Electrical performance test method and system based on big data

Abstract

The invention discloses an electrical performance testing method and system based on big data, and relates to the field of electrical performance testing, and the main scheme is that the method comprises the steps: obtaining the health state data of a battery, and calculating a dynamic health factor; calculating a working condition stress coefficient through the working condition data, and calculating a life attenuation comprehensive index in combination with the dynamic health factor; calculating a dynamic security risk index through the security risk data; calculating a risk degree value of interface-working condition comprehensive failure through the stability data and the battery health state data; respectively comparing the life attenuation comprehensive index, the dynamic safety risk index and the failure risk degree value of the interface-working condition combined failure with corresponding threshold sets, and judging the risk levels of different indexes of the battery; the technical problems that an existing automobile battery electrical performance testing method is insufficient in comprehensiveness, dynamism, accuracy and safety and the like are solved.

Description

Technical Field

[0001] The present invention relates to the field of electrical performance testing, and in particular to an electrical performance testing method based on big data. Background Art

[0002] With the booming automotive industry, new energy vehicles are becoming a mainstream market trend. As a core component of new energy vehicles, the electrical performance of automotive batteries directly impacts key vehicle indicators such as range, power output, service life, and safety. This performance is crucial for automakers to enhance product competitiveness, meet consumer demand, and promote the continued advancement of the new energy vehicle industry. Accurate and comprehensive electrical performance testing of automotive batteries throughout their R&D, production, use, and maintenance processes provides critical data support for battery design optimization, quality control, fault diagnosis, and performance prediction, ensuring batteries are always in good working condition to adapt to the complex and ever-changing automotive operating environment and the growing demand for high performance.

[0003] However, existing methods for testing the electrical performance of automotive batteries have numerous drawbacks. For one thing, traditional testing methods often focus on single-dimensional performance evaluations, such as basic parameters like battery capacity, internal resistance, or voltage. These methods ignore the complex and complex stress factors that affect batteries under actual operating conditions. As a result, test results fail to truly reflect the battery's dynamic electrical performance and health during actual use.

[0004] On the other hand, existing technologies also have shortcomings in battery safety risk assessment. They typically only test certain specific battery safety indicators and lack a comprehensive, dynamic monitoring and analysis system for battery safety risks. This makes it difficult to promptly detect potential battery safety hazards, and thus cannot effectively prevent safety accidents caused by battery failures. Furthermore, existing testing methods are not accurate enough in predicting battery life degradation and fail to comprehensively consider the impact of multiple factors on battery life, such as battery health, operating stress, and stability.

[0005] To address the shortcomings of these existing technologies, this solution aims to provide an electrical performance testing method based on big data. By comprehensively analyzing multi-dimensional information such as battery health status data, operating condition data, safety risk data, and stability data, it can achieve accurate assessment and prediction of battery life degradation, safety risks, and comprehensive failure risks, thereby effectively solving the technical problems of existing automotive battery electrical performance testing methods in terms of comprehensiveness, dynamism, accuracy, and safety, and meeting the higher requirements of the new energy vehicle industry for battery electrical performance testing. Summary of the Invention

[0006] (1) Technical problems solved

[0007] In response to the shortcomings of the existing technology, the present invention provides an electrical performance testing method based on big data, which at least solves one of the technical problems of the existing automotive battery electrical performance testing method, such as the lack of comprehensiveness, dynamism, accuracy and safety.

[0008] (2) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for electrical performance testing based on big data, comprising:

[0010] Step 1: Obtain battery health status data and analyze it to obtain the dynamic health factor DHF;

[0011] Step 2: Obtain and analyze the operating condition data through the battery management system to obtain the operating condition stress coefficient OSC, and calculate the life decay comprehensive index LDT by combining the operating condition stress coefficient OSC and the dynamic health factor DHF;

[0012] Step 3: Obtain safety risk data through the battery management system and scanning equipment, conduct comprehensive analysis, and calculate the dynamic safety risk index (SRI);

[0013] Step 4: Obtain battery stability data and calculate the risk level of interface-operating condition comprehensive failure based on the stability data and battery health status data;

[0014] Step 5: Set the life decay comprehensive threshold, dynamic safety risk threshold set, and interface-operating condition joint risk threshold set; and compare the failure risk degree values ​​of the life decay comprehensive index, dynamic safety risk index, and interface-operating condition joint failure with the corresponding threshold sets to determine the risk level of different battery indicators;

[0015] Step 6: Generate a battery test report based on the different indicators calculated in steps 1 to 5 and the risk levels of different battery indicators.

[0016] In the preferred embodiment of the electrical performance test method based on big data, the battery health status data includes the current actual capacity Cactual, the rated capacity Crated and the charge transfer impedance Rct i , specifically:

[0017] The battery capacity is measured by using a 1C constant current discharge method. The amount of electricity discharged from the battery from a fully charged state to the cut-off voltage is recorded as the current actual capacity Cactual. The rated capacity Crated is then obtained.

