Distinguishing method for thermal runaway of lithium ion battery

Abstract

The invention discloses a method for judging thermal runaway of a lithium ion battery. The method comprises the following steps: 1, synchronously collecting gas concentration, light intensity value and temperature and humidity data; 2, constructing a baseline by adopting exponential moving average and carrying out baseline removal processing; 3, temperature and humidity abrupt change is detected, and first-order linear self-adaptive correction is carried out; 4, performing z-score standardization on the corrected concentration and change rate; 5, performing combination trigger statistics based on a multi-parameter combination criterion; 6, confirming effective triggering through a continuity and de-jitter mechanism; and 7, calculating a risk score R according to the sensor weight vector and the combined score adding coefficient, and mapping the score into a multi-level risk level in combination with a hysteresis interval threshold. The method effectively solves the problems that a traditional single threshold method is high in false alarm rate and delayed in response, is suitable for scenes of electric automobiles, industrial energy storage systems, battery production tests and the like, and achieves accurate and reliable early warning of lithium ion battery thermal runaway.

Description

Technical Field

[0001] This invention relates to the field of lithium-ion battery safety monitoring technology, specifically to a method for identifying thermal runaway in lithium-ion batteries. It is applicable to scenarios such as electric vehicle power batteries, industrial energy storage systems, UPS backup power supplies, and battery production testing. It is used to collect, fuse, analyze, and provide graded early warnings of multimodal information such as gas, smoke, temperature, and humidity during the operation and resting process of lithium-ion batteries, thereby reducing the safety risks caused by thermal runaway. Background Technology

[0002] Lithium-ion batteries are widely used in industrial production fields such as electric vehicles, rail transportation, industrial energy storage, power peak shaving, and backup power for communication base stations due to their high energy density, long lifespan, and gradually decreasing cost. However, lithium-ion batteries may experience thermal runaway under the influence of factors such as overcharging, short circuits, external impacts, or high temperatures. This can lead to electrolyte decomposition, the release of flammable gases and smoke, and ultimately, fires or explosions, causing significant personal injury and property damage.

[0003] Current thermal runaway monitoring primarily relies on cell voltage, temperature, and gas pressure signals, but this approach suffers from limitations such as limited information, delayed response, and poor adaptability. Numerous studies have shown that lithium-ion batteries preferentially release various gases, including H2, CH4, CO, and VOCs, in the early stages of thermal runaway, and these changes typically precede signals from smoke, temperature, and pressure. Therefore, utilizing gas sensors in conjunction with smoke optical characteristics and temperature and humidity data can provide faster and more reliable early warnings in the early stages of thermal runaway.

[0004] In recent years, some studies have introduced gas sensors into the field of lithium battery thermal runaway monitoring. For example, SnO2-type metal oxide gas sensors are used to monitor changes in the concentration of combustible gases such as CH4, C3H8, and CO; optical smoke sensors are used to detect changes in the concentration of particulate matter released during thermal runaway. However, these studies generally suffer from problems such as insufficient fusion of multi-source information, overly simplistic discrimination methods, lack of early warning classification, and limited application scope. Summary of the Invention This invention addresses the shortcomings of existing technologies by proposing a method for identifying thermal runaway in lithium-ion batteries. The aim is to construct an intelligent early warning system that simulates expert decision-making processes, thereby improving the accuracy of early, reliable, and robust identification and early warning of initial signs of thermal runaway in lithium-ion batteries. This ensures the safety of battery operation, reduces the risk of accidents caused by thermal runaway, and enhances the reliability and safety of lithium-ion batteries used in electric vehicles, energy storage systems, and other applications.

