Battery low temperature pulse heating control device and equipment with risk perception of warm-up fusion

Abstract

This invention belongs to the field of lithium-ion batteries. To address the complexity of existing lineage data analysis processes, it provides a battery low-temperature pulse heating control device and equipment with temperature rise fusion risk perception. The device includes: a time-series input matrix construction module; an embedded sequence vector formation module; a sequence feature extraction module; a time-series converter model training module; and a transient pulse current control module. The transient pulse current control module calculates the first and second difference terms of the predicted internal temperature sequence based on the real-time predicted internal temperature sequence values ​​from the trained time-series converter model. Based on this, it constructs a temperature rise trend regulation factor, combines it with the temperature rise momentum term and risk perception indicators, to achieve multi-factor correction control of the transient pulse current.

Description

Technical Field

[0001] This invention belongs to the field of lithium-ion batteries, and particularly relates to a battery low-temperature pulse heating control device and equipment for temperature rise fusion risk perception. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Under low-temperature conditions, the internal temperature of a battery rises slowly and unevenly. Inaccurate heating control can easily lead to temperature overshoot, localized overheating, or structural stress concentration, potentially causing thermal damage or even safety accidents, affecting battery life and system stability. Traditional low-temperature heating control often relies on preset heating strategies or simple temperature feedback adjustments, making it difficult to accurately capture the dynamic temperature rise within the battery and the thermal shock effect caused by short-term pulse currents. Some methods trigger heating based on static temperature thresholds, ignoring the time-varying characteristics of temperature rise trends and structural stress risks, resulting in delayed or excessive heating control response and an inability to achieve precise regulation.

[0004] In addition, existing heating control schemes lack a joint sensing mechanism for the internal temperature change trend and structural risks of the battery, making it difficult to effectively identify thermal shock risks and dynamically adjust heating pulse parameters, thus limiting the level of intelligence in battery thermal safety management under low-temperature environments. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a battery low-temperature pulse heating control device and equipment with integrated temperature rise risk perception. It can combine limited external measurement data to predict internal temperature dynamics in real time, perceive structural risks, and intelligently adjust the heating pulse output to achieve precise control and effective protection of the battery thermal state under low-temperature conditions, thereby improving the safety and reliability of the battery system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of the present invention provides a battery low-temperature pulse heating control device with temperature rise fusion risk perception.

[0007] In one or more embodiments, a battery low-temperature pulse heating control device with temperature rise fusion risk perception is provided, comprising: The timing input matrix construction module is used to construct a timing input matrix of the battery operating state based on multi-dimensional battery sensing data. The embedded sequence vector forming module is used to perform high-dimensional linear mapping on the time-series input matrix of the battery operating state and inject time position information to form an embedded sequence vector. The sequence feature extraction module is used to introduce a temperature-stress coupling gating mechanism to gating and modulate the embedded vector to obtain a gated and modulated embedded sequence vector. Then, a multi-layer stacked converter encoder is used to perform deep temporal feature extraction and nonlinear enhancement on the gated and modulated embedded sequence vector to obtain sequence features. The time series converter model training module is used to aggregate sequence features into a single feature vector, then map the aggregated feature vector to the internal temperature prediction scalar, and construct a weighted loss function that integrates smoothing and constraint terms to train the time series converter model. The transient pulse current control module is used to calculate the first and second difference terms of the predicted internal temperature sequence of the battery based on the real-time predicted internal temperature sequence value of the trained time-series converter model. Based on this, a temperature rise trend regulation factor is constructed, which is combined with the temperature rise momentum term and risk perception index to achieve multi-factor correction control of transient pulse current.

[0008] As one implementation method, the expression for multi-factor correction control of transient pulse current is:

[0009] in, The output value is the pulse current value executed by outputting the pulse current. This is the standard current reference value; and These are the first and second difference terms of the internal temperature prediction sequence; The real-time predicted internal temperature sequence of the battery by the time-series converter model; It is an inhibitory factor; As a risk adjustment factor; , Construct adjustment factors for absolute values.

