A lithium battery monitoring and early warning method and system

Abstract

The application discloses a lithium battery monitoring and early warning method and system, and the method comprises the following steps: pasting and installing a fiber grating temperature sensor and a fiber grating strain sensor on the surface of a battery; applying a preset pre-tightening force to the whole battery; performing adhesion and curing, temperature balance, reference wavelength collection, temperature compensation, pre-tightening force coupling correction and elimination of the influence of thin plate temperature deformation additional strain items on the fiber grating strain sensor and the fiber grating temperature sensor, and obtaining a final strain value; extracting a thermal runaway feature, wherein the thermal runaway feature comprises a strain change rate and a strain accumulation amount; performing multi-stage early warning on the battery based on the thermal runaway feature; and the application has the advantages of high monitoring accuracy and reliability.

Description

Technical Field

[0001] This invention relates to the field of lithium battery safety monitoring technology, specifically to a lithium battery monitoring and early warning method and system. Background Technology

[0002] With the rapid development of the new energy industry, large-capacity lithium iron phosphate batteries are widely used in energy storage power stations, electric vehicles, and other fields. However, lithium batteries are prone to thermal runaway under abnormal operating conditions such as overcharging, short circuits, and high temperatures, which can lead to fires, explosions, and other safety accidents, seriously threatening the safety of people and property. Therefore, real-time monitoring and early warning of lithium battery status are of great significance.

[0003] Currently, lithium battery monitoring mainly employs traditional electrical signal sensors and thermocouples. Traditional electrical signal sensors are susceptible to electromagnetic interference, and their measurement accuracy and reliability are difficult to guarantee in high-voltage, strong electromagnetic environments. Temperature sensors such as thermocouples have slow response speeds and can only measure local surface temperatures, failing to capture subtle changes in the early stages of thermal runaway within the battery.

[0004] Fiber Bragg Grating (FBG) sensors offer advantages such as electromagnetic interference resistance, small size, light weight, corrosion resistance, and distributed measurement capabilities, and have been increasingly applied in lithium battery monitoring in recent years. However, existing FBG lithium battery monitoring solutions mainly suffer from the following three types of technical shortcomings: The first type of defect is its overemphasis on temperature monitoring and insufficient utilization of mechanical parameters. Existing solutions primarily focus on temperature monitoring, neglecting the monitoring of mechanical parameters such as deformation and expansion forces that accompany battery thermal runaway. In reality, before thermal runaway occurs in large-capacity lithium iron phosphate batteries, a series of internal chemical reactions occur, generating gases and causing the battery volume to expand and the casing to deform. This mechanical deformation often precedes a significant increase in temperature and is a crucial characteristic parameter for early warning of thermal runaway. Current technologies fail to fully utilize strain sensing information for early warning of thermal runaway, resulting in insufficient warning time windows and a need to improve warning accuracy.

[0005] The second type of defect: Lack of standardized design in sensor deployment and installation processes. The existing methods for arranging fiber optic sensors on the battery surface lack systematic design. Key technical issues such as sensor placement, preload control, and temperature compensation have not been adequately addressed, affecting the accuracy and reliability of monitoring. Specifically, this manifests as: arbitrary sensor placement without considering the deformation distribution patterns of large-capacity square batteries; inconsistent bonding processes leading to significant differences in strain transfer efficiency; and a lack of quantitative basis for preload application, resulting in mismatches with the battery's factory clamping force and poor comparability of measurement results between different batteries.

[0006] The third type of defect: The temperature-strain decoupling method is simplistic and does not consider the mechanical properties of large-capacity square aluminum shells. Chinese Patent Publication No. CN121324947A discloses a battery pack thermal runaway early warning method and system based on an enhanced fiber optic sensor, employing a time-series network early warning scheme combining fiber optic temperature and strain monitoring. However, it has three key defects for 314Ah large-capacity square lithium iron phosphate batteries: 1) The temperature compensation model is coarse: It only uses a general linear temperature compensation method and does not take into account the additional strain caused by the thermal bending of the large-capacity square aluminum shell thin plate. This results in a decoupling error of up to 28με (micro-stress) in the range from room temperature to 60℃, which affects the accuracy and reliability of monitoring. 2) Rigid temporal interpolation method: The use of linear temporal interpolation with fixed weights cannot adapt to the nonlinear working condition of instantaneous strain change during thermal runaway. The data alignment error at the moment of change is large, which affects the accuracy and reliability of monitoring. 3) Lack of preload coupling effect: No coupling calibration model for preload and shell deformation was established, and the system offset caused by the superposition of sensor installation preload and battery factory clamping force was ignored. The measurement consistency among large-capacity batteries was poor, which affected the accuracy and reliability of monitoring.