[0006] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: The present invention provides a method for determining thermal runaway in lithium-ion batteries, characterized by the following steps: Step 1: Collect the concentration C of the i-th gas released by the lithium-ion battery at time t. i (t), light intensity L(t), temperature T(t), relative humidity RH(t), where i=1,2,3,4, when i=1, it represents hydrogen H2, when i=2, it represents methane CH4, when i=3, it represents carbon monoxide CO, and when i=4, it represents volatile organic compound gas VOC; Step 2: Based on the data from Step 1, calculate the baseline-free concentration of the i-th gas released by the lithium-ion battery at time t. The change in light intensity outside the lithium-ion battery at time t Temperature change of lithium-ion battery at time t The relative humidity change at time t ; Step 3: Based on the data from Step 2, calculate the corrected concentration of the i-th gas released by the lithium-ion battery at time t. and its concentration change rate ; Step 4: Set the gas concentration threshold to Th i ,when ≥ Th i If , then it means that the warning value for the i-th gas is triggered at time t; Set the light intensity change threshold to Th L ,when ≤ Th L , which represents the warning value for triggering smoke at time t; Set the temperature change threshold to Th T Set the relative humidity change threshold to Th RH ,like ≥ Th T or ≥ Th RH If , then it represents the temperature and humidity warning value triggered at time t; Step 5: For each , , , , Standardization is performed to obtain the baseline value of the standardized gas concentration of the lithium-ion battery at time t. The rate of change of the concentration of the i-th gas in a lithium-ion battery after standardization at time t. The normalized change in light intensity of a lithium-ion battery at time t The standardized temperature change of a lithium-ion battery at time t The relative humidity change of a lithium-ion battery at time t after standardization. ; Step 6: If the warning value of the i-th gas is triggered at time t, and the warning value of temperature and humidity is also triggered at time t, it means that the lithium-ion battery triggers the "gas-temperature combination" situation at time t. If the warning value for the i-th gas is triggered at time t, and the warning value for smoke is also triggered at time t, then it means that the lithium-ion battery triggers the "gas-smoke combination" situation at time t. If the warning values ​​for any two gases are triggered at time t, and the warning values ​​for temperature and humidity are also triggered at time t, then it means that the lithium-ion battery has triggered a "gas-gas-temperature combination" situation at time t. If the warning value for any two gases is triggered at time t, and the warning value for smoke is triggered at time t, then it means that the lithium-ion battery triggers a "gas-gas-smoke combination" situation at time t. If the warning value of any three gases is triggered at time t, it means that the lithium-ion battery has triggered the "three-gas linkage combination" at time t. If the warning values ​​for any two or more gases are triggered at time t, and the warning values ​​for smoke and temperature and humidity are triggered at time t, then it means that the lithium-ion battery has triggered the "all-element linkage combination" situation at time t. Step 7: Set the trigger threshold number of times to N. If the total number of times any of the following combinations is triggered within a period of time T reaches N, then the combination trigger count value M is incremented by 1. If the total number of times the "all-element linkage combination" is triggered reaches N within a period of time T, the combination trigger count value M will be incremented by 2. Step 8: Based on the data from M and Step 5, calculate the total score R of the lithium-ion battery state over a period of time T; Step 9: Define the early warning baseline threshold as θ1 and the alarm baseline threshold as θ2, where θ1 < θ2; Define the hysteresis offset as ε, and ε <θ1; Define the upgrade / downgrade hold time as T. hold ; Calculate the uplink warning threshold θ respectively up,1 =θ1+ ε、Downlink security threshold θ down,1 =θ1- ε、Uplink alarm threshold θ up,2 =θ² + ε; Downlink warning threshold θ down,2 =θ² - ε; and θ down,1 < θ up,1 <θ down,2 <θ up,2 ; Step 10, if R ≥ θ up,2 If M ≥ 2, it indicates that the lithium-ion battery state is seriously abnormal within a period of time T, and is judged as a level 2 risk level, and step 11 is executed; If R ≥ θ up,1 If M = 1, it indicates that there is a significant abnormality in the state of the lithium-ion battery within a period of time T, which is judged as a level 1 risk level, and step 12 is executed. If R ≤ θ down,1 If M=0, it means that the lithium-ion battery status has not shown any obvious abnormalities within a period of time T, and is judged to be at a risk-free level, and step 13 is executed. Step 11, if θ down,1 ≤ R ≤ θ down,2 And the duration of M=1 reaches T. hold If so, the risk level will be reduced to Level 1; If R ≤θ down,1 And the duration of M=0 reaches T. hold If so, it will be downgraded to a risk-free level; Otherwise, maintain the risk level as Level 2 and proceed to step 21; Step 12: If R ≤ θ down,1 And the duration of M=0 reaches T. hold If so, it will be downgraded to a risk-free level; If θ down,2 ≤ R ≤θ up,2 And the duration of M=1 reaches T. hold If so, the risk level will remain at Level 1. Otherwise, proceed to step 21; Step 13, if R≤θ up,1 And the duration of M=0 reaches T. hold If so, the risk level remains unchanged; Otherwise, proceed to step 21; Step 14: After assigning T+t to t, return to Step 1 and execute sequentially until the judgment time ends.

