on the first equivalent temperature, the first confidence level, the second equivalent temperature, and the second confidence level. ,as follows: ; in, The error covariance can be determined based on the weighted least squares method.

[0108] In this step, the fusion weights of the first and second equivalent temperatures are determined by estimating the reliability of the predictions and the confidence of the measured data, thus generating the process equivalent temperature. When the sensor is operating stably and the measurement noise is low, the second confidence level is increased to enhance the decision weight of the measured data; when drastic changes in operating conditions lead to model parameter mismatch or the sensor is subjected to electromagnetic interference, the first confidence level is increased to rely on the smooth prediction capability of the physical model. This adaptive confidence level adjustment mechanism achieves a dynamic game balance between the estimated prediction and the measured feedback: it avoids both over-reliance on noise disturbances that may be introduced by the sensor and the accumulation of errors caused by long-term model operation, ensuring that the process equivalent temperature maintains optimal estimation accuracy across the entire operating range. This fusion strategy significantly improves the robustness and reliability of the thermal management system under sensor failure, extreme environments, and dynamic loads.

[0109] S404. Perform a safety verification on the process equivalent temperature and obtain the global equivalent temperature. .

[0110] In this step, multi-level safety verification can be implemented on the process equivalent temperature to ensure that the final output global equivalent temperature meets the safety operating constraints of each key heat source. An example of this multi-level safety verification is as follows: First, compare the process equivalent temperature with the measured temperature of each target heat source minus the safety margin, and take the maximum value as the lower safety limit constraint to prevent the risk of overheating of key components from being ignored due to setting the equivalent temperature too low. Second, predictive safety correction is introduced, based on the state model to predict the future temperature evolution trend of key heat sources. If it is predicted that a heat source will approach the safety upper limit, the current equivalent temperature is proactively increased. This verification mechanism takes into account both short-term optimization and long-term safety, ensuring both energy-saving optimization space for the global equivalent temperature and a solid thermal safety baseline, avoiding short-sighted behavior such as sacrificing component lifespan or triggering overheat protection in pursuit of optimal energy consumption.

[0111] This embodiment constructs a vehicle thermal management control system that balances energy efficiency optimization and thermal safety reliability through a four-level progressive architecture: prediction-driven prior estimation, measured data posterior correction, dynamic confidence fusion, and multi-level safety verification. Compared to traditional independent control strategies, this embodiment first utilizes a multi-heat source coupling model to achieve system-level temperature state prediction, overcoming the limitations of local optima in single-point control; then, it focuses on measured data from key heat sources to ensure rapid response to extreme risks; next, it achieves a dynamic balance between model accuracy and data reliability through adaptive confidence fusion; finally, it outputs a globally equivalent temperature after safety verification, using a unified target temperature to coordinate the collaborative actions of various actuators, effectively reducing redundant heat exchange.

[0112] In some embodiments, the process for determining the first confidence level a(k) is as follows: S501. Determine the measurement noise matrix R at the current moment based on the temperature measurement noise of each heat source, and determine the first error matrix from the measurement noise matrix R at the current moment. ; This step is essentially the error analysis stage of the Weighted Least Squares (WLS) method. First error matrix. It is a theoretical measure of the accuracy of WLS estimation, used for subsequent covariance calculations with Kalman filtering. Perform a fusion comparison.

[0113] The estimated solution obtained by the weighted least squares method in step S203 above is: Among them, the weight matrix ; The covariance matrix of the estimation error in this process is: ; The first error matrix in this embodiment is: .

[0114] This step transforms the accuracy characteristics of each heat source sensor into state-space information metrics through the synergistic effect of the measurement noise matrix and the observation matrix, enabling high-precision sensors to play a greater role in global temperature estimation. At the same time, it provides a quantifiable accuracy comparison benchmark for subsequent information fusion with extended Kalman filtering.

[0115] S502. Based on the predicted temperature of each heat source at the previous moment and the actual temperature of each heat source at the previous moment, determine the error covariance matrix of the previous moment; based on the error covariance matrix of the previous moment... And the gain matrix at the current moment is determined by the measurement noise matrix R, which is determined by the temperature measurement noise of each heat source. The error covariance matrix of the previous time step is updated based on the gain matrix at the current time step to obtain the error covariance matrix at the current time step. The second error matrix is ​​determined by the error covariance matrix at the current time.