[0007] For large-capacity batteries with a 314Ah square lithium iron phosphate dual-core structure, existing technologies have not yet formed a complete and systematic solution from sensor deployment, pre-tightening force calibration, temperature decoupling to graded early warning, resulting in low accuracy and reliability of early warning. Summary of the Invention

[0008] The technical problem to be solved by this invention is that the accuracy and reliability of existing methods for real-time monitoring and early warning of lithium battery status are not high.

[0009] This invention solves the above-mentioned technical problems through the following technical means: a lithium battery monitoring and early warning method, comprising the following steps: S1. Attach and install a fiber optic temperature sensor and a fiber optic strain sensor onto the surface of the battery; S2. Apply a preset preload force to the entire battery; S3. The fiber optic strain sensor and fiber optic temperature sensor are bonded, cured, and allowed to stand at a constant temperature. The reference wavelength is then acquired, temperature is compensated, preloaded, coupled, and the influence of the additional strain term due to temperature deformation of the thin plate is eliminated to obtain the final strain value. S4. Extract thermal runaway features, including strain change rate and strain accumulation. S5. Provides multi-level early warning for batteries based on thermal runaway characteristics.

[0010] Further, S1 includes: A fiber optic temperature sensor and a fiber optic strain sensor are installed near the positive electrode side at the geometric center of the battery side. The fiber optic strain sensor is in the form of bare optical fiber, and the fiber optic temperature sensor is in the form of a sleeved package.

[0011] Furthermore, the bonding and curing stand is to stand at room temperature for a first preset time, and the temperature equilibration stand is to place the battery and sensor together in a constant temperature environment for a second preset time; the reference wavelength acquisition is to continuously acquire the wavelength data of the fiber optic strain sensor and the fiber optic temperature sensor for 30 seconds after the temperature is completely balanced, and take the average value as the reference wavelength of the fiber optic strain sensor and the fiber optic temperature sensor.

[0012] Furthermore, data synchronization processing is required during the acquisition of the reference wavelength. This data synchronization processing includes: Interpolate the temperature data to the same time point as the strain data to achieve point-to-point synchronization:

[0013] in, For the i-th strain sampling time of the strain data, and For two adjacent temperature sampling times, For the k-th temperature sampling time The corresponding temperature value, For the (k+1)th temperature sampling time The corresponding temperature value, For the i-th strain sampling time The corresponding temperature value, and These are all weighting coefficients, and the dynamic weighting calculation formula is as follows:

[0014] In the formula, This represents the temperature rise rate between adjacent sampling intervals.

[0015] Furthermore, the temperature compensation process is as follows: The reference wavelength of the fiber optic strain sensor is denoted as The reference wavelength of the fiber optic grating temperature sensor is denoted as Bragg wavelength shift of fiber optic strain sensors = - , The wavelength of the fiber optic strain sensor at the current moment is given by the temperature compensation formula. ,in, The actual strain value obtained after temperature compensation. This represents the strain sensitivity coefficient of the fiber Bragg grating strain sensor. The temperature sensitivity coefficient of the fiber Bragg grating strain sensor; The temperature change measured by the fiber Bragg grating temperature sensor. , The wavelength shift of the fiber optic temperature sensor and , This represents the wavelength of the fiber Bragg grating temperature sensor at the current moment.

[0016] Furthermore, the process of preload coupling correction is as follows: Constructing the preload coupling correction formula In the formula, To correct the decoupling strain value; Apply preload to the sensor; The preload coupling correction coefficient and ,in, For aluminum shell thickness, For shell bending stiffness, The elastic modulus of the battery casing. This refers to the elastic modulus of the optical fiber.

[0017] Furthermore, the preload coupling correction coefficient The calibration method involves collecting data on the aluminum shell thickness, shell bending stiffness, shell bending stiffness, and fiber elastic modulus under four preload levels of 2KN, 2.5KN, 3KN, and 3.5KN. These values ​​are then substituted into the preload coupling correction coefficient formula for calculation, and a linear curve is fitted to obtain the final calibration result. .

[0018] Furthermore, the elimination of the effect of the additional strain term due to temperature deformation of the thin plate includes: Additional strain term due to temperature deformation of thin plate In the formula, For the fourth-order Laplace deflection term of the shell thin plate; The coefficient of thermal expansion of the aluminum shell; The increased temperature causes the aluminum shell sheet to expand, resulting in additional strain. The final strain value .

