Lithium battery temperature prediction method and device based on multi-source information time-frequency combination

Abstract

The invention discloses a lithium battery temperature prediction method and device based on multi-source information time-frequency combination. The method comprises the following steps: acquiring current time sequence data, voltage time sequence data, temperature time sequence data and capacity time sequence data of a target lithium battery; inputting the current time sequence data, the voltage time sequence data, the temperature time sequence data and the capacity time sequence data into a pre-trained lithium battery temperature change prediction model to obtain a temperature change prediction value at the next moment; correcting the temperature change predicted value according to a preset heat balance constraint to obtain a temperature change corrected value, and determining a target lithium battery temperature at the next moment according to the current lithium battery temperature and the temperature change corrected value; the lithium battery temperature change prediction model comprises a multi-channel time convolution network, a frequency domain feature extraction module, a feature fusion module and a multi-layer perceptron. The method improves the accuracy of lithium battery temperature prediction, and can be applied to the technical field of power batteries.

Description

Technical Field

[0001] This invention relates to the field of power battery technology, and in particular to a method and apparatus for predicting lithium battery temperature based on the combination of multi-source information and time-frequency data. Background Technology

[0002] Lithium battery temperature prediction is a core technology for ensuring battery safety, optimizing performance, and extending lifespan. Real-time monitoring and temperature warnings can reduce the risk of thermal runaway and improve system efficiency.

[0003] Lithium-ion battery temperature prediction technology includes two main approaches: physical model-based and data-driven. Among them, the computational complexity of the physical model-based lithium-ion battery temperature prediction approach is relatively high, while the data-driven lithium-ion battery temperature prediction approach is mostly limited to time-domain data, which cannot capture the electrochemical processes inside the lithium-ion battery. It also lacks physical constraints, resulting in insufficient prediction accuracy.

[0004] The above problems urgently need to be addressed. Summary of the Invention

[0005] The purpose of this invention is to at least partially solve one of the technical problems existing in the prior art.

[0006] Therefore, one objective of this invention is to provide a lithium battery temperature prediction method based on the combination of multi-source information and time-frequency data, which improves the accuracy of lithium battery temperature prediction.

[0007] Another objective of this invention is to provide a lithium battery temperature prediction device based on the combination of multi-source information and time and frequency.

[0008] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of the present invention include: On one hand, embodiments of the present invention provide a lithium battery temperature prediction method based on multi-source information time-frequency combination, including the following steps: Acquire the current timing data, voltage timing data, temperature timing data, and capacity timing data of the target lithium battery; The current time series data, the voltage time series data, the temperature time series data, and the capacity time series data are input into a pre-trained lithium battery temperature change prediction model to obtain the predicted temperature change value at the next moment. The predicted temperature change is corrected according to the preset thermal balance constraint to obtain the corrected temperature change value, and the target lithium battery temperature at the next moment is determined according to the current lithium battery temperature and the corrected temperature change value. The lithium battery temperature change prediction model includes a multi-channel temporal convolutional network, a frequency domain feature extraction module, a feature fusion module, and a multilayer perceptron. The multi-channel temporal convolutional network is used to extract multimodal time domain features based on the current time series data, the voltage time series data, the temperature time series data, and the capacity time series data. The frequency domain feature extraction module is used to extract multimodal frequency domain features based on the current time series data and the voltage time series data. The feature fusion module is used to fuse the multimodal time domain features and the multimodal frequency domain features to obtain time-frequency fusion features. The multilayer perceptron is used to map the time-frequency fusion features to the predicted temperature change value.

[0009] Furthermore, in one embodiment of the present invention, the step of extracting multimodal time-domain features based on the current time-series data, the voltage time-series data, the temperature time-series data, and the capacity time-series data specifically includes: Determine the current rate of change time data based on the current time series data; Power timing data is determined based on the current timing data and the voltage timing data; Determine the state of charge timing data based on the capacity timing data; The time-series data of the current rate of change, the power time-series data, the state of charge time-series data, and the temperature time-series data are input into the multi-channel temporal convolutional network for convolution processing to obtain the multimodal temporal features.