[0116] This step is essentially the error analysis stage of the Extended Kalman Filter (EKF). The complete Kalman filtering process is as follows: S601, Based on the predicted temperature of each heat source at the previous moment. and the actual temperature of each heat source at the previous moment. Determine the error covariance matrix of the previous time step. Example formula is as follows: ; The posterior error covariance matrix of the previous time step represents the degree of uncertainty of the algorithm's temperature estimation of the previous time step. ( ): Expectation operator, indicating that the statistical average of a random variable is taken. : The state vector of the true temperature (or actual temperature) at the previous moment, which is the theoretical true value; The posterior optimal estimate of the previous time step, i.e., the optimal temperature estimate obtained based on all information from time step k-1 and before. Matrix transpose operator.

[0117] In this step, the uncertainty of the estimate is quantified by calculating the expected square of the difference between the actual temperature and the predicted temperature at the previous moment. This error covariance reflects the algorithm's confidence level in the temperature estimate at the previous moment. A smaller covariance indicates a more accurate estimate, while a larger covariance indicates a larger bias in the model or measurement, providing an uncertainty benchmark for subsequent dynamic adjustment of the gain coefficient.

[0118] S602. Based on the error covariance matrix from the previous time step and the measurement noise matrix determined by the temperature measurement noise of each heat source, determine the gain matrix for the current time step. An example formula is as follows: ; in, The Kalman gain matrix at the current moment represents the ratio of confidence in the measured data to the predicted data. The prior error covariance matrix at the current time step is obtained from the posterior covariance matrix at the previous time step through state transition. ,in, : Process noise covariance matrix, characterizing model uncertainty; : Measurement noise matrix.

[0119] In this step, the optimal fusion weights for the current moment are calculated by comprehensively considering the relative magnitudes of prediction uncertainty and measurement noise. When the prediction error is large but the measurement is relatively reliable, the gain coefficient is automatically increased to prioritize the adoption of measured data; when the prediction accuracy is high but the sensor noise is large, the gain coefficient is decreased to rely on model deduction, thereby achieving adaptive trust allocation between prediction and observation.

[0120] S603. Based on the gain matrix at the current moment, the measured temperature of each heat source at the current moment, and the predicted temperature at the previous moment, determine the predicted temperature of each heat source at the current moment. An example formula is as follows: ; The best posterior estimate at the current moment, that is, the best estimate of the current temperature after correction by measured data, or the predicted temperature of each heat source; The prior estimate at the current moment is obtained by predicting the state equation: ; The Kalman gain matrix at the current moment is determined by S502; : The measured temperature vector at the current moment; The predicted output is the prior estimate mapped to the measurement space through the observation matrix; Innovation refers to the residual between the measured value and the predicted output.

[0121] In this step, the deviation between the model prediction and the sensor measurement is weighted and corrected using a gain coefficient to generate the optimal temperature estimate for the current moment. This estimate preserves the physical continuity of the model deduction while incorporating real-time information from the measured data, effectively suppressing noise interference from a single source, and making the predicted temperature both smooth and track-oriented.

[0122] S604. Update the error covariance matrix of the previous time step based on the gain matrix at the current time step to obtain the error covariance matrix at the current time step. An example formula is as follows: ; in, = The posterior error covariance matrix at the current time. : Identity matrix; The Kalman gain matrix at the current moment is determined by S602; Observation matrix; The prior error covariance matrix at the current time step is obtained from the posterior covariance matrix at the previous time step through state transition. ,in, : Process noise covariance matrix, characterizing model uncertainty.

[0123] The essence of this update process is to use Kalman gain to correct prior uncertainty: when the gain is large (high reliability of measured data), the posterior covariance decreases significantly, indicating improved estimation accuracy; when the gain is small (more reliable model prediction), the posterior covariance is close to the prior value, maintaining the smoothness of model inference. Finally, this posterior error covariance is extracted. The inverse matrix is ​​used as the second error matrix. The first error matrix used in subsequent weighted least squares methods By comparing the amount of information, the confidence weights of the two estimation methods in global fusion are dynamically determined.

[0124] S503, Based on the first error matrix and the second error matrix Determine the first confidence level The first confidence level increases with the increase of the first error matrix and decreases with the increase of the second error matrix.