[0007] The method for determining thermal runaway in lithium-ion batteries described in this invention is also characterized in that step 2 includes the following steps: Step 2.1: Calculate the baseline concentration B of the i-th gas released by the lithium-ion battery at time t using equation (1). i (t): (1) In equation (1), λ is the smoothing coefficient. Let be the baseline concentration of the i-th gas released by the lithium-ion battery at time t-1, where t ≥ 1. When t=1, initialize... ; Step 2.2: Use equation (2) to obtain the baseline concentration of the i-th gas released by the lithium-ion battery at time t. : (2) Step 2.3: Calculate the change in light intensity outside the lithium-ion battery at time t using equation (3). : (3) In equation (3), For sliding time windows; For the external of lithium-ion batteries Light intensity at any given moment; Step 2.4: Calculate the temperature change of the lithium-ion battery at time t according to equations (4) and (5). The relative humidity change at time t : (4) (5) In equations (4) and (5), For lithium-ion batteries Temperature at any moment; For lithium-ion batteries Relative humidity at any given time.

[0008] Furthermore, step 3 includes the following steps: Step 3.1: Use equation (6) to obtain the corrected concentration of the i-th gas released by the lithium-ion battery at time t. : (6) In equation (6), T ref The ambient reference temperature under standard conditions, RH ref The ambient reference humidity is given under standard conditions; α is the sensitivity of the gas sensor to temperature deviations; and β is the sensitivity of the gas sensor to humidity deviations. Step 3.2: Use equation (7) to obtain the corrected concentration change rate of the i-th gas released by the lithium-ion battery at time t. : (7) In equation (7), For lithium-ion batteries The corrected concentration of the i-th gas released at time i.

[0009] Furthermore, step 8 includes the following steps: Step 8.1: Use equation (8) to obtain the lithium-ion battery state combination technology score S over a time period T. m : (8) In equation (8), γ is the weighting coefficient of the combination, and M is the final combination trigger count value; Step 8.2: Calculate the x-th variable after nonnegativity constraint at time t using equation (9). , x∈{C i L, T, RH, r i}; (9) In equation (9), Let x represent the x-th variable after standardization at time t, where x = 1, 2, 3, 4, 5; when x = 1, ... express When x=2, express When x=3, express When x=4, express When x=5, express ; Step 8.3: Calculate the continuous score of the lithium-ion battery state at time t using equation (10). : (10) In equation (10), express The weights are given by x = 1, 2, 3, 4, 5; when x = 1, express When x=2, express When x=3, express When x=4, express When x=5, express ; Step 8.4: Use equation (11) to obtain the total score R of the lithium-ion battery state over a time period T: (11) The present invention provides an electronic device, including a memory and a processor, characterized in that the memory is used to store a program that supports the processor in executing the discrimination method, and the processor is configured to execute the program stored in the memory.

[0010] The present invention provides a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, performs the steps of the discrimination method.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention integrates multimodal data from gas concentration, smoke shading, and temperature and humidity signals, combined with baseline correction, dynamic threshold determination, and a sliding window weighted scoring mechanism, to form a complete multi-parameter risk assessment method for lithium-ion battery thermal runaway. This method effectively overcomes the problems of single-sensor detection being susceptible to interference, poor adaptability of fixed thresholds, and unstable risk assessment, thereby achieving earlier, more accurate, and more stable early warning of lithium-ion battery thermal runaway. It also improves the detection sensitivity, accuracy, real-time performance, and robustness of lithium-ion battery thermal runaway, reduces the risk of false alarms and missed alarms, and greatly enhances the applicability of this invention in various application scenarios.

