Method for detecting state of battery and electronic device supporting same

The electronic device uses temperature and current measurements, spectrogram analysis, and a CNN to predict battery swelling, addressing the challenge of accurate swelling prediction and ensuring safe battery operation.

WO2026235399A1PCT designated stage Publication Date: 2026-11-12SAMSUNG ELECTRONICS CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/KR2026/095512
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-06-02
Filing Date
2026-04-03
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing battery technologies lack effective methods to accurately predict and prevent battery swelling, which can be caused by internal pressure from vaporized lithium-ion electrolyte, posing safety risks.

Method used

An electronic device employs temperature sensors to monitor battery temperature, measures voltage and current changes, generates a spectrogram in the time-frequency domain, and uses a convolutional neural network (CNN) to analyze this data alongside additional battery information to predict swelling risk.

Benefits of technology

Enables early detection of battery swelling with high accuracy across various charging environments, ensuring safe and efficient battery operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure KR2026095512_12112026_PF_FP_ABST
    Figure KR2026095512_12112026_PF_FP_ABST
Patent Text Reader

Abstract

An electronic device according to an embodiment is configured to: measure the temperature of a battery using at least one temperature sensor adjacent to the battery; if it is determined that the temperature of the battery is at least a reference temperature for at least a threshold time, confirm the battery voltage of the battery and the battery current of the battery; obtain, for a set time, the change rate of the battery voltage and the change rate of the battery current, on the basis of the confirmation result; on the basis of the obtained change rate of the battery voltage and the obtained change rate of the battery current, generate a spectrogram in a time-frequency domain; obtain additional information collected in relation to the state of the battery; and obtain, from an artificial intelligence model, an output value related to the swelling risk of the battery by applying the spectrogram and the additional information to the artificial intelligence model.
Need to check novelty before this filing date? Find Prior Art

Description

Method for detecting battery status and electronic device supporting the same

[0001] The present disclosure relates to a method for detecting the state of a battery and an electronic device supporting the same.

[0002] With the proliferation of mobile devices and the growth of the electric vehicle industry, battery technology is being utilized as a key energy storage method across various industrial sectors. In particular, lithium-ion batteries are widely used in smartphones, laptops, and electric vehicles due to characteristics such as high energy density, long cycle life, and low self-discharge rates. Batteries installed in electronic devices can be charged via wired or wireless methods using an external charger.

[0003] The information described above may be provided as related art for the purpose of aiding understanding of the present disclosure. No claim or determination is made as to whether any of the foregoing may be applied as prior art in relation to the present disclosure.

[0004] An electronic device according to one embodiment may include a battery, at least one temperature sensor adjacent to the battery, at least one processor, and a memory for storing instructions. The instructions may be executed individually or collectively by the at least one processor so that the electronic device measures the temperature of the battery through the at least one temperature sensor, and, upon determining that the temperature of the battery has a value greater than or equal to a reference temperature for a period of time greater than a threshold time, checks the battery voltage and the battery current of the battery, obtains the rate of change of the battery voltage and the rate of change of the battery current based on the check result for a set period of time, generates a spectrogram in the time-frequency domain based on the obtained rate of change of the battery voltage and the obtained rate of change of the battery current, obtains additional information collected in relation to the state of the battery, and applies the spectrogram and the additional information to an artificial intelligence model to obtain an output value related to the swelling risk of the battery from the artificial intelligence model.

[0005] A method of operation of an electronic device according to one embodiment may include: measuring the temperature of the battery through at least one temperature sensor adjacent to the battery; determining that the temperature of the battery has a value greater than or equal to a reference temperature for a period of time greater than a threshold time, and checking the battery voltage and the battery current of the battery for a set period of time; obtaining the rate of change of the battery voltage and the rate of change of the battery current based on the result of the check; generating a spectrogram in the time-frequency domain based on the obtained rate of change of the battery voltage and the obtained rate of change of the battery current; obtaining additional information collected in relation to the state of the battery; and applying the spectrogram and the additional information to an artificial intelligence model to obtain an output value related to the swelling risk of the battery from the artificial intelligence model.

[0006] In relation to the description of the drawings, the same or similar reference numerals may be used for identical or similar components.

[0007] FIG. 1 is a diagram illustrating the overall flow and overview of a method for an electronic device to detect the risk of swelling of a battery according to one embodiment.

[0008] FIG. 2 is an exploded perspective view of an electronic device according to one embodiment.

[0009] FIG. 3 is a flowchart illustrating the process of an electronic device according to one embodiment detecting the risk of swelling of a battery.

[0010] FIG. 4 is a flowchart illustrating the operation of repeatedly obtaining the rate of change of battery voltage and the rate of change of battery current in an electronic device according to one embodiment.

[0011] FIG. 5 is a flowchart illustrating the process of converting to time-frequency domain data and generating a spectrogram based on changes in battery voltage and current in an electronic device according to one embodiment.

[0012] FIG. 6 is a diagram showing an example of a spectrogram generated in an electronic device according to one embodiment.

[0013] FIG. 7 is a flowchart illustrating the process of an electronic device according to one embodiment performing normalization on data included in a spectrogram.

[0014] FIG. 8 is a diagram showing a specific example of an electronic device according to one embodiment performing a normalization process.

[0015] FIG. 9 is a diagram illustrating the process of obtaining an output value related to the risk of swelling of a battery by inputting normalized spectrogram data into an artificial intelligence model of an electronic device according to one embodiment.

[0016] FIG. 10 is a diagram illustrating the process of obtaining an output value related to the risk of swelling of a battery by inputting additional information along with a spectrogram into an artificial intelligence model of an electronic device according to one embodiment.

[0017] FIG. 11 is a diagram illustrating a CNN-based learning and inference process performed by an artificial intelligence model according to one embodiment to generate output values ​​related to battery swelling risk.

[0018] FIG. 12 is a block diagram of an electronic device in a network environment according to various embodiments.

[0019] FIG. 13 is a diagram showing an example of a system including a generative artificial intelligence model according to one embodiment.

[0020] Embodiments of the present disclosure are described below in detail with reference to the attached drawings so that those skilled in the art can easily implement them. However, the present disclosure may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present disclosure in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0021] The terms used in this disclosure are described in their current, general form considering the functions mentioned herein; however, they may refer to various other terms depending on the intent of those skilled in the art, case law, or the emergence of new technologies. Accordingly, the terms used in this disclosure should not be interpreted solely by their names, but should be interpreted based on the meaning of the terms and the overall content of this disclosure.

[0022] Additionally, terms such as the first, second, third, ..., Nth may be used to describe various components, but the components should not be limited by these terms. These terms are used for the purpose of distinguishing one component from another.

[0023] Throughout the specification, when a part is described as being "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "electrically connected" with other components interposed between them. Furthermore, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0024] Phrases such as "in one embodiment" appearing in various places in this disclosure do not necessarily refer to the same embodiment.

[0025] One embodiment of the present disclosure may be represented by functional block configurations and various processing steps. Some or all of these functional blocks may be implemented by various numbers of hardware and / or software configurations that execute specific functions. For example, the functional blocks of the present disclosure may be implemented by one or more microprocessors or by circuit configurations for a specific function. Additionally, for example, the functional blocks of the present disclosure may be implemented in various programming or scripting languages. The functional blocks may be implemented as algorithms executed on one or more processors. Furthermore, the present disclosure may employ prior art for electronic configuration, signal processing, and / or data processing. Terms such as "mechanism," "element," "means," and "configuration" may be used broadly and are not limited to mechanical and physical configurations.

[0026] Furthermore, the connecting lines or connecting members between the components depicted in the drawings are merely illustrative of functional connections and / or physical or circuit connections. In the actual device, connections between components may be represented by various alternative or added functional connections, physical connections, or circuit connections.

[0027] Artificial intelligence technology consists of machine learning (e.g., deep learning) and elemental technologies utilizing machine learning.

[0028] Machine learning is an algorithmic technology that classifies and learns the features of input data on its own, and the elemental technology is a technology that mimics functions such as cognition and judgment of the human brain by utilizing deep learning machine learning algorithms, and consists of the fields of linguistic understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control.

[0029] The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology that recognizes, applies, and processes human language and text, and includes natural language processing, machine translation, dialogue systems, question answering, and speech recognition / synthesis. Visual understanding is a technology that perceives and processes objects like human vision, and includes object recognition, object tracking, image search, person recognition, scene understanding, spatial understanding, and image enhancement. Inference and prediction is a technology that judges information to logically infer and predict, and includes knowledge / probability-based inference, optimization prediction, preference-based planning, and recommendation. Knowledge representation is a technology that automatically processes human experiential information into knowledge data, and includes knowledge construction (data generation / classification) and knowledge management (data utilization). Motion control is a technology that controls the autonomous driving of vehicles and the movement of robots, and includes motion control (navigation, collision, driving) and manipulation control (behavior control).

[0030] In the present disclosure, battery swelling refers to a phenomenon in which a battery expands. For example, battery swelling may occur due to the pressure generated by the vaporized lithium-ion electrolyte within the battery.

[0031] In the present disclosure, the temperature of the battery refers to an estimated value of the temperature inside the battery of an electronic device. In one embodiment, the temperature of the battery may be obtained based on a plurality of temperature data obtained through a plurality of sensors disposed adjacent to the battery of the electronic device. For example, the temperature of the battery may be obtained based on the temperature measured at the battery cell or a region adjacent to the cell. For example, the temperature of the battery may be measured through a temperature sensor disposed on the outer surface of the battery cell, a frame adjacent to the battery cell, or an internal space of a module. For example, the temperature sensor may include at least one of a negative temperature coefficient (NTC) thermistor, an RTD, a temperature probe, or a sensor for detecting PCB temperature. For example, temperature data of the outer surface of the battery cell may be obtained through a thermistor (NTC thermistor) disposed on the outer surface of the battery cell. For example, data related to the internal temperature of the battery cell may be obtained through an RTD disposed inside the battery cell. For example, data related to the internal ambient temperature of the battery module may be obtained through a temperature probe within the battery module. For example, data related to the temperature of the external frame of the battery module can be obtained through an external temperature sensor placed on the external frame of the battery module. For example, data related to the temperature of the BMS (battery management system) circuit can be obtained through a PCB temperature sensing sensor placed in a location adjacent to the BMS circuit. An electronic device according to one embodiment can analyze a plurality of temperature data obtained through a plurality of temperature sensors placed on the electronic device to estimate the internal temperature of a battery cell and determine the estimated internal temperature of the battery cell as the temperature of the battery.

