Method and system for cascade detection of thermal runaway of lithium battery based on multiple characteristic parameters

Abstract

The invention relates to the technical field of lithium battery safety early warning, in particular to a method and a system for cascade detection of lithium battery thermal runaway based on multiple characteristic parameters. The method is adaptive to the unmanned use scene of a single battery, the real-time voltage value, the stress change rate, the hydrogen concentration, the ultrasonic transmission amplitude, the temperature and the temperature change rate of the lithium battery are detected in sequence and compared with corresponding threshold values one by one, and if any parameter exceeds the limit, the system automatically triggers autonomous safety treatment measures such as shutdown and isolation; and if all the parameters do not exceed the limit, judging that the single lithium battery is in a normal state and continuously supplies power. Therefore, the technical pain points of misjudgment and missed judgment of thermal runaway early warning in the scene are effectively solved, accurate identification and autonomous prevention and control of thermal runaway risks are achieved, battery resource waste caused by excessive evaluation is avoided, it is guaranteed that a single lithium battery operates safely, stably and economically in an unattended environment, and the safety and reliability of thermal runaway early warning are improved. And the system is suitable for the actual application requirements of remote environment monitoring, large-scale low-power-consumption terminals and the like.

Description

A method and system for cascade detection of lithium battery thermal runaway based on multiple characteristic parameters Technical Field

[0001] This invention relates to the field of lithium battery safety early warning technology, specifically to a method and system for detecting thermal runaway of lithium batteries based on multiple characteristic parameters in a tiered manner. Background Technology

[0002] Lithium batteries, with their excellent energy density, ultra-long cycle life and environmental friendliness, have become the core power source for new energy vehicles, large-scale energy storage power stations, portable electronic devices and other fields.

[0003] Thermal runaway is the most serious safety hazard of lithium batteries. Once it occurs, it can quickly lead to serious accidents such as battery fires and explosions. Therefore, building a precise and efficient thermal runaway early warning system has become a core technical challenge that the industry urgently needs to solve.

[0004] Currently, research on lithium battery thermal runaway early warning technology mainly focuses on risk identification through single-parameter monitoring or coupled analysis of multiple characteristic parameters. Single-parameter monitoring methods often rely on easily obtainable physical quantities such as temperature and voltage, determining whether the battery is abnormal by setting fixed thresholds. Multi-parameter coupled analysis methods further introduce parameters such as gas concentration and stress-strain, comprehensively assessing thermal runaway risk by establishing weighted models or algorithmic models. Some technologies have attempted to combine sensor arrays to achieve multi-dimensional data acquisition, trying to improve the comprehensiveness of the early warning. These technologies have made some progress in laboratory environments or specific operating conditions, providing a basic framework for thermal runaway early warning.

[0005] However, thermal runaway early warning methods designed based on the above ideas still have certain limitations: single-parameter monitoring methods, which only focus on the characteristic signals of a certain stage of thermal runaway, are easily affected by environmental interference, resulting in problems such as early warning lag or misjudgment; although multi-parameter coupling analysis methods consider a variety of influencing factors, they do not pay enough attention to the temporal correlation and stepwise evolution law between parameters, and often use a unified weight model for overall evaluation, which makes it difficult to accurately match the risk characteristics of different stages of thermal runaway, resulting in problems such as over-evaluation of lithium battery risks, failure judgment in key stages, or insufficient judgment accuracy. They are difficult to meet the comprehensive requirements of accuracy, timeliness, and economy in practical applications, and there is an urgent need for a precise early warning technology solution that can fit the evolution law of thermal runaway. Summary of the Invention

[0006] To address the technical problems of inaccurate risk assessment, misjudgment, or missed detection in lithium battery thermal runaway early warning processes, this invention provides a method for detecting lithium battery thermal runaway based on a tiered approach using multiple characteristic parameters. By sequentially detecting multiple characteristic parameters such as the real-time voltage value and stress change rate of the lithium battery and comparing them with corresponding thresholds, the method achieves accurate identification and timely prevention of lithium battery thermal runaway risks, ensuring safe battery use. Based on this detection method, this invention also provides a system for tiered detection of lithium battery thermal runaway using multiple characteristic parameters.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters in a tiered manner, comprising the following detection steps: sequentially determining whether the real-time voltage value, real-time stress change rate, real-time hydrogen concentration, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate of the lithium battery exceed their respective thresholds; if the current characteristic parameter exceeds its corresponding threshold, the battery is stopped from use; if all characteristic parameters do not exceed their corresponding thresholds, the lithium battery is determined to be normal and can continue to be used.