[0012] 2. The significance of this invention lies in promoting the evolution of lithium-ion battery thermal safety monitoring technology from the traditional, isolated, and simple judgment of static thresholds to an integrated, dynamic, and intelligent early warning paradigm, providing core technical support for active safety protection in the field of battery applications. Attached Figure Description

[0013] Figure 1 This is a block diagram of the data acquisition system. Figure 2 This is a flowchart of the overall method for judging thermal runaway in lithium batteries; Figure 3 This is a logic diagram for risk scoring and threshold determination. Detailed Implementation

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] In this example, at the system hardware level, the method of this invention uses four types of gas sensors—hydrogen, methane, carbon monoxide, and VOCs—in conjunction with an ambient light sensor and a temperature and humidity sensor to collect multi-source signals from the lithium-ion battery in real time during operation and resting. Sudden changes in the concentration of key gases, accompanied by variations in temperature, humidity, and smoke, are often early signs of thermal runaway in lithium-ion batteries. The hydrogen sensor has a range of 0–1000 ppm, the methane sensor 0–10000 ppm, the carbon monoxide sensor 0–1000 ppm, and the VOC sensor 0–500 ppm. The main data acquisition system structure is as follows: Figure 1 As shown.

[0016] In this example, based on Figure 1 A method for detecting thermal runaway in a lithium-ion battery, as shown in the figure, is described in [reference needed]. Figure 2 It includes the following steps: Step 1: Collect the concentration C of the i-th gas released by the lithium-ion battery at time t. i The data includes light intensity L(t), temperature T(t), and relative humidity RH(t), where i = 1, 2, 3, 4. When i = 1, it represents hydrogen (H2); when i = 2, it represents methane (CH4); when i = 3, it represents carbon monoxide (CO); and when i = 4, it represents volatile organic compounds (VOCs). This step involves multi-source data acquisition, which is the basis for subsequent discrimination. The hardware sampling time interval is 0.5s.

[0017] Step 2: Considering the zero-point drift of the gas sensor, the baseline concentration value B of the i-th gas released by the lithium-ion battery at time t is calculated using equation (1). i (t): (1) In equation (1), λ is the smoothing coefficient. In this embodiment, λ is 0.95, which can effectively suppress the slow drift of data while remaining sensitive to sudden signal increases. Let be the baseline concentration of the i-th gas released by the lithium-ion battery at time t-1, where t ≥ 1. When t=1, initialize... .

[0018] Step 3: Use equation (2) to obtain the baseline-free concentration of the i-th gas released by the lithium-ion battery at time t. : (2) This step yields the baseline concentration, retaining only the anomalous components of the gas concentration changes during lithium battery operation and rest. Experiments have shown that the baseline concentration of hydrogen often increases significantly earliest. Step 4: Calculate the change in light intensity outside the lithium-ion battery at time t using equation (3). : (3) In equation (3), In this embodiment, the sliding time window is set to two sampling time intervals, i.e. =1s; For the external of lithium-ion batteries The intensity of light at any given moment, if If the value suddenly exceeds 150 lux, it indicates that smoke is obscuring the view.

[0019] Step 5: Calculate the temperature change of the lithium-ion battery at time t according to equations (4) and (5). The relative humidity change at time t : (4) (5) In equations (4) and (5), For lithium-ion batteries Temperature at any moment; For lithium-ion batteries The relative humidity at any given time, if or This indicates an abnormality in temperature and humidity.

[0020] Step 6: To eliminate the sensitivity of the gas sensor to ambient temperature and humidity, the baseline concentration of the i-th gas released by the lithium-ion battery at time t needs to be calculated. A first-order linear correction is performed, and the corrected concentration of the i-th gas released by the lithium-ion battery at time t is obtained using equation (6). : (6) In equation (6), T ref The ambient reference temperature under standard conditions is set to 25℃, RH. ref The ambient reference humidity under standard conditions is set to 45%. α represents the sensitivity of the gas sensor to temperature deviations, set to 0.02 / ℃. β represents the sensitivity of the gas sensor to humidity deviations, set to 0.01 / %. This correction effectively distinguishes between "ambient fluctuations" and "real gas release," effectively reducing the impact of the environment on the gas sensor.

[0021] Step 7: In order to detect abnormal trends earlier, the concentration change rate of the i-th gas released by the lithium-ion battery at time t is obtained after correction using equation (7). : (7) In equation (7), For lithium-ion batteries The corrected concentration of the i-th gas released at time i.