[0032] In the present disclosure, battery voltage may refer to a potential formed at both ends of a battery cell or battery pack. In one embodiment, battery voltage may be obtained through a voltage sensing circuit or a fuel gauge IC included in a battery management system (BMS). For example, battery voltage may be detected in real time through a sensing circuit and may change according to a plurality of factors including charge state, load conditions, or temperature.

[0033] In the present disclosure, battery current may refer to the magnitude of current flowing into or out of a battery. An electronic device according to one embodiment may obtain battery current by measuring the voltage difference across a shunt resistor connected to a fuel gaze IC and calculating the current based on the corresponding resistance value. For example, battery current may be measured periodically at predetermined time intervals. For example, battery current may be used as an indicator representing the flow of a charging or discharging process.

[0034] In the present disclosure, the artificial intelligence model may include a convolutional neural network (CNN). A CNN refers to a neural network structure based on deep learning that can automatically extract features from input data and perform tasks such as classification, regression, or anomaly detection based thereon. An electronic device according to one embodiment may use a CNN to learn visual or time-frequency patterns of input data and classify battery status or determine whether there is an anomaly based on the learned patterns. A CNN according to one embodiment may include a plurality of convolutional layers, pooling layers, activation function layers, or fully connected layers. For example, each layer may generate a meaningful representation while gradually abstracting the features of the input data. For example, a convolutional layer may extract partial features within the input data through a filter (kernel), and a pooling layer may improve computational efficiency by reducing the spatial size of the feature map. Activation functions enhance the model's expressiveness by introducing non-linearity, while fully connected layers can perform the role of deriving the final output value. For example, CNNs can use images or time-series data as input. For instance, electronic devices can analyze the time-frequency variation of a battery signal by applying a spectrogram generated through the Short-Time Fourier Transform (STFT) to a CNN.

[0035] The present disclosure will be described in detail below with reference to the attached drawings.

[0036] FIG. 1 is a diagram illustrating the overall flow and overview of a method for an electronic device (100) according to one embodiment to detect the risk of swelling of a battery.

[0037] An electronic device (100) according to one embodiment monitors the battery temperature through a temperature sensor adjacent to the battery, and when it is determined that the battery temperature maintains a value above a reference temperature for a certain period of time or longer, periodically measures the battery voltage (111) and battery current (112) to calculate the rate of change thereof, generates a spectrogram in the time-frequency domain based on the rate of change of each of the battery voltage (111) and battery current (112), and then applies the generated spectrogram to an artificial intelligence model (101) along with additional information to obtain an output value related to the risk of battery swelling, thereby providing a method for detecting the risk of battery swelling.

[0038] According to one embodiment, the electronic device (100) can measure the temperature in real time through at least one temperature sensor placed adjacent to the battery. At this time, if it is determined that the temperature remains above a set reference temperature for a certain period of time (e.g., 3 minutes or more), a subsequent procedure to perform a more precise analysis of the battery can be entered. For example, the voltage and current of the battery can be measured periodically, and their temporal rates of change (dv / dt, di / dt) can be calculated. Referring to identification number 110, data related to the rate of change of battery voltage and battery current over time and the rate of change of charging capacity under CC (constant current) mode can be obtained. According to one embodiment, the battery can be charged or discharged, and at least one of the battery voltage or battery current can be measured in the charging state or the discharging state of the battery. Based on at least one of the measured voltage data or current data, the electronic device (100) can verify information related to the temporal rate of change of at least one of the battery voltage or battery current. According to one embodiment, the electronic device (100) can detect the charge capacity in a constant current (CC) charging mode based on at least one of the measured voltage data or current data when charging the battery. The charging and / or discharging of the battery may be performed repeatedly. The electronic device (100) can measure data related to the charge capacity based on the number of repeated operations (cycle). Data related to the rate of change of battery voltage and the rate of change of battery current can be converted into data in the time-frequency domain through a short-time Fourier transform (STFT). The result of the conversion can be obtained in the form of a spectrogram (120). The spectrogram (120) may be composed of a time axis and a frequency axis.

[0039] For example, the electronic device (100) can obtain additional information related to the state of the battery. For example, the additional information may include not only state information such as the battery's State of Health (SOH), battery type, or temperature change history, but also at least one of the battery's charging speed (fast charging or slow charging), charging method (e.g., wired charging or wireless charging), cumulative charge / discharge cycles, depth of discharge (DOD), or charging history information. The additional information can complementarily reflect the long-term usage environment and history conditions of the battery, which are difficult to reflect solely by the spectrogram (120), thereby improving the prediction accuracy of the artificial intelligence model.

[0040] The generated spectrogram and acquired additional information can be input into an artificial intelligence model (101). For example, the artificial intelligence model (101) may include a convolutional neural network (CNN). The artificial intelligence model (101) can generate an output value (140) related to the swelling risk of the battery by analyzing the pattern of the spectrogram (120) and the additional information. For example, the output value (140) related to the swelling risk of the battery may include binary classification results of normal and abnormal. Alternatively, the output value (140) related to the swelling risk of the battery may be a probability-based continuous value or a value provided based on a multi-level rating.

[0041] According to one embodiment, signs of swelling can be detected in advance with high accuracy even in various charging environments and battery conditions, and safe and efficient battery operation can be enabled.

[0042] FIG. 2 is an unfolded perspective view of an electronic device (100) according to one embodiment.

[0043] Referring to FIG. 2, the electronic device (100) may include a side bezel structure (210), a first support member (211) (e.g., a bracket), a front plate (220), a display (230), an electromagnetic induction panel (290), a printed circuit board (240), a battery (250), a second support member (260) (e.g., a rear case), an antenna (270), a pen input device (201), and a rear plate (280). In some embodiments, the electronic device (100) may omit at least one of the components (e.g., the first support member (211), or the second support member (260)) or additionally include other components. At least one of the components of the electronic device (100) may be identical or similar to at least one of the components of the electronic device (100) of FIG. 1 or FIG. 2, and redundant descriptions are omitted below.

[0044] An electromagnetic induction panel (290) (e.g., a digitizer) may be a panel for detecting input from a pen input device (201). For example, the electromagnetic induction panel (290) may include a printed circuit board (PCB) (e.g., a flexible printed circuit board (FPCB)) and a shielding sheet. The shielding sheet may prevent interference between said components caused by electromagnetic fields generated from said components (e.g., a display module, a printed circuit board, an electromagnetic induction panel, etc.) included in the electronic device (100). By blocking the electromagnetic fields generated from the components, the shielding sheet may ensure that input from the pen input device (201) is accurately transmitted to a coil included in the electromagnetic induction panel (240). According to various embodiments, the electromagnetic induction panel (240) may include an opening formed in at least a portion of the area corresponding to a biosensor mounted in the electronic device (100).

[0045] The first support member (211) may be disposed inside the electronic device (100) and connected to the side bezel structure (210), or may be formed integrally with the side bezel structure (210). The first support member (211) may be formed, for example, from a metal material and / or a non-metal (e.g., polymer) material. The first support member (211) may have a display (230) attached to one side and a printed circuit board (240) attached to the other side. The printed circuit board (240) may be equipped with a processor, memory, and / or an interface. The processor may include, for example, one or more of a central processing unit, an application processor, a graphics processing unit, an image signal processor, a sensor hub processor, or a communication processor.

[0046] Memory may include, for example, volatile memory or non-volatile memory.

[0047] The interface may include, for example, an HDMI (high definition multimedia interface), a USB (universal serial bus) interface, an SD card interface, and / or an audio interface. The interface may, for example, electrically or physically connect the electronic device (100) to an external electronic device and may include a USB connector, an SD card / MMC connector, or an audio connector.

[0048] The battery (250) is a device for supplying power to at least one component of the electronic device (100) and may include, for example, a non-rechargeable primary battery, a rechargeable secondary battery, or a fuel cell. At least a portion of the battery (250) may be disposed substantially coplanar with, for example, the printed circuit board (240). The battery (250) may be disposed integrally within the electronic device (100) or may be disposed detachably from the electronic device (100). For example, the battery (250) may include a micro-controller unit (MCU). The micro-controller included in the battery (250) can process various data related to the state of the battery (250) and execute various commands related to the battery (250). For example, a microcontroller included in the battery (250) may control other configurations so that at least one operation described in the present disclosure is performed to detect the risk of swelling of the battery (250).

[0049] An antenna (270) may be positioned between the rear plate (280) and the battery (250). The antenna (270) may include, for example, a near field communication (NFC) antenna, a wireless charging antenna, and / or a magnetic secure transmission (MST) antenna. The antenna (270) may, for example, communicate near-field with an external device or wirelessly transmit and receive power required for charging. In other embodiments, the antenna structure may be formed by a part or combination thereof of the side bezel structure (210) and / or the first support member (211). For example, the antenna (270) may be positioned to overlap with the battery (250).

[0050] A pen input device (201) (e.g., a stylus pen) can be guided into the interior of the housing through a hole (202) formed on the side of the housing of the electronic device (100) and inserted or removed, and may include a button to facilitate removal. The pen input device (201) may have a separate resonant circuit built in and may be coupled with an electromagnetic induction panel (290) (e.g., a digitizer) included in the electronic device (100). The pen input device (201) may include an EMR (electro-magnetic resonance) method, an AES (active electrical stylus) method, and an ECR (electric coupled resonance) method.