[0018] The charge transfer impedance Rct of the battery at different cycle times is obtained using electrochemical impedance spectroscopy testing equipment. i ;

[0019] Through the current actual capacity C actual , Rated capacity C rated and charge transfer impedance Rct i The dynamic health factor DHF is calculated based on the following formula:

[0020]

[0021] Where, e is the base of natural logarithm; λ is the attenuation coefficient; Rct i is the charge transfer impedance of the i-th cycle, i is the number of cycles, and its value is a positive integer, and N represents the maximum number of cycles; Rct ref is the charge transfer impedance reference value.

[0022] In the preferred embodiment of the electrical performance test method based on big data, the operating condition data includes the effective value of the current I rns , monitor the battery temperature T t and the activation energy Ea of the battery material;

[0023] According to the effective value of current I rns , temperature T t The operating stress coefficient OSC is calculated based on the activation energy Ea, and the formula is:

[0024]

[0025] Among them, T ref represents the temperature reference value; R is the gas constant; t0 is the starting time, and t is the end time.

[0026] In the preferred embodiment of the electrical performance test method based on big data, the life attenuation comprehensive index is calculated based on the dynamic health factor DHF and the operating stress coefficient OSC, and the formula is as follows:

[0027] LDT=k·OSC·[1+k T ·(α LLI -0.5%)]·DHF -0.8 .

[0028] Where k is the decay rate constant; z is the decay coefficient; ts is the battery life; α LLI is the loss rate of lithium ions.

[0029] In the preferred embodiment of the electrical performance test method based on big data, the safety risk data include voltage fluctuation entropy (V), temperature rise rate ΔT / Δt, wettability change rate I wct and terminal voltage V;

[0030] The BMS continuously collects the voltage value of each battery cell through the voltage sensor, records it by timestamp, and forms a voltage time series data sequence, which is recorded as [t1, V1-1], [t2, V1-2], ..., [t m ,Vn-m], where n is the number of monomers and m is the time series length. The voltage fluctuation entropy Entropy (V) in the window is calculated by the Shannon entropy algorithm;

[0031] The BMS obtains the temperature rise rate ΔT / Δt by monitoring the data of the temperature sensor;

[0032] The battery is scanned by a scanning device to obtain the scan data. The collected data is processed and the wettability is calculated by counting the number of foreground pixels. The wettability change rate I is obtained by comparing the wettability values ​​at adjacent time points. wct ;

[0033] The terminal voltage V is measured by the distributed voltage acquisition module.

[0034] In the preferred embodiment of the electrical performance testing method based on big data, the dynamic safety risk index (SRI) is calculated using the safety risk data according to the following formula:

[0035]

[0036] Among them, β1 is the weight coefficient of voltage fluctuation entropy; β2 is The weight coefficient of The weight coefficient of The weight coefficient is ΔT, which is the temperature change, and Δt, which is the time change.

[0037] In the preferred embodiment of the above-mentioned electrical performance test method based on big data, the stability data includes a critical interphase overpotential threshold value CIOP;

[0038] The stability data and battery health status data are input into the interface-operating condition combined failure risk model to calculate the risk level of the interface-operating condition combined failure. The formula is:

[0039]

[0040] In the preferred embodiment of the above-mentioned electrical performance testing method based on big data, the comprehensive life decay threshold includes a life decay warning threshold and a life decay danger threshold, the dynamic safety risk threshold set includes a dynamic safety warning threshold and a dynamic safety danger threshold, and the interface-operating condition joint failure risk threshold set includes an interface-operating condition joint failure warning threshold and an interface-operating condition joint failure danger threshold.

[0041] In the preferred embodiment of the above-mentioned electrical performance testing method based on big data, the life decay comprehensive index is compared with the life decay warning threshold and the life decay danger threshold respectively. When the life decay comprehensive index is less than the life decay warning threshold, it indicates that the battery life status is qualified. When the life decay warning threshold is less than or equal to the life decay comprehensive index and less than the life decay danger threshold, it indicates that the battery life status has a warning level risk. When the life decay danger threshold is less than the life decay comprehensive index, it indicates that the battery life status has a danger level risk.

[0042] The dynamic safety risk index is compared with the dynamic safety warning threshold and the dynamic safety danger threshold respectively. When the dynamic safety risk index is less than the dynamic safety warning threshold, it indicates that the dynamic safety status of the battery is qualified. When the dynamic safety warning threshold ≤ the dynamic safety risk index and less than the dynamic safety danger threshold, it indicates that the dynamic safety status of the battery has a warning-level risk. When the dynamic safety danger threshold is less than the dynamic safety risk index, it indicates that the dynamic safety status of the battery has a danger-level risk.

[0043] The failure risk degree value is compared with the interface-operating condition joint failure warning threshold and the interface-operating condition joint failure danger threshold respectively. When the failure risk degree value is less than the interface-operating condition joint failure warning threshold, it indicates that the comprehensive failure state of the battery is qualified. When the interface-operating condition joint failure warning threshold ≤ the failure risk degree value < the interface-operating condition joint failure danger threshold, it indicates that the comprehensive failure state of the battery has a warning level risk. When the interface-operating condition joint failure danger threshold is less than the failure risk degree value, it indicates that the comprehensive failure state of the battery has a danger level risk.