[0125]

[0126] The confidence determination mechanism in this embodiment achieves adaptive optimal fusion of two heterogeneous estimation methods. Specifically, the first error matrix represents the information content of the weighted least squares estimation; a larger value indicates higher accuracy and lower uncertainty in fitting the measured data, in which case the first confidence should be increased to prioritize the data-driven fitting result. The second error matrix represents the information content of the extended Kalman filter estimation; a larger value indicates higher accuracy after fusing the model deduction and the measured data, in which case the first confidence should be decreased to prioritize the model-driven prediction result. This dynamic weight allocation based on information content comparison enables the algorithm to automatically select a more reliable estimation source according to real-time operating conditions: increasing the data fitting weight when the sensor is stable and the measurement noise is low, and increasing the model prediction weight when the operating conditions change drastically and the model parameters are accurate, thereby ensuring the optimal accuracy and robustness of the global equivalent temperature estimation across the entire operating range.

[0127] In some embodiments, the equivalent temperature of the process in S404 is... Perform security checks and obtain the global equivalent temperature. ,include: Based on the measured temperature of the battery Measured temperature of motor controller and preset safety margin The corrected measured battery temperature was determined respectively. and the corrected measured temperature of the motor controller The equivalent temperature of the process Corrected battery predicted temperature and the corrected measured temperature of the motor controller The maximum value in the value is determined as the equivalent temperature of the verification process. Determine the predicted future battery temperature after a preset time period. Based on the predicted future temperature of the battery The equivalent temperature of the verification process is corrected, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature. .

[0128] Wherein, the prediction of the future temperature of the battery Equivalent temperature of the process The correction is performed, and the corrected process equivalent temperature is determined as the global equivalent temperature. This includes: responding to determining the predicted future temperature of the battery. The upper limit of battery safety temperature (which can be the difference between the upper limit of battery temperature and the safety margin, i.e., expressed as...) If the temperature difference during verification is greater than a preset value (0 in this example), then the equivalent temperature of the verification process is adjusted based on the temperature difference. The correction is performed, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature. In response to determining that the verification temperature difference is less than or equal to a preset value (0 in this example), the equivalent temperature of the verification process is corrected based on the preset value, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature. .

[0129] The above process can be represented by the following formula: ; ; in, These are the measured temperatures of the battery and the motor controller, respectively. These are all preset parameters, representing the safety margin of battery temperature (3-5℃ in the example), the safety margin of motor controller temperature (5-10℃ in the example), and the safety margin of the upper limit of battery temperature (2-4℃ in the example). : Preset parameters (example is 0.5-0.8), empirical compensation values ​​are used for system offset correction; The future temperature prediction of a battery can be derived using Kalman filtering, and is used to predict the battery's future temperature. The temperature at any given moment, for example, 5-10, corresponding to 5-10 seconds; : Preset parameter, upper limit of battery temperature, example is 55℃.

[0130] It should be noted that in this embodiment, only the predicted future battery temperature is used to correct the equivalent temperature during the calibration process. This is mainly based on the following engineering considerations: As the core component and energy carrier of the vehicle's thermal management system, the thermal safety of the power battery directly determines the vehicle's operational safety and service life. Furthermore, the battery has high thermal inertia and a significant temperature response lag, making overheating risks cumulative and irreversible. In contrast, while the motor controller is equally critical, it has a fast thermal response, good heat dissipation, and short-term overload capability; temperature exceeding limits can be quickly mitigated by reducing power. Therefore, by predicting the battery's temperature evolution trend after a preset time, and proactively compensating for the current equivalent temperature, preventative control can be initiated before the battery temperature reaches the safety boundary, thus mitigating thermal risks in advance.

[0131] In this embodiment, the corrected measured temperature of the key heat source is first used as the safety lower limit to ensure that the equivalent temperature does not fall below the safety baseline of any core component. Then, a forward-looking assessment is conducted using the predicted future temperature of the battery. When it is predicted that the battery will approach the safety upper limit, the equivalent temperature is proactively increased, driving the cooling system to enhance its heat dissipation capacity in advance. This mechanism overcomes the lag limitation of traditional control strategies that only respond to the current temperature, shifting the time window for thermal management decisions from post-event remediation to pre-event prevention. This significantly reduces the risk of battery thermal runaway while avoiding short-sighted behavior that sacrifices long-term safety for short-term energy efficiency, achieving a dynamic balance between thermal safety and energy efficiency.