[0051] FIG. 3 is a flowchart illustrating the process of an electronic device (100) according to one embodiment detecting the risk of swelling of a battery.

[0052] In the following embodiments, each operation may be performed sequentially, but is not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. Additionally, some operations may be omitted.

[0053] Referring to operation 310, an electronic device (100) according to one embodiment may measure the temperature of a battery through at least one temperature sensor that is in contact with or adjacent to the battery. The sensor for measuring the temperature of the battery may be attached to or placed adjacent to the surface or interior of the battery. Details regarding the temperature of the battery have been explained in detail prior to the detailed description of FIG. 1, so they will be omitted here.

[0054] Referring to operation 320, the electronic device (100) can determine at least one of the battery voltage or battery current as it is determined that the battery temperature has a value greater than or equal to the reference temperature for a period longer than the threshold time. The electronic device (100) can obtain a value for at least one of the battery voltage or battery current as it is determined that the battery temperature has a value greater than or equal to the reference temperature for a period longer than the threshold time. For example, the electronic device (100) can measure at least one of the battery voltage or battery current as it is determined that the battery temperature has a value greater than or equal to the reference temperature for a period longer than the threshold time. For example, the electronic device (100) can periodically measure at least one of the battery voltage or battery current. In one embodiment, when the electronic device (100) determines that the battery temperature has a value greater than or equal to the reference temperature for a period longer than the threshold time, it can obtain information related to at least one of the battery voltage or battery current along with information related to the battery temperature. For example, the electronic device (100) may periodically measure the battery voltage and battery current for a set time when it is determined that the battery has entered a swelling risk condition. For example, the reference temperature and the threshold time may be set values. For example, the reference temperature may be 35-45 degrees Celsius, and preferably approximately 40 degrees.

[0055] Referring to operation 330, the electronic device (100) can obtain the rate of change over time of the battery voltage and battery current, respectively, from battery voltage data and battery current data obtained during a set time. For example, the electronic device (100) can calculate the rate of change over time of the battery voltage (dv / dt) and the rate of change over time of the battery current (di / dt). Since the details regarding the operation of obtaining the rate of change over time for the battery voltage and battery current, respectively, have been explained in detail in the detailed description section of FIG. 4, they will be omitted here.

[0056] Referring to operation 340, the electronic device (100) can generate a spectrogram (120) in the time-frequency domain based on the rate of change over time of the battery voltage and battery current, respectively. The spectrogram (120) can be generated based on data obtained through a short-time Fourier transform (STFT). The spectrogram (120) may be data that visually represents the change in frequency components over time of the battery voltage and battery current. Details regarding the operation of generating the spectrogram (120) have been explained in detail in the detailed description section of FIGS. 5 and 6, so they will be omitted here.

[0057] Referring to operation 350, the electronic device (100) may obtain at least one of additional supplementary information related to the state of the battery, such as the battery charging method (e.g., wired charging, or wireless charging), charging speed, State of Health (SOH), or a state indicator collected within the battery system. The obtained supplementary information may be applied to an artificial intelligence model (101) together with a spectrogram (120) to increase the accuracy of the battery anomaly analysis.

[0058] Referring to operation 360, the electronic device (100) can obtain an output value related to the swelling risk of the battery by inputting the generated spectrogram (120) and additional information into the artificial intelligence model (101). For example, the artificial intelligence model (101) may be configured based on CNN and can classify normal / abnormal states through learning or quantify and output the swelling risk. Details regarding the operation of obtaining the output value related to the swelling risk of the battery are explained in detail in the detailed description section of FIGS. 9 and 10, so they will be omitted here.

[0059] According to one embodiment, the electronic device (100) may include at least one of a processor or a microcontroller. The electronic device (100) may obtain an output value related to the risk of battery swelling from an artificial intelligence model by applying a spectrogram or additional information to the artificial intelligence model under the control of at least one of the processor or the microcontroller. For example, the microcontroller may correspond to a processor (e.g., the processor (1220) of FIG. 12). For example, the microcontroller may be at least part of the processor. For example, the microcontroller may be included as a circuit that is at least part of the processor (e.g., the processor (1220) of FIG. 12). For example, the microcontroller may be included in a power management module (e.g., the power management module (1228) of FIG. 12). For example, the microcontroller may be at least part of the control circuit within the power management module (e.g., the power management module (1228) of FIG. 12). However, the entity performing the operations according to one embodiment is not limited thereto.

[0060] FIG. 4 is a flowchart illustrating the operation of repeatedly obtaining the rate of change of battery voltage and the rate of change of battery current in an electronic device (100) according to one embodiment.

[0061] The description related to operation 410 in Fig. 4 may correspond to the description related to operation 330 in Fig. 3. Below, in relation to this, the details regarding the process of repeatedly obtaining the rate of change of battery voltage and the rate of change of battery current will be explained in more detail.

[0062] Referring to operation 410, the electronic device (100) can obtain the rate of change of battery voltage and the rate of change of battery current by repeating them a set number of times. The rate of change of battery voltage and the rate of change of battery current may represent amounts of change over time, and may represent the rate of change of battery voltage over time (dv / dt) and the rate of change of battery current over time (di / dt), respectively.

[0063] In one embodiment, the electronic device (100) can acquire data related to battery voltage and data related to battery current at regular time intervals. The electronic device (100) can acquire a plurality of battery voltage values ​​and a plurality of battery current values ​​at regular time intervals. The electronic device (100) can sample the acquired data. For example, the electronic device (100) can calculate the difference between two battery voltage values ​​acquired in two adjacent time intervals and acquire the result of the calculation as the rate of change of battery voltage. For example, the electronic device (100) can calculate the difference between two battery current values ​​acquired in two adjacent time intervals and acquire the result of the calculation as the rate of change of battery current. The electronic device (100) can acquire a plurality of values ​​related to the rate of change of battery voltage by repeating this calculation of the rate of change of battery voltage a certain number of times. Based on the plurality of values ​​related to the rate of change of battery voltage, the electronic device (100) can acquire time domain data related to the rate of change of battery voltage. The electronic device (100) can obtain multiple values ​​related to the rate of change of battery current by repeating the calculation of the rate of change of battery current a certain number of times. Based on the multiple values ​​related to the rate of change of battery current, the electronic device (100) can obtain time domain data related to the rate of change of battery current.

[0064] For example, the set number of times can be set according to the setting value of the electronic device (100), and the balance of analysis resolution or processing time can be adjusted according to the increase or decrease in the number of repetitions.

[0065] According to one embodiment, the period or unit time for measuring the voltage and current of a battery may be set differently depending on at least one of the state or type of the electronic device (100). For example, when the application processor (AP) (e.g., the processor (1220) of FIG. 12) is in an active state (awaken state), since changes in battery voltage and battery current are likely to occur relatively quickly, periodic measurements may be performed at first cycles. In this case, the unit time for obtaining the rate of change of each of the battery voltage and battery current may be the first cycle. In one embodiment, the state of the electronic device (100) may be a low-activity state, such as a sleep state. In this case, measurements for each of the battery voltage and battery current may be performed at second cycles, which are relatively longer than the first cycle. In one embodiment, by adjusting the measurement cycle of the battery voltage and battery current in this way, a battery swelling detection method that is advantageous in terms of optimizing power consumption and securing data efficiency can be provided. Alternatively, the unit time or measurement cycle may be set in various ways based on at least one of the charging speed of the battery of the electronic device (100), the charge state of the battery, the SOH of the battery, or the state of the processor of the electronic device.

[0066] FIG. 5 is a flowchart illustrating the process of converting to time-frequency domain data and generating a spectrogram based on changes in battery voltage and current in an electronic device (100) according to one embodiment.

[0067] Referring to operation 510, the electronic device (100) can convert data in the time domain into data in the time-frequency domain through a short-time fourier transform (STFT) based on data of the rate of change of battery voltage and the rate of change of battery current obtained in the previous step. The electronic device (100) can obtain data that simultaneously represents time and frequency information by performing a fourier transform on the data obtained at each unit time through the STFT. In one embodiment, the electronic device (100) can obtain data that simultaneously represents time and frequency information related to the rate of change of battery voltage and the rate of change of battery current by performing a fourier transform on the data related to the rate of change of battery voltage and the rate of change of battery current obtained at each unit time through the STFT.

[0068] STFT is the defined window function η(t- Based on ), it can be performed for continuous signals or discrete signals according to the following Equation 1 or Equation 2, respectively. At this time, η can represent the time position or time center to be analyzed, and can represent the center time of a sliding window. For example, in a window function, η can include a Hamming window, a Gaussian window, or a rectangular window.

[0069] STFT for a continuous signal can be performed based on Equation 1 below.

[0070] [Formula 1]

[0071]

[0072] STFT for discrete signals can be performed based on Equation 2 below.

[0073] [Equation 2]

[0074]

[0075] Through a transformation such as Equation 1 or Equation 2, data related to the rate of change of battery voltage and battery current over time, respectively, can be processed into data based on two axes called the time axis and the frequency axis.

[0076] For example, the input data used when performing STFT consists of 2 channels (battery voltage and battery current), and since each channel consists of 128 time-series data points for each of the two measurement times, the total input can have a dimension of 2 x (2 x 128). For this time-series data, the window function η(t- Local analysis by interval can be performed by applying ). Depending on the measurement time or measurement point, the size, composition, or dimensions of the entire input data may vary.

[0077] Referring to operation 520, the electronic device (100) can generate a spectrogram based on time-frequency domain data transformed by STFT. The spectrogram is two-dimensional data, where the horizontal axis represents time, the vertical axis represents frequency, and the color or shade value represents the magnitude of the signal strength. In one embodiment, the output data resulting from the STFT execution may include a 2 x (2 x 16 x 8) dimensional spectrogram (e.g., the spectrogram (120) of FIG. 6). The generated spectrogram (120) may be configured along the time and frequency axes.