[0044] (3) Beneficial effects

[0045] The present invention provides an electrical performance testing method based on big data, which has the following beneficial effects:

[0046] (1) By acquiring battery health status data and analyzing it to obtain dynamic health factors, the health status of the battery under different operating conditions can be accurately grasped. Combined with the working condition stress coefficient, the comprehensive life attenuation index is calculated, and the impact of actual working conditions on battery life is fully considered, making the life prediction closer to the actual usage scenario, providing an accurate basis for battery maintenance and replacement, and effectively avoiding sudden failures caused by deviations in life estimation.

[0047] (2) Utilize the battery management system and scanning equipment to obtain safety risk data, calculate the dynamic safety risk index, and achieve comprehensive and dynamic monitoring of battery safety risks. Compared with existing methods that only detect some safety indicators, this method can promptly identify potential safety hazards of batteries under complex working conditions, take preventive measures in advance, reduce the probability of safety accidents, and ensure the safety of batteries during use.

[0048] (3) The risk level of interface-operating condition combined failure is calculated based on stability data and battery health status data, comprehensively considering the impact of the battery's internal interface characteristics and operating conditions on failure, thus overcoming the one-sidedness of existing technologies in failure risk assessment. By setting a threshold set and comparing the risk level with the calculated results, a clear and intuitive quantitative basis is provided for battery risk management.

[0049] (4) Generate a test report based on the different indicators and risk levels calculated in each step, providing comprehensive and detailed data support for battery R&D, production, use and maintenance. R&D personnel can optimize battery design based on the report, manufacturers can strengthen quality control, and users can reasonably arrange battery use and maintenance plans, thereby improving the management level of the battery's entire life cycle, extending battery service life, reducing operating costs, and enhancing the market competitiveness of products. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a schematic diagram of the steps of an electrical performance testing method based on big data of the present invention. DETAILED DESCRIPTION

[0051] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0052] Example 1

[0053] See also Figure 1 The present invention provides an electrical performance testing method based on big data, comprising:

[0054] Step 1: Obtain battery health status data and analyze it to obtain a dynamic health factor.

[0055] It should be noted that in the process of calculating the dynamic health factor, all parameters involved in each calculation step need to be normalized and preprocessed to eliminate the dimensions of different parameters to facilitate subsequent formula calculations.

[0056] Specifically, the battery health status data includes the current actual capacity Cactual, rated capacity Crated and charge transfer impedance Rct i .

[0057] Step 101: Measure the battery capacity using a 1C constant current discharge method. This involves discharging the battery at a current rate of 1C. The amount of electricity discharged from a fully charged state to a cutoff voltage is recorded as the actual capacity (Cactual). The rated capacity (Crated) is then calculated based on the battery capacity calibrated by the vehicle manufacturer.

[0058] Step 102: Use electrochemical impedance spectroscopy testing equipment to test the battery at different cycle times. During the test, by applying an AC signal of unit amplitude and measuring the impedance frequency response of the battery, various impedance information including charge transfer impedance can be obtained. The test data is processed and fitted by electrochemical analysis software to extract the charge transfer impedance Rct of each cycle. i ; Obtain the charge transfer impedance reference value Rct through the battery data provided by the car manufacturer ref You can also place the battery under the same test conditions when it is new or after a certain number of cycles, and use the EIS test method to obtain the charge transfer impedance at this time as Rct ref .

[0059] Step 103: Using the current actual capacity C actual , Rated capacity C rated , charge transfer impedance Rct i , charge transfer impedance as Rct ref The dynamic health factor DHF is calculated based on the following formula:

[0060]

[0061] Where, e is the base of natural logarithm; λ is the attenuation coefficient; Rct i is the charge transfer impedance of the i-th cycle, i is the number of cycles, and its value is a positive integer, and N represents the maximum number of cycles; Rct ref is the charge transfer impedance reference value.

[0062] It should be noted that the attenuation coefficient λ is determined by data fitting based on a large amount of historical battery data, including battery capacity and impedance data at different usage stages, such as different number of cycles and different degrees of aging.

[0063] It should be noted that The ratio of the two directly quantifies the "remaining capacity percentage," reflecting the degree of degradation in the battery's energy output. This capacity reduction is due to performance loss caused by internal chemical or physical degradation (such as electrolyte loss and electrode corrosion). The calculation logic is simple and straightforward: a higher ratio indicates higher capacity retention and better health (the ideal value is 1, indicating no degradation). It represents the average deviation of the cycle, and calculates the average value of the charge transfer impedance deviation of all cycles. The impedance deviation of a single cycle is susceptible to interference. By taking the absolute value of the deviation of N cycles and averaging it, the noise is filtered out, the stable trend of long-term impedance fluctuation is highlighted, and the "stability decay" of the electrochemical process inside the battery is reflected. -λ It converts the average deviation into a decay multiplier based on an exponential decay model: larger impedance deviations result in smaller overall multipliers (stronger decay), indicating that impedance changes exacerbate health deterioration. Logically, the average deviation is fed into an exponential function: when the deviation is 0, the multiplier is 1 (no effect), and as the deviation increases, the multiplier approaches 0 (complete degradation). DHF is a product that combines the capacity ratio (representing static performance) with the impedance deviation decay multiplier (representing dynamic performance). The final DHF value ranges from 0 to 1: 1 indicates optimal health, and 0 indicates complete failure. During battery testing, even if capacity remains temporarily stable, an increase in impedance can provide an early warning through a decrease in the DHF value (the decay effect of e), avoiding the delayed response of relying solely on capacity. In short, it accurately characterizes "performance degradation," including rate (controlled by λ), amplitude, and comprehensive factors, making health assessment no longer static but rather evolving dynamically with the number of cycles.