[0132] In some embodiments, the determination of the predicted future battery temperature after a preset time period ,include: S701. Based on the predicted temperature of the battery at the previous moment and the actual temperature of the battery at the previous moment, determine the error covariance at the previous moment. Steps and S Similarly, specifically, extracting S The battery temperature data is sufficient and will not be elaborated further.

[0133] S702. Based on the error covariance and battery temperature measurement noise of the previous moment, determine the gain coefficient at the current moment. Steps and S Similarly, specifically, extracting S The battery temperature data is sufficient and will not be elaborated further.

[0134] S703. Based on the gain coefficient at the current moment, the measured temperature of the battery at the current moment, and the predicted temperature at the previous moment, determine the predicted temperature of the battery at the current moment. Steps and S Similarly, specifically, extracting S The battery temperature data is sufficient and will not be elaborated further.

[0135] S704. Based on the current predicted battery temperature and the multi-heat source temperature coupling state model, determine the predicted future battery temperature after a preset time period. Example formula is as follows:

[0136]

[0137] in, Prediction based on information at time k The temperature state at time j is the recursive predicted value at step j. Prediction based on information at time k Temperature status at any time; The state transition matrix is ​​obtained by discretizing the heat capacity matrix and the thermal conductivity matrix. ; Input matrix, ; : The input vector at any given time includes cooling flow rate, heat generation power, ambient temperature, etc. : Preset prediction steps, corresponding to a preset duration (e.g., =10, sampling period =1 s, then the prediction duration is 10 seconds). The predicted future battery temperature after a preset duration, i.e., the battery temperature component. Step-by-step predicted value; Extract the battery temperature row vector and extract the battery temperature component from the state vector.

[0138] In this step, the aforementioned prediction-update process is iteratively executed. Using the optimal estimate at the current moment as the initial value, the state transition model (i.e., the multi-heat source temperature coupling state model) is continuously recursively pushed forward for a preset time to obtain the battery's temperature evolution trend at future moments. This cyclical mechanism enables the algorithm to continuously track the dynamic changes in the battery's thermal state and perform rolling optimization based on the latest information, ensuring that the future predicted temperature always reflects the most realistic temperature state of the current system.

[0139] This embodiment constructs a rolling optimization mechanism for battery temperature estimation-correction-prediction through a cyclic recursive architecture of extended Kalman filtering, achieving full-time-domain thermal management decision support from the current state to future trends. This mechanism first quantifies the estimation uncertainty using error covariance, enabling the algorithm to have self-awareness of its own confidence level. Then, through dynamic adjustment of the gain coefficient, it achieves optimal fusion of model prediction and measured feedback, avoiding the cumulative drift of pure model inference and overcoming the noise sensitivity of purely data-driven approaches. Finally, it generates the predicted future temperature through multiple recursive steps. Compared to traditional control strategies based solely on the current temperature, this embodiment can identify the evolution trend of battery thermal risks in advance, triggering preventative regulation before the temperature reaches the safety boundary. This significantly improves the active safety capabilities and dynamic adaptability of the vehicle's thermal management system, while reducing excessive cooling energy consumption to cope with sudden overheating, achieving synergistic optimization of thermal safety and energy efficiency.

[0140] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0141] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0142] Based on the same inventive concept, corresponding to any of the above embodiments, this application also provides a thermal management control device.

[0143] refer to Figure 2 The thermal management control method includes: The acquisition module 201 is configured to acquire measured state parameters of multiple heat sources of the vehicle; the measured state parameters include the measured temperature of each heat source and the temperature measurement noise. The determination module 202 is configured to determine the predicted temperature of each heat source based on the measured temperature, temperature measurement noise and model estimated temperature, and to determine the temperature weighting coefficient corresponding to each heat source based on the measured state parameters. The execution module 203 is configured to determine the global equivalent temperature based on the measured state parameters, predicted temperature and temperature weighting coefficient of the heat source, so as to perform thermal management control based on the global equivalent temperature. The model estimates the temperature based on a pre-constructed multi-heat-source temperature coupling state model, which is constructed based on the heat generation power of each heat source, the cooling power acting on each heat source, and the pre-stored heat capacity and thermal conductivity parameters of each heat source.