[0010] As one implementation method, the expression for the risk adjustment factor is:

[0011]

[0012] in, It is the suppression magnitude hyperparameter for controlling risk perception; m is the number of external temperature monitoring points, and n is the number of stress monitoring points; Indicates the first Temperature at one external monitoring point; Indicates the first The rate of temperature change over time at each external monitoring point Indicates the first The rate of change of stress over time at each stress monitoring point It is the local structural stress state obtained through monitoring; It is the hyperbolic tangent function.

[0013] As one implementation method, the expression for the inhibition factor is:

[0014]

[0015] in, The momentum smoothing factor, This is a hyperparameter sensitive to temperature rise. and This represents the momentum term.

[0016] As one implementation method, a weighted loss function that integrates smoothing and constraint terms is constructed. for:

[0017]

[0018]

[0019]

[0020] in, As weight; This is the weighted mean square error loss function with weight adjustment; For time smoothing constraints; To add penalties; , , For loss weights; and This is a predicted temperature value; This represents the maximum temperature. It is a function for maximizing the value; This represents the total number of monitoring points.

[0021] As one implementation method, the embedded sequence vector is set as : ; This is a position-encoded vector; The vector is the result of a high-dimensional linear mapping of the state vector at each time step of the input matrix; L is the total number of time steps.

[0022] As one implementation method, a stress-temperature coupled sensing gating mechanism is introduced to generate a gated modulation vector that is consistent with the embedding dimension of the converter input; this gated modulation vector is then applied to the embedding sequence vector to obtain the gated modulation embedding sequence vector.

[0023] A second aspect of the present invention provides a battery low-temperature pulse heating control method based on temperature rise fusion risk perception.

[0024] In one or more embodiments, a battery low-temperature pulse heating control method based on temperature rise fusion risk perception includes: Based on multi-dimensional battery sensing data, a time-series input matrix of battery operating status is constructed; The time-series input matrix of the battery operating status is subjected to high-dimensional linear mapping and injected with time position information to form an embedded sequence vector; A temperature-stress coupling gating mechanism is introduced to perform gating modulation on the embedded vector to obtain a gated modulated embedded sequence vector. Then, a multi-layer stacked converter encoder is used to perform deep temporal feature extraction and nonlinear enhancement on the gated modulated embedded sequence vector to obtain sequence features. The sequence features are aggregated into a single feature vector, and then the aggregated feature vector is mapped to the internal temperature prediction scalar. Based on this, a weighted loss function that integrates smoothing and constraint terms is constructed to train the time series transformer model. Based on the real-time predicted battery internal temperature sequence value based on the trained time-series converter model, the first-order difference term and the second-order difference term of the predicted internal temperature value sequence are calculated, and a temperature rise trend regulation factor is constructed accordingly. Combined with the temperature rise momentum term and risk perception index, a multi-factor correction control for transient pulse current is achieved.

[0025] A third aspect of the present invention provides a computer storage medium.

[0026] A computer storage medium storing a computer program that, when executed by a processor, implements the steps in the battery low-temperature pulse heating control method for temperature rise fusion risk perception as described above.

[0027] A fourth aspect of the present invention provides an electronic device.

[0028] An electronic device includes 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 steps of the battery low-temperature pulse heating control method for temperature rise fusion risk perception as described above.

[0029] Compared with the prior art, the beneficial effects of the present invention are: (1) This invention proposes for the first time a battery low-temperature pulse heating control device based on temperature rise trend prediction and risk perception. It integrates time-series temperature rise change characteristics and multi-source structural response information to construct a multi-factor regulation mechanism, which can achieve dynamic adaptive adjustment of pulse heating behavior under limited monitoring conditions. This method effectively identifies potential thermal shock risks by modeling the first-order and second-order differences of the battery internal temperature prediction sequence, introducing temperature rise momentum and structural risk indicators, and correcting the control output accordingly, thus achieving the goal of safe and rapid battery heating in low-temperature environments.