[0064] Step 2: Obtain and analyze the operating condition data through the battery management system to obtain the operating condition stress coefficient OSC, and calculate the life decay comprehensive index LDT by combining the operating condition stress coefficient OSC and the dynamic health factor DHF.

[0065] It should be noted that in the process of calculating the comprehensive index of life attenuation, all parameters involved in each calculation step need to be normalized and preprocessed separately to eliminate the dimensions of different parameters in order to facilitate subsequent formula calculations.

[0066] Working condition data includes current effective value I rns , monitor the battery temperature T t and the activation energy Ea of the battery material;

[0067] Step 201: The current sensor in the battery management system collects the effective value of the battery current I in real time. rns The battery temperature T is monitored in real time by the temperature sensor installed in the battery pack. t .

[0068] Step 202: Obtain activation energy Ea of the battery material using battery parameters provided by the battery manufacturer;

[0069] Step 203: According to the effective value of the current I rns , temperature T t The operating stress coefficient OSC is calculated based on the activation energy Ea, and the formula is:

[0070]

[0071] Among them, T ref represents the temperature reference value, which can refer to industry standards; R is the gas constant, which can be taken as 8.314J / (mol·K); t0 is the starting time, and t is the end time.

[0072] It should be noted that Represents the time accumulation of operating condition stress. Battery aging is a time-accumulated damage process: the "current thermal stress × temperature acceleration factor" at each instant will continue to damage the battery (for example, the heat + temperature conditions every second are promoting the aging reaction). The role of integration is to superimpose the instantaneous stress in the continuous time dimension to obtain the total stress OSC from the starting time t0 to the end time t, reflecting the comprehensive damage degree of the operating condition to the battery during this time period; is the exponential term of the temperature acceleration factor. The higher the temperature (T t The larger the The smaller the value, the greater the difference, the exponential term is significantly amplified, the higher the Ea, the more sensitive the material is to temperature, the temperature difference is scaled by Ea / R times and input into the exponential function to obtain the temperature stress acceleration factor. When T t =T ref When T t >T ref When , the index term > 1 (temperature accelerated aging).

[0073] Step 204: Calculate the comprehensive life attenuation index based on the dynamic health factor DHF and the operating stress coefficient OSC, using the following formula:

[0074] LDT=k·OSC·[1+k T ·(α LLI -0.5%)]·DHF -0.8 .

[0075] Where k is the decay rate constant, which can be determined according to the type of battery and the corresponding industry standard. For example, for ternary lithium batteries, the value is 1.2×10 -5 ;k T represents the temperature correction factor, which can be determined based on the battery type and the corresponding industry standard; ts is the battery usage time; z is the attenuation coefficient. A large amount of battery life attenuation data is collected and fitted using the SVR algorithm to obtain the optimal attenuation index z; is the dynamic weight factor of lithium plating risk; α LLI is the loss rate of lithium ions.

[0076] It should be noted that the battery is tested by electrochemical impedance spectroscopy (EIS) equipment, the impedance spectrum at different cycle times is recorded, and the spectrum is input into the lithium-ion battery analysis software. By fitting the capacity decay curve, impedance change curve, etc., the lithium ion loss rate α can be calculated. LLI .

[0077] It should be noted that [1+k T ·(α LLI -0.5%)] represents the temperature-lithium ion loss correction term. Temperature changes the risk of lithium plating and the degree of lithium ion loss by affecting the lithium ion migration rate and the activation energy of the side reaction (at low temperatures, lithium ions migrate slowly and are prone to lithium plating on the negative electrode surface, resulting in α LLI The “temperature-lithium deposition correlation” of different batteries is different, so the industry standard defines k T ; 0.5% is the "lithium ion loss critical value" (the industry has conducted a large number of experimental statistics: when α LLI When ≤0.5%, irreversible losses such as lithium deposition can be ignored; if it exceeds, the risk of lithium deposition will increase significantly and the decay will be accelerated). LLI >0.5%, correction term>1, amplifying the lifespan attenuation (reflecting the accelerated effect of lithium deposition); if α LLI <0.5%, correction term <1, weakening the impact (reflecting normal attenuation). First pass k T Introducing the difference in sensitivity of temperature to lithium deposition, and then using α LLI The difference from the critical value quantifies the accelerated effect of "excess lithium ion loss" on life, and finally outputs the comprehensive correction coefficient of temperature-lithium ion loss. -0.8 Represents the dynamic health factor correction term. There is a nonlinear negative correlation between battery health and lifespan attenuation: the worse the health state (the smaller the DHF), the faster the subsequent attenuation rate (due to the accumulation of side reactions and increased structural damage, such as accelerated side reactions after SEI membrane rupture). The exponent -0.8 is the "attenuation correlation optimal index" obtained by fitting a large amount of battery attenuation data (such as the SVR algorithm). It quantifies this nonlinear relationship and can be set according to specific circumstances.