[0022] Step 8: Set the gas concentration threshold to Th i ,when ≥ Th i If , then it means that the warning value of the i-th gas is triggered at time t, where Th1 is 500ppm, Th2 is 300ppm, Th3 is 150ppm, and Th4 is 200ppm. Set the light intensity change threshold to Th L =150 lux, when ≤ Th L , which represents the warning value for triggering smoke at time t; Set the temperature change threshold to Th T =3℃, set the relative humidity change threshold to Th RH =5%, if ≥ Th T or ≥ Th RH If , then it represents the temperature and humidity warning value triggered at time t.

[0023] Step 9: For each , , , , Standardization is performed to obtain the baseline value of the standardized gas concentration of the lithium-ion battery at time t. The rate of change of the concentration of the i-th gas in a lithium-ion battery after standardization at time t. The normalized change in light intensity of a lithium-ion battery at time t The standardized temperature change of a lithium-ion battery at time t The relative humidity change of a lithium-ion battery at time t after standardization. In this example, the normalization method uses threshold normalization, which scales each variable to the [0,1] interval according to its corresponding warning threshold: when the variable is less than or equal to 0, it is normalized to 0, and when the variable reaches or exceeds the threshold, it is normalized to 1, with a linear mapping in between.

[0024] Step 10: If the warning value of the i-th gas is triggered at time t, and the warning value of temperature and humidity is also triggered at time t, it means that the lithium-ion battery triggers the "gas-temperature combination" situation at time t. If the warning value for the i-th gas is triggered at time t, and the warning value for smoke is also triggered at time t, then it means that the lithium-ion battery triggers the "gas-smoke combination" situation at time t. If the warning values ​​for any two gases are triggered at time t, and the warning values ​​for temperature and humidity are also triggered at time t, then it means that the lithium-ion battery has triggered a "gas-gas-temperature combination" situation at time t. If the warning value for any two gases is triggered at time t, and the warning value for smoke is triggered at time t, then it means that the lithium-ion battery triggers a "gas-gas-smoke combination" situation at time t. If the warning value of any three gases is triggered at time t, it means that the lithium-ion battery has triggered the "three-gas linkage combination" at time t. If the warning values ​​for any two or more gases are triggered at time t, and the warning values ​​for smoke and temperature and humidity are triggered at time t, then it means that the lithium-ion battery has triggered the "all-element linkage combination" situation at time t.

[0025] Step 11: Set the trigger threshold number to N. If the total number of triggers of any combination among "gas-temperature combination", "gas-smoke combination", "gas-gas-temperature combination", "gas-gas-smoke combination", and "three-gas linkage combination" reaches N within a period of time T, then increment the combination trigger count value M by 1. If the total number of times the "all-element linkage combination" is triggered within a period of time T reaches N, the combination trigger count value M is incremented by 2. In this embodiment, the time window is set to T=30s and the trigger threshold number of times N=10, that is, the same combination is triggered 10 times within half a minute to be counted as one valid count.

[0026] Step 12: Use equation (8) to obtain the lithium-ion battery state combination technology score S over a period of time T. m : (8) In equation (8), γ is the weight coefficient of the combination, which is 5, and M is the final combination trigger count value. This design makes a strong combination equivalent to a risk score of 5, thereby increasing the weight of the combination evidence.

[0027] Step 13: To ensure that the risk score only considers positive contributions, this invention uses non-negativity constraints in subsequent calculations, and uses equation (9) to calculate the x-th variable after non-negativity constraints at time t. , x∈{C i L, T, RH, r i}; (9) In equation (9), Let x represent the x-th variable after standardization at time t, where x = 1, 2, 3, 4, 5; when x = 1, ... express When x=2, express When x=3, express When x=4, express When x=5, express .

[0028] Step 14: Calculate the continuous score of the lithium-ion battery state at time t using equation (10). : (10) In equation (10), express The weights of the baseline value of the standardized gas concentration of the lithium-ion battery at time t. The value is 0.3, representing the weight of the normalized, corrected concentration change rate of the i-th gas in the lithium-ion battery at time t. The value is 0.25, representing the weight of the standardized change in light intensity at time t for the lithium-ion battery. The value is 0.2, representing the weight of the standardized temperature change of the lithium-ion battery at time t. The value is 0.15, representing the weight of the standardized change in relative humidity at time t for the lithium-ion battery. The value is 0.1; this weighted value comprehensively considers the gas concentration, gas change rate, smoke concentration, and the temperature change of the lithium-ion battery itself under abnormal conditions, as well as the relative humidity change caused by the electrolyte released under abnormal conditions; x=1,2,3,4,5; when x=1, express When x=2, express When x=3, express When x=4, express When x=5, express .