[0030] (2) Based on the dynamic characteristics of the internal temperature prediction sequence, this invention constructs a multi-level temperature rise trend adjustment factor, which realizes precise control of the pulse heating amplitude and frequency, and effectively suppresses the thermal shock problem caused by excessive transient temperature rise rate. This invention integrates external temperature and stress sensing data and proposes a structural response-driven risk control mechanism, so that the heating behavior can adapt to the dynamic changes of the battery structure state, thereby enhancing the safety robustness of the system under complex working conditions.

[0031] (3) The present invention introduces temperature rise momentum modeling into pulse control strategy design, which significantly improves the heating system’s ability to perceive and respond to short-term cumulative thermal effects and avoids the risk of stress change during heating. The control method proposed in this invention can be deployed in conjunction with the internal temperature estimation model without adding extra hardware resources. It has good embedded implementation capabilities and is suitable for online operation and control in various battery application scenarios. Attached Figure Description

[0032] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0033] Figure 1 This is a schematic diagram of the battery low-temperature pulse heating control device with temperature rise fusion risk perception according to an embodiment of the present invention; Figure 2 This is a schematic flowchart of the battery low-temperature pulse heating control method based on temperature rise fusion risk perception according to an embodiment of the present invention. Figure 3 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0034] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0035] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0036] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0037] Figure 1 This is a schematic diagram of a battery low-temperature pulse heating control device with temperature rise fusion risk perception, as described in an embodiment of the present invention. Figure 1 As shown, the battery low-temperature pulse heating control device 100 with temperature rise fusion risk perception in this embodiment may include the following: The timing input matrix construction module 101 is used to construct a timing input matrix of the battery operating state based on multi-dimensional battery sensing data. The embedded sequence vector forming module 102 is used to perform high-dimensional linear mapping on the time-series input matrix of the battery operating state and inject time position information to form an embedded sequence vector. The sequence feature extraction module 103 is used to introduce a temperature-stress coupling gating mechanism to perform gating modulation on the embedded vector to obtain a gated modulated embedded sequence vector. Then, a multi-layer stacked converter encoder is used to perform deep temporal feature extraction and nonlinear enhancement on the gated modulated embedded sequence vector to obtain sequence features. The time series converter model training module 104 is used to aggregate sequence features into a single feature vector, then map the aggregated feature vector to an internal temperature prediction scalar, and construct a weighted loss function that integrates smoothing and constraint terms to train the time series converter model. The transient pulse current control module 105 is used to calculate the first-order and second-order difference terms of the predicted internal temperature sequence of the battery based on the real-time predicted internal temperature sequence value of the trained time-series converter model, thereby constructing a temperature rise trend regulation factor, and combining the temperature rise momentum term and risk perception index to achieve multi-factor correction control of transient pulse current.

[0038] Figure 2 This is a schematic diagram of a battery low-temperature pulse heating control method based on temperature rise fusion risk perception, as described in an embodiment of the present invention. Figure 2 As shown, the battery low-temperature pulse heating control method with temperature rise fusion risk perception includes steps S101 to S105.

[0039] It should be noted here that, Figure 1 The various modules and Figure 2The steps in the battery low-temperature pulse heating control method with temperature rise fusion risk perception correspond one-to-one, and their specific implementation processes are the same, so they will not be described in detail here.

[0040] The specific implementation process of steps S101 to S105 is as follows: Step S101: Construct a time-series input matrix of battery operating status based on multi-dimensional battery sensing data.

[0041] For example, sensor data such as temperature, stress, and electrical signals can be fused to construct a time-series input matrix for battery operating status. The fused data includes, but is not limited to, parameters such as current, voltage, stress, strain, battery surface temperature, battery internal temperature, and data acquisition time.

[0042] The specific steps of step S101 are as follows: Step S1011: Obtain the surface temperature distribution of the battery by using m temperature sensors and n stress sensors deployed on the surface of the battery casing. and stress distribution Time-series data. Simultaneously, an internal battery temperature sensor is deployed to acquire the battery's internal temperature data at time t. .

[0043]

[0044]

[0045] The sampling frequency of the sensor can be selected according to the control accuracy and response requirements, and its layout can be appropriately optimized according to the specific battery casing shape.