[0078] The scheme is through “intrinsic rate (k) → operating condition stress (OSC) → microscopic mechanism (k T , α LLI)→health status (DHF)→big data fitting (z)”, and constructed a battery life attenuation model “from microscopic mechanism to macroscopic life, from static parameters to dynamic correction”, which solves the problems of traditional models uniformly treating different batteries, ignoring the differences in chemical systems, and only deducing life from macroscopic parameters, ignoring microscopic mechanisms such as lithium plating and lithium ion loss. The microscopic electrochemical mechanism of “lithium plating-lithium ion loss” is quantified into the life model, so that the attenuation calculation fits the chemical nature of battery aging; through the nonlinear correction of DHF-0.8 and the operating condition mapping of OSC, the dynamic life assessment of “full life cycle + full scenario operating condition” is realized; relying on SVR fitting and industry standard parameters, it ensures “type adaptation” (different batteries use different k, k T ), and achieved "scenario universality" (adaptation to complex working conditions) through big data optimization, which ultimately greatly improved the life prediction accuracy compared with traditional models (especially in low temperature and fast charging scenarios).

[0079] Step 3: Obtain safety risk data through the battery management system and scanning equipment, conduct comprehensive analysis, and calculate the dynamic safety risk index (SRI).

[0080] It should be noted that in the process of calculating the dynamic security risk index, all parameters involved in each calculation step need to be normalized and preprocessed to eliminate the dimensions of different parameters to facilitate subsequent formula calculations.

[0081] Safety risk data includes voltage fluctuation entropy (V), temperature rise rate ΔT / Δt, and wettability change rate I wct and terminal voltage V;

[0082] Step 301: The BMS continuously collects the voltage value of each cell (single battery) through the voltage sensor and records it by timestamp to form a voltage time series data sequence, such as: collecting once every 100ms and recording it as [t1, V1-1], [t2, V1-2], ..., [t m ,Vn-m], where n is the number of cells and m is the time series length. The software algorithm presets the sliding window length and intercepts the voltage data matrix segment by segment. For example, a window length of m corresponds to the voltage at m time points, and a window width of n corresponds to n cells. Each interception forms a local voltage matrix. The intercepted window data is then processed using the Shannon entropy algorithm, such as calculating the voltage fluctuation entropy (V) within the window.

[0083] Step 302: The BMS is equipped with a temperature sensor, typically installed at the battery module terminal or inside the battery pack. The BMS monitors the temperature sensor data and obtains the temperature rise rate ΔT / Δt.

[0084] Traditional temperature monitoring only focuses on the absolute value of temperature (such as an alarm when it is >60°C), but the core of thermal runaway is the "sudden change in temperature rise rate". In the early stage of thermal runaway, the hotspot temperature may be only 50°C (below the threshold), but the temperature rise rate has reached 0.5°C / s (much higher than the normal 0.05°C / s). Traditional methods cannot provide early warning; by capturing the "sudden change in temperature rise rate" (such as ΔT / Δt from 0.05 to 0.5°C / s), the warning time is greatly advanced.

[0085] Step 303: Select a terahertz imaging scanning system, an ultrasonic scanning imaging device, or an optical coherence tomography device, and fix the battery sample on the corresponding test platform to scan the battery. The terahertz imaging scanning system can obtain image information inside the battery by emitting and receiving terahertz waves; the ultrasonic scanning imaging device uses an ultrasonic probe to emit and receive ultrasonic waves to form ultrasonic imaging; the optical coherence tomography device uses the principle of low-coherence light interference to perform optical sectioning imaging of the battery, and uses the corresponding image processing and analysis algorithms to process the collected data, and calculate the infiltration by counting the number of foreground pixels. By comparing the infiltration values ​​at adjacent time points, the infiltration change rate I is obtained. wct .

[0086] Step 304: Measure the terminal voltage V through a distributed voltage acquisition module, such as a battery management dedicated IC.

[0087] Step 305: Calculate the dynamic safety risk index SRI using the voltage fluctuation entropy (V), the temperature rise rate ΔT / Δt, the terminal voltage V, and the wettability change rate Iwet, using the following formula:

[0088]

[0089] Among them, β1 is the weight coefficient of voltage fluctuation entropy; β2 is The weight coefficient of The weight coefficient of The weight coefficients are: β1+β2+β3+β4=1, each taking a value of 0.25, and can be adjusted according to specific needs; ΔT is the temperature change, and Δt is the time change.

[0090] It should be noted that β1·Entropy(V) quantifies the contribution of “voltage disorder between cells” to safety risk—the higher the entropy value, the more serious the local anomaly, and the higher the risk weight; You can focus on the "temperature rise acceleration of the local hottest area" - the larger the max(ΔT / Δt), the more urgent the thermal runaway risk, and the higher the weight; It is the second-order derivative of the terminal voltage, reflecting the "acceleration" of voltage change (for example, when an internal short circuit occurs, the terminal voltage first drops rapidly, and the absolute value of the second-order derivative increases sharply); It is the first derivative of the wettability change rate, reflecting the "speed" of wettability deterioration (e.g., when the electrolyte leaks, will increase rapidly in a negative direction).