[0029] Step 15: Taking into account the above steps, use equation (11) to obtain the total score R of the lithium-ion battery state over a period of time T: (11) Step 16: Define the early warning baseline threshold as θ1 and the alarm baseline threshold as θ2, and θ1 < θ2. In this example, θ1 = 20 and θ2 = 35. Define the hysteresis offset as ε, and ε <θ1, in this example, ε is set to 1; Define the upgrade / downgrade hold time as T. hold In this example, T is set hold= 10s; Calculate the uplink warning threshold θ up,1 =θ1 + ε; Downlink safety threshold θ down,1 =θ1 - ε; Uplink alarm threshold θ up,2 =θ² + ε; Downlink warning threshold θ down,2 =θ² - ε; and θ down,1 < θ up,1 <θ down,2 <θ up,2 In this example, the uplink warning threshold θ is set. up,1 = 21, downlink security threshold θ down,1 = 19, Uplink alarm threshold θ up,2 = 36, downlink warning threshold θ down,2 =34.

[0030] Step 17: Next, we will make the final risk level determination. The specific logic is as follows: Figure 3 As shown. If R ≥ θ up,2 If M ≥ 2, it indicates that the lithium-ion battery is in a serious abnormal state within a period of time T, and is judged to be at risk level 2. When the lithium-ion battery is judged to be at risk level 2, the hardware system controls the red LED to flash rapidly at a frequency of 3 Hz and the host computer triggers an audible and visual alarm, and generates a high-risk event package, including the trigger combination name and sensor type, raw data and temperature and humidity corrected values, trigger timestamp, risk scoring trajectory within the monitoring window, alarm data frame sent via CAN bus, and can also trigger a relay to cut off the battery output to ensure safety, and execute step 18.

[0031] If R ≥ θ up,1 If M = 1, it indicates that the lithium-ion battery status is significantly abnormal within a period of time T, and is judged as a level 1 risk level. When the lithium-ion battery status is judged to be a level 1 risk level, the hardware system controls the yellow LED to flash at a frequency of 1 Hz, and the host computer pops up a prompt and records the warning log, including the timestamp, the specific indicators triggered, the current risk score, and sends a warning data frame through the CAN bus, and executes step 19.

[0032] If R ≤ θ down,1 If M=0, it means that the lithium-ion battery status has not shown any obvious abnormalities within a period of time T, and is judged to be at the risk level. When the lithium-ion battery status is judged to be at the risk level, the hardware system controls the green LED to be constantly lit and the CAN bus to send regular data frames (low priority), and the host computer status displays "normal", without triggering a warning, and executes step 20.

[0033] Step 18, if θ down,1 ≤ R ≤ θ down,2 And the duration of M=1 reaches T. hold If so, the risk level will be reduced to Level 1; If R ≤θ down,1 And the duration of M=0 reaches T. hold If so, it will be downgraded to a risk-free level; Otherwise, maintain the risk level as Level 2 and proceed to step 21; Step 19: If R ≤ θ down,1 And the duration of M=0 reaches T. hold If so, it will be downgraded to a risk-free level; If θ down,2 ≤ R ≤θ up,2 And the duration of M=1 reaches T. hold If so, the risk level will remain at Level 1. Otherwise, proceed to step 21; Step 20: If R ≤ θ up,1 And the duration of M=0 reaches T. hold If so, the risk level remains unchanged; Otherwise, proceed to step 21.

[0034] Step 21: After assigning T+t to t, return to Step 1 and execute sequentially, that is, advance the time window forward by a duration T and enter the next judgment cycle. Repeat this process until the judgment time ends.

[0035] In this embodiment, an electronic device includes a memory and a processor. The memory stores a program that supports the processor in executing the above-described method, and the processor is configured to execute the program stored in the memory.