[0019] Further, S5 includes: When the strain change rate is ≥0.25με / s for 10 s continuously and the cumulative strain is ≥350με, a Level 1 warning is issued, corresponding to an ultra-early gas production warning. When the cumulative strain within 30 seconds is ≥500με and the strain change rate is ≥5με / s, a level two warning is issued, corresponding to the self-acceleration warning of the side reaction. When the absolute value of the strain change rate within 1 second is ≥300με / s, a Level 3 warning is issued, corresponding to the critical chain reaction warning.

[0020] The present invention also provides a lithium battery monitoring and early warning system, which performs the above-described method, including: The sensor deployment module is used to attach and install fiber Bragg grating temperature sensors and fiber Bragg grating strain sensors on the battery surface. The preload application module is used to apply a preset preload to the entire battery. The calibration module is used to perform bonding, curing, and static setting, temperature equilibration, reference wavelength acquisition, temperature compensation, preload coupling correction, and elimination of the influence of additional strain terms due to temperature deformation of thin plates on fiber optic strain sensors and fiber optic temperature sensors, so as to obtain the final strain value. The feature extraction module is used to extract thermal runaway features, including strain change rate and strain accumulation. The early warning module is used to provide multi-level early warnings for batteries based on thermal runaway characteristics.

[0021] The advantages of this invention are: (1) The present invention applies a preset preload to the fiber Bragg grating strain sensor and the fiber Bragg grating temperature sensor, thereby taking into account the system offset caused by the superposition of the sensor installation preload and the battery factory clamping force, and improving the accuracy and reliability of the early warning. The fiber Bragg grating strain sensor and the fiber Bragg grating temperature sensor are bonded, cured and left to stand to ensure that the adhesive layer is completely cured, avoiding baseline drift caused by adhesive layer creep. Temperature balancing and standing ensure that the battery temperature is completely consistent with the ambient temperature, eliminating the influence of temperature gradient on the initial wavelength. Temperature compensation eliminates temperature coupling interference. Preload coupling correction eliminates the influence of individual sensor differences on the measurement results, as well as the influence of the additional strain term of thin plate temperature deformation, to obtain the final strain value, thereby further improving the accuracy and reliability of the early warning.

[0022] (2) This invention establishes a complete standardized sensor deployment process through a four-layer bonding process consisting of side geometric center + position optimization near the positive electrode (combined with deformation distribution law), sanding + anhydrous ethanol cleaning + epoxy resin adhesive + high-temperature tape, and a decoupled combination scheme of bare optical fiber strain + sleeved temperature. This improves the consistency of strain measurement between different batteries and solves the problems of arbitrary deployment and poor data comparability in existing technologies.

[0023] (3) The present invention adopts a 3KN pre-tightening force consistent with the factory clamping force of the 314Ah battery, and constructs a pre-tightening force coupling correction formula based on Kirchhoff thin plate theory. Through the four-level pre-tightening force calibration fitting correction coefficient, the system error caused by the superposition of the clamp pre-tightening force and the battery factory clamping force is eliminated.

[0024] (4) The present invention constructs a systematic temperature compensation system from three dimensions: spatial consistency (sensor spacing ≤ 5mm), time synchronization (same demodulator acquisition + dynamic weighted interpolation), and algorithm accuracy (reference temperature compensation + thin plate deformation correction term). The full temperature range strain measurement error is reduced and the accuracy is improved compared with the prior art.

[0025] (5) The present invention constructs a three-level early warning model. The first level of early warning corresponds to the gas production start-up stage, the second level corresponds to the side reaction self-acceleration stage, and the third level corresponds to the critical chain reaction stage. Compared with the traditional single temperature threshold early warning, the first level early warning (ultra-early gas production early warning) can advance the early warning time window by 6 to 12 minutes, which buys valuable time for emergency response.