[0078] For example, the generated spectrogram can be input into an artificial intelligence model in a subsequent step and utilized as a feature map for determining the risk of battery swelling.

[0079] FIG. 6 is a drawing showing an example of a spectrogram (120) generated in an electronic device (100) according to one embodiment.

[0080] FIG. 6 shows an example of a spectrogram (120) that visually represents data in the time-frequency domain generated based on the rate of change of battery voltage and the rate of change of battery current. As described in FIG. 5, the electronic device (100) can generate data in the time-frequency domain by performing a Short-Time Fourier Transform (STFT) based on the rate of change of battery voltage and the rate of change of battery current, and can generate a spectrogram (120) based thereon. Referring to the spectrogram (120) shown in FIG. 6, for example, the spectrogram (120) has a horizontal axis representing time (time, [sec]) and a vertical axis representing frequency (frequency, [Hz]), and can represent the signal strength of the frequency band corresponding to each time interval by the intensity of color or shading. For example, in the spectrogram (120) shown in FIG. 6, a light color may represent a relatively low signal strength, and a dark color may represent a relatively high signal strength.

[0081] The spectrogram (120) can be an object of analysis that can be input into an artificial intelligence-based classification model, such as a convolutional neural network (CNN). In one embodiment, the CNN processes the spectrogram (120) as a single two-dimensional image to learn the spatial features of the image (e.g., a pattern of repeated increase in intensity in a specific frequency band, or irregular fluctuations in amplitude), and based on this, can classify or predict the risk of battery swelling. For example, the CNN can detect anomalies at a specific point in time that appear repeatedly in low-frequency or high-frequency regions through filters in a manner similar to that used in general image recognition fields (e.g., a continuous increase in high-frequency components) and use this as a classification criterion.

[0082] According to one embodiment, an artificial intelligence model (101) can be trained to obtain an output value related to the swelling risk of a battery by learning the time-by-time / frequency-by-frequency spectral distribution based on spectrograms (120) obtained for a plurality of types of batteries through multi-layer accumulation. Accordingly, according to one embodiment, by using the frequency configuration of a time-series signal as an input value instead of a quantitative value, a method can be provided to determine the swelling risk of a battery more accurately and quickly despite changes in the external environment or measurement noise.

[0083] FIG. 7 is a flowchart illustrating the process of an electronic device (100) according to one embodiment performing normalization on data included in a spectrogram (120).

[0084] As described in FIG. 6, the electronic device (100) can generate data in the time-frequency domain based on the voltage and current change rates of the battery and obtain a spectrogram (120) that visualizes this data. FIG. 7 shows the content related to the normalization process as a preprocessing process for effectively inputting the data included in the spectrogram (120) into a neural network-based artificial intelligence model.

[0085] Referring to operation 710, the electronic device (100) can calculate statistical values ​​for the entire data distribution, such as the mean and variance, for the data included in the spectrogram (120). The mean and variance are calculated based on the intensity of the frequency components at each time point constituting the spectrogram and may be global statistical values ​​for the entire spectrogram image.

[0086] Referring to operation 720, the electronic device (100) can perform normalization on each data value of the spectrogram (120) using the calculated mean and variance values. In this case, normalization can generally be performed through the following formula.

[0087] The normalization process can be performed based on the mean and variance at the data sample level. For example, each sample Average for and variance It can be obtained by the following Equations 3 and 4.

[0088] [Equation 3]

[0089]

[0090] [Equation 4]

[0091]

[0092] Afterwards, each sample can be normalized according to Equation 5 below.

[0093] [Formula 5]

[0094]

[0095] Normalized values ​​are learning parameters and Using this, it can be scaled and shifted according to Equation 6 below.

[0096] [Equation 6]

[0097]

[0098] Input As the output for, can be obtained. is the normalized input value The scaling factor is Multiply by, and the shifting coefficient is to the result It can be obtained by adding. For example, scaling coefficient and shifting coefficients can be a parameter learned depending on the situation. For example, scaling coefficient and shifting coefficients is a parameter learned based on the type of battery, and may have different values ​​depending on the type of battery.

[0099] Accordingly, learning stability can be ensured and the distribution of input data can be made uniform. In other words, through the normalization process, the distribution of input data is transformed into a normal distribution with a mean of 0 and a variance of 1, which can improve the learning efficiency and convergence speed of the artificial intelligence model.

[0100] As a data preprocessing for battery condition analysis, normalizing the time-frequency domain data included in the spectrogram (120) based on statistics can provide the effect of improving the quality and interpretability of the data to be input into the artificial intelligence model.

[0101] This normalized data can be input into a convolutional neural network (CNN) or other classification models to effectively learn patterns within spectrograms and be used to classify or predict the risk of battery swelling.

[0102] FIG. 8 is a diagram showing an example in which an electronic device (100) according to one embodiment performs a normalization process.

[0103] As described in FIG. 7, the electronic device (100) can calculate the mean and variance of the data included in the spectrogram (120) and perform normalization based thereon. FIG. 8 shows an example to explain how this normalization process is performed.

[0104] Referring to FIG. 8, input data for normalization can be organized in batch units. Referring to the example in FIG. 8, input data organized in the form of a 6-row, 3-column array is used, and each column may correspond to a different frequency channel or filter of the spectrogram.

[0105] For each column, the electronic device (100) can calculate the mean and standard deviation of the corresponding column, respectively. For example, referring to FIG. 8, the first column consists of [1, 2, 0, 4, 5, 1], and the mean and standard deviation are calculated as 2. In the same way, the mean and standard deviation of the second column are 3 and 2, respectively, and the mean and standard deviation of the third column are 3 and 2, respectively.

[0106] Afterwards, the electronic device (100) can perform normalization for each element according to Equation 3, Equation 4 and Equation 5.

[0107] Referring to Figure 8, for example, the first value in the first column is 1, and the normalized value by applying a mean of 2 and a standard deviation of 2 can be (1-2) / 2=0.5(1-2) / 2=-0.5(1-2) / 2=-0.5. When the entire data is normalized in this way, each column will have a distribution with a mean of 0 and a standard deviation of 1.

[0108] The normalization process illustrated in FIG. 8 can be performed as a preprocessing step to uniformly scale the data input to a CNN or other artificial intelligence model by an electronic device (100) according to one embodiment, thereby increasing learning stability, improving the convergence speed of the model, and improving the accuracy of battery state prediction.

[0109] Meanwhile, the normalization according to one embodiment can be implemented as a type of layer normalization or batch normalization performed on a batch basis, and can be applied in various ways depending on the characteristics of the input data and the structure of the applied artificial intelligence model.

[0110] In the subsequent steps, classification or regression analysis for predicting swelling risk can be performed based on the normalized data. The operations related to this will be described in detail in the description section of Figures 9 and 10.

[0111] FIG. 9 is a diagram illustrating the process of obtaining an output value related to the swelling risk of a battery by inputting normalized spectrogram data into an artificial intelligence model (101) by an electronic device (100) according to one embodiment.

[0112] Referring to FIG. 9, an electronic device (100) according to one embodiment may provide data of a spectrogram (120) normalized in FIG. 8 as input to an artificial intelligence model (101). The spectrogram (120) is data converted into the time-frequency domain based on the rate of change of battery voltage and current as described in FIG. 6 and FIG. 7, and is preprocessed into a form that facilitates feature extraction in a CNN-based artificial intelligence model.

[0113] In one embodiment, the artificial intelligence model (101) may be a deep learning model based on a convolutional neural network (CNN). The spectrogram (120) is composed of a multidimensional array, and since each pixel of this array represents signal energy (or its magnitude) in a specific time and frequency interval, it has a structure similar to a two-dimensional image used in image classification or object recognition. Due to these structural characteristics, the CNN model can effectively learn local patterns on the spectrogram.

[0114] The input layer of the artificial intelligence model (101) is configured to accept normalized spectrogram data as is and includes one or more convolution layers, pooling layers, normalization layers and fully connected layers, and can extract and synthesize features that have a high correlation with the risk of swelling within the input data.

[0115] The artificial intelligence model (101) can generate an output value corresponding to the risk of swelling of the battery as a result of analyzing input spectrogram data. The output value may include, for example, a binary classification value regarding whether the battery is swelling (e.g., 0: normal / 1: risk of swelling), a continuous value regarding the risk of swelling of the battery (e.g., a risk index within the range of 0.0-1.0), or numerical information regarding the predicted time or expected time until swelling occurs.

[0116] An artificial intelligence model (101) according to one embodiment may be trained in advance based on a plurality of battery data. The training process of the artificial intelligence model (101) will be described in detail in FIG. 11, which will be described later. In addition, in addition to spectrograms, battery state information may be additionally provided as input to the artificial intelligence model (101). This will be described in detail in FIG. 10, which will be described later.

[0117] FIG. 10 is a diagram illustrating the process of obtaining an output value related to the swelling risk of a battery by inputting additional information along with a spectrogram into an artificial intelligence model (101) of an electronic device (100) according to one embodiment.

[0118] For example, with reference to FIG. 10, an electronic device (100) according to one embodiment may provide additional information related to the state of the battery as input to an artificial intelligence model (101), in addition to the spectrogram (120) described in FIG. 9. In this way, by considering other additional information together with the spectrogram (120), a more precise and reliable judgment can be made through the artificial intelligence model (101).

[0119] In one embodiment, additional information may include various information related to the overall physical and environmental conditions of the battery, such as information related to the state of the battery or charging history information. For example, the additional information may include information related to the state of health (SOH) of the battery. For example, the additional information may include indicators representing the remaining lifespan or degradation state of the battery. In addition, the type of battery (e.g., lithium-ion, lithium-polymer, etc.) or the history of temperature changes of the battery may also be included as state information. Since such information can reflect the degree of battery aging and sensitivity under specific conditions, it can contribute to improving the accuracy of the assessment of swelling risk.