[0144] Furthermore, the execution module 203 is also configured to: Based on the temperature weighting coefficient of each heat source and the predicted temperature of each heat source, the first equivalent temperature of each heat source is determined. At least one target heat source is determined among the multiple heat sources, and a target weight coefficient corresponding to each target heat source is determined based on the measured state parameters of each target heat source, so as to determine the second equivalent temperature of each target heat source based on the measured temperature of each target heat source and the corresponding target weight coefficient. Determine a first confidence level corresponding to the first equivalent temperature and a second confidence level corresponding to the second equivalent temperature, and determine the process equivalent temperature based on the first equivalent temperature, the first confidence level, the second equivalent temperature and the second confidence level; The equivalent temperature of the process is verified for safety, and the global equivalent temperature is obtained.

[0145] Furthermore, the target heat source is a motor controller and a battery, and the measured state parameters corresponding to the motor controller also include the motor output torque; Furthermore, the execution module 203 is also configured to: The target heat source temperature difference is determined based on the measured temperature of the motor controller and the measured temperature of the battery. Call the pre-stored weighted sensitivity coefficient corresponding to the temperature difference of the target heat source, and the pre-stored load weight coefficient corresponding to the output torque of the motor; Based on the target heat source temperature difference, weighted sensitivity coefficient, motor output torque, and load weight coefficient, the target weight coefficients corresponding to the motor controller and battery are determined respectively. The target heat source temperature difference and the motor output torque are both positively correlated with the target weight coefficient corresponding to the battery.

[0146] Furthermore, the execution module 203 is also configured to: Based on the measured temperature of the battery, the measured temperature of the motor controller, and the preset safety margin, the corrected measured temperature of the battery and the corrected measured temperature of the motor controller are determined respectively. The maximum value among the process equivalent temperature, the corrected predicted battery temperature, and the corrected measured motor controller temperature is determined as the verification process equivalent temperature. Determine the predicted future temperature of the battery after a preset time period; The equivalent temperature of the verification process is corrected based on the predicted future temperature of the battery, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature.

[0147] Furthermore, the execution module 203 is also configured to: In response to the determination that the difference between the predicted future temperature of the battery and the upper limit of the battery safe temperature is greater than a preset value, the equivalent temperature of the verification process is corrected based on the verification temperature difference, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature. In response to determining that the temperature difference of the verification process is less than or equal to a preset value, the equivalent temperature of the verification process is corrected based on the preset value, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature.

[0148] Furthermore, the execution module 203 is also configured to: Based on the predicted temperature of the battery at the previous moment and the actual temperature of the battery at the previous moment, the error covariance at the previous moment is determined. Based on the error covariance and battery temperature measurement noise from the previous moment, determine the gain coefficient for the current moment. Based on the gain coefficient at the current moment, the measured temperature of the battery at the current moment, and the predicted temperature at the previous moment, the predicted temperature of the battery at the current moment is determined. Based on the current battery predicted temperature and the multi-heat source temperature coupling state model, the future predicted battery temperature after a preset time period is determined.

[0149] Furthermore, the determining module 202 is also configured to: Construct an observation function to reflect the relationship between the measured temperature and the model-estimated temperature of each heat source; A measurement noise matrix is ​​determined based on the temperature measurement noise of each heat source, and a weighted residual cost function is constructed based on the observation function and the weight matrix using the inverse matrix of the measurement noise matrix as the weight matrix. The predicted temperature of each heat source is obtained by minimizing the weighted residual cost function.

[0150] Furthermore, the execution module 203 is also configured to: The measurement noise matrix at the current moment is determined based on the temperature measurement noise of each heat source at the current moment, and the first error matrix is ​​determined based on the measurement noise matrix at the current moment. Based on the predicted temperature and actual temperature of each heat source at the previous moment, the error covariance matrix at the previous moment is determined. Based on the error covariance matrix at the previous moment and the measurement noise matrix determined by the temperature measurement noise of each heat source, the gain matrix at the current moment is determined. Based on the gain matrix, the error covariance matrix at the previous moment is updated to obtain the error covariance matrix at the current moment, so as to determine the second error matrix through the error covariance matrix at the current moment. The first confidence level is determined based on the first error matrix and the second error matrix; The first confidence level increases with the increase of the first error matrix and decreases with the increase of the second error matrix.