[0010] Furthermore, in one embodiment of the present invention, the step of extracting multimodal frequency domain features based on the current time-series data and the voltage time-series data specifically includes: The current spectrum is obtained by performing a fast Fourier transform on the current time-series data; The voltage spectrum is obtained by performing a fast Fourier transform on the voltage time-series data; Calculate the frequency domain impedance information based on the current spectrum and the voltage spectrum; The low-frequency impedance amplitude, mid-frequency phase angle, and dynamic internal resistance are determined based on the frequency domain impedance information. The multimodal frequency domain characteristics are determined based on the low-frequency impedance amplitude, the mid-frequency phase angle, and the dynamic internal resistance.

[0011] Furthermore, in one embodiment of the present invention, the frequency domain impedance information is calculated using the following formula:

[0012] in, Represents frequency The corresponding frequency domain impedance, and These represent sliding windows. Voltage timing data and current timing data within, τ , Indicates time, Indicates the window length. Indicates Fourier transform; The low-frequency impedance amplitude is calculated using the following formula:

[0013] in, Indicates the low-frequency impedance amplitude. Indicates in Frequency domain impedance within the frequency band The maximum value of the modulus; The intermediate frequency phase angle is calculated using the following formula:

[0014] in, Indicates the intermediate frequency phase angle. Indicates in Time-frequency domain impedance The phase angle; The dynamic internal resistance is calculated using the following formula:

[0015] in, Indicates dynamic internal resistance. This indicates the internal resistance value of a lithium battery with a state of charge (SOC) of 100% at 25℃. This represents the learnable scaling factor.

[0016] Furthermore, in one embodiment of the present invention, the lithium battery temperature change prediction model is trained through the following steps: Acquire current time-series sample data, voltage time-series sample data, temperature time-series sample data, and capacity time-series sample data of the lithium battery samples; Training samples are constructed based on the current time series sample data, the voltage time series sample data, the temperature time series sample data, and the capacity time series sample data, and the temperature change labels corresponding to the training samples are determined by manual annotation. The training samples are input into the initialized lithium battery temperature change prediction model for training, resulting in the trained lithium battery temperature change prediction model.

[0017] Furthermore, in one embodiment of the present invention, the step of inputting the training samples into an initialized lithium battery temperature change prediction model for training to obtain a trained lithium battery temperature change prediction model specifically includes: Multimodal temporal sample features are extracted from the training samples using an initialized multichannel temporal convolutional network. The frequency domain feature extraction module, after initialization, extracts multimodal frequency domain sample features based on the current time series sample data and the voltage time series sample data. The time-frequency fusion sample features are obtained by fusing the multimodal time-domain sample features and the multimodal frequency-domain sample features through the initialized feature fusion module. The time-frequency fusion sample features are mapped to temperature change prediction results using an initialized multilayer perceptron; The loss value is determined based on the temperature change prediction results and the temperature change label; Based on the loss value, the parameters of the multi-channel temporal convolutional network, the frequency domain feature extraction module, the feature fusion module, and the multilayer perceptron are updated using the backpropagation algorithm to obtain the trained lithium battery temperature change prediction model.

[0018] Furthermore, in one embodiment of the present invention, the temperature change correction value is determined by the following formula:

[0019] in, This indicates the correction value for temperature changes. This represents the predicted value of temperature change. Indicates time The current, Indicates time The dynamic internal resistance, Indicates the time interval between adjacent moments. Indicates the quality of lithium batteries. Indicates the specific heat capacity of a lithium battery. Indicates the tolerance for heat dissipation; The target lithium battery temperature is determined by the following formula:

[0020] in, Indicates time The target lithium battery temperature, Indicates time The current lithium battery temperature, Indicates the convective heat transfer coefficient. This indicates the surface area of ​​the lithium battery. Indicates ambient temperature.