[0120] In one embodiment, additional information may also include battery charging history information. For example, charging speed (e.g., fast charging, normal charging, or slow charging), charging method (e.g., wired charging, or wireless charging), the cumulative number of charges, number of discharges, or DOD (depth of discharge) of the battery may be used as additional information. Such charging history information is closely related to the accumulation of internal chemical stress in the battery and can serve as a useful indicator for inferring the risk of swelling during long-term use.

[0121] According to one embodiment, additional information may be provided to an artificial intelligence model (101) as an input vector separate from the spectrogram, and the artificial intelligence model (101) may accept this as an auxiliary input and reflect it in the learning and prediction process. For example, even when the same spectrogram is input, the artificial intelligence model may be trained to calculate a higher output value for the swelling risk when the SOH value provided as additional information is low, the number of charging cycles is excessively high, or the DOD is large. In this way, the time-frequency characteristics captured in the spectrogram and the state context of the battery provided in the additional information are comprehensively considered, making it possible to make a more situation-adaptive judgment.

[0122] According to one embodiment, an electronic device (100) may communicate with a server (e.g., server (1208) of FIG. 12) to receive an artificial intelligence model (101) trained to suit the type of battery or state of health (SOH). In this case, the artificial intelligence model (101) previously stored in the electronic device (100) may be replaced with an updated model. For example, the electronic device (100) may selectively receive a model pre-trained to suit the capacity of the battery or a battery with a specific charging pattern.

[0123] According to one embodiment, the risk of swelling can be accurately predicted by considering additional environmental or historical information of the battery together with the spectrogram (120). Accordingly, the possibility of early detection of the risk of swelling can be increased, and a highly reliable battery management system can be implemented.

[0124] FIG. 11 is a diagram illustrating a convolutional neural network (CNN)-based learning and inference process performed by an artificial intelligence model (101) according to one embodiment to generate an output related to battery swelling risk.

[0125] Figure 11 visually illustrates a series of structures in which a CNN performs normalization and learning using, for example, spectrograms and additional information as inputs, and finally generates an output that determines the swelling risk. Referring to Figure 11, for example, the CNN structure may be composed of a total of three convolutional layers.

[0126] In one embodiment, operation 1110 may represent a processing operation for an input layer. For example, spectrogram data obtained through FIGS. 6 to 10 above may be input to the input layer in the form of a multidimensional matrix. In one embodiment, two-dimensional data may be input to the input layer. For example, the data input to the input layer may be data from multiple time points arranged based on frequency, and may include a two-dimensional numerical data structure for at least one of signal strength or energy density. For example, two-dimensional numerical data composed of a time-frequency domain may be included in the spectrogram data. For example, the spectrogram representing the time-frequency domain for data from multiple time points may generally have a structure similar to a two-dimensional image. Each data element (e.g., a value corresponding to the brightness or color of each pixel in the image) may represent at least one of the energy density or the magnitude of the STFT coefficient in a specific time interval and frequency band. In one embodiment, a region corresponding to each coordinate in a spectrogram having the form of a two-dimensional image may correspond to each data element included in the spectrogram. For example, each data element included in the spectrogram may represent the brightness or color of the region corresponding to each coordinate in the spectrogram having the form of a two-dimensional image. For example, each data element may correspond to at least one value among the energy density or the magnitude of the STFT coefficient in each time interval and frequency band. For example, the input layer may be configured in the form of a single channel (gray scale) or a multi-channel image (RGB, etc.). For example, the input layer may be configured to have the voltage change rate and current change rate of the battery as separate channels, or one channel may include voltage, another channel may include current, yet another channel may include the temperature change rate, etc.The first convolutional layer (1110) uses 60 kernels of size 2x2, and zero-padding may be applied. For example, the output of the first convolutional layer (1110) may have dimensions of 60 x 7 x 15. For example, a ReLU activation function may be applied, and the Dropout rate may be set to 0.6.

[0127] In one embodiment, operation 1120 may represent a feature extraction operation including multiple convolutional layers. In operation 1120, features related to at least one of a local pattern, edge, frequency transition, or repetitive oscillation pattern of an input spectrogram image may be repeatedly extracted through a filter. For example, in the first convolutional layer (1110), mainly low-dimensional features (e.g., linear patterns, boundaries, or frequency band distributions, etc.) may be extracted, and in subsequent layers, more complex and high-dimensional patterns (e.g., specific current noise patterns, specific periodicity changes, or unique frequency distributions appearing during fast charging, etc.) may be extracted hierarchically. These filters may be automatically adjusted through learning and may be trained to respond sensitively to various battery conditions. For example, the second convolutional layer (1120) may use the same 2x2 kernel but may have 30 filters applied, may have the same ReLU activation function applied as the first convolutional layer (1110), and may have the Dropout rate set to 0.6.

[0128] In one embodiment, operation 1130 may represent a feature compression operation including a pooling layer and an optional normalization layer. For example, the pooling layer may reduce computational load by reducing the size of the feature map through max pooling or avg pooling, and may provide invariance to local positional changes. For example, compression may be performed so that the CNN can stably recognize changes occurring in a specific frequency band even if they shift slightly over time. For example, the third convolutional layer (1130) may have a kernel of size 1x2 and may be composed of five filters. For example, a ReLU activation function may be applied to the third convolutional layer (1130), and the Dropout rate may be set to 0.6. Accordingly, overfitting may be prevented.

[0129] In one embodiment, operation 1140 may represent a fully connected layer operation. In the fully connected layer, high-dimensional features extracted in the previous step are unfolded into a vector form and then processed into a form suitable for final classification or regression tasks through a multilayer perceptron structure. At this time, the weights of the artificial neurons can be iteratively updated through a backpropagation algorithm based on the error between the output swelling risk prediction value and the correct value. For example, while learning from a large amount of normal or abnormal battery data, it may learn on its own that the spectrogram shape showing excessive current oscillation during fast charging is correlated with battery swelling.

[0130] Operation 1150 may correspond to an operation of creating an output layer and may represent an operation of obtaining a final classification result. In one embodiment, it may operate in a binary classification manner that outputs one of a "Normal" class (1151) and an "Abnormal" class (1152) in relation to the risk of swelling of the battery. Alternatively, depending on the implementation method, the risk level may be classified into multi-classes such as "low," "medium," and "high," and the output layer may be configured to calculate probability values ​​for each class through a softmax function, etc. Alternatively, it may be configured to predict the risk level as a continuous value between 0 and 1 as a regression-based output.

[0131] FIG. 12 is a block diagram of an electronic device (1501) in a network environment (1500) according to various embodiments.

[0132] The electronic device (100) of the present disclosure may correspond to the electronic device (1201) of FIG. 12 and may include components that may be included in the electronic device (1201) disclosed in FIG. 12. However, the electronic device (100) of the present disclosure is not required to include all components of FIG. 12, and may include only some components of FIG. 12.

[0133] The battery of the present disclosure may correspond to the battery (250) of FIG. 2 or the battery (1289) of FIG. 12, and may include configurations that may be included in the battery (250) of FIG. 2 or the battery (1289) of FIG. 12. However, the battery disclosed throughout the present disclosure must not include all configurations of FIG. 2 or FIG. 12, nor must it have all features disclosed in FIG. 2 or FIG. 12. The battery disclosed throughout the present disclosure may include only some configurations of the battery (250) and battery (1289) disclosed in FIG. 2 or FIG. 12.

[0134] Referring to FIG. 12, in a network environment (1200), an electronic device (1201) may communicate with an electronic device (1202) through a first network (1298) (e.g., a short-range wireless communication network) or with at least one of an electronic device (1204) or a server (1208) through a second network (1299) (e.g., a long-range wireless communication network). According to one embodiment, the electronic device (1201) may communicate with the electronic device (1204) through a server (1208). According to one embodiment, the electronic device (1201) may include a processor (1220), memory (1230), input module (1250), sound output module (1255), display module (1260), audio module (1270), sensor module (1276), interface (1277), connection terminal (1278), haptic module (1279), camera module (1280), power management module (1288), battery (1289), communication module (1290), subscriber identification module (1296), or antenna module (1297). In some embodiments, at least one of these components (e.g., connection terminal (1278)) may be omitted from the electronic device (1201), or one or more other components may be added. In some embodiments, some of these components (e.g., sensor module (1276), camera module (1280), or antenna module (1297)) may be integrated into a single component (e.g., display module (1260)).

[0135] The processor (1220) can, for example, execute software (e.g., program (1240)) to control at least one other component (e.g., hardware or software component) of the electronic device (1201) connected to the processor (1220) and perform various data processing or operations. According to one embodiment, as at least part of the data processing or operations, the processor (1220) can store commands or data received from other components (e.g., sensor module (1276) or communication module (1290)) in volatile memory (1232), process the commands or data stored in volatile memory (1232), and store the resulting data in non-volatile memory (1234). According to one embodiment, the processor (1220) may include a main processor (1221) (e.g., a central processing unit or an application processor) or an auxiliary processor (1223) that can operate independently or together with it (e.g., a graphics processing unit, a neural processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor). For example, if the electronic device (1201) includes a main processor (1221) and an auxiliary processor (1223), the auxiliary processor (1223) may be configured to use less power than the main processor (1221) or to be specialized for a specified function. The auxiliary processor (1223) may be implemented separately from the main processor (1221) or as part thereof.

[0136] The auxiliary processor (1223) may control at least some of the functions or states associated with at least one component of the electronic device (1201) (e.g., display module (1260), sensor module (1276), or communication module (1290)) on behalf of the main processor (1221) while the main processor (1221) is in an inactive (e.g., sleep) state, or together with the main processor (1221) while the main processor (1221) is in an active (e.g., application execution) state. According to one embodiment, the auxiliary processor (1223) (e.g., image signal processor or communication processor) may be implemented as part of another functionally related component (e.g., camera module (1280) or communication module (1290)). According to one embodiment, the auxiliary processor (1223) (e.g., neural network processing unit) may include a hardware structure specialized for processing an artificial intelligence model. The artificial intelligence model may be generated through machine learning. Such learning may be performed, for example, on the electronic device (1201) itself where the artificial intelligence model is executed, or through a separate server (e.g., server (1208)). The learning algorithm may include, for example, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but is not limited to the examples described above. The artificial intelligence model may include a plurality of artificial neural network layers.An artificial neural network may be a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted Boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), a deep Q-network, or a combination of two or more of the above, but is not limited to the examples described above. In addition to the hardware structure, the artificial intelligence model may include a software structure, either additionally or substantially.