[0036] In this embodiment, a computer-readable storage medium stores a computer program, which is executed by a processor to perform the steps of the above method.

Claims

1. A method for determining thermal runaway in lithium-ion batteries, characterized in that, Includes the following steps: Step 1: Collect the concentration C of the i-th gas released by the lithium-ion battery at time t. i (t), light intensity L(t), temperature T(t), relative humidity RH(t), where i=1,2,3,4, when i=1, it represents hydrogen H2, when i=2, it represents methane CH4, when i=3, it represents carbon monoxide CO, and when i=4, it represents volatile organic compound gas VOC; Step 2: Based on the data from Step 1, calculate the baseline-free concentration of the i-th gas released by the lithium-ion battery at time t. The change in light intensity outside the lithium-ion battery at time t Temperature change of lithium-ion battery at time t The relative humidity change at time t ; Step 3: Based on the data from Step 2, calculate the corrected concentration of the i-th gas released by the lithium-ion battery at time t. and its concentration change rate ; Step 4: Set the gas concentration threshold to Th i ,when ≥ Th i If , then it means that the warning value for the i-th gas is triggered at time t; Set the light intensity change threshold to Th L ,when ≤ Th L , which represents the warning value for triggering smoke at time t; Set the temperature change threshold to Th T Set the relative humidity change threshold to Th RH ,like ≥ Th T or ≥ Th RH If , then it represents the temperature and humidity warning value triggered at time t; Step 5: For each , , , , Standardization is performed to obtain the baseline value of the standardized gas concentration of the lithium-ion battery at time t. The rate of change of the concentration of the i-th gas in a lithium-ion battery after standardization at time t. The normalized change in light intensity of a lithium-ion battery at time t The standardized temperature change of a lithium-ion battery at time t The relative humidity change of a lithium-ion battery at time t after standardization. ; Step 6: If the warning value of the i-th gas is triggered at time t, and the warning value of temperature and humidity is also triggered at time t, it means that the lithium-ion battery triggers the "gas-temperature combination" situation at time t. If the warning value for the i-th gas is triggered at time t, and the warning value for smoke is also triggered at time t, then it means that the lithium-ion battery triggers the "gas-smoke combination" situation at time t. If the warning values ​​for any two gases are triggered at time t, and the warning values ​​for temperature and humidity are also triggered at time t, then it means that the lithium-ion battery has triggered a "gas-gas-temperature combination" situation at time t. If the warning value for any two gases is triggered at time t, and the warning value for smoke is triggered at time t, then it means that the lithium-ion battery triggers a "gas-gas-smoke combination" situation at time t. If the warning value of any three gases is triggered at time t, it means that the lithium-ion battery triggers the "three-gas linkage combination" at time t. If the warning values ​​for any two or more gases are triggered at time t, and the warning values ​​for smoke and temperature and humidity are triggered at time t, then it means that the lithium-ion battery has triggered the "all-element linkage combination" situation at time t. Step 7: Set the trigger threshold number of times to N. If the total number of times any of the following combinations is triggered within a period of time T reaches N, then the combination trigger count value M is incremented by 1. If the total number of times "all-element linkage combination" is triggered reaches N within a period of time T, the combination trigger count value M will be incremented by 2. Step 8: Based on the data from M and Step 5, calculate the total score R of the lithium-ion battery state over a period of time T; Step 9: Define the early warning baseline threshold as θ1 and the alarm baseline threshold as θ2, where θ1 < θ2; Define the hysteresis offset as ε, where ε < θ1; Define the upgrade / downgrade hold time as T. hold ; Calculate the uplink warning threshold θ respectively up,1 =θ1+ ε、Downlink security threshold θ down,1 =θ1- ε、Uplink alarm threshold θ up,2 =θ² + ε; Downlink warning threshold θ down,2 =θ² - ε; and θ down,1 <θ up,1 <θ down,2 <θ up,2 ; Step 10: If R ≥ θ up,2 If M ≥ 2, it indicates that the lithium-ion battery state is seriously abnormal within a period of time T, and is judged as a level 2 risk level, and step 11 is executed; If R ≥ θ up,1 If M = 1, it indicates that there is a significant abnormality in the state of the lithium-ion battery within a period of time T, which is judged as a level 1 risk level, and step 12 is executed. If R ≤ θ down,1 If M=0, it means that the lithium-ion battery status has not shown any obvious abnormalities within a period of time T, and is judged to be at a risk-free level, and step 13 is executed. Step 11, if θ down,1 ≤ R≤ θ down,2 And the duration of M=1 reaches T. hold If so, the risk level will be reduced to Level 1; If R≤θ down,1 And the duration of M=0 reaches T. hold If so, it will be downgraded to a risk-free level; Otherwise, maintain the risk level as Level 2 and proceed to step 21; Step 12: If R ≤ θ down,1 And the duration of M=0 reaches T. hold If so, it will be downgraded to a risk-free level; If θ down,2 ≤ R≤θ up,2 And the duration of M=1 reaches T. hold If so, the risk level will remain at Level 1. Otherwise, proceed to step 21; Step 13, if R≤θ up,1 And the duration of M=0 reaches T. hold If so, the risk level remains unchanged; Otherwise, proceed to step 21; Step 14: After assigning T+t to t, return to Step 1 and execute sequentially until the judgment time ends.