[0151] Furthermore, the measured state parameters also include: water pump speed, pedal opening value, and motor output torque; the determining module 202 is also configured to: For each of the aforementioned heat sources, perform the following: The basic weights are determined based on the temperature measurement noise of the heat source; The heat exchange dynamic factor is determined based on the real-time thermal resistance from the heat source to the coolant, the water pump speed, and the pre-stored coolant flow rate influence coefficient. Dynamic operating condition factors are determined based on pedal opening value, motor output torque, pre-stored pedal dynamic influence coefficient, and pre-stored torque influence coefficient. The safety boundary factor is determined based on the measured temperature of the heat source, the pre-stored safety enhancement coefficient, the pre-stored upper limit of the safe temperature, and the pre-stored safety margin. The temperature weight coefficient corresponding to the heat source is determined based on the basic weight, heat exchange dynamic factor, dynamic operating condition factor, and safety boundary factor.

[0152] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle thermal management control method described in any of the above embodiments.

[0153] Figure 3 This embodiment illustrates a more specific hardware structure of an electronic device. The device may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0154] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0155] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0156] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0157] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WFI, Bluetooth, etc.).

[0158] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0159] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0160] The electronic devices described above are used to implement the corresponding vehicle thermal management control methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0161] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a vehicle, including an electronic device, which is used to execute the vehicle thermal management control method described in any of the above embodiments.

[0162] The vehicles described in the above embodiments are used to implement the corresponding vehicle thermal management control methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0163] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the vehicle thermal management control method as described in any of the above embodiments.

[0164] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0165] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the vehicle thermal management control method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0166] It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0167] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0168] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0169] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0170] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0171] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0172] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0173] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A thermal management control method, characterized in that, include: The measured state parameters of multiple heat sources in the vehicle are obtained; the measured state parameters include the measured temperature and temperature measurement noise of each heat source. The predicted temperature of each heat source is determined based on the measured temperature, temperature measurement noise, and model estimated temperature, and the temperature weighting coefficient corresponding to each heat source is determined based on the measured state parameters. Based on the measured state parameters, predicted temperature, and temperature weighting coefficient of the heat source, the global equivalent temperature is determined, and thermal management control is performed based on the global equivalent temperature. The model estimates the temperature based on a pre-constructed multi-heat-source temperature coupling state model, which is constructed based on the heat generation power of each heat source, the cooling power acting on each heat source, and the pre-stored heat capacity and thermal conductivity parameters of each heat source.

2. The method according to claim 1, characterized in that, The determination of the global equivalent temperature based on the measured state parameters, predicted temperature, and temperature weighting coefficient of the heat source includes: Based on the temperature weighting coefficient of each heat source and the predicted temperature of each heat source, the first equivalent temperature of each heat source is determined. At least one target heat source is determined among the multiple heat sources, and a target weight coefficient corresponding to each target heat source is determined based on the measured state parameters of each target heat source, so as to determine the second equivalent temperature of each target heat source based on the measured temperature of each target heat source and the corresponding target weight coefficient. Determine a first confidence level corresponding to the first equivalent temperature and a second confidence level corresponding to the second equivalent temperature, and determine the process equivalent temperature based on the first equivalent temperature, the first confidence level, the second equivalent temperature and the second confidence level; The equivalent temperature of the process is verified for safety, and the global equivalent temperature is obtained.

3. The method according to claim 2, characterized in that, The target heat source is a motor controller and a battery, and the measured state parameters corresponding to the motor controller also include the motor output torque; The determination of the target weight coefficient corresponding to each target heat source based on the measured state parameters of each target heat source includes: The target heat source temperature difference is determined based on the measured temperature of the motor controller and the measured temperature of the battery. Call the pre-stored weighted sensitivity coefficient corresponding to the temperature difference of the target heat source, and the pre-stored load weight coefficient corresponding to the output torque of the motor; Based on the target heat source temperature difference, weighted sensitivity coefficient, motor output torque, and load weight coefficient, the target weight coefficients corresponding to the motor controller and battery are determined respectively. The target heat source temperature difference and the motor output torque are both positively correlated with the target weight coefficient corresponding to the battery.

