KR-20260062308-A - Method For Diagnosing Fire Safety And Health Status Of Electric Vehicles Using Battery Internal Short Circuit Detection Algorithm And Health Status Diagnosis Algorithm

KR20260062308AKR 20260062308 AKR20260062308 AKR 20260062308AKR-20260062308-A

Abstract

The present invention relates to a method for fire safety and health diagnosis of an electric vehicle to which a battery internal short circuit detection algorithm and a health diagnosis algorithm are applied, more specifically comprising: a data measurement and collection step of measuring, collecting, and storing data including the voltage, current, and ambient temperature of a battery through a sensor; a comparison step with measured battery characteristic data, which compares the data of an electric vehicle battery to be diagnosed with the data stored through the data measurement and collection step; a numerical analysis step of outputting numerical analysis data by performing numerical analysis on deviation values derived by comparing with characteristic values such as the internal resistance value, positive electrode hole density, negative electrode charge density, electrolyte concentration, internal temperature, and the amount of change therein at the cell level of the measured battery, and expanding the numerical analysis data to calculate the temperature rise value, state of charge (SOC), and state of health (SOH) at the battery pack level and outputting numerical analysis data; and a battery internal short circuit detection and health diagnosis step of applying the numerical analysis data output from the numerical analysis step to a battery internal short circuit, battery degradation degree, and remaining life prediction algorithm to output battery internal short circuit data and battery degradation degree and predicted remaining life diagnosis data. The present invention relates to a method for fire safety and health status diagnosis of an electric vehicle to which a battery internal short circuit detection algorithm and a health status diagnosis algorithm are applied, comprising: a detection and diagnosis information output step for transmitting each data information to external communication devices configured including the vehicle manager's user terminal and the charging station manager's terminal when the diagnosis result of the battery internal short circuit detection and health status diagnosis step indicates that an internal short circuit of the battery is predicted or that a battery failure is predicted according to a high degree of battery degradation. This project (result) is the outcome of the Local Government-University Cooperation-based Regional Innovation Project, conducted in 2024 with funding from the Ministry of Education and support from the National Research Foundation of Korea. (Project Management No.: 2021RIS-002)

Inventors

정대원
황보승

Assignees

호남대학교 산학협력단

Dates

Publication Date: 20260507
Application Date: 20241029

Claims (5)

Method for diagnosing fire safety and health status of electric vehicle batteries. A data measurement and collection step of measuring, collecting, and storing data including the battery voltage, current, and ambient temperature through a sensor; A comparison step with measured battery characteristic data, which compares the data of the electric vehicle battery to be diagnosed with the data stored through the data measurement and collection step; A numerical analysis step that outputs numerical analysis data by performing numerical analysis on deviation values derived by comparison with characteristic values such as the internal resistance value, positive electrode hole density, negative electrode charge density, electrolyte concentration, internal temperature, and change amount at the cell unit of the measured battery, and extends the numerical analysis data to calculate the temperature rise, state of charge (SOC), and state of health (SOH) at the battery pack unit and outputs numerical analysis data; A battery internal short circuit detection and health status diagnosis step that applies numerical analysis data output from the above numerical analysis step to an algorithm for predicting battery internal short circuit, battery degradation degree, and remaining life, and outputs battery internal short circuit data and battery degradation degree and predicted remaining life diagnosis data; and A method for fire safety and health status diagnosis of an electric vehicle to which a battery internal short circuit detection algorithm and a health status diagnosis algorithm are applied, comprising: a detection and diagnosis information output step for transmitting each data information to external communication devices configured including the vehicle manager's user terminal and the charging station manager's terminal when the diagnosis result of the battery internal short circuit detection and health status diagnosis step above predicts a battery internal short circuit or predicts a battery failure according to a high degree of battery degradation.
In claim 1, An electric vehicle fire safety and soundness diagnosis method in which a battery internal short circuit detection algorithm and a soundness diagnosis algorithm are applied, characterized in that the battery internal short circuit algorithm, the battery degradation degree diagnosis algorithm, and the battery remaining life prediction algorithm applied in the above-mentioned battery internal short circuit detection and soundness diagnosis step undergo a process of being applied simultaneously in parallel.
In claim 2, The battery internal short circuit detection algorithm is A step of comparing the measured battery cell voltage and the rate of change data with the corresponding voltage and rate of change threshold of the battery to be measured; A step of comparing the measured battery cell current and the rate of change data with the corresponding current and rate of change threshold of the battery to be measured; A step of comparing the measured battery cell electrolyte potential and the rate of change data with the corresponding electrolyte potential and rate of change threshold of the battery to be measured; and An electric vehicle fire safety and soundness diagnosis method incorporating a battery internal short circuit detection algorithm and a soundness diagnosis algorithm, comprising the step of comparing measured battery cell irreversible heat change rate data with a corresponding heat change rate threshold of the battery to be measured, wherein the comparison in each step can be performed chronologically or simultaneously in parallel.
In claim 2, The battery degradation level diagnosis algorithm is A step of analyzing changes in the battery cell solid electrolyte interface layer according to the battery cell electrolyte state using the measured and calculated rate of change in internal resistance of the battery cell, the amount of decrease in battery capacity (SOH), and the open-circuit voltage value; An electric vehicle fire safety and soundness diagnosis method incorporating a battery internal short circuit detection algorithm and a soundness diagnosis algorithm, comprising a step of measuring and analyzing the charge/discharge cycle and number of times of a specific battery to be measured; wherein the comparison in each step can be performed chronologically or simultaneously in parallel.
In claim 2, The battery remaining life prediction algorithm is A step of detecting the amount of change in the battery capacity measured in the past and present; Step of applying the following remaining life prediction formula; and SOH: The estimated change in battery capacity, calculated as [Current battery capacity / (New (initial) battery rated capacity] * 100 (%), SOH start : Initial SOH value, SOH end: The SOH value at the end of the battery's life, Cycle total : The time it takes for the battery SOH to reach 70%, i.e., the period (lifespan) until the battery is replaced. A method for fire safety and health status diagnosis of an electric vehicle with an applied battery internal short-circuit detection algorithm and health status diagnosis algorithm, characterized by comprising a step of outputting a remaining life prediction result and completing a remaining life diagnosis.