[0046] Step S1012: Synchronously record battery voltage Current Ambient temperature With data collection timestamp.

[0047] Step S1013: Align the collected data with timestamps to construct a d-dimensional vector (d=m+n+3) representing the current battery operating state. :

[0048] Step S1014: Select a sliding window length of L (L is set according to the sampling period), and construct an input matrix containing L time steps: .

[0049] Step S102: Perform high-dimensional linear mapping on the time-series input matrix of the battery operating state and inject time position information to form an embedded sequence vector.

[0050] In the specific implementation process, step S102 is as follows: Step S1021: Use a linear mapping function to transform the input matrix. The state vector at each time step Perform high-dimensional linear mapping:

[0051] in, The dimension is , and These are trainable parameters. At this point, the sequence input... Turn to :

[0052] Step S1022: To inject time series information and enable the model to distinguish the order of time steps, standard sine and cosine position coding is adopted, and a position coding vector is introduced. :

[0053]

[0054] in, .

[0055] S23: Construct an input encoding representation based on the time-series converter. This vector contains both sensor information and clearly defines the time step sequence position, providing sufficient contextual information for subsequent converter processing. .

[0056] Step S103: Introduce a temperature-stress coupling gating mechanism to perform gating modulation on the embedded vector to obtain a gated modulated embedded sequence vector. Then, use a multi-layer stacked converter encoder to perform deep temporal feature extraction and nonlinear enhancement on the gated modulated embedded sequence vector to obtain sequence features.

[0057] The specific process of step S103 is as follows: Step S1031: To enhance the response capability to transient thermal stress behavior characteristics, extract the temperature-stress sensing sub-vector from the battery operating state vector at the current time t:

[0058] By introducing a stress-temperature coupled sensing gating mechanism, a gated modulation vector with the same embedding dimension as the converter input is generated:

[0059] in, For at any time The extracted temperature-stress sensing sub-vector is derived from the battery surface temperature distribution. Stress distribution and ambient temperature Composition, used to characterize the instantaneous thermo-mechanical coupling state of the battery; It is a trainable gate weight matrix used to learn the contribution relationship of different sensor information in gate modulation; It is a bias vector used to correct the feature benchmark and improve adaptability under different working conditions; For nonlinear activation functions (such as Sigmoid), the modulation factor is compressed to the [0,1] interval, allowing it to be interpreted as gating strengths of different dimensions. The resulting... This is the temperature-stress sensing gated modulation vector.

[0060] The gated modulation vector Acting on the input encoded sequence The gated and modulated embedded sequence is obtained as follows:

[0061] in, This represents element-wise multiplication, where each dimension of the gated vector is multiplied by the corresponding dimension of the input embedding vector, achieving dimension-wise weighted modulation of the input features. Its physical meaning is that in the input embedding sequence, feature dimensions deemed important under the current temperature-stress state are amplified, while those with insignificant effects are weakened, thus achieving adaptive feature selection for environmental perception.

[0062] Step S1032: The gated modulated input sequence is processed using a multi-layer stacked converter encoder. Deep feature extraction is performed. For any layer transformer encoder structure, firstly, through a multi-head self-attention module (h heads), for the ... Layer input Calculate query ,key ,value matrix:

[0063]

[0064]

[0065] in, , and This is the weight matrix.

[0066] Step S1033: Calculate the attention weight matrix, representing the attention weight of each time step in the sequence to all other time steps:

[0067]

[0068] Step S1034: Calculate any single-head output And concatenate the outputs of h heads:

[0069]

[0070] Step S1035: Project the output mapping matrix back into the model's hidden dimension space, after the first step... The layer outputs:

[0071] Step S1036: Concatenate the multi-head self-attention output with the input residual, and obtain the result after layer normalization. :

[0072] Step S1037: ... The input is a feedforward fully connected network, consisting of two fully connected layers, which further abstracts the features:

[0073] To facilitate gradient flow and feature fusion, a residual connection is made between the FFN output and input:

[0074] right After performing layer normalization, the output of the last encoder layer is obtained: .