[0021] On the other hand, embodiments of the present invention provide a lithium battery temperature prediction device based on multi-source information time-frequency combination, comprising: The data acquisition module is used to acquire the current time-series data, voltage time-series data, temperature time-series data, and capacity time-series data of the target lithium battery. The temperature change prediction module is used to input the current time series data, the voltage time series data, the temperature time series data and the capacity time series data into a pre-trained lithium battery temperature change prediction model to obtain the predicted temperature change value at the next moment. The temperature change correction module is used to correct the predicted temperature change value according to the preset thermal balance constraint to obtain the temperature change correction value, and to determine the target lithium battery temperature at the next moment according to the current lithium battery temperature and the temperature change correction value. The lithium battery temperature change prediction model includes a multi-channel temporal convolutional network, a frequency domain feature extraction module, a feature fusion module, and a multilayer perceptron. The multi-channel temporal convolutional network is used to extract multimodal time domain features based on the current time series data, the voltage time series data, the temperature time series data, and the capacity time series data. The frequency domain feature extraction module is used to extract multimodal frequency domain features based on the current time series data and the voltage time series data. The feature fusion module is used to fuse the multimodal time domain features and the multimodal frequency domain features to obtain time-frequency fusion features. The multilayer perceptron is used to map the time-frequency fusion features to the predicted temperature change value.

[0022] On the other hand, embodiments of the present invention provide an electronic device, including: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above-described method for predicting lithium battery temperature based on multi-source information time-frequency combination.

[0023] On the other hand, embodiments of the present invention also provide a computer-readable storage medium storing a processor-executable computer program that, when executed by a processor, implements the above-described method for predicting lithium battery temperature based on multi-source information time-frequency combination.

[0024] On the other hand, embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the above-described method for predicting lithium battery temperature based on multi-source information time-frequency combination.

[0025] The advantages and beneficial effects of the present invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention: This invention acquires current time-series data, voltage time-series data, temperature time-series data, and capacity time-series data of a target lithium battery. These data are then input into a pre-trained lithium battery temperature change prediction model to obtain a predicted temperature change value for the next time step. The predicted temperature change value is corrected according to a preset thermal balance constraint to obtain a corrected temperature change value. Finally, the target lithium battery temperature for the next time step is determined based on the current lithium battery temperature and the corrected temperature change value. The lithium battery temperature change prediction model includes a multi-channel temporal convolutional network, a frequency domain feature extraction module, a feature fusion module, and a multilayer perceptron. The multi-channel temporal convolutional network extracts multimodal temporal features from the current, voltage, temperature, and capacity time-series data. The frequency domain feature extraction module extracts multimodal frequency domain features from the current and voltage time-series data. The feature fusion module fuses the multimodal temporal and frequency domain features to obtain time-frequency fusion features. The multilayer perceptron maps the time-frequency fusion features to the predicted temperature change value. This invention captures the long-sequence dependent time-domain features of multi-source information in lithium batteries through a multi-channel temporal convolutional network. It converts time-domain current and time-domain voltage into frequency-domain impedance through fast Fourier transform, thereby capturing the frequency-domain features characterizing the internal electrochemical state of the battery. After fusing the time-domain and frequency-domain features, it maps them to a predicted temperature change value through a multilayer perceptron. Then, it corrects the predicted temperature change value through thermal balance constraints to satisfy the heat generation-heat dissipation balance. Finally, it determines the lithium battery temperature at the next moment based on the corrected temperature change value, thus improving the accuracy of lithium battery temperature prediction. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the embodiments of the present invention are described below. It should be understood that the drawings described below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A flowchart illustrating the steps of a lithium battery temperature prediction method based on multi-source information time-frequency combination provided in an embodiment of the present invention; Figure 2 A structural block diagram of a lithium battery temperature prediction device based on multi-source information time-frequency combination provided in an embodiment of the present invention; Figure 3 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of this invention; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this invention as detailed in the appended claims.

[0029] Unless otherwise defined, 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. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0030] The lithium battery temperature prediction method based on multi-source information time-frequency combination provided in this invention can be applied to terminals, servers, or software running on terminals or servers. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application that implements the lithium battery temperature prediction method based on multi-source information time-frequency combination, but is not limited to the above forms.

[0031] This invention can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0032] It should be noted that in various specific embodiments of the present invention, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of the present invention require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to a confirmation page. Only after obtaining the user's separate permission or consent is the necessary user-related data for the normal operation of the embodiments of the present invention acquired.

[0033] Existing data-driven lithium battery temperature prediction schemes have the following drawbacks: 1) They only focus on the use of time-domain data and lack consideration for the use of frequency-domain data; 2) They lack physical constraints and have insufficient model interpretability; 3) They lack consideration for hardware platforms and are not suitable for online prediction by battery management systems (BMS).