[0026] (6) The present invention adopts a non-invasive monitoring scheme with surface bonding, which does not damage the internal structure of the battery and does not affect the electrochemical performance of the battery; based on wavelength division multiplexing technology, multiple grating points can be connected in series; it is easy to extend to distributed safety monitoring at the battery module and battery pack level, and adapts to the application needs of energy storage systems of different scales. Attached Figure Description

[0027] Figure 1 This is a schematic flowchart of a lithium battery monitoring and early warning method disclosed in an embodiment of the present invention; Figure 2 This is a schematic diagram of the sensor arrangement on the battery surface in a lithium battery monitoring and early warning method disclosed in an embodiment of the present invention; Figure 3 This is a schematic diagram of the fiber optic strain sensor and temperature sensor in a lithium battery monitoring and early warning method disclosed in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Example 1 like Figure 1 As shown, Embodiment 1 of the present invention provides a lithium battery monitoring and early warning method, comprising the following steps: S1, Sensor Deployment A 314Ah square lithium iron phosphate battery (rated voltage range 2.5V-3.65V) was selected as the monitoring object. The core of this step lies in designing a standardized sensor deployment scheme based on the deformation distribution law of large-capacity square lithium iron phosphate batteries to ensure the consistency and reliability of strain measurements.

[0030] Location selection criteria: Sensors are placed near the positive electrode side at the geometric center of the battery side, with fiber Bragg grating temperature sensors and fiber Bragg grating strain sensors attached respectively. This location selection is not arbitrary, but based on the following physical mechanisms: (1) The geometric center of the side is the area with the most uniform battery deformation, which can avoid strain measurement deviation caused by edge effects; (2) It is near the positive electrode side because the volume change of the positive electrode material is the main source of battery expansion, and thermal runaway often starts from the positive electrode side first. Placing sensors at this location can capture the early signs of thermal runaway.

[0031] Please see Figure 2 , Figure 2 The left side shows a schematic diagram of the front of the battery. Figure 2 The diagram on the right shows the side of the battery. The top surface is the location of the positive and negative terminals, and the opposite side is the bottom surface. A fiber Bragg grating temperature sensor and a fiber Bragg grating strain sensor are arranged side-by-side along the battery side and perpendicular to the bottom surface, with the distance between the two sensors controlled within 5mm. The grating length is 8mm for both sensors, ensuring they are in identical temperature fields. This close-proximity side-by-side arrangement is the spatial basis of the temperature compensation scheme in this invention—only by ensuring that the temperature sensor and strain sensor sense exactly the same temperature field can the subsequent reference temperature compensation algorithm achieve optimal results.

[0032] Pre-treatment process for bonding: Before bonding, lightly sand the bonding area on the battery surface with sandpaper to remove the surface oxide layer and stains, increasing the roughness of the bonding surface; then wipe it clean with anhydrous ethanol cotton balls to remove dust and oil generated from sanding, and let it dry before use. This pre-treatment process can improve the bonding strength and prevent the sensor from falling off under high temperature conditions.

[0033] Adhesion and Fixing Process: High-temperature resistant epoxy resin adhesive is used for bonding to ensure a tight fit between the sensor and the battery casing. After bonding, high-temperature tape is used for secondary fixation, forming a dual-protection mechanism of adhesive bonding and tape fixation. This dual-fixing scheme ensures effective strain transmission and prevents sensor displacement caused by softening of the adhesive layer at high temperatures.

[0034] Please see Figure 3The fiber optic strain sensor uses bare optical fiber to directly sense the deformation of the battery casing. The fiber optic temperature sensor uses a sheathed enclosure. The difference between sheathed and unsheathed sensors lies in whether the grating area of ​​the sensor is sheathed. If the grating area is sheathed, it is a sheathed enclosure; if the grating area is unsheathed, it is a bare optical fiber. The sheath does not simply provide protection, but rather achieves mechanical decoupling—the sheath prevents the temperature sensor from being affected by the deformation of the battery casing, allowing it to sense only temperature changes. This enables independent measurement of temperature and strain, providing a clean temperature reference signal for subsequent temperature compensation.

[0035] S2. Preload Application: Apply a preload of 3KN to the battery assembly with the fiber Bragg grating strain sensor and fiber Bragg grating temperature sensor attached to ensure a tight fit between the sensors and the battery casing. The preload is applied using a dedicated steel clamp, consistent with the battery's factory clamping force, eliminating the influence of preload differences on strain measurement results.

[0036] Technical principle of preload: Preload is a key parameter affecting the accuracy of strain measurement. If the preload is too small, the sensor will not fit tightly with the housing, resulting in low strain transfer efficiency and large measurement error; if the preload is too large, it may damage the fiber optic grating or cause baseline drift.

[0037] The determination of the 3KN preload force: The 3KN preload force in this invention is not an empirical value, but is consistent with the factory clamping force of the 314Ah square lithium iron phosphate battery. The technical effect of this design is that it eliminates the superposition effect of the additional stress introduced by the sensor installation and the original clamping force of the battery, so that the measured strain increment can directly reflect the real deformation caused by the change in the internal state of the battery, rather than the coupling effect between the sensor installation process and the battery structure.