[0075] Step S104: Aggregate the sequence features into a single feature vector, then map the aggregated feature vector to the internal temperature prediction scalar, and construct a weighted loss function that integrates smoothing and constraint terms to train the time series transformer model.

[0076] Step S104 outputs the internal temperature through feature aggregation and fully connected mapping, constructs a weighted loss function to fuse smoothing and constraint terms, and improves prediction accuracy and physical plausibility. The specific steps are as follows: S41: Employ the time-step feature convergence method to integrate the sequence features output by the encoder. Aggregate into a single vector:

[0077] in, Indicates the first Encoder output features at each time step, The time step length is used. The deep features from all time steps are averaged to obtain a comprehensive feature vector that can represent the current battery operating state as a whole. This avoids over-reliance on data from a single instant and improves the overall ability to grasp temporal behavior.

[0078] S42: A fully connected layer is used to map the aggregated feature vector to the internal temperature prediction scalar output.

[0079] in, This represents the battery's internal temperature predicted by the model. The weight matrix is ​​trainable and determines the contribution of different aggregate features to temperature prediction. This is the bias vector used to correct the prediction baseline. The physical significance of this step is to transform the time-series feature aggregation results into specific internal temperature prediction values, realizing the mapping from "multi-dimensional operating state features" to "single physical quantity output".

[0080] S43: Utilizing a label with actual internal temperature The training data is used for supervised learning. A weighted mean square error loss function with weighted adjustment is constructed, and weights are applied to different prediction error intervals to enhance the sensitivity to extreme temperature deviations.

[0081] Among them, weight , This is the adjustment factor. To prevent unreasonable time jumps in the prediction results, a first-order difference smoothing regularization is introduced to construct a time smoothing constraint term:

[0082] If the predicted temperature is outside the reasonable range Add a penalty item:

[0083] The total loss function is:

[0084] Introducing hyperparameters The three loss weights are defined as follows:

[0085]

[0086]

[0087] Among them, the first-order difference variance based on the stationarity of the predicted sequence is introduced as a confidence factor. Using the current predicted sequence The normalized first-order difference variance characterizes the smoothness of the prediction results:

[0088] in, The standard deviation of the training temperature is used for normalization. All weight matrices are optimized using the backpropagation algorithm, allowing the model to gradually learn the complex nonlinear mapping relationship between input features and the internal temperature of the battery, achieving high-precision online temperature prediction. .

[0089] Step S105: Based on the real-time predicted battery internal temperature sequence values ​​from the trained time-series converter model, calculate the first-order and second-order difference terms of the predicted internal temperature sequence. Construct a temperature rise trend control factor accordingly, combining it with the temperature rise momentum term and risk perception indicators to achieve multi-factor correction control of transient pulse current. The specific steps are as follows: S51: Based on the current-temperature mapping table calibrated at the battery's factory, determine the pulse current reference value set for the current battery temperature within a specific temperature range. .

[0090] S52: Real-time prediction of the battery's internal temperature sequence using the aforementioned time-series converter model. Calculate the first and second differences between the predicted internal temperature sequence:

[0091]

[0092] Construct adjustment factors based on their absolute values:

[0093]

[0094] in, , To balance the hyperparameter of temperature rise response sensitivity.

[0095] S53: To reflect the short-term cumulative effect of the temperature rise trend, a momentum term is introduced as a moving weighted average:

[0096] And based on this, an inhibitory factor is constructed:

[0097] in, The momentum smoothing factor, It is a hyperparameter sensitive to the temperature rise.

[0098] S54: Given m external temperature monitoring points and n stress monitoring points, calculate the risk perception index based on the predicted sequence:

[0099] in, Indicates the first The rate of temperature change over time at each external monitoring point Indicates the first The rate of change of stress over time at each stress monitoring point It is the monitored local structural stress state, and its rate of change can reflect the potential risk evolution of the battery casing or key connection parts under thermo-mechanical coupling. Through a comprehensive weighted average of temperature and stress, It can simultaneously capture transient risk levels under the influence of multiple electric, thermal, and mechanical fields.