[0008] As a further improvement to the above solution, the real-time voltage value of the lithium battery is calculated as follows: In the formula, V real V represents the real-time voltage value; k is the sensitivity coefficient of the voltage sensor used to detect the voltage; V out is the output voltage value of the voltage sensor; b is the zero-point offset.

[0009] As a further improvement to the above scheme, the real-time stress change rate is calculated as follows: In the formula, dσ / dt represents the real-time stress change rate; E is the elastic modulus of the lithium battery material; Δt is the time change; ε1 and ε2 are the strain values ​​before and after Δt, respectively.

[0010] As a further improvement to the above scheme, the real-time stress calculation is as follows: In the formula, σ represents the real-time stress; R0 represents the initial resistance of the strain gauge in the strain sensor attached to the surface of the lithium battery; ΔR represents the change in resistance of the strain gauge after deformation; and K represents the sensitivity coefficient of the strain gauge.

[0011] As a further improvement to the above scheme, the real-time ultrasonic transmission amplitude is calculated as follows: In the formula, B represents the real-time ultrasonic transmission amplitude; U p The peak output voltage of the ultrasonic sensor when it receives the echo is represented by M; M represents the sensitivity of the ultrasonic sensor, which is the output voltage per unit sound pressure.

[0012] As a further improvement to the above scheme: the threshold corresponding to the real-time voltage value is V. cr The threshold corresponding to the real-time stress change rate is σ.cr The threshold corresponding to the real-time hydrogen concentration is C. cr The threshold corresponding to the real-time ultrasonic transmission amplitude is B. cr .

[0013] As a further improvement to the above scheme: when the real-time temperature T of the battery surface is within the set temperature range τ or the real-time temperature change rate dT / dt is greater than or equal to the set temperature change rate threshold ΔT', it is determined that the real-time temperature and real-time temperature change rate of the lithium battery exceed the corresponding threshold, and the lithium battery is in the thermal runaway stage.

[0014] As a further improvement to the above scheme: before performing characteristic parameter detection, it is necessary to detect the overcharge or overheat state of the lithium battery; when the lithium battery is in an overheated state, it is determined in sequence whether the real-time stress change rate, real-time hydrogen concentration, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate of the lithium battery exceed the corresponding threshold; when the battery is in an overcharged state, it is determined in sequence whether the real-time voltage value, real-time stress change rate, real-time hydrogen concentration, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate of the lithium battery exceed the corresponding threshold.

[0015] As a further improvement to the above scheme: overcharging refers to the act of applying a charging voltage to a lithium battery that exceeds its design cutoff voltage; overheating refers to the act of the lithium battery body temperature exceeding its safe operating temperature range due to external heat sources or abnormal internal heating.

[0016] This invention also provides a system for cascade detection of lithium battery thermal runaway based on multiple characteristic parameters. This system is used to execute a method for cascade detection of lithium battery thermal runaway based on multiple characteristic parameters, comprising: a data acquisition module for real-time acquisition of multiple characteristic parameters of the lithium battery; a data transmission module electrically connected to the data acquisition module for transmitting the multiple characteristic parameters; and a data processing module communicatively connected to the data transmission module for receiving and processing the multiple characteristic parameters. The data acquisition module includes: a voltage sensor for acquiring the real-time voltage value of the lithium battery; a strain sensor for acquiring strain signals on the surface of the lithium battery and calculating the stress change rate based on the strain signals; and a hydrogen sensor for acquiring the thermal runaway of the lithium battery. The system includes: a real-time hydrogen concentration in the surrounding environment; an ultrasonic sensor assembly, comprising an ultrasonic transmitter, receiver, and signal generator, for acquiring the amplitude of ultrasonic waves penetrating the lithium battery; a temperature sensor for acquiring the real-time temperature of the lithium battery surface and calculating the rate of temperature change; and a data processing module configured to perform tiered early warning judgments, the judgment logic of which is as follows: sequentially judging whether the real-time voltage value, real-time stress change rate, real-time hydrogen concentration value, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate exceed their respective thresholds; if any judgment result is yes, then a corresponding level of early warning signal is generated and a battery shutdown command is triggered; if all judgment results are no, then the battery status is determined to be normal.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention integrates a variety of characteristic parameters that are highly correlated with the thermal runaway evolution process of lithium batteries, such as voltage, stress change rate, hydrogen concentration, ultrasonic transmission amplitude, temperature and temperature change rate, to construct a comprehensive and realistic multi-dimensional monitoring system. This effectively makes up for the shortcomings of traditional single characteristic parameter monitoring, which is susceptible to environmental interference and difficult to capture early hidden signals, and greatly improves the comprehensiveness and accuracy of thermal runaway risk identification. The sequential, tiered detection logic employed in this invention can progressively screen for risks based on the characteristics of different stages of thermal runaway corresponding to the feature parameters. This allows for timely detection of the initial stages of thermal runaway through early feature parameters, avoiding delayed warnings. Furthermore, the multi-feature parameter layer-by-layer verification reduces the risk of misjudgment caused by a single parameter anomaly, ensuring the reliability of the warning results. Simultaneously, the invention achieves rapid response and timely prevention of risks through the judgment rule of "stopping battery use if any feature parameter exceeds the limit," minimizing the risk of safety accidents caused by thermal runaway. The setting of "continuing use if all feature parameters are normal" avoids the impact of over-evaluation on the normal use of the battery, balancing safety and economic efficiency, and is applicable to the actual needs of various lithium battery application scenarios. Attached Figure Description