[0038] The preload is applied using a specialized steel clamp to ensure its uniformity and repeatability. Extensive experimental verification has shown that the strain transfer efficiency is stable under this preload parameter, and the sensor's lifespan remains unaffected.

[0039] S3. Zero-point calibration and sensor initialization: Perform system initialization calibration of the sensor in a constant temperature environment at room temperature (25±2℃). The specific process is as follows: (1) Bonding and curing: After the sensor is bonded, it is left to stand at room temperature for 2 hours to ensure that the bonding layer is fully cured and to avoid baseline drift caused by the creep of the bonding layer. Experiments show that if sufficient curing and standing are not carried out, the baseline drift can reach 50~80με (micro-stress) in the first 2 hours, which seriously affects the accuracy of subsequent measurements.

[0040] (2) Temperature equilibration: Place the battery and the test device together in a constant temperature environment for 2 hours to ensure that the battery temperature is completely consistent with the ambient temperature and eliminate the influence of temperature gradient on the initial wavelength. Since the battery has a large heat capacity, if sufficient temperature equilibration is not carried out, the temperature difference between the inside and the surface of the battery will lead to inaccurate initial wavelength, which will introduce systematic measurement error.

[0041] (3) Reference wavelength acquisition: After the temperature is completely balanced, wavelength data is continuously acquired for 30 seconds, and the average value is taken as the reference wavelength of the sensor. The reference wavelength of the fiber optic strain sensor is denoted as . The reference wavelength of the fiber optic grating temperature sensor is denoted as In this embodiment, the reference wavelength of the fiber optic strain sensor is 1550.8150 nm, and the reference wavelength of the fiber optic temperature sensor is 1549.7150 nm. By using 30-second averaging instead of single-point acquisition, the random noise of the demodulator can be effectively suppressed, reducing the uncertainty of the reference wavelength by an order of magnitude.

[0042] In this embodiment, two optical fibers (a fiber Bragg grating temperature sensor and a fiber Bragg grating strain sensor) are synchronously connected to the same fiber Bragg grating demodulator. Wavelength division multiplexing (WDM) technology is used to synchronously acquire wavelength data from both sensors. The demodulator's sampling frequency is 2Hz, ensuring complete consistency of the timestamps for both signals. The two signals are acquired through different channels of the same demodulator, and a unified internal clock within the demodulator ensures strict synchronization of the timestamps. This hardware-level synchronization scheme offers significantly higher accuracy than software synchronization between two independent devices, providing the time basis for achieving high-precision temperature compensation. The data synchronization process is as follows: Timestamp Alignment: Since the two signals are acquired through different channels of the same demodulator, the unified clock inside the demodulator ensures strict synchronization of the timestamps of the two data streams, with a synchronization error ≤10μs. For data with different sampling rates, an autonomous dynamic weighted interpolation formula is used to interpolate the temperature data to the same time node as the strain data, achieving point-to-point synchronization.

[0043] in, For the i-th strain sampling time of the strain data, and For two adjacent temperature sampling times, For the k-th temperature sampling time The corresponding temperature value, For the (k+1)th temperature sampling time The corresponding temperature value, For the i-th strain sampling time The corresponding temperature value, and These are all weighting coefficients. For any strain sampling time... It is necessary to find two adjacent temperature sampling points in the temperature sampling sequence that satisfy... ,Right now It falls between the k-th and (k+1)-th temperature sampling times. Let the temperature sampling rate be... The sampling interval is = Then: if the strain sampling rate is N times the temperature ( = Therefore, there is only one temperature sampling point for every N strain sampling points. Interpolation requires using two consecutive temperature points to calculate the temperature value at each intermediate strain moment.

[0044] The formula for dynamic weight calculation is as follows:

[0045] In the formula, The weighting of the sampling rate is dynamically adjusted based on the temperature change rate between adjacent sampling intervals; during periods of sudden temperature changes, the weighting of the most recent sampling time is automatically increased, while the weighting is balanced for gradual temperature rises. Dynamic weighted interpolation adapts to the abrupt temperature change conditions during thermal runaway.