[0100] Constructing risk adjustment factors:

[0101] in, It is the hyperparameter for controlling the suppression magnitude of risk perception.

[0102] S55: Based on the internal temperature change trend, the cumulative effect of heating momentum, and the structural response risk, at the standard current reference value... Based on the above, the output pulse current value is modified to produce the final output. This enables proactive avoidance of transient thermal shock risks. .

[0103] The structure of the electronic device according to an embodiment of the present invention will be described in detail below. Figure 3 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of the present invention. It can be understood that... Figure 3 The diagram shows only an exemplary structure of the electronic device, not the entire structure. Some or all of the structures shown may be implemented as needed.

[0104] The electronic device provided in this embodiment of the invention includes: at least one processor 301, a memory 302, a user interface 303, and at least one network interface 304. The various components in the temperature rise fusion risk perception battery low-temperature pulse heating control device are coupled together via a bus system 305. It can be understood that the bus system 305 is used to realize the connection and communication between these components. In addition to a data bus, the bus system 305 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 3 The general designated all buses as Bus System 305.

[0105] The user interface 303 may include a monitor, keyboard, mouse, trackball, click wheel, buttons, touchpad, or touch screen.

[0106] It is understood that memory 302 can be volatile memory or non-volatile memory, or both. In this embodiment of the invention, memory 302 is capable of storing data to support the operation of the terminal. Examples of this data include any computer programs used to operate on the terminal, such as operating systems and applications. The operating system includes various system programs, such as framework layers, core library layers, driver layers, etc., used to implement various basic services and handle hardware-based tasks. Applications can include various applications.

[0107] In some embodiments, the battery low-temperature pulse heating control device 300 with temperature rise fusion risk perception provided in this invention can be implemented using a combination of hardware and software. For example, the battery low-temperature pulse heating control device 300 with temperature rise fusion risk perception provided in this invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the battery low-temperature pulse heating control method with temperature rise fusion risk perception provided in this invention. For example, the processor in the form of a hardware decoding processor can employ one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0108] As an example, processor 301 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., wherein the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0109] As an example of the hardware implementation of the battery low-temperature pulse heating control device 300 for temperature rise fusion risk perception provided in this embodiment of the invention, the device provided in this embodiment of the invention can be directly executed by a processor 301 in the form of a hardware decoding processor. For example, it can be executed by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components to implement the battery low-temperature pulse heating control method for temperature rise fusion risk perception provided in this embodiment of the invention.

[0110] The memory 302 in this embodiment of the invention is used to store various types of data to support the operation of the battery low-temperature pulse heating control device for temperature rise fusion risk perception, or to store data for execution. Figure 1 The program code for the method shown. Examples of this data include: any executable instructions for operation on a battery cryogenic pulse heating control device for temperature rise fusion risk perception, such as executable instructions that can be included in the executable instructions to implement the battery cryogenic pulse heating control method for temperature rise fusion risk perception of embodiments of the present invention.

[0111] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including functions for executing... Figure 1 The program code for the method shown. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by the central processing unit, it performs the various functions defined in the apparatus of this application.