[0091] Step 4: Obtain the stability data of the battery, and calculate the risk value Risk of the interface-operating condition comprehensive failure based on the stability data and battery health status data.

[0092] It should be noted that in the process of calculating the risk level of the interface-operating condition comprehensive failure, all parameters involved in each calculation step need to be normalized and preprocessed to eliminate the dimensions of different parameters for the convenience of subsequent formula calculations.

[0093] Stability data includes the critical interphase overpotential threshold CIOP.

[0094] Step 401: The overpotential test of lithium-ion batteries usually refers to the common electrochemical standards to obtain the critical interphase overpotential threshold CIOP, such as ISO 15118 (electric vehicle and grid interface) or GB / T36276-2023 (lithium-ion batteries for power storage); the value can be 0.1-1.5mA / cm 2 .

[0095] Step 402: Input the stability data and battery health status data into the interface-operating condition combined failure risk model to calculate the risk level of the interface-operating condition combined failure, Risk, based on the formula:

[0096]

[0097] It should be noted that OSC (operating stress coefficient) quantifies the degree of damage to the battery caused by the operating conditions (e.g. OSC during fast charging > OSC during slow charging), and its time derivative Reflects the speed of change of working condition stress (such as the fast charging power suddenly jumps from 1C to 2C, Sudden increase in working conditions). Sudden changes in working conditions will instantly increase overpotential (e.g., sudden change in large current will cause a sudden increase in polarization, overpotential will instantly break through CIOP, and trigger lithium deposition). This sub-item specifically captures this "dynamic coercion risk". By achieving "millisecond-level response to sudden changes in working conditions": the risk identification time for scenarios such as fast charging power jumps and load mutations is greatly shortened; temperature correction index item When the high temperature (T t >T ref ): Index term > 1, amplifying risk (high temperature accelerates side reactions, and overpotential is more likely to exceed CIOP); low temperature (T t <T ref ): The index item is <1, but the CIOP itself has decreased (the standard low-temperature CIOP threshold is lower, such as ternary lithium low-temperature CIOP = 0.8mA / cm 2 ), the two complement each other and avoid "underestimation of risks at low temperatures".

[0098] The traditional model separates the coupling relationship between "temperature" and "operating stress", and only considers the impact of temperature on overpotential (such as power reduction at low temperature), without quantifying how temperature changes the operating stress → risk path of overpotential breakthrough; this solution realizes the coupled quantification of "temperature-operating condition-overpotential", reducing the risk prediction error in high temperature (45℃) and low temperature (-10℃) scenarios; at high temperatures, the exponential term amplifies the risk and provides early warning of the synergistic damage of "sudden operating condition change + high temperature" (such as fast charging + 45℃, the risk value instantly doubles); at low temperatures, the CIOP threshold is reduced and complemented by the exponential term to avoid "underestimation of low temperature risk" (such as fast charging in winter, the risk value can still accurately reflect the hidden dangers of lithium plating).

[0099] Through static criticality (CIOP), a safety margin of overpotential is provided to quantify the "maximum overpotential allowed"; dynamic stress Capture the mutation rate of working condition stress and quantify the "dynamic impact of coercion"; temperature correction (exponential term) describes the acceleration / deceleration of temperature on the "coercion → breakthrough" process, quantifies the "amplification effect of the environment", and realizes three-dimensional risk coupling, covering the three major risk dimensions of static criticality, dynamic coercion, and environmental coupling, and identifies the synergistic risk of "working condition mutation + extreme temperature" that has been traditionally missed.

[0100] Step 5: Set the life attenuation comprehensive threshold, dynamic safety risk threshold set, and interface-operating condition joint risk threshold set; and compare the life attenuation comprehensive index, dynamic safety risk index, and interface-operating condition joint failure failure risk degree values ​​with the corresponding threshold sets to determine the risk level of different battery indicators.

[0101] Step 501: Set a comprehensive life decay threshold, a dynamic safety risk threshold set, and an interface-operating condition joint failure risk threshold set; wherein, the comprehensive life decay threshold includes a life decay warning threshold and a life decay danger threshold, the dynamic safety risk threshold set includes a dynamic safety warning threshold and a dynamic safety danger threshold, and the interface-operating condition joint failure risk threshold set includes an interface-operating condition joint failure warning threshold and an interface-operating condition joint failure danger threshold.

[0102] It should be noted that the life attenuation comprehensive threshold set can be set with reference to the failure marking reverse method or the statistical distribution method. For example, the failure marking reverse method can be set by clarifying the failure criteria of the industry / scenario, such as: the power battery capacity drops to 80% of the initial capacity as the retirement standard; the energy storage battery capacity drops to 70% to trigger maintenance, etc.; through accelerated life tests, simulating different OSC (fast charging, high load), DHF (different aging stages), α LLI (lithium loss rate) working conditions, record the LDT value when the battery reaches the failure criterion as the "failure critical threshold"; if LDT ≥ 80% of the failure critical value is set as the life decay warning threshold, and LDT ≥ the failure critical value is set as the life decay danger threshold; for example, when the ternary lithium battery is retired, the capacity is 80% and after accelerated testing, the corresponding LDT is 1.5×10 -4 , then the lifespan attenuation warning threshold can be set as 1.2×10 -4 (80% critical value), life decay risk threshold 1.5×10 -4 .