[0018] Figure 1 is a schematic diagram of the detection device of the present invention.

[0019] Figure 2 is a flowchart of the detection method in this invention.

[0020] In the diagram: A, Lithium battery under test; 10, Battery clamp; 13, Oscilloscope; 141, Thermistor; 15, Data analysis terminal; 16, Power supply; 17, Heating base plate; 20, Base plate; 30, Top plate; 40, Integrator; 50, Strain sensor; 60, Hydrogen detector; 70, Ultrasonic transmitter; 80, Ultrasonic receiver; 90, Ultrasonic signal generator. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and 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.

[0022] I. Detection Device

[0023] As shown in Figure 1, the detection device of this invention, through the coordinated operation of basic support components, a data acquisition module, signal processing and transmission elements, and a data processing and power unit, completes the real-time acquisition, conversion, transmission, and cascade judgment of multiple characteristic parameters of lithium batteries, ultimately achieving accurate early warning of thermal runaway risk. The device has a compact overall structure and precise component positioning, making it suitable for use as a laboratory testing device.

[0024] (a) Basic support components

[0025] 1. Battery clip

[0026] The lithium battery A to be tested is placed in the slot of the battery clamp 10, and the top plate 30 is mounted above the battery clamp 10, with the top plate 30 and the lithium battery A to be tested maintaining a set vertical distance.

[0027] The battery clamp 10 not only fixes the lithium battery A under test, but also provides support for the top plate 30 and the hydrogen detector 60 suspended below the top plate 30, ensuring the accurate relative position of the hydrogen detector 60 and the lithium battery A under test, and avoiding data distortion caused by component displacement during the test.

[0028] The lithium battery under test, A, can be a high-capacity picocell, such as a 392Ah, 472Ah, 587Ah, 625Ah, or 688Ah picocell.

[0029] 2. Base plate

[0030] The base plate 20 is placed horizontally on the experimental table or inside the equipment compartment, ensuring a flat surface. The battery clamp 10 and the heating base plate 17 are then fixed onto the base plate 20 in sequence. The base plate 20 provides a stable mounting foundation, supporting the lithium battery A under test, the heating base plate 17, and some sensors, ensuring the overall stability of the device.

[0031] 30. Top Slab

[0032] The top plate 30 is fixed to the top of the battery clamp 10, and a hydrogen detector 60 is suspended below it by a suspension line. The height of the hydrogen detector 60 probe is controlled by adjusting the length of the suspension line so that the probe is directly above the gas release valve of the lithium battery A under test. The top plate 30 can fix the hydrogen detector 60, ensuring the relative position of its probe and the battery gas release valve, and guaranteeing the timeliness and accuracy of hydrogen concentration collection.

[0033] (ii) Data Acquisition Module

[0034] The data acquisition module is the core detection unit of the device and must be installed strictly according to the location requirements to ensure the accuracy of parameter acquisition.

[0035] 1. Voltage sensor

[0036] The positive and negative terminals of the voltage sensor are connected to the positive and negative terminals of the lithium battery A under test, respectively. The terminals are wrapped with insulating tape to prevent short circuits. The voltage sensor collects its own output voltage value V in real time. out (Unit: V) This data will be transmitted to the data analysis terminal 15 to calculate the real-time voltage value of the lithium battery A under test.