[0137] The memory (1230) can store various data used by at least one component of the electronic device (1201) (e.g., processor (1220) or sensor module (1276)). The data may include, for example, software (e.g., program (1240)) and input or output data for related commands. The memory (1230) may include volatile memory (1232) or non-volatile memory (1234).

[0138] The program (1240) may be stored as software in memory (1230) and may include, for example, an operating system (1242), middleware (1244), or an application (1246).

[0139] The input module (1250) can receive commands or data to be used for a component of the electronic device (1201) (e.g., processor (1220)) from outside the electronic device (1201) (e.g., user). The input module (1250) may include, for example, a microphone, a mouse, a keyboard, a key (e.g., a button), or a digital pen (e.g., a stylus pen).

[0140] The sound output module (1255) can output a sound signal to the outside of the electronic device (1201). The sound output module (1255) may include, for example, a speaker or a receiver. The speaker may be used for general purposes, such as multimedia playback or recording playback. The receiver may be used to receive incoming calls. According to one embodiment, the receiver may be implemented separately from the speaker or as part thereof.

[0141] The display module (1260) can visually provide information to an external (e.g., user) of the electronic device (1201). The display module (1260) may include, for example, a display, a holographic device, or a projector and a control circuit for controlling said device. According to one embodiment, the display module (1260) may include a touch sensor configured to detect a touch, or a pressure sensor configured to measure the intensity of the force generated by said touch.

[0142] The audio module (1270) can convert sound into an electrical signal or, conversely, convert an electrical signal into sound. According to one embodiment, the audio module (1270) can acquire sound through the input module (1250) or output sound through the sound output module (1255) or an external electronic device (e.g., electronic device (1202)) (e.g., speaker or headphones) connected directly or wirelessly to the electronic device (1201).

[0143] The sensor module (1276) can detect the operating state of the electronic device (1201) (e.g., power or temperature) or the external environmental state (e.g., user state) and generate an electrical signal or data value corresponding to the detected state. According to one embodiment, the sensor module (1276) may include, for example, a gesture sensor, a gyroscope sensor, a barometric pressure sensor, a magnetic sensor, an accelerometer sensor, a grip sensor, a proximity sensor, a color sensor, an IR (infrared) sensor, a biosensor, a temperature sensor, a humidity sensor, or an illuminance sensor.

[0144] The interface (1277) may support one or more specified protocols that can be used for the electronic device (1201) to be connected directly or wirelessly to an external electronic device (e.g., electronic device (1202)). According to one embodiment, the interface (1277) may include, for example, a high definition multimedia interface (HDMI), a universal serial bus (USB) interface, an SD card interface, or an audio interface.

[0145] The connection terminal (1278) may include a connector through which the electronic device (1201) can be physically connected to an external electronic device (e.g., electronic device (1202)). According to one embodiment, the connection terminal (1278) may include, for example, an HDMI connector, a USB connector, an SD card connector, or an audio connector (e.g., a headphone connector).

[0146] The haptic module (1279) can convert an electrical signal into a mechanical stimulus (e.g., vibration or movement) or an electrical stimulus that the user can perceive through tactile or kinesthetic senses. According to one embodiment, the haptic module (1279) may include, for example, a motor, a piezoelectric element, or an electric stimulation device.

[0147] The camera module (1280) can capture still images and video. According to one embodiment, the camera module (1280) may include one or more lenses, image sensors, image signal processors, or flashes.

[0148] The power management module (1288) can manage the power supplied to the electronic device (1201). According to one embodiment, the power management module (1288) can be implemented, for example, as at least part of a power management integrated circuit (PMIC).

[0149] The battery (1289) can supply power to at least one component of the electronic device (1201). According to one embodiment, the battery (1289) may include, for example, a non-rechargeable primary battery, a rechargeable secondary battery, or a fuel cell.

[0150] The communication module (1290) can support the establishment of a direct (e.g., wired) communication channel or a wireless communication channel between an electronic device (1201) and an external electronic device (e.g., electronic device (1202), electronic device (1204), or server (1208)), and the performance of communication through the established communication channel. The communication module (1290) may include one or more communication processors that operate independently of the processor (1220) (e.g., application processor) and support direct (e.g., wired) communication or wireless communication. According to one embodiment, the communication module (1290) may include a wireless communication module (1292) (e.g., cellular communication module, short-range wireless communication module, or GNSS (global navigation satellite system) communication module) or a wired communication module (1294) (e.g., LAN (local area network) communication module, or power line communication module). The corresponding communication module among these communication modules can communicate with an external electronic device (1204) through a first network (1298) (e.g., a short-range communication network such as Bluetooth, WiFi (wireless fidelity) direct, or IrDA (infrared data association)) or a second network (1299) (e.g., a legacy cellular network, a 5G network, a next-generation communication network, the Internet, or a computer network (e.g., a LAN or WAN)). These various types of communication modules may be integrated into a single component (e.g., a single chip) or implemented as multiple separate components (e.g., multiple chips). The wireless communication module (1292) can identify or authenticate the electronic device (1201) within a communication network such as the first network (1298) or the second network (1299) using subscriber information (e.g., International Mobile Subscriber Identifier (IMSI)) stored in the subscriber identification module (1296).

[0151] The wireless communication module (1292) can support 5G networks and next-generation communication technologies following 4G networks, for example, new radio access technology. NR access technology can support high-speed transmission of high-capacity data (enhanced mobile broadband (eMBB)), minimization of terminal power and connection of multiple terminals (massive machine type communications (mMTC)), or high reliability and low latency (ultra-reliable and low-latency communications (URLLC)). The wireless communication module (1292) can support a high-frequency band (e.g., mmWave band) to achieve a high data transmission rate, for example. The wireless communication module (1292) can support various technologies for securing performance in the high-frequency band, such as beamforming, massive MIMO (multiple-input and multiple-output), full-dimensional MIMO (FD-MIMO), array antenna, analog beam-forming, or large-scale antenna. The wireless communication module (1292) can support various requirements specified in the electronic device (1201), external electronic device (e.g., electronic device (1204)), or network system (e.g., second network (1299)). According to one embodiment, the wireless communication module (1292) can support a Peak data rate (e.g., 20 Gbps or more) for realizing eMBB, loss coverage (e.g., 164 dB or less) for realizing mMTC, or U-plane latency (e.g., downlink (DL) and uplink (UL) each 0.5 ms or less, or round trip 1 ms or less) for realizing URLLC.

[0152] An antenna module (1297) can transmit a signal or power to or from an external source (e.g., an external electronic device). According to one embodiment, the antenna module (1297) may include an antenna comprising a radiator made of a conductor or a conductive pattern formed on a substrate (e.g., a PCB). According to one embodiment, the antenna module (1297) may include a plurality of antennas (e.g., an array antenna). In this case, at least one antenna suitable for a communication method used in a communication network, such as a first network (1298) or a second network (1299), may be selected from the plurality of antennas, for example, by a communication module (1290). A signal or power may be transmitted or received between the communication module (1290) and an external electronic device through the selected at least one antenna. According to some embodiments, in addition to the radiator, other components (e.g., a radio frequency integrated circuit (RFIC)) may be additionally formed as part of the antenna module (1297).

[0153] According to various embodiments, the antenna module (1297) may form a mmWave antenna module. According to one embodiment, the mmWave antenna module may include a printed circuit board, an RFIC disposed on or adjacent to a first surface (e.g., bottom surface) of the printed circuit board and capable of supporting a specified high frequency band (e.g., mmWave band), and a plurality of antennas (e.g., array antennas) disposed on or adjacent to a second surface (e.g., top surface or side surface) of the printed circuit board and capable of transmitting or receiving a signal of the specified high frequency band.

[0154] At least some of the above components can be connected to each other via a communication method between peripheral devices (e.g., bus, GPIO (general purpose input and output), SPI (serial peripheral interface), or MIPI (mobile industry processor interface)) and exchange signals (e.g., commands or data) with each other.

[0155] According to one embodiment, commands or data may be transmitted or received between the electronic device (1201) and an external electronic device (1204) through a server (1208) connected to a second network (1299). Each of the external electronic devices (1202, or 1204) may be the same or a different type of device as the electronic device (1201). According to one embodiment, all or part of the operations performed on the electronic device (1201) may be performed on one or more of the external electronic devices (1202, 1204, or 1208). For example, if the electronic device (1201) needs to perform a function or service automatically or in response to a request from a user or another device, the electronic device (1201) may request one or more external electronic devices to perform at least part of the function or service instead of performing the function or service itself or additionally. One or more external electronic devices that receive the above request may execute at least part of the requested function or service, or additional function or service related to the request, and transmit the result of the execution to the electronic device (1201). The electronic device (1201) may provide the result as is or additionally processed as at least part of the response to the request. For this purpose, for example, cloud computing, distributed computing, mobile edge computing (MEC), or client-server computing technology may be used. The electronic device (1201) may provide ultra-low latency services using, for example, distributed computing or mobile edge computing. In another embodiment, the external electronic device (1204) may include an Internet of Things (IoT) device. The server (1208) may be an intelligent server using machine learning and / or neural networks.According to one embodiment, an external electronic device (1204) or server (1208) may be included within the second network (1299). The electronic device (1201) may be applied to intelligent services (e.g., smart home, smart city, smart car, or healthcare) based on 5G communication technology and IoT-related technology.