[0112] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0113] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A battery low-temperature pulse heating control device with temperature rise fusion risk perception, characterized in that, include: The timing input matrix construction module is used to construct a timing input matrix of the battery operating state based on multi-dimensional battery sensing data. The embedded sequence vector forming module is used to perform high-dimensional linear mapping on the time-series input matrix of the battery operating state and inject time position information to form an embedded sequence vector. The sequence feature extraction module is used to introduce a temperature-stress coupling gating mechanism to perform gating modulation on the embedded sequence vector to obtain the gated modulated embedded sequence vector. Then, a multi-layer stacked converter encoder is used to perform deep temporal feature extraction and nonlinear enhancement on the gated modulated embedded sequence vector to obtain sequence features. The time series converter model training module is used to aggregate sequence features into a single feature vector, then map the aggregated feature vector to the internal temperature prediction scalar, and construct a weighted loss function that integrates smoothing and constraint terms to train the time series converter model. The transient pulse current control module is used to calculate the first and second difference terms of the predicted internal temperature sequence of the battery based on the trained time-series converter model, and construct the temperature rise trend regulation factor accordingly. Combined with the temperature rise momentum term and risk perception index, it can realize multi-factor correction control of transient pulse current. The expression for achieving multi-factor correction control of transient pulse current is as follows: in, The output value is the pulse current value executed by outputting the pulse current. This is the standard current reference value; and These are the first and second difference terms of the internal temperature prediction sequence; The real-time predicted internal temperature sequence of the battery by the time-series converter model; It is an inhibitory factor; As a risk adjustment factor; , Construct adjustment factors for absolute values; The expression for the inhibitory factor is: in, The momentum smoothing factor, This is a hyperparameter sensitive to temperature rise. and This represents the momentum term.

2. The battery low-temperature pulse heating control device with temperature rise fusion risk perception as described in claim 1, characterized in that, The expression for the risk adjustment factor is: in, It is the suppression magnitude hyperparameter for controlling risk perception; m is the number of external temperature monitoring points, and n is the number of stress monitoring points; Indicates the first Temperature at one external monitoring point; Indicates the first The rate of temperature change over time at each external monitoring point Indicates the first The rate of change of stress over time at each stress monitoring point It is the local structural stress state obtained through monitoring; It is the hyperbolic tangent function.

3. The battery low-temperature pulse heating control device with temperature rise fusion risk perception as described in claim 1, characterized in that, Constructed weighted loss function of fusion smoothing and constraint terms for: in, As weight; This is the weighted mean square error loss function with weight adjustment; For time smoothing constraints; This is an additional penalty item; , , For loss weights; and This is a predicted temperature value; This represents the maximum temperature. It is a function for maximizing the value; This represents the total number of monitoring points.

4. The battery low-temperature pulse heating control device with temperature rise fusion risk perception as described in claim 1, characterized in that, Embedded sequence vectors are set as : ; This is a position-encoded vector; The vector is the result of a high-dimensional linear mapping of the state vector at each time step of the input matrix; L is the total number of time steps.

5. The battery low-temperature pulse heating control device with temperature rise fusion risk perception as described in claim 1, characterized in that, By introducing a stress-temperature coupled sensing gating mechanism, a gated modulation vector with the same embedding dimension as the converter input is generated; this gated modulation vector is then applied to the embedding sequence vector to obtain the gated modulated embedding sequence vector.

6. A battery low-temperature pulse heating control method integrating temperature rise risk perception, characterized in that, A battery low-temperature pulse heating control device based on temperature rise fusion risk perception as described in any one of claims 1-5, comprising: Based on multi-dimensional battery sensing data, a time-series input matrix of battery operating status is constructed; The timing input matrix of the battery operating status is subjected to high-dimensional linear mapping and injected with time position information to form an embedded sequence vector; A temperature-stress coupling gating mechanism is introduced to perform gating modulation on the embedded sequence vector to obtain a gated modulated embedded sequence vector. Then, a multi-layer stacked converter encoder is used to perform deep temporal feature extraction and nonlinear enhancement on the gated modulated embedded sequence vector to obtain sequence features. The sequence features are aggregated into a single feature vector, and then the aggregated feature vector is mapped to the internal temperature prediction scalar. Based on this, a weighted loss function that integrates smoothing and constraint terms is constructed to train the time series transformer model. Based on the real-time predicted battery internal temperature sequence value based on the trained time-series converter model, the first-order difference term and the second-order difference term of the predicted internal temperature value sequence are calculated, and a temperature rise trend regulation factor is constructed accordingly. Combined with the temperature rise momentum term and risk perception index, a multi-factor correction control for transient pulse current is achieved.

7. A computer storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the battery low-temperature pulse heating control method for temperature rise fusion risk perception as described in claim 6.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the battery low-temperature pulse heating control method for temperature rise fusion risk perception as described in claim 6.

Images