[0037] 2. Strain sensor

[0038] The sensitive surface of the strain sensor 50 is bonded to the middle region of the surface of the lithium battery A under test using thermally conductive adhesive, with no gap between the sensitive surface and the surface of the lithium battery A. The strain sensor 50 acquires the strain signal from the surface of the lithium battery A under test in real time. This strain signal is converted into a strain value by a strain display instrument and then transmitted to the data analysis terminal 15 for calculating the real-time stress change rate.

[0039] 3. Hydrogen detector

[0040] The hydrogen detector 60 is suspended below the top plate 30 by a suspension line. The probe direction can be adjusted to face the gas release valve of the lithium battery. The hydrogen detector 60 collects the hydrogen concentration value of the environment in which the lithium battery under test A is located in real time. When the lithium battery under test A abnormally releases hydrogen, the hydrogen detector 60 can quickly capture the signal and transmit it to the hydrogen detector host, and then to the data analysis terminal 15.

[0041] 4. Ultrasonic sensor assembly

[0042] Ultrasonic transmitter 70: Installed on one side (near the edge) of the large surface of the lithium battery A under test, and fixed by a bracket so that the transmitting surface is perpendicular to the surface of the hydrogen detector 60.

[0043] Ultrasonic receiver 80: Installed on the other side of the large surface of the lithium battery A under test (symmetrical to ultrasonic transmitter 70), ensuring that the receiving surface and the transmitting surface are completely aligned.

[0044] Ultrasonic signal generator 90: connected to ultrasonic transmitter 70 via cable.

[0045] The ultrasonic signal generator 90 generates a high-frequency ultrasonic signal, which penetrates the lithium battery A under test through the ultrasonic transmitter 70 and is received by the ultrasonic receiver 80. After the signal is processed by the oscilloscope 13, the ultrasonic transmission amplitude data is extracted and transmitted to the data analysis terminal 15.

[0046] 5. Thermistor

[0047] At least two thermistors 141 are attached to the electrode area of ​​the lithium battery A under test (one for the positive electrode and one for the negative electrode) using thermally conductive adhesive. The thermistors 141 are then connected to the temperature sensor after being organized with insulating sleeves.

[0048] Thermistor 141 collects the temperature value of the surface of the lithium battery A under test in real time. After the data is transmitted to the temperature sensor, the temperature change rate is calculated and then synchronized to the data analysis terminal 15.

[0049] 6. Heating base plate

[0050] The heating base plate 17 is placed between the base plate 20 and the lithium battery A under test, with its surface in contact with the bottom of the lithium battery A under test, and is connected to the power supply 16 via a cable.

[0051] The heating base plate 17 receives power input from the power supply 16 to generate heat, which is used to simulate the overheating phenomenon that lithium batteries may encounter in real-world application scenarios. For example, passive overheating caused by environmental factors such as high temperature weather, outdoor exposure, and poor equipment ventilation, or active heating caused by internal abnormalities such as long-term high-load operation of the battery, internal short circuit, electrode material aging, and electrolyte decomposition. This provides verification and application support for the effectiveness of the device in detecting real overheating risk scenarios.

[0052] (III) Signal Processing and Transmission Components

[0053] Signal processing and transmission components act as "signal relay stations" between sensors and data analysis terminals, responsible for converting raw signals into analyzable standard data.

[0054] 1. Strain gauge

[0055] The strain display instrument is connected to the strain sensor 50 via a shielded cable. It receives the raw strain signal from the strain sensor 50, converts it into a digital strain value, and then transmits it to the data analysis terminal 15 via a serial port.

[0056] 2. Hydrogen detector main unit

[0057] The hydrogen detector host is connected to the hydrogen detector 60 via a dedicated cable, receives the raw gas signal from the hydrogen detector 60, converts it into a hydrogen concentration value in ppm, and then transmits it to the data analysis terminal 15.

[0058] 3. Oscilloscope

[0059] The oscilloscope 13 is connected to the ultrasonic receiver 80 to receive ultrasonic signals and display waveforms. At the same time, it extracts the peak voltage of the waveform, calculates the ultrasonic transmission amplitude, and transmits it to the data analysis terminal 15.

[0060] 4. Temperature sensor

[0061] The temperature sensor is connected to the thermistor 141, and the temperature data of multiple thermistors 141 are collected to calculate the real-time temperature value T and the temperature change rate, which is then transmitted to the data analysis terminal 15.

[0062] 5. Integrator

[0063] The hydrogen detector main unit, temperature sensor, and voltage sensor are combined to form an integrator 40.