[0103] Furthermore, the dynamic safety risk threshold set can be set by referring to the failure case inference method, statistical distribution method or machine learning optimization method. For example, the failure case inference method can collect historical battery safety accident data (such as thermal runaway, short circuit, bulge), extract the Entropy (V), ΔT / Δt, Calculate the corresponding SRI and define the "life decay danger threshold", such as taking the average SRI at the moment of the accident (such as SRI = 100 during thermal runaway), and define the "life decay warning threshold", such as taking the average SRI in the critical warning stage before the accident (such as 1 hour before thermal runaway) (such as SRI = 80, reserving time for disposal). For example: when a ternary lithium battery has thermal runaway, the SRI = 95, then the warning threshold is set to 85 (triggering the warning 10% in advance), and the danger threshold is set to 95.

[0104] Furthermore, the interface-operating condition joint risk threshold set can be set by referring to the failure case inference method, statistical distribution method or machine learning optimization method. For example, the failure case inference method can collect parameters of historical interface failure events (such as dendrite short circuit and interface delamination): CIOP (overpotential at that time), (operating condition change rate), temperature T t Substitute the corresponding Risk value into the formula to calculate the "dynamic safety danger threshold" (failure threshold). Take 80% to 90% of the danger threshold as the "dynamic safety warning threshold" (to allow for action and identify risks in advance). For example: If a dendrite short-circuit occurs in a ternary lithium battery, the Risk = 100. Therefore, the warning threshold is set to 85 and the danger threshold is set to 100.

[0105] Step 502: Compare the life decay comprehensive index with the life decay warning threshold and the life decay danger threshold respectively. When the life decay comprehensive index is less than the life decay warning threshold, it indicates that the battery life status is qualified. When the life decay warning threshold ≤ the life decay comprehensive index < the life decay danger threshold, it indicates that there is a warning level risk in the battery life status. When the life decay danger threshold is less than the life decay comprehensive index, it indicates that there is a danger level risk in the battery life status.

[0106] Step 503: Compare the dynamic safety risk index with the dynamic safety warning threshold and the dynamic safety danger threshold respectively. When the dynamic safety risk index is less than the dynamic safety warning threshold, it indicates that the dynamic safety status of the battery is qualified. When the dynamic safety warning threshold ≤ the dynamic safety risk index < the dynamic safety danger threshold, it indicates that there is a warning-level risk in the dynamic safety status of the battery. When the dynamic safety danger threshold is less than the dynamic safety risk index, it indicates that there is a danger-level risk in the dynamic safety status of the battery.

[0107] Step 504: Compare the failure risk degree value with the interface-operating condition combined failure warning threshold and the interface-operating condition combined failure danger threshold respectively. When the failure risk degree value is less than the interface-operating condition combined failure warning threshold, it indicates that the comprehensive failure state of the battery is qualified. When the interface-operating condition combined failure warning threshold ≤ the failure risk degree value < the interface-operating condition combined failure danger threshold, it indicates that the comprehensive failure state of the battery has a warning level risk. When the interface-operating condition combined failure danger threshold is less than the failure risk degree value, it indicates that the comprehensive failure state of the battery has a danger level risk.

[0108] Step 6: Generate a battery test report based on the different indicators calculated in steps 1 to 5 and the risk levels of the different indicators of the battery.

[0109] Example 2

[0110] An electrical performance testing system based on big data is used to implement any of the above electrical performance testing methods.

[0111] When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithmic steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented using electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0112] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0113] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. An electrical performance testing method based on big data, characterized in that: include: Step 1: Obtain battery health status data and analyze it to obtain the dynamic health factor DHF; Step 2: Obtain and analyze the operating condition data through the battery management system to obtain the operating condition stress coefficient OSC, and calculate the life decay comprehensive index LDT by combining the operating condition stress coefficient OSC and the dynamic health factor DHF; Step 3: Obtain safety risk data through the battery management system and scanning equipment, conduct comprehensive analysis, and calculate the dynamic safety risk index (SRI); Step 4: Obtain battery stability data and calculate the risk level of interface-operating condition comprehensive failure based on the stability data and battery health status data; Step 5: Set the life decay comprehensive threshold, dynamic safety risk threshold set, and interface-operating condition joint risk threshold set; and compare the failure risk degree values ​​of the life decay comprehensive index, dynamic safety risk index, and interface-operating condition joint failure with the corresponding threshold sets to determine the risk level of different battery indicators; Step 6: Generate a battery test report based on the different indicators calculated in steps 1 to 5 and the risk levels of different battery indicators.