[0034] Reference Figure 1 This invention provides a method for predicting lithium battery temperature based on the combination of multi-source information and time-frequency data, specifically including the following steps: S101. Acquire the current timing data, voltage timing data, temperature timing data, and capacity timing data of the target lithium battery; S102. Input the current time series data, voltage time series data, temperature time series data and capacity time series data into the pre-trained lithium battery temperature change prediction model to obtain the temperature change prediction value at the next moment. S103. Correct the predicted temperature change value according to the preset thermal balance constraint to obtain the corrected temperature change value, and determine the target lithium battery temperature at the next moment according to the current lithium battery temperature and the corrected temperature change value. The lithium battery temperature change prediction model includes a multi-channel temporal convolutional network, a frequency domain feature extraction module, a feature fusion module, and a multilayer perceptron. The multi-channel temporal convolutional network is used to extract multimodal time domain features based on current time series data, voltage time series data, temperature time series data, and capacity time series data. The frequency domain feature extraction module is used to extract multimodal frequency domain features based on current time series data and voltage time series data. The feature fusion module is used to fuse the multimodal time domain features and multimodal frequency domain features to obtain time-frequency fusion features. The multilayer perceptron is used to map the time-frequency fusion features into predicted temperature change values.

[0035] This invention captures the long-sequence dependent time-domain features of multi-source information in lithium batteries through a multi-channel temporal convolutional network. It converts time-domain current and time-domain voltage into frequency-domain impedance through fast Fourier transform, thereby capturing the frequency-domain features characterizing the internal electrochemical state of the battery. After fusing the time-domain and frequency-domain features, it maps them to a predicted temperature change value through a multilayer perceptron. Then, it corrects the predicted temperature change value through thermal balance constraints to satisfy the heat generation-heat dissipation balance. Finally, it determines the lithium battery temperature at the next moment based on the corrected temperature change value, thus improving the accuracy of lithium battery temperature prediction.

[0036] It is understood that the embodiments of the present invention reflect the changes in battery internal resistance through the characteristic changes in frequency domain impedance, and embed a correction mechanism based on thermal balance constraints in the model output layer, making the lithium battery temperature prediction results more reliable. At the same time, using frequency domain data for assistance and combining it with physical hard constraints can provide earlier fault warnings, improve robustness under extreme conditions, enhance model interpretability, and make the predicted temperature meet the heat generation-heat dissipation balance. In addition, the multi-channel temporal convolutional network and the frequency domain feature extraction module can be deployed in the battery management system, which has more user-friendly hardware deployment conditions.

[0037] As a further optional implementation, multimodal time-domain features are extracted based on current time-series data, voltage time-series data, temperature time-series data, and capacity time-series data, specifically including: S201. Determine the current rate of change time series data based on the current time series data; S202. Determine the power timing data based on the current timing data and voltage timing data; S203. Determine the state-of-charge timing data based on the capacity timing data; S204. Input the time series data of current rate of change, power, state of charge, and temperature into a multi-channel temporal convolutional network for convolution processing to obtain multimodal time-domain features.

[0038] First, collect the current time series data I(t), voltage time series data V(t), temperature time series data T(t), and capacity time series data C(t) during battery operation.

[0039] The state of charge (SOC) is calculated using the capacity calculation formula: The rate of change of current is calculated using I(t) and time t to reflect sudden changes in operating conditions. The calculation process is as follows: It can identify charge / discharge steps; by calculating instantaneous power P(t) = I(t)V(t) using I(t) and V(t), it directly relates to Joule heating. The above... P(t), And T(t) constitutes the initial time-domain features.

[0040] The initial temporal features are modeled using an adaptive multi-channel temporal convolutional network (TCN). The feature dimension is 4×t, where 4 corresponds to four temporal features and t is the time step. First, a dynamically adjusted dilation factor d is set. ,default k=10. The initial temporal features are input into the TCN model, and the convolution calculation process is as follows: It represents the gated causal convolution of the l-th layer. The resulting multimodal temporal features can be represented as h, which is a causal convolution with an inflation rate of d. time =TCN[ P(t), , (t)].