[0046] (4) Sensitivity coefficient calibration: The strain sensitivity coefficient of the fiber optic grating strain sensor was determined through static calibration experiments. (Unit: pm / με) and temperature sensitivity coefficient (Unit: pm / ℃), and the temperature sensitivity coefficient of the fiber Bragg grating temperature sensor. In this embodiment, the values ​​of each sensitivity coefficient are directly given through experiments. For example, the strain sensitivity coefficient of the fiber optic strain sensor. The temperature sensitivity coefficient of the fiber Bragg grating strain sensor is 1.20 pm / με. and the temperature sensitivity coefficient of the fiber Bragg grating temperature sensor All values ​​were 9.9 pm / ℃. Temperature calibration was performed using a high and low temperature test chamber in a 10℃ step range from -40℃ to 40℃, with each temperature point held for 30 minutes to ensure thermal equilibrium. The measurement was repeated three times. Strain calibration was performed by loading weights in stages in a range from -1900 to 1750με. Each load point was held for 5 minutes until the wavelength stabilized, and the measurement was recorded. The measurement was repeated three times.

[0047] The function of the reference wavelength: reference wavelength It serves as the reference benchmark for all subsequent strain calculations. Based on the fiber optic grating sensing principle, the Bragg wavelength shift... = - It includes both strain and temperature terms. The wavelength of the fiber optic strain sensor at the current moment can be used to eliminate temperature coupling interference using the temperature compensation formula:

[0048] in, The actual strain value obtained after temperature compensation. This represents the actual wavelength drift of the strain sensor. The temperature change measured by the fiber Bragg grating temperature sensor. , The wavelength shift of the temperature sensor and , This is the wavelength of the fiber Bragg grating temperature sensor at the current moment. The reference wavelength is the starting point for calculating all strain data.

[0049] (5) Existing calibration only obtains , The prior art patents did not consider the additional strain shift caused by the coupling between the 3KN preload of the 314Ah battery and the sensor preload, and completely ignored the preload coupling error. This invention, based on Kirchhoff's thin-plate theory, constructs a preload coupling correction formula:

[0050] In the formula, To correct the decoupling strain value; This solution applies a preload force of 3 kN to the sensor. This is the preload coupling correction factor, in με / KN;

[0051] in, For the thickness of the aluminum shell, in this embodiment The value is 1.5mm; For the shell bending stiffness, this embodiment The value is 22.1N. m; The elastic modulus of the battery casing is given in this embodiment. The value is 70 GPa; This refers to the elastic modulus of the optical fiber.

[0052] The calibration method involves repeated strain calibration under four levels of preload: 2KN, 2.5KN, 3KN, and 3.5KN. This means that under various stress conditions, the aluminum shell thickness, shell bending stiffness, shell bending stiffness, and fiber elastic modulus are collected, substituted into the preload coupling correction coefficient formula, and a linear curve is fitted to obtain the final value. This embodiment .

[0053] The technical advantage of this calibration process is that each sensor has an independent preload coupling correction coefficient, instead of using the nominal value provided by the manufacturer, thereby eliminating the influence of individual sensor differences on the measurement results and improving the measurement accuracy by one level.

[0054] In addition, to eliminate spurious strain caused by the physical expansion of the aluminum shell due to heat, and to avoid the sensor misinterpreting it as an increase in battery internal pressure, thus eliminating thermally induced baseline drift, this embodiment calculates the additional strain term due to temperature deformation of the thin plate:

[0055] In the formula, For the fourth-order Laplace deflection term of the shell thin plate; The coefficient of thermal expansion of the aluminum shell; The additional strain caused by the expansion of the aluminum shell sheet due to temperature rise is calculated. By calculating the additional strain term of the sheet sheet under temperature deformation, the measured values ​​can accurately reflect the internal stress state of the battery, significantly improving the long-term stability and data reliability of strain monitoring under high-temperature conditions, thereby improving the accuracy of thermal runaway early warning and effectively reducing the risk of false alarms.

[0056] The final strain value .

[0057] Existing technologies only statically calibrate sensitivity, without considering the additional strain caused by the bending effect of the thin plate of the large-capacity square aluminum battery casing, which leads to baseline drift. This invention eliminates the systematic error caused by the superposition of the clamp preload and the battery factory clamping force through a preload coupling correction formula.

[0058] S4. Thermal runaway feature extraction: Real-time analysis of the acquired strain data to extract the following multi-dimensional thermal runaway feature parameters: (1) Exhaustive enumeration of characteristic parameters: Strain change rate dε / dt: The amount of strain change per unit time, with units of με / s, reflects the rate of battery expansion and is a core indicator for identifying the start of gas production. Under normal operating conditions, the strain change rate is slow and stable. In the early stage of thermal runaway, the internal chemical reaction accelerates, the gas production rate increases, and the strain change rate increases significantly.