[0064] (iv) Data processing and power unit

[0065] 1. Data Analysis Terminal

[0066] The data analysis terminal 15 communicates with components such as the strain gauge, hydrogen detection device host, oscilloscope 13, and temperature sensor via serial port, USB, and other interfaces. The data analysis terminal 15 receives all characteristic parameter data and performs tiered detection according to the logic of "sequentially judging whether each parameter exceeds its corresponding threshold": if any parameter exceeds the limit, an early warning signal is immediately generated and a battery shutdown command is triggered; if all parameters are normal, the battery is determined to be safe, and monitoring continues.

[0067] 2. Power supply

[0068] Power supply 16 is connected to the lithium battery under test (BUT) A and the heating base plate 17 via different cables. Power supply 16 provides charging and discharging power to the BUT A to simulate overcharging behavior that may occur in real-world application scenarios. For example, situations such as aging charging equipment circuitry, battery management system (BMS) failure, charging protocol mismatch, or abnormal external power supply may cause the battery to be charged with a voltage exceeding its design cutoff voltage. At the same time, it supplies power to the heating base plate 17 (simulating overheating conditions in real-world scenarios). It is the core power source for the device to reproduce real-world risk scenarios and verify the reliability of the testing method.

[0069] II. Detection Methods

[0070] As shown in Figure 2, the detection method of the present invention is based on the above-mentioned detection device and adopts a cascade detection logic. First, the working condition of the lithium battery A to be tested is determined, and then the characteristic parameters are detected in sequence.

[0071] (I) Operating Conditions

[0072] Before performing characteristic parameter testing, the operating state of the lithium battery A under test must first be described to determine whether it is in an overcharged or overheated state. Overcharging refers to applying a charging voltage exceeding the designed cutoff voltage to the lithium battery A through power supply 16. This can be explained by comparing the output voltage of power supply 16 with the battery's designed cutoff voltage.

[0073] Overheating refers to the behavior of the tested lithium battery A exceeding its safe operating temperature range due to external heat sources or abnormal internal heating. This can be explained by comparing the initial temperature collected by the temperature sensor with the battery's safe operating temperature range.

[0074] The purpose of the operating condition description is to determine the starting point for characteristic parameter detection. Under overcharge conditions, voltage parameter detection needs to be added, while under overheat conditions, the detection should start directly from the stress change rate parameter.

[0075] (II) Calculation and detection of characteristic parameters

[0076] According to the working condition description, the corresponding characteristic parameters are calculated in sequence and it is determined whether they exceed the corresponding threshold. The calculation method, detection sequence and threshold requirements of each characteristic parameter are as follows: 1. Detection sequence and requirements under overcharge condition Under overcharge condition, the detection is carried out in the order of "real-time voltage value → real-time stress change rate → real-time hydrogen concentration → real-time ultrasonic transmission amplitude → real-time temperature and real-time temperature change rate": (1) Real-time voltage value detection calculation method: According to the formula V real =k V out +b calculates the real-time voltage value of the lithium battery, where V real The voltage value is in real-time, measured in V; k is the sensitivity coefficient of the voltage sensor, dimensionless; V out is the output voltage value of the voltage sensor, in V; b is the zero offset, in V; k and b are determined through prior calibration using a standard voltage source.

[0077] Threshold determination: Calculate the V... real With the design cutoff voltage V of the lithium battery cr (Unit: V;) Compare, if V real ≥V cr If the real-time voltage value exceeds the threshold, then it is determined that the voltage value exceeds the threshold; if V real <V cr If so, proceed to the next feature parameter detection.

[0078] (2) Real-time stress change rate detection

[0079] First, the strain value ε (dimensionless) is calculated based on the strain signal acquired by strain sensor 50. The formula for calculating the strain value is ε = (ΔR / R0) / K0, where R0 is the initial resistance of the strain gauge in Ω; ΔR is the change in resistance of the strain gauge after deformation in Ω; and K0 is the strain gauge sensitivity coefficient, which is dimensionless. Then, the stress σ (in kPa) is calculated based on the strain value, σ = E. ε, where E is the elastic modulus of the lithium battery material, in Pa; finally, according to the formula dσ / dt=E The real-time stress change rate is calculated using (ε2-ε1) / Δt, where Δt is the time change in seconds; ε1 and ε2 are the strain values ​​before and after Δt, respectively, both of which are dimensionless.

[0080] The threshold for the real-time stress change rate is 0.45 kPa / s. If the calculated dσ / dt ≥ σ crIf dσ / dt = 0.45 kPa / s, it is determined that the threshold is exceeded; if dσ / dt < 0.45 kPa / s, then proceed to the next feature parameter detection.