[0041] As a further optional implementation, multimodal frequency domain features are extracted based on current time-series data and voltage time-series data, specifically including: S301. Perform a fast Fourier transform on the current time-series data to obtain the current spectrum; S302. Perform a fast Fourier transform on the voltage time-series data to obtain the voltage spectrum; S303. Calculate the frequency domain impedance information based on the current spectrum and voltage spectrum; S304. Determine the low-frequency impedance amplitude, mid-frequency phase angle, and dynamic internal resistance based on the frequency domain impedance information. S305. Determine the multi-mode frequency domain characteristics based on the low-frequency impedance amplitude, mid-frequency phase angle, and dynamic internal resistance.

[0042] As a further optional implementation, the frequency domain impedance information is calculated using the following formula:

[0043] in, Represents frequency The corresponding frequency domain impedance, and These represent sliding windows. Voltage timing data and current timing data within, τ , Indicates time, Indicates the window length. Indicates Fourier transform; The low-frequency impedance amplitude is calculated using the following formula:

[0044] in, Indicates the low-frequency impedance amplitude. Indicates in Frequency domain impedance within the frequency band The maximum value of the modulus; The intermediate frequency phase angle is calculated using the following formula:

[0045] in, Indicates the intermediate frequency phase angle. Indicates in Time-frequency domain impedance The phase angle; The dynamic internal resistance is calculated using the following formula:

[0046] in, Indicates dynamic internal resistance. This indicates the internal resistance value of a lithium battery with a state of charge (SOC) of 100% at 25℃. This represents the learnable scaling factor.

[0047] Specifically, the frequency domain impedance characteristics are calculated using a sliding window FFT. The calculation process is as follows: Z(ω) = ,in and These are the instantaneous data for voltage and current, τ, respectively. [tW, t], where W is the window length. This is a Fourier transform. Frequency domain impedance reflects the battery's response to currents of different frequencies.

[0048] Obtain the low-frequency impedance amplitude and correlate it with the lithium-ion diffusion process: .

[0049] Obtaining the mid-frequency phase angle reflects the charge transfer reaction activity: .

[0050] The calculation process for dynamic internal resistance estimation is shown below: in This is the resistance value at 25℃ with a state of charge (SOC) of 100%. .

[0051] The resulting multimodal frequency domain features can be expressed as h freq ={ , , }

[0052] Then, by performing time-frequency feature fusion through the feature fusion module, the time-frequency fused feature h can be obtained. fusion =Concat(h freq h time ).

[0053] Finally, temperature prediction is performed using a multilayer perceptron (MLP). The calculation process is shown below: .

[0054] It should be noted that the multi-channel temporal convolutional network, frequency domain feature extraction module, feature fusion module, and multilayer perceptron in the embodiments of the present invention are all pre-trained and their network parameters are adjusted. After training, they form a lithium battery temperature change prediction model to achieve real-time lithium battery temperature change prediction.

[0055] As an optional further implementation, the lithium battery temperature change prediction model is trained through the following steps: S301. Obtain the current time-series sample data, voltage time-series sample data, temperature time-series sample data, and capacity time-series sample data of the sample lithium battery; S302. Construct training samples based on current time series sample data, voltage time series sample data, temperature time series sample data and capacity time series sample data, and determine the temperature change label corresponding to the training samples through manual annotation. S303. Input the training samples into the initialized lithium battery temperature change prediction model for training, and obtain the trained lithium battery temperature change prediction model.

[0056] As a further optional implementation, training samples are input into an initialized lithium battery temperature change prediction model for training, resulting in a trained lithium battery temperature change prediction model, which specifically includes: S3031. Extract multimodal temporal sample features from training samples using an initialized multichannel temporal convolutional network; S3032. Extract multi-mode frequency domain sample features based on current time series sample data and voltage time series sample data through the initialized frequency domain feature extraction module; S3033. The feature fusion module initializes the multimodal time-domain sample features and the multimodal frequency-domain sample features to obtain time-frequency fusion sample features; S3034. The time-frequency fusion sample features are mapped to temperature change prediction results through an initialized multilayer perceptron; S3035. Determine the loss value based on the temperature change prediction results and temperature change labels; S3036. Based on the loss value, update the parameters of the multi-channel temporal convolutional network, frequency domain feature extraction module, feature fusion module, and multilayer perceptron using the backpropagation algorithm to obtain the trained lithium battery temperature change prediction model.