[0059] Cumulative strain Δε: The total change in strain within a specified time window, reflecting the cumulative extent of battery expansion; where Δε = ε(t end )-ε(t start ), t start t is the start time of the time window. end This parameter represents the end time of the time window. It effectively filters out interference from short-term fluctuations and improves the stability of early warnings.

[0060] S5. Multi-parameter fusion three-level thermal runaway early warning model: Construct a three-level early warning model corresponding to three risk levels: gas production start-up, side reaction self-acceleration, and critical chain reaction.

[0061] Level 1 warning (ultra-early gas production warning) – corresponds to the inflection point where the electrolyte evolution changes from slow growth to accelerated growth, marking the start of electrolyte oxidation and decomposition and the initiation of internal gas production.

[0062] Judgment criteria: The strain change rate is ≥0.25με / s for 10 s continuously, and the cumulative strain is ≥350με. The 0.25 με / s is the average strain rate during the slow growth phase, which can effectively identify the first deviation from the strain growth trend; the 350 με strain accumulation auxiliary threshold is lower than the minimum strain at the inflection point of 0.5C, 0.75C, and 1C (360.8 με), ensuring that the early warning is triggered as early as possible at the full scale and avoiding missed reports.

[0063] Level 2 warning (side reaction self-acceleration warning) – This corresponds to the rapid growth stage before the change in the corresponding strain evolution, marking the entry of the internal side reaction into the self-acceleration stage, with the gas production rate increasing exponentially.

[0064] Judgment criteria: The cumulative strain within 30 seconds is ≥500με, and the strain change rate is ≥5με / s. This threshold matches the strain growth rate before the mutation at 0.5C, 0.75C, and 1C rates, which can effectively distinguish between normal lithium intercalation expansion and abnormal gas production expansion, thus avoiding false triggering.

[0065] Level 3 warning (critical chain reaction warning) – corresponds to the instantaneous abrupt change stage before the safety valve opens, marking the entry of the internal chain exothermic reaction into an irreversible stage.

[0066] Judgment criterion: Absolute value of strain change rate within 1 second ≥ 300 με / s This value is 28% (less than 30%) of the minimum measured peak strain change rate, which meets the 3 times safety redundancy requirement. It can be determined without continuous triggering, matches the instantaneous physical characteristics of strain mutation, solves the defect of traditional models in determining continuous high change rates, and ensures no missed reports under all working conditions.

[0067] Example 2 Embodiment 2 of the present invention also provides a lithium battery monitoring and early warning system, which executes the method of Embodiment 1, including: The sensor deployment module is used to attach and install fiber Bragg grating temperature sensors and fiber Bragg grating strain sensors on the battery surface. The preload application module is used to apply a preset preload to the entire battery. The calibration module is used to perform bonding, curing, and static setting, temperature equilibration, reference wavelength acquisition, temperature compensation, preload coupling correction, and elimination of the influence of additional strain terms due to temperature deformation of thin plates on fiber optic strain sensors and fiber optic temperature sensors, so as to obtain the final strain value. The feature extraction module is used to extract thermal runaway features, including strain change rate and strain accumulation. The early warning module is used to provide multi-level early warnings for batteries based on thermal runaway characteristics.

[0068] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A lithium battery monitoring and early warning method, characterized in that, Includes the following steps: S1. Attach and install a fiber optic temperature sensor and a fiber optic strain sensor onto the surface of the battery; S2. Apply a preset preload force to the entire battery; S3. The fiber optic strain sensor and fiber optic temperature sensor are bonded, cured, and allowed to stand at a constant temperature. The reference wavelength is then acquired, temperature is compensated, preloaded, coupled, and the influence of the additional strain term due to temperature deformation of the thin plate is eliminated to obtain the final strain value. S4. Extract thermal runaway features, including strain change rate and strain accumulation. S5. Provides multi-level early warning for batteries based on thermal runaway characteristics.

2. The lithium battery monitoring and early warning method according to claim 1, characterized in that, S1 includes: A fiber optic temperature sensor and a fiber optic strain sensor are installed near the positive electrode side at the geometric center of the battery side. The fiber optic strain sensor is in the form of bare optical fiber, and the fiber optic temperature sensor is in the form of a sleeved package.