[0081] (3) Real-time hydrogen concentration detection

[0082] The real-time hydrogen concentration value is calculated using the formula Cx=(Ix-I0) / Kx, where Cx is the concentration of the gas to be measured (in ppm), Ix is the operating current corresponding to the gas to be measured (in nA), I0 is the zero-point current (the sensor output current in clean air) (in nA), and Kx is the sensitivity coefficient of the hydrogen detector (obtained through standard gas calibration, in nA / ppm).

[0083] The threshold for real-time hydrogen concentration is 500 ppm. If Cx ≥ C cr =500ppm, then it is determined that the threshold C is exceeded. cr (Unit: ppm); if Cx < 500 ppm, proceed to the next characteristic parameter detection.

[0084] (4) Real-time ultrasonic transmission amplitude detection

[0085] The real-time ultrasonic transmission amplitude is calculated using the formula B=Up / M, where B is the transmission amplitude (in %), Up is the peak output voltage of the ultrasonic sensor when receiving the echo (in V), and M is the sensor sensitivity (representing the output voltage corresponding to a unit sound pressure level) (in V).

[0086] The threshold corresponding to the real-time ultrasonic transmission amplitude is 30%. If B ≥ B cr If the percentage is 30%, then it is determined that the threshold B has been exceeded. cr (Unit: V); if B < 30%, proceed to the next feature parameter detection.

[0087] (5) Real-time temperature and real-time temperature change rate detection

[0088] The real-time temperature value T (in °C) is directly collected by the temperature sensor; the real-time temperature change rate is calculated according to the formula dT / dt=(T2-T1) / (t2-t1), where T1 is the temperature value corresponding to the starting time t1 and T2 is the temperature value corresponding to the ending time t2.

[0089] If the real-time surface temperature of the battery is T∈τ=[120℃,150℃] or the real-time temperature change rate dT / dt≥ΔT'=1℃ / s, then the real-time temperature and the real-time temperature change rate are determined to exceed the corresponding threshold, and the lithium battery is in the thermal runaway stage; if none of the above conditions are met, then it is determined that the threshold has not been exceeded.

[0090] 2. Testing sequence and requirements under overheated conditions

[0091] Under overheating conditions, real-time voltage detection is skipped, and detection is performed in the following order: "real-time stress change rate → real-time hydrogen concentration → real-time ultrasonic transmission amplitude → real-time temperature and real-time temperature change rate". The calculation methods and threshold requirements for each parameter are the same as those for the parameters under overcharge conditions, and will not be repeated here.

[0092] (III) Result Determination and Processing

[0093] During the aforementioned tiered detection process, if any characteristic parameter exceeds its corresponding threshold, a warning signal of the corresponding level will be generated immediately, triggering a battery shutdown command to stop the use of the battery. At the same time, corresponding handling measures will be taken according to the warning level: If the real-time voltage value exceeds the threshold (under overcharge condition): the power supply will be immediately cut off, and relevant departments will be notified to send personnel to check the battery status.

[0094] If the real-time stress change rate exceeds the threshold (level 1 warning detected): Continuous monitoring will be conducted to prevent false alarms. Power will be immediately cut off, and relevant departments will be notified to send personnel to check the battery status.

[0095] If the real-time hydrogen concentration exceeds the threshold (a level 2 warning is detected): continue monitoring to prevent false alarms, immediately cut off the power supply, activate ventilation or exhaust devices to reduce the local concentration of flammable gas, and remind relevant departments to send personnel to check the battery status.

[0096] If the real-time ultrasonic transmission amplitude exceeds the threshold (level 3 warning detected): immediately cut off the power supply, isolate the individual unit from the system's main power circuit, close the relevant compartments or isolation valves to prevent heat and gas from spreading to adjacent areas, and implement thermal runaway suppression measures to reduce the risk of subsequent fire and reignition.

[0097] If the real-time temperature and real-time temperature change rate exceed the threshold (detection of thermal runaway): If an open flame is generated: cut off the power supply, immediately activate the fire extinguishing system, and completely terminate battery use; If no open flame is generated: immediately cut off the power supply, isolate the individual cells from the system's main power circuit, close the relevant compartments or isolation valves to prevent heat and gas from spreading to adjacent areas, and implement thermal runaway suppression measures to reduce the risk of subsequent fire and reignition.

[0098] If all characteristic parameters do not exceed the corresponding thresholds, the lithium battery is determined to be in normal condition and can continue to be used. At the same time, the detection system continues to run and monitor parameter changes in real time.