[0057] Specifically, in the test scenario, current time-series sample data, voltage time-series sample data, temperature time-series sample data, and capacity time-series sample data of the lithium battery are acquired to construct training samples and determine the corresponding temperature change labels. These are then input into an initialized lithium battery temperature change prediction model. An initialized multi-channel temporal convolutional network extracts multimodal time-domain sample features from the training samples. An initialized frequency-domain feature extraction module extracts multimodal frequency-domain sample features from the current and voltage time-series sample data. An initialized feature fusion module fuses the multimodal time-domain and multimodal frequency-domain sample features to obtain time-frequency fusion sample features. An initialized multilayer perceptron maps the time-frequency fusion sample features to temperature change prediction results. Based on these temperature change prediction results and temperature change labels, a training loss value is determined. Based on this loss value, the parameters of the multi-channel temporal convolutional network, frequency-domain feature extraction module, feature fusion module, and multilayer perceptron are updated using a backpropagation algorithm. When the number of iterations reaches a preset threshold or the loss value falls below a preset threshold, the trained lithium battery temperature change prediction model is obtained.

[0058] As a further optional implementation, the temperature change correction value is determined by the following formula:

[0059] in, This indicates the correction value for temperature changes. This represents the predicted value of temperature change. Indicates time The current, Indicates time The dynamic internal resistance, Indicates the time interval between adjacent moments. Indicates the quality of lithium batteries. Indicates the specific heat capacity of a lithium battery. Indicates the tolerance for heat dissipation; The target lithium battery temperature is determined by the following formula:

[0060] in, Indicates time The target lithium battery temperature, Indicates time The current lithium battery temperature, Indicates the convective heat transfer coefficient. This indicates the surface area of ​​the lithium battery. Indicates ambient temperature.

[0061] Specifically, after predicting the temperature change at the next moment based on the lithium battery temperature change prediction model, the predicted temperature change is corrected based on a preset thermal balance constraint. The specific formula is as follows: , This indicates the tolerance for heat dissipation, to avoid violating the first law of thermodynamics.

[0062] The final target lithium battery temperature is It takes into account the convective heat transfer process between the lithium battery and the environment, further improving the accuracy of lithium battery temperature prediction.

[0063] The method steps of the embodiments of the present invention have been described above. It can be understood that the embodiments of the present invention capture the long-sequence dependent time-domain features of multi-source information of lithium batteries through a multi-channel temporal convolutional network, convert time-domain current and time-domain voltage into frequency-domain impedance through a fast Fourier transform, thereby capturing the frequency-domain features characterizing the internal electrochemical state of the battery. After fusing the time-domain features and frequency-domain features, the results are mapped to a predicted temperature change value through a multilayer perceptron. Then, the predicted temperature change value is corrected through thermal balance constraints to satisfy the heat generation-heat dissipation balance. Finally, the lithium battery temperature at the next moment is determined based on the corrected temperature change value, thus improving the accuracy of lithium battery temperature prediction.

[0064] Reference Figure 2 This invention provides a lithium battery temperature prediction device based on multi-source information time-frequency combination, comprising: The data acquisition module is used to acquire the current time-series data, voltage time-series data, temperature time-series data, and capacity time-series data of the target lithium battery. The temperature change prediction module is used to input current time series data, voltage time series data, temperature time series data and capacity time series data into a pre-trained lithium battery temperature change prediction model to obtain the temperature change prediction value at the next moment. The temperature change correction module is used to correct the predicted temperature change value according to the preset thermal balance constraint to obtain the temperature change correction value, and to determine the target lithium battery temperature at the next moment based on the current lithium battery temperature and the temperature change correction value. The lithium battery temperature change prediction model includes a multi-channel temporal convolutional network, a frequency domain feature extraction module, a feature fusion module, and a multilayer perceptron. The multi-channel temporal convolutional network is used to extract multimodal time domain features based on current time series data, voltage time series data, temperature time series data, and capacity time series data. The frequency domain feature extraction module is used to extract multimodal frequency domain features based on current time series data and voltage time series data. The feature fusion module is used to fuse the multimodal time domain features and multimodal frequency domain features to obtain time-frequency fusion features. The multilayer perceptron is used to map the time-frequency fusion features into predicted temperature change values.