2. The electrical performance testing method based on big data according to claim 1, characterized in that: Battery health status data includes the current actual capacity Cactual, rated capacity Crated and charge transfer impedance Rct i , specifically: The battery capacity is measured by using a 1C constant current discharge method. The amount of electricity discharged from the battery from a fully charged state to the cut-off voltage is recorded as the current actual capacity Cactual. The rated capacity Crated is then obtained. The charge transfer impedance Rct of the battery at different cycle times is obtained using electrochemical impedance spectroscopy testing equipment. i ; Through the current actual capacity C actual , Rated capacity C rated and charge transfer impedance Rct i The dynamic health factor DHF is calculated based on the following formula: Where, e is the base of natural logarithm; λ is the attenuation coefficient; Rct i is the charge transfer impedance of the i-th cycle, i is the number of cycles, and its value is a positive integer, and N represents the maximum number of cycles; Rct ref is the charge transfer impedance reference value.

3. The electrical performance testing method based on big data according to claim 2, characterized in that: Working condition data includes current effective value I rns , monitor the battery temperature T t and the activation energy Ea of the battery material; According to the effective value of current I rns , temperature T t The operating stress coefficient OSC is calculated based on the activation energy Ea, and the formula is: Among them, T ref represents the temperature reference value; R is the gas constant; t0 is the starting time, and t is the end time.

4. The electrical performance testing method based on big data according to claim 3, characterized in that: The comprehensive life attenuation index is calculated based on the dynamic health factor DHF and the operating stress coefficient OSC. The formula is: LDT=k·OSC·[1+k T ·(α LLI -0.5%)]·DHF -0.8 。 Where k is the decay rate constant; z is the decay coefficient; ts is the battery life; α LLI is the loss rate of lithium ions.

5. The electrical performance testing method based on big data according to claim 4, characterized in that: Safety risk data includes voltage fluctuation entropy (V), temperature rise rate ΔT / Δt, and wettability change rate I wct and terminal voltage V; The BMS continuously collects the voltage value of each battery cell through the voltage sensor, records it by timestamp, and forms a voltage time series data sequence, which is recorded as [t1, V1-1], [t2, V1-2], ..., [t m ,Vn-m], where n is the number of monomers and m is the time series length. The voltage fluctuation entropy Entropy (V) in the window is calculated by the Shannon entropy algorithm; The BMS obtains the temperature rise rate ΔT / Δt by monitoring the data of the temperature sensor; The battery is scanned by a scanning device to obtain the scan data. The collected data is processed and the wettability is calculated by counting the number of foreground pixels. The wettability change rate I is obtained by comparing the wettability values ​​at adjacent time points. wct ; The terminal voltage V is measured by the distributed voltage acquisition module.

6. The electrical performance testing method based on big data according to claim 5, characterized in that: The dynamic safety risk index (SRI) is calculated using safety risk data according to the following formula: Among them, β1 is the weight coefficient of voltage fluctuation entropy; β2 is The weight coefficient of The weight coefficient of The weight coefficient is ΔT, which is the temperature change, and Δt, which is the time change.

7. The electrical performance testing method based on big data according to claim 6, characterized in that: Stability data include the critical interphase overpotential threshold CIOP; The stability data and battery health status data are input into the interface-operating condition combined failure risk model to calculate the risk level of the interface-operating condition combined failure. The formula is:

8. The electrical performance testing method based on big data according to claim 1, characterized in that: The comprehensive life decay threshold includes the life decay warning threshold and the life decay danger threshold, the dynamic safety risk threshold set includes the dynamic safety warning threshold and the dynamic safety danger threshold, and the interface-operating condition joint failure risk threshold set includes the interface-operating condition joint failure warning threshold and the interface-operating condition joint failure danger threshold.

9. The electrical performance testing method based on big data according to claim 8, characterized in that: The life decay comprehensive index is compared with the life decay warning threshold and the life decay danger threshold respectively. When the life decay comprehensive index is less than the life decay warning threshold, it indicates that the battery life status is qualified. When the life decay warning threshold ≤ the life decay comprehensive index < the life decay danger threshold, it indicates that the battery life status has a warning level risk. When the life decay danger threshold is less than the life decay comprehensive index, it indicates that the battery life status has a danger level risk. The dynamic safety risk index is compared with the dynamic safety warning threshold and the dynamic safety danger threshold respectively. When the dynamic safety risk index is less than the dynamic safety warning threshold, it indicates that the dynamic safety status of the battery is qualified. When the dynamic safety warning threshold ≤ the dynamic safety risk index and less than the dynamic safety danger threshold, it indicates that the dynamic safety status of the battery has a warning-level risk. When the dynamic safety danger threshold is less than the dynamic safety risk index, it indicates that the dynamic safety status of the battery has a danger-level risk. The failure risk degree value is compared with the interface-operating condition joint failure warning threshold and the interface-operating condition joint failure danger threshold respectively. When the failure risk degree value is less than the interface-operating condition joint failure warning threshold, it indicates that the comprehensive failure state of the battery is qualified. When the interface-operating condition joint failure warning threshold ≤ the failure risk degree value < the interface-operating condition joint failure danger threshold, it indicates that the comprehensive failure state of the battery has a warning level risk. When the interface-operating condition joint failure danger threshold is less than the failure risk degree value, it indicates that the comprehensive failure state of the battery has a danger level risk.

10. An electrical performance testing system based on big data, characterized in that: Used to implement the electrical performance testing method of any one of claims 1-9 above.

Images