3. The lithium battery monitoring and early warning method according to claim 1, characterized in that, The bonding and curing stand is to stand at room temperature for a first preset time, and the temperature equilibration stand is to place the battery and sensor together in a constant temperature environment for a second preset time; the reference wavelength acquisition is to continuously acquire the wavelength data of the fiber optic strain sensor and the fiber optic temperature sensor for 30 seconds after the temperature is completely balanced, and take the average value as the reference wavelength of the fiber optic strain sensor and the fiber optic temperature sensor.

4. The lithium battery monitoring and early warning method according to claim 1, characterized in that, The reference wavelength acquisition process requires data synchronization processing, which includes: Interpolate the temperature data to the same time point as the strain data to achieve point-to-point synchronization: in, For the i-th strain sampling time of the strain data, and For two adjacent temperature sampling times, For the k-th temperature sampling time The corresponding temperature value, For the (k+1)th temperature sampling time The corresponding temperature value, For the i-th strain sampling time The corresponding temperature value, and These are all weighting coefficients, and the dynamic weighting calculation formula is as follows: In the formula, This represents the temperature rise rate between adjacent sampling intervals.

5. A lithium battery monitoring and early warning method according to claim 1, characterized in that, The temperature compensation process is as follows: The reference wavelength of the fiber optic strain sensor is denoted as The reference wavelength of the fiber optic grating temperature sensor is denoted as Bragg wavelength shift of fiber optic strain sensors = - , The wavelength of the fiber optic strain sensor at the current moment is given by the temperature compensation formula. ,in, The actual strain value obtained after temperature compensation. This represents the strain sensitivity coefficient of the fiber Bragg grating strain sensor. The temperature sensitivity coefficient of the fiber Bragg grating strain sensor; The temperature change measured by the fiber Bragg grating temperature sensor. , The wavelength shift of the fiber optic temperature sensor and , This represents the wavelength of the fiber Bragg grating temperature sensor at the current moment.

6. The lithium battery monitoring and early warning method according to claim 5, characterized in that, The process of preload coupling correction is as follows: Constructing the preload coupling correction formula In the formula, To correct the decoupling strain value; Apply preload to the sensor; The preload coupling correction coefficient and ,in, For aluminum shell thickness, For shell bending stiffness, The elastic modulus of the battery casing. This refers to the elastic modulus of the optical fiber.

7. A lithium battery monitoring and early warning method according to claim 6, characterized in that, The preload coupling correction coefficient The calibration method involves collecting data on the aluminum shell thickness, shell bending stiffness, shell bending stiffness, and fiber elastic modulus under four preload levels of 2KN, 2.5KN, 3KN, and 3.5KN. These values ​​are then substituted into the preload coupling correction coefficient formula for calculation, and a linear curve is fitted to obtain the final calibration result. .

8. A lithium battery monitoring and early warning method according to claim 6, characterized in that, The elimination of the effect of additional strain term due to temperature deformation of thin plates includes: Additional strain term due to temperature deformation of thin plate In the formula, For the fourth-order Laplace deflection term of the shell thin plate; The coefficient of thermal expansion of the aluminum shell; The increased temperature causes the aluminum shell sheet to expand, resulting in additional strain. The final strain value .

9. A lithium battery monitoring and early warning method according to claim 1, characterized in that, S5 includes: When the strain change rate is ≥0.25με / s for 10 s continuously and the cumulative strain is ≥350με, a Level 1 warning is issued, corresponding to an ultra-early gas production warning. When the cumulative strain within 30 seconds is ≥500με and the strain change rate is ≥5με / s, a level two warning is issued, corresponding to the self-acceleration warning of the side reaction. When the absolute value of the strain change rate within 1 second is ≥300με / s, a Level 3 warning is issued, corresponding to the critical chain reaction warning.

10. A lithium battery monitoring and early warning system, characterized in that, The method of any one of claims 1-9 comprises: The sensor deployment module is used to attach and install fiber Bragg grating temperature sensors and fiber Bragg grating strain sensors on the battery surface. The preload application module is used to apply a preset preload to the entire battery. The calibration module is used to perform bonding, curing, and static setting, temperature equilibration, reference wavelength acquisition, temperature compensation, preload coupling correction, and elimination of the influence of additional strain terms due to temperature deformation of thin plates on fiber optic strain sensors and fiber optic temperature sensors, so as to obtain the final strain value. The feature extraction module is used to extract thermal runaway features, including strain change rate and strain accumulation. The early warning module is used to provide multi-level early warnings for batteries based on thermal runaway characteristics.

Images