[0065] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0066] Reference Figure 3 This invention provides an electronic device, comprising: At least one processor; At least one memory for storing at least one program; When the above-mentioned at least one program is executed by the above-mentioned at least one processor, the above-mentioned at least one processor implements the above-mentioned lithium battery temperature prediction method based on multi-source information time-frequency combination.

[0067] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0068] This invention also provides a computer-readable storage medium storing a processor-executable computer program that, when executed by a processor, implements the aforementioned lithium battery temperature prediction method based on multi-source information time-frequency combination.

[0069] This invention provides a computer-readable storage medium that can execute a lithium battery temperature prediction method based on multi-source information time-frequency combination provided in the method embodiment of this invention. It can execute any combination of the implementation steps of the method embodiment and has the corresponding functions and beneficial effects of the method.

[0070] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for predicting lithium battery temperature based on the combination of multi-source information and time and frequency.

[0071] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0072] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0073] The embodiments described in this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.

[0074] The terms "first," "second," "third," "fourth," etc. (if present) in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0075] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the aforementioned blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0076] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the aforementioned functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0077] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0078] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0079] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the aforementioned program can be printed, because the aforementioned program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0080] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0081] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0082] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0083] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for predicting lithium battery temperature based on multi-source information time-frequency combination, characterized in that, Includes the following steps: Acquire the current timing data, voltage timing data, temperature timing data, and capacity timing data of the target lithium battery; The current time series data, the voltage time series data, the temperature time series data, and the capacity time series data are input into a pre-trained lithium battery temperature change prediction model to obtain the predicted temperature change value at the next moment. The predicted temperature change is corrected according to the preset thermal balance constraint to obtain the corrected temperature change value, and the target lithium battery temperature at the next moment is determined according to the current lithium battery temperature and the corrected temperature change value. The lithium battery temperature change prediction model includes a multi-channel temporal convolutional network, a frequency domain feature extraction module, a feature fusion module, and a multilayer perceptron. The multi-channel temporal convolutional network is used to extract multimodal time domain features based on the current time series data, the voltage time series data, the temperature time series data, and the capacity time series data. The frequency domain feature extraction module is used to extract multimodal frequency domain features based on the current time series data and the voltage time series data. The feature fusion module is used to fuse the multimodal time domain features and the multimodal frequency domain features to obtain time-frequency fusion features. The multilayer perceptron is used to map the time-frequency fusion features to the predicted temperature change value.

2. The lithium battery temperature prediction method based on multi-source information time-frequency combination according to claim 1, characterized in that, The step of extracting multimodal time-domain features based on the current time-series data, the voltage time-series data, the temperature time-series data, and the capacity time-series data specifically includes: Determine the current rate of change time data based on the current time series data; Power timing data is determined based on the current timing data and the voltage timing data; Determine the state of charge timing data based on the capacity timing data; The time-series data of the current rate of change, the power time-series data, the state of charge time-series data, and the temperature time-series data are input into the multi-channel temporal convolutional network for convolution processing to obtain the multimodal temporal features.

3. The lithium battery temperature prediction method based on multi-source information time-frequency combination according to claim 1, characterized in that, The step of extracting multimodal frequency domain features based on the current time-series data and the voltage time-series data specifically includes: The current spectrum is obtained by performing a fast Fourier transform on the current time-series data; The voltage spectrum is obtained by performing a fast Fourier transform on the voltage time-series data; Calculate the frequency domain impedance information based on the current spectrum and the voltage spectrum; The low-frequency impedance amplitude, mid-frequency phase angle, and dynamic internal resistance are determined based on the frequency domain impedance information. The multimodal frequency domain characteristics are determined based on the low-frequency impedance amplitude, the mid-frequency phase angle, and the dynamic internal resistance.

4. The lithium battery temperature prediction method based on multi-source information time-frequency combination according to claim 3, characterized in that, The frequency domain impedance information is calculated using the following formula: in, Represents frequency The corresponding frequency domain impedance, and These represent sliding windows. Voltage timing data and current timing data within, τ , Indicates time, Indicates the window length. Indicates Fourier transform; The low-frequency impedance amplitude is calculated using the following formula: in, Indicates the low-frequency impedance amplitude. Indicates in Frequency domain impedance within the frequency band The maximum value of the modulus; The intermediate frequency phase angle is calculated using the following formula: in, Indicates the intermediate frequency phase angle. Indicates in Time-frequency domain impedance The phase angle; The dynamic internal resistance is calculated using the following formula: in, Indicates dynamic internal resistance. This indicates the internal resistance value of a lithium battery with a state of charge (SOC) of 100% at 25℃. This represents the learnable scaling factor.