2. The method for determining thermal runaway of a lithium-ion battery according to claim 1, characterized in that, Step 2 includes the following steps: Step 2.1: Calculate the baseline concentration B of the i-th gas released by the lithium-ion battery at time t using equation (1). i (t): (1) In equation (1), λ is the smoothing coefficient. Let be the baseline concentration of the i-th gas released by the lithium-ion battery at time t-1, where t ≥ 1. When t=1, initialize... ; Step 2.2: Use equation (2) to obtain the baseline concentration of the i-th gas released by the lithium-ion battery at time t. : (2) Step 2.3: Calculate the change in light intensity outside the lithium-ion battery at time t using equation (3). : (3) In equation (3), For sliding time windows; For the external of lithium-ion batteries Light intensity at any given moment; Step 2.4: Calculate the temperature change of the lithium-ion battery at time t according to equations (4) and (5). The relative humidity change at time t : (4) (5) In equations (4) and (5), For lithium-ion batteries Temperature at any moment; For lithium-ion batteries Relative humidity at any given time.

3. The method for determining thermal runaway of a lithium-ion battery according to claim 2, characterized in that, Step 3 includes the following steps: Step 3.1: Use equation (6) to obtain the corrected concentration of the i-th gas released by the lithium-ion battery at time t. : (6) In equation (6), T ref The ambient reference temperature under standard conditions, RH ref The ambient reference humidity is given under standard conditions; α is the sensitivity of the gas sensor to temperature deviations; and β is the sensitivity of the gas sensor to humidity deviations. Step 3.2: Use equation (7) to obtain the corrected concentration change rate of the i-th gas released by the lithium-ion battery at time t. : (7) In equation (7), For lithium-ion batteries The corrected concentration of the i-th gas released at time i.

4. The method for determining thermal runaway of a lithium-ion battery according to claim 3, characterized in that, Step 8 includes the following steps: Step 8.1: Use equation (8) to obtain the lithium-ion battery state combination technology score S over a time period T. m : (8) In equation (8), γ is the weighting coefficient of the combination, and M is the final combination trigger count value; Step 8.2: Calculate the x-th variable after nonnegativity constraint at time t using equation (9). , x∈{C i L, T, RH, r i }; (9) In equation (9), Let x represent the x-th variable after standardization at time t, where x = 1, 2, 3, 4, 5; when x = 1, ... express When x=2, express ; When x=3 express When x=4, express ; When x=5 express ; Step 8.3: Calculate the continuous score of the lithium-ion battery state at time t using equation (10). : (10) In equation (10), express The weights are given by x = 1, 2, 3, 4, 5; when x = 1, express When x=2, express ; When x=3 express ; When x=4 express ; When x=5 express ; Step 8.4: Use equation (11) to obtain the total score R of the lithium-ion battery state over a time period T: (11)。 5. An electronic device, comprising a memory and a processor, characterized in that, The memory is used to store a program that supports the processor in executing the discrimination method according to any one of claims 1-4, and the processor is configured to execute the program stored in the memory.

6. A computer-readable storage medium storing a computer program thereon, characterized in that, The computer program is executed by the processor to perform the steps of the discrimination method according to any one of claims 1-4.

Images