[0156] The electronic device according to the various embodiments disclosed in this document may be of various forms. The electronic device may include, for example, a portable communication device (e.g., a smartphone), a computer device, a portable multimedia device, a portable medical device, a camera, a wearable device, or a consumer electronics device. The electronic device according to the embodiments of this document is not limited to the devices described above.

[0157] The various embodiments of this document and the terms used therein are not intended to limit the technical features described in this document to specific embodiments, and should be understood to include various modifications, equivalents, or substitutions of said embodiments. In connection with the description of the drawings, similar reference numerals may be used for similar or related components. The singular form of a noun corresponding to an item may include one or more of said items unless the relevant context clearly indicates otherwise. In this document, phrases such as "A or B," "at least one of A and B," "at least one of A or B," "A, B or C," "at least one of A, B and C," and "at least one of A, B, or C" may each include any one of the items listed together in the corresponding phrase, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used simply to distinguish said components from other said components and do not limit said components in any other aspect (e.g., importance or order). Where any (e.g., 1st) component is referred to as “coupled” or “connected” to another (e.g., 2nd) component, with or without the terms “functionally” or “communicationly,” it means that said any component may be connected to said other component directly (e.g., via a wire), wirelessly, or through a third component.

[0158] The term “module” as used in the various embodiments of this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit, for example. A module may be a component formed integrally, or a minimum unit of said component or a part thereof that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0159] Various embodiments of the present document may be implemented as software (e.g., program (1240)) comprising one or more instructions stored in a storage medium (e.g., internal memory (1236) or external memory (1238)) readable by a machine (e.g., electronic device (1201)). For example, a processor (e.g., processor (1220)) of the machine (e.g., electronic device (1201)) may call at least one of the one or more instructions stored from the storage medium and execute it. This enables the machine to be operated to perform at least one function according to the at least one called instruction. The one or more instructions may include code generated by a compiler or code that can be executed by an interpreter. The storage medium readable by the machine may be provided in the form of a non-transitory storage medium. Here, 'non-temporary' simply means that the storage medium is a tangible device and does not contain a signal (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily.

[0160] According to one embodiment, the method according to the various embodiments disclosed herein may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or an application store (e.g., Play Store). TM It can be distributed online (e.g., downloaded or uploaded) through ) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created on a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0161] According to various embodiments, each component (e.g., module or program) of the components described above may include a singular or multiple entities, and some of the multiple entities may be separated and placed in other components. According to various embodiments, one or more of the components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Generally or additionally, multiple components (e.g., module or program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the multiple components in the same or similar manner as those performed by the corresponding component among the multiple components prior to integration. According to various embodiments, operations performed by the module, program, or other components may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0162] FIG. 13 is a diagram showing an example of a system including a generative artificial intelligence model according to one embodiment.

[0163] Throughout the present disclosure, the artificial intelligence model (101) may correspond to the generative artificial intelligence (AI) model (1350) of FIG. 13. However, the artificial intelligence model (101) of the present disclosure is not limited to the generative AI model (1350) described in FIG. 13. Furthermore, the electronic device (100) of the present disclosure does not have to include all the components of FIG. 13, but may include only some of the components of FIG. 13.

[0164] The artificial intelligence system may include a user query / response interface (1310), an AI framework (1320), an application / service component (1330), knowledge repositories (1340), and / or a generative AI model (1350).

[0165] Referring to FIG. 13, a user query / response interface (1310) may receive input. The input may include user input and / or data obtained or generated by an electronic device (e.g., electronic device (101, 201, 401, 601, 701, 901)). The data may include images, videos, and / or sensor data generated by at least one processor of the electronic device (e.g., at least one processor (120)) (e.g., illuminance data around the electronic device obtained from a sensor or sensor hub (e.g., auxiliary processor (123), attitude data (or orientation data) of the electronic device, temperature inside the electronic device (e.g., display module (130)) or temperature of at least one processor (120)), size information of the display area of ​​the display module (130), and / or images obtained through an image sensor of the electronic device (e.g., included in a camera module (180)). For example, user input may take the form of biosignals, natural language, touch data acquired through touch circuits included within the display module (e.g., used to identify input from a finger and / or stylus), images, and / or videos. Additionally, context information may be transmitted along with the user input. Context information may include various additional information at the time of user input. For example, this could include information about the application currently being used by the user or the user's location information. Furthermore, user input may take the form of a mixture of the aforementioned biosignals, natural language, images, sounds, and context information. Additionally, user input may take the form of non-natural language input, such as input for selecting a menu.

[0166] The user query / response interface (1310) can output results from a generative artificial intelligence system to the user. The output may include results (or result information) generated or obtained by the artificial intelligence system (1300) based on at least part of the input. The output may be in the form of natural language or specific content, and may also be provided in a form such as an action requested by the user. For example, the output may have a format according to the user settings of the electronic device. For example, the output may include information regarding an emergency situation in which the user is facing. The output may include detailed information about the emergency situation, for example, information indicating the type of emergency situation, information regarding the user's surrounding circumstances, personal status (health status) information, location information, information regarding the user's request, and information indicating the truthfulness of the user's request.

[0167] The AI ​​framework (1320) can receive input from the user and coordinate and control each component necessary to perform the user's intent based on the user's query.

[0168] User input received from the user query / response interface (1310) can be transmitted to a prompt design component (1321). The prompt design component (1321) can be used to generate a prompt suitable for inputting the user input into a generative AI model (1350) (e.g., a large language model (LLM), a large vision model (LVM), and / or large multimodal models (LMM)).

[0169] The prompt design component (1321) may be an AI component that uses machine learning algorithms or neural networks to develop better prompts over time. The prompt design component (1321) may generate prompts by accessing a knowledge component (e.g., knowledge repository (1340)) containing user preference data, a prompt library, and prompt examples based on user input, and may pass the generated prompts to a generative AI model (1350) (e.g., LLM, LVM, and / or LMM).

[0170] The APIs / Plugins management component (1323) can perform the role of communicating with external information when there is a request for additional information when user input is passed as input to the generative AI model (1350). The APIs / Plugins management component (1323) establishes a channel to communicate with the outside of the AI ​​interface via APIs, and can enable access to various data sources (e.g., knowledge repository (1340)) through the established channel. For example, the APIs / Plugins management component (1323) can be used to request another component (e.g., application / service component (1330)) that performs feedback (or response) according to the prompt. If the application or service needs to perform an action that ultimately executes the user's input rather than an intermediate result, the APIs / Plugins management component (1323) can request that action from the application / service component (1330) via APIs. Information obtained from the outside may be used to generate a prompt in the prompt design component (1321) along with user input, or it may be passed as input to the generative model.

[0171] A refiner component (e.g., output modification component (1325)) can fine-tune (or adjust) (or modify) the output produced by a generative AI model (1350) (e.g., LLM, LVM, and / or LMM). For example, the refiner component can verify whether the content generated by the generative AI model (1350) (e.g., LLM, LVM, and / or LMM) is irrelevant, contains biased content, or contains harmful content. Additionally, the refiner component can determine the extent to which the output matches the desired result and, if necessary, proceed with the additional process. Furthermore, the refiner component can configure and provide hints to the user to help avoid unwanted outputs.

[0172] A generative AI model (1350) generally refers to an artificial intelligence neural network that generates new forms of data based on user input information. A generative AI model (1350) may include a model that generates images and / or a model that generates language. Models that generate images include, but are not limited to, GANs (generative adversarial networks) and VAEs (variational autoencoders), and examples include diffusion-based generative models that use VAEs and Transformer structures. Models that generate language are models trained to output the most statistically appropriate output value based on input values, and examples include models such as CHAT-GPT 3 and CHAT-GPT 4. There are also LMMs that can recognize various forms of data input, such as sensing data, biosignals, text, images, and voice, and generate new data corresponding to them.

[0173] In one embodiment, the AI ​​framework (1320) and / or generative AI model (1350) may be included in a server communicating with the electronic device, but is not limited thereto, and may be included in an AI module (e.g., including a processing circuit) within the electronic device. For example, the AI ​​module may be operatively coupled with at least one processor of the electronic device (e.g., at least one processor (120)). For example, the AI ​​module may be operatively coupled with a sensor hub of the electronic device for one or more sensors within the electronic device.

[0174] According to one embodiment, the electronic device may be configured to include at least some of the user query / response interface (1310) AI framework (1320), application / service component (1330), knowledge repository (1340), or generative AI model (1350) of FIG. 10. According to one embodiment, at least some of the user query / response interface (1310) AI framework (1320), application / service component (1330), knowledge repository (1340), or generative AI model (1350) of FIG. 10 may be included in another electronic device (e.g., a server, a wearable electronic device, or another user's electronic device) that communicates with the electronic device.

[0175] An electronic device (100) according to one embodiment may include a battery (250), at least one temperature sensor adjacent to the battery (250), at least one processor (1220), and a memory (1230) for storing instructions. The above instructions are executed individually or collectively by the at least one processor (1220), so that the electronic device (100) measures the temperature of the battery (250) through the at least one temperature sensor, and, as it is determined that the temperature of the battery (250) has a value greater than or equal to a reference temperature for a period of time greater than a threshold time, periodically measures the battery voltage and battery current of the battery (250) for a set period of time, obtains the rate of change of the battery voltage and the rate of change of the battery current based on the measurement results, generates a spectrogram (120) in the time-frequency domain based on the obtained rate of change of the battery voltage and the obtained rate of change of the battery current, obtains additional information collected in relation to the state of the battery (250), and applies the spectrogram (120) and the additional information to the artificial intelligence model (101), thereby obtaining from the artificial intelligence model (101) the It is possible to obtain an output value related to the swelling risk of the battery (250).

[0176] The above additional information may include state information of the battery (250). The state information may include information related to at least one of the state of health (SOH) of the battery (250), the type of the battery (250), or the temperature change history of the battery (250).