5. The lithium battery temperature prediction method based on multi-source information time-frequency combination according to claim 1, characterized in that, The lithium battery temperature change prediction model is trained through the following steps: Acquire current time-series sample data, voltage time-series sample data, temperature time-series sample data, and capacity time-series sample data of the lithium battery samples; Training samples are constructed based on the current time series sample data, the voltage time series sample data, the temperature time series sample data, and the capacity time series sample data, and the temperature change labels corresponding to the training samples are determined by manual annotation. The training samples are input into the initialized lithium battery temperature change prediction model for training, resulting in the trained lithium battery temperature change prediction model.

6. The lithium battery temperature prediction method based on multi-source information time-frequency combination according to claim 5, characterized in that, The step of inputting the training samples into the initialized lithium battery temperature change prediction model for training, to obtain the trained lithium battery temperature change prediction model, specifically includes: Multimodal temporal sample features are extracted from the training samples using an initialized multichannel temporal convolutional network. The frequency domain feature extraction module, after initialization, extracts multimodal frequency domain sample features based on the current time series sample data and the voltage time series sample data. The time-frequency fusion sample features are obtained by fusing the multimodal time-domain sample features and the multimodal frequency-domain sample features through the initialized feature fusion module. The time-frequency fusion sample features are mapped to temperature change prediction results using an initialized multilayer perceptron; The loss value is determined based on the temperature change prediction results and the temperature change label; Based on the loss value, the parameters of the multi-channel temporal convolutional network, the frequency domain feature extraction module, the feature fusion module, and the multilayer perceptron are updated using the backpropagation algorithm to obtain the trained lithium battery temperature change prediction model.

7. A lithium battery temperature prediction method based on multi-source information time-frequency combination according to any one of claims 1 to 6, characterized in that, The temperature change correction value is determined by the following formula: in, This indicates the correction value for temperature changes. This represents the predicted value of temperature change. Indicates time The current, Indicates time The dynamic internal resistance, Indicates the time interval between adjacent moments. Indicates the quality of lithium batteries. Indicates the specific heat capacity of a lithium battery. Indicates the tolerance for heat dissipation; The target lithium battery temperature is determined by the following formula: in, Indicates time The target lithium battery temperature, Indicates time The current temperature of the lithium battery, Indicates the convective heat transfer coefficient. This represents the surface area of ​​a lithium battery. Indicates ambient temperature.

8. A lithium battery temperature prediction device based on multi-source information time-frequency combination, characterized in that, include: The data acquisition module is used to acquire the current time-series data, voltage time-series data, temperature time-series data, and capacity time-series data of the target lithium battery. The temperature change prediction module is used to input the current time series data, the voltage time series data, the temperature time series data and the capacity time series data into a pre-trained lithium battery temperature change prediction model to obtain the predicted temperature change value at the next moment. The temperature change correction module is used to correct the predicted temperature change value according to the preset thermal balance constraint to obtain the temperature change correction value, and to determine the target lithium battery temperature at the next moment according to the current lithium battery temperature and the temperature change correction value. The lithium battery temperature change prediction model includes a multi-channel temporal convolutional network, a frequency domain feature extraction module, a feature fusion module, and a multilayer perceptron. The multi-channel temporal convolutional network is used to extract multimodal time domain features based on the current time series data, the voltage time series data, the temperature time series data, and the capacity time series data. The frequency domain feature extraction module is used to extract multimodal frequency domain features based on the current time series data and the voltage time series data. The feature fusion module is used to fuse the multimodal time domain features and the multimodal frequency domain features to obtain time-frequency fusion features. The multilayer perceptron is used to map the time-frequency fusion features to the predicted temperature change value.

9. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a lithium battery temperature prediction method based on multi-source information time-frequency combination as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements a lithium battery temperature prediction method based on the combination of multi-source information and time and frequency as described in any one of claims 1 to 7.

Images