[0099] As shown in Figure 2, the operation logic of this process is a closed-loop dynamic detection and control mechanism for lithium battery thermal runaway. The specific details are as follows: First, the detection system and device are started. The system automatically determines the current operating condition of the battery: If it is identified as an overcharged state, it prioritizes detecting whether the real-time voltage value exceeds the battery's designed cutoff voltage threshold. Once the limit is exceeded, the system will immediately trigger a power cut-off command and push a battery abnormality prompt to relevant personnel, and simultaneously start a deep detection of the battery's physical state; if the voltage does not exceed the limit, it enters the tiered early warning detection stage.

[0100] Next, the system sequentially checks whether a Level 1 warning (corresponding to excessive stress change rate), a Level 2 warning (corresponding to excessive hydrogen concentration), and a Level 3 warning (corresponding to excessive ultrasonic transmission amplitude) are triggered. When a Level 1 warning is triggered, the system continues to monitor to prevent misjudgment, and immediately cuts off the power supply, sends a dispatch message to the relevant departments, and notifies personnel to come to the site to check the battery status.

[0101] When a Level 2 warning is triggered, the system continuously monitors to prevent false alarms, immediately cuts off the power supply and activates ventilation or exhaust devices to reduce the local concentration of combustible gases, and simultaneously sends a dispatch notification to relevant departments to notify personnel to check the battery status.

[0102] When a Level 3 warning is triggered, the system immediately executes measures such as cutting off power, isolating individual units from the system's main power circuit, closing relevant compartments or isolation valves to prevent heat and gas from spreading to adjacent areas, and suppressing thermal runaway, in order to reduce the risk of subsequent fire and reignition.

[0103] After completing the early warning detection, the system further determines whether the battery has entered the thermal runaway stage. If it is detected that the battery is in this stage, the system will simultaneously detect whether an open flame has been generated. If an open flame is generated, the fire extinguishing system will be activated immediately to put out the fire and the battery will be taken out of service. If no open flame is generated, the system will also perform power cut-off, isolation of the individual battery from the main power circuit of the system, closure of relevant compartments or isolation valves to prevent heat and gas from spreading to adjacent areas, and thermal runaway suppression measures to reduce the risk of subsequent fire and reignition.

[0104] Once the battery has been repaired and restored to a safe state through appropriate processing, the system will not directly restore its use. Instead, it will restart the entire testing process from the "start the detection system" stage, repeating all the steps of condition determination, voltage detection, cascade warning detection, and thermal runaway determination. The closed-loop process of "detection-early warning processing-re-retest after recovery" will be executed in a loop until, in a certain round of testing, the battery does not experience any parameter over-limits, warning triggers, or thermal runaway in any of the stages. Only then will the system determine that the battery is in normal condition and finally end the current testing process. This achieves precise and dynamic control over the battery's safety status.

[0105] The fire suppression system adopts PLC automatic control and uses a graded safety state machine control strategy to classify accidents into corresponding levels based on the real-time operating status of individual batteries (normal / Level 1 warning / Level 2 warning / Level 3 warning / thermal runaway stage) and execute differentiated automatic safety response actions. Among them, the flame signal has the highest control priority. Regardless of the current accident level, once an open flame is detected, the highest level of fire suppression is immediately triggered to ensure rapid risk response.

[0106] The specific PLC action logic and the coordinated handling of each early warning / accident stage are as follows: (1) Level 1 early warning (stress change rate exceeds the limit), Level 2 early warning (hydrogen concentration exceeds the limit): The PLC executes the action of "setting early warning signs, recording event logs, and not starting the fire extinguishing system", which coordinates with the handling measures of "continuous detection to prevent misjudgment, cutting off power supply, and pushing manual inspection instructions" mentioned above. This not only avoids excessive fire extinguishing and waste of resources, but also provides a basis for subsequent manual investigation through early warning records; (2) Level 3 warning (ultrasonic transmission amplitude exceeds limit), thermal runaway without open flame stage: PLC simultaneously executes the actions of "cutting off charging and discharging circuit, battery electrical isolation, starting thermal runaway suppression device, and prohibiting the system from autonomously restoring power supply", which is linked with the physical control measures of "isolation of individual units and closing relevant compartments / isolation valves to prevent heat and gas diffusion", thus curbing the risk escalation from both electrical and physical dimensions; (3) Level 3 warning, thermal runaway with open flame stage: PLC immediately executes the actions of "outputting fire extinguishing start signal, opening fire extinguishing solenoid valve, prohibiting midway stop spraying, entering 'accident lock state', prohibiting automatic reset (only supports manual or remote authorized reset), opening exhaust / pressure relief device to prevent deflagration or structural damage, closing isolation valve, starting adjacent area protection, and locking compartment", realizing full-process automated control of fire extinguishing, preventing reignition and preventing diffusion, and minimizing accident losses without manual intervention.

[0107] The PLC-controlled fire suppression system is deeply integrated with the closed-loop process of "detection-early warning processing-recovery retesting" mentioned earlier: After the battery is dealt with, if the PLC is in an "accident lockout state", it must be manually or remotely authorized to reset before it can be reconnected to the full process detection; if the lockout state is not triggered, the PLC enters the retesting stage synchronously with the system, continuously monitoring the status of individual batteries to ensure that the safety control closed loop is complete and fully adapts to the automation and high reliability requirements of unattended single-battery scenarios.

[0108] III. Detection System

[0109] The detection system of this invention uses a data acquisition module as its front-end core. It acquires output voltage via voltage sensors connected to the positive and negative electrodes of the lithium battery under test (BAT), acquires strain signals via strain sensors 50 attached to the surface of BAT, acquires hydrogen concentration via a hydrogen detector 60 positioned opposite the gas valve of BAT, acquires transmission amplitude via an ultrasonic sensor assembly (with transmitter and receiver symmetrically arranged on both sides of BAT), and acquires temperature data via a temperature sensor located in the electrode area. Simultaneously, it uses components such as a strain gauge and oscilloscope 13 to convert the raw signals into standard parameters. The data transmission module stably transmits these multi-dimensional parameters to the data processing module via wired methods such as shielded cables and serial ports. The data processing module, based on a preset tiered judgment logic, sequentially compares real-time voltage values, stress change rate, hydrogen concentration, ultrasonic transmission amplitude, temperature, and temperature change rate with corresponding thresholds. If any parameter exceeds the limit, an early warning signal is immediately generated and a battery shutdown command is triggered. If all parameters are normal, the battery is determined to be safe and continuously monitored. Ultimately, this achieves precise and real-time prevention and control of lithium battery thermal runaway risks, adapting to the safety requirements of new energy vehicles, energy storage devices, and other scenarios.

[0110] IV. Core Application Targets

[0111] The cascade detection method and system of this invention are primarily applied to scenarios where a single battery is used unattended: these are high-value, high-importance applications where real-time human intervention is lacking during equipment operation, and battery malfunctions can only be addressed through the battery itself or its associated monitoring system for status identification and safety procedures. Specifically, these scenarios include the following two core categories: 1. Long-term data monitoring in remote environments. Typical applications include forest fire risk monitoring stations (powering temperature, humidity, and smoke sensors 24 / 7), plateau glacier displacement observation points (powering GNSS positioning modules), and ocean buoys (powering seawater salinity sensors). The core characteristics of these scenarios are the lack of mains power supply, extremely high manual inspection costs (typically only 1-2 times per year), and the need for long-term stable battery power. The compatible individual batteries are mostly high-capacity, long-life models, such as the 21700 high-capacity lithium battery (capacity 4000~5000mAh) or lithium thionyl chloride battery (ER14505, capacity 2700mAh, life 10 years). They are usually combined with 5~10W micro solar panels to form a "photovoltaic-storage integrated" system. The BMS automatically switches the charging and discharging state to achieve an unattended power supply mode of charging during the day and discharging at night / on cloudy days.

[0112] 2. Large-scale low-power terminal monitoring scenarios

[0113] Typical applications include metering modules for smart water / electricity meters, tracking tags for logistics packages, soil pH sensors for agricultural greenhouses, and tilt monitors for urban manhole covers. The core characteristics of these scenarios are a massive deployment of terminals (up to millions), the need for "maintenance-free" operation for 3-5 years, and extremely high requirements for ultra-low power consumption and low self-discharge rates in batteries. The compatible single batteries are ultra-low power models. Button batteries (such as CR2450, 600mAh capacity) are suitable for terminals that are only occasionally woken up (such as logistics tags that send a signal once a day), while lithium fluorocarbon batteries (BR series, annual self-discharge rate <1%) are suitable for terminals that operate continuously for extended periods (such as smart water meters with a battery life of over 5 years).

[0114] This invention precisely matches the core requirements of the aforementioned unattended single-battery usage scenarios through miniaturized sensor adaptation, low-power operation design, and autonomous tiered processing logic. It not only ensures the safe and stable operation of the battery in long-term unattended environments, but also avoids the waste of battery resources caused by over-evaluation, and can effectively support the continuous and reliable operation of