[0177] The above additional information may include charging history information related to the charging of the battery (250). The charging history information may include at least one of information related to the charging speed of the battery (250), information related to the charging method of the battery (250), the number of times the battery (250) is charged, the number of times the battery (250) is discharged, or information related to the DOD (depth of discharge) of the battery (250).

[0178] The artificial intelligence model (101) may include a convolutional neural network (CNN) model trained to output an output value related to the swelling risk of the battery (250) as input the time-frequency domain spectrogram (120) generated based on the rate of change of the battery voltage and the rate of change of the battery current.

[0179] The spectrogram (120) may be generated by converting time domain data based on the rate of change of the battery voltage and the rate of change of the battery current into a time-frequency domain spectrogram (120).

[0180] The spectrogram (120) may be generated based on the time-frequency domain data obtained as a result of converting time-domain data based on the rate of change of the battery voltage and the rate of change of the battery current into time-frequency domain data through a short-term Fourier transform (STFT).

[0181] The above instructions may be executed individually or collectively by the at least one processor (1220) to enable the electronic device (100) to determine the rate of change of the battery voltage and the rate of change of the battery current for each unit time while the unit time is repeated a first time.

[0182] The above commands may be executed individually or collectively by the at least one processor (1220) to enable the electronic device (100) to repeatedly obtain the rate of change of the battery voltage and the rate of change of the battery current a set number of times.

[0183] Each of the above-mentioned multiple battery voltage change rate values ​​and the above-mentioned battery current change rate values ​​may correspond to each set unit time included within the above-mentioned set time.

[0184] The above unit time is 30 seconds, and the above set number of times may be 128 times.

[0185] The above unit time may be set based on at least one of the charging speed of the battery (250), the charging state of the battery (250), the SOH of the battery (250), or the state of the processor (1220) of the electronic device (100).

[0186] The battery (250) may include a microcontroller unit (MCU). The electronic device (100) may apply the spectrogram (120) and the additional information to the artificial intelligence model (101) under the control of the microcontroller, and obtain the output value related to the swelling risk of the battery (250) from the artificial intelligence model (101).

[0187] The above reference temperature may be 35 to 45 degrees Celsius.

[0188] The above instructions may be executed individually or collectively by the at least one processor (1220) to enable the electronic device (100) to obtain the mean and variance of the data included in the spectrogram (120) based on the spectrogram (120), to perform normalization on the data included in the spectrogram (120) based on the obtained mean and variance, and to apply the spectrogram (120) obtained as a result of the normalization to the artificial intelligence model (101).

[0189] The artificial intelligence model (101) may include a first layer, a second layer, and a third layer arranged sequentially. The first layer may include a first number of first kernels having a first size. The second layer may include a second number of second kernels having the first size, and the second number may be equal to or smaller than half of the first number. The third layer may include a third number of third kernels having a second size equal to or smaller than half of the first size.

[0190] A method of operating an electronic device (100) according to one embodiment includes: measuring the temperature of the battery (250) through at least one temperature sensor adjacent to the battery (250); periodically measuring the battery voltage and battery current of the battery (250) for a set time period as it is determined that the temperature of the battery (250) has a value greater than or equal to a reference temperature for a period of time greater than a threshold time; obtaining the rate of change of the battery voltage and the rate of change of the battery current based on the measurement results; generating a spectrogram (120) in the time-frequency domain based on the obtained rate of change of the battery voltage and the obtained rate of change of the battery current; obtaining additional information collected in relation to the state of the battery (250); and applying the spectrogram (120) and the additional information to an artificial intelligence model (101) to obtain an output value related to the swelling risk of the battery (250) from the artificial intelligence model (101). It is possible.

[0191] The above additional information may include status information of the battery (250).

[0192] The above state information may include information related to at least one of the state of health (SOH) of the battery (250), the type of the battery (250), or the temperature change history of the battery (250).

[0193] The above additional information may include charging history information related to the charging of the battery (250).

[0194] The above charging history information may include at least one of information related to the charging speed of the battery (250), information related to the charging method of the battery (250), the number of times the battery (250) is charged, the number of times the battery (250) is discharged, or information related to the DOD (depth of discharge) of the battery (250).

[0195] The artificial intelligence model (101) may include a convolutional neural network (CNN) model trained to output an output value related to the swelling risk of the battery (250) as input the time-frequency domain spectrogram (120) generated based on the rate of change of the battery voltage and the rate of change of the battery current.

[0196] The spectrogram (120) may be generated by converting time domain data based on the rate of change of the battery voltage and the rate of change of the battery current into a time-frequency domain spectrogram (120).

[0197] The spectrogram (120) may be generated based on the time-frequency domain data obtained as a result of converting time-domain data based on the rate of change of the battery voltage and the rate of change of the battery current into time-frequency domain data through a short-term Fourier transform (STFT).

[0198] Various embodiments of the present document may be implemented as software (e.g., program (140)) comprising one or more instructions stored in a storage medium (e.g., internal memory (136) or external memory (138)) readable by a machine (e.g., electronic device (100)). For example, a processor (e.g., processor (120)) of the machine (e.g., electronic device (100)) may call at least one of the one or more instructions stored from the storage medium and execute it. This enables the machine to be operated to perform at least one function according to the at least one called instruction. The one or more instructions may include code generated by a compiler or code that can be executed by an interpreter. The storage medium readable by the machine may be provided in the form of a non-transitory storage medium. Here, 'non-temporary' simply means that the storage medium is a tangible device and does not contain a signal (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily.

[0199] According to one embodiment, the method according to the various embodiments disclosed herein may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or an application store (e.g., Play Store). TMIt can be distributed online (e.g., downloaded or uploaded) through ) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created on a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0200] According to various embodiments, each component (e.g., module or program) of the components described above may include a singular or multiple entities, and some of the multiple entities may be separated and placed in other components. According to various embodiments, one or more of the components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Generally or additionally, multiple components (e.g., module or program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the multiple components in the same or similar manner as those performed by the corresponding component among the multiple components prior to integration. According to various embodiments, operations performed by the module, program, or other components may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

Claims

1. In an electronic device, battery; At least one temperature sensor adjacent to the battery; At least one processor; and It includes memory for storing instructions, The above instructions are executed individually or collectively by the at least one processor, and the electronic device: The temperature of the battery is measured through at least one temperature sensor, and As it is determined that the temperature of the battery has a value greater than or equal to the reference temperature for a period exceeding the critical time, the battery voltage and battery current of the battery are checked for a set period, and Based on the above verification results, the rate of change of the battery voltage and the rate of change of the battery current are obtained, and Based on the rate of change of the battery voltage and the rate of change of the battery current obtained above, a spectrogram in the time-frequency domain is generated, and Acquire additional information collected in relation to the state of the above battery, and Applying the spectrogram and the additional information to an artificial intelligence model to obtain an output value related to the swelling risk of the battery from the artificial intelligence model. Electronic device.

2. In Claim 1, The above additional information includes the status information of the battery, and The above state information includes information related to at least one of the state of health (SOH) of the battery, the type of the battery, or the history of temperature changes of the battery. Electronic device.

3. In Claim 1, The above additional information includes charging history information related to the charging of the battery, and The above charging history information includes at least one of information related to the charging speed of the battery, information related to the charging method of the battery, the number of times the battery is charged, the number of times the battery is discharged, or information related to the DOD (depth of discharge) of the battery. Electronic device.

4. In Claim 1, The above artificial intelligence model is a convolutional neural network (CNN) model trained to output an output value related to the swelling risk of the battery as input the time-frequency domain spectrogram generated based on the rate of change of the battery voltage and the rate of change of the battery current. Electronic device.

5. In Claim 1, The above spectrogram is generated by converting time domain data based on the rate of change of the battery voltage and the rate of change of the battery current into a time-frequency domain spectrogram. Electronic device.

6. In Claim 1, The spectrogram above is generated based on the time-frequency domain data obtained as a result of converting time-domain data based on the rate of change of the battery voltage and the rate of change of the battery current into time-frequency domain data through a short-term Fourier transform (STFT). Electronic device.

7. In Claim 1, The above instructions are executed individually or collectively by the at least one processor, so that the electronic device determines the rate of change of the battery voltage and the rate of change of the battery current for each unit time while the unit time is repeated a first time. Electronic device.

8. In Claim 1, The above instructions are executed individually or collectively by the at least one processor so as to cause the electronic device to repeatedly acquire the rate of change of the battery voltage and the rate of change of the battery current a set number of times. Electronic device.

9. In Claim 8, Each of the above-mentioned multiple battery voltage change rate values ​​and the above-mentioned battery current change rate values ​​corresponds to each set unit time included within the above-mentioned set time, Electronic device.

10. In Claim 9, The above unit time is 30 seconds, and the above set number of times is 128, Electronic device.

11. In Claim 9, The above unit time is set based on at least one of the charging speed of the battery, the charge state of the battery, the SOH of the battery, or the state of the processor of the electronic device. Electronic device.

12. In Claim 1, The above battery includes a microcontroller (MCU, micro-controller unit), and By controlling the microcontroller, the spectrogram and the additional information are applied to the artificial intelligence model to obtain the output value related to the swelling risk of the battery from the artificial intelligence model. Electronic device.

13. In Claim 1, The above reference temperature is 35 to 45 degrees Celsius, Electronic device.

14. In Claim 1, The above instructions are executed individually or collectively by the at least one processor, and the electronic device: Based on the above spectrogram, the mean and variance of the data included in the above spectrogram are obtained, and Based on the mean and variance obtained above, normalization is performed on the data included in the spectrogram, and Applying the spectrogram obtained as a result of performing the above normalization to the above artificial intelligence model, Electronic device.

15. In Claim 4, The above artificial intelligence model includes a first layer, a second layer, and a third layer arranged sequentially, and The above first layer includes a first number of first kernels having a first size, and The second layer comprises a second number of second kernels having the first size, wherein the second number is equal to or smaller than half of the first number, The third layer comprises a third number of third kernels having a second size that is equal to or smaller than half of the first size. Electronic device.