4. The method according to claim 3, characterized in that, The process equivalent temperature is verified for safety, and a global equivalent temperature is obtained, including: Based on the measured temperature of the battery, the measured temperature of the motor controller, and the preset safety margin, the corrected measured temperature of the battery and the corrected measured temperature of the motor controller are determined respectively. The maximum value among the process equivalent temperature, the corrected predicted battery temperature, and the corrected measured motor controller temperature is determined as the verification process equivalent temperature. Determine the predicted future temperature of the battery after a preset time period; The equivalent temperature of the verification process is corrected based on the predicted future temperature of the battery, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature.

5. The method according to claim 4, characterized in that, The step of correcting the process equivalent temperature based on the predicted future temperature of the battery, and determining the corrected process equivalent temperature as the global equivalent temperature, includes: In response to the determination that the difference between the predicted future temperature of the battery and the upper limit of the battery safe temperature is greater than a preset value, the equivalent temperature of the verification process is corrected based on the verification temperature difference, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature. In response to determining that the temperature difference of the verification process is less than or equal to a preset value, the equivalent temperature of the verification process is corrected based on the preset value, and the corrected equivalent temperature of the verification process is determined as the global equivalent temperature.

6. The method according to claim 4, characterized in that, Determining the predicted future temperature of the battery after a preset time period includes: Based on the predicted temperature of the battery at the previous moment and the actual temperature of the battery at the previous moment, the error covariance at the previous moment is determined. Based on the error covariance and battery temperature measurement noise from the previous moment, determine the gain coefficient for the current moment. Based on the gain coefficient at the current moment, the measured temperature of the battery at the current moment, and the predicted temperature at the previous moment, the predicted temperature of the battery at the current moment is determined. Based on the current battery predicted temperature and the multi-heat source temperature coupling state model, the future predicted battery temperature after a preset time period is determined.

7. The method according to claim 1, characterized in that, The determination of the predicted temperature of each heat source based on the measured temperature, temperature measurement noise, and the pre-constructed multi-heat source temperature coupling state model includes: Construct an observation function to reflect the relationship between the measured temperature and the model-estimated temperature of each heat source; A measurement noise matrix is ​​determined based on the temperature measurement noise of each heat source, and a weighted residual cost function is constructed based on the observation function and the weight matrix using the inverse matrix of the measurement noise matrix as the weight matrix. The predicted temperature of each heat source is obtained by minimizing the weighted residual cost function.

8. The method according to claim 2, characterized in that, Determining the first confidence level corresponding to the first equivalent temperature includes: The measurement noise matrix at the current moment is determined based on the temperature measurement noise of each heat source at the current moment, and the first error matrix is ​​determined based on the measurement noise matrix at the current moment. Based on the predicted temperature and actual temperature of each heat source at the previous moment, the error covariance matrix at the previous moment is determined; based on the error covariance matrix at the previous moment and the measurement noise matrix determined by the temperature measurement noise of each heat source, the gain matrix at the current moment is determined. The error covariance matrix of the previous time step is updated based on the gain matrix to obtain the error covariance matrix of the current time step, so as to determine the second error matrix through the error covariance matrix of the current time step. The first confidence level is determined based on the first error matrix and the second error matrix; The first confidence level increases with the increase of the first error matrix and decreases with the increase of the second error matrix.

9. The method according to claim 1, characterized in that, The measured state parameters also include: water pump speed, pedal opening value, and motor output torque; the determination of temperature weighting coefficients corresponding to each heat source based on the measured state parameters includes: For each of the aforementioned heat sources, perform the following: The basic weights are determined based on the temperature measurement noise of the heat source; The heat exchange dynamic factor is determined based on the real-time thermal resistance from the heat source to the coolant, the water pump speed, and the pre-stored coolant flow rate influence coefficient. Dynamic operating condition factors are determined based on pedal opening value, motor output torque, pre-stored pedal dynamic influence coefficient, and pre-stored torque influence coefficient. The safety boundary factor is determined based on the measured temperature of the heat source, the pre-stored safety enhancement coefficient, the pre-stored upper limit of the safe temperature, and the pre-stored safety margin. The temperature weight coefficient corresponding to the heat source is determined based on the basic weight, heat exchange dynamic factor, dynamic operating condition factor, and safety boundary factor.

10. A vehicle, characterized in that, The device includes an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor, when executing the program, implements the method as claimed in any one of claims 1 to 9.

Images