Description

Method for Diagnosing Fire Safety and Health Status of Electric Vehicles Using Battery Internal Short Circuit Detection Algorithm and Health Status Diagnosis Algorithm The present invention relates to a method for fire safety and health diagnosis of an electric vehicle to which a battery internal short circuit detection algorithm and a health diagnosis algorithm are applied, more specifically comprising: a data measurement and collection step of measuring, collecting, and storing data including the voltage, current, and ambient temperature of a battery through a sensor; a comparison step with measured battery characteristic data, which compares the data of an electric vehicle battery to be diagnosed with the data stored through the data measurement and collection step; a numerical analysis step of outputting numerical analysis data by performing numerical analysis on deviation values derived by comparing with characteristic values such as the internal resistance value, positive electrode hole density, negative electrode charge density, electrolyte concentration, internal temperature, and the amount of change therein at the cell level of the measured battery, and expanding the numerical analysis data to calculate the temperature rise value, state of charge (SOC), and state of health (SOH) at the battery pack level and outputting numerical analysis data; and a battery internal short circuit detection and health diagnosis step of applying the numerical analysis data output from the numerical analysis step to a battery internal short circuit, battery degradation degree, and remaining life prediction algorithm to output battery internal short circuit data and battery degradation degree and predicted remaining life diagnosis data. The present invention relates to a method for fire safety and health status diagnosis of an electric vehicle to which a battery internal short circuit detection algorithm and a health status diagnosis algorithm are applied, comprising: a detection and diagnosis information output step for transmitting each data information to external communication devices configured including the vehicle manager's user terminal and the charging station manager's terminal when the diagnosis result of the battery internal short circuit detection and health status diagnosis step indicates that an internal short circuit of the battery is predicted or that a battery failure is predicted according to a high degree of battery degradation. In applications requiring high-speed charging and discharging characteristics for lithium-ion batteries, such as electric vehicles, the movement of lithium ions is prone to heat generation and volume expansion. If such high-temperature heat generation and volume expansion persist, the internal separator may rupture, causing an internal short circuit where the positive and negative electrode materials mix. This leads to thermal runaway, which in turn causes a fire. Lithium-ion battery fires originate from thermal runaway caused by damage to the internal separator. Since separator damage can be effectively identified by diagnosing internal short circuits, early detection of such phenomena is crucial; however, the reality is that effective diagnosis and detection are difficult because it is impossible to physically detect (measure) them directly from the outside of the battery. Therefore, in order to prevent fires caused by thermal runaway in battery packs installed in electric vehicles, a real-time diagnostic method that effectively detects thermal runaway phenomena is required. In this regard, according to Korean Published Patent No. 10-2021-0108076 (Title of Invention: Machine Learning-based Battery Charge State Estimation Method Using Thermal Distribution Images, Publication Date: September 2, 2021), the invention relates to a machine learning-based battery charge state estimation method using thermal distribution images, wherein thermal distribution images by battery state captured through a thermal imaging camera are constructed as a training dataset, and a charging state estimation learning model and a discharging state estimation learning model are generated using the constructed training dataset to estimate the charging state or discharging state of a target battery. The method comprises: a thermal distribution image acquisition step (S100) for acquiring a plurality of battery surface thermal distribution images by capturing the battery surface; and a charge/discharge amount information acquisition step (S200) performed simultaneously with the thermal distribution image acquisition step (S100), wherein charge amount information or discharge amount information of the captured battery is acquired at each moment of capture. The present invention relates to a machine learning-based battery charge state estimation method utilizing thermal distribution images, characterized by including: a learning data set configuration step (S300) for configuring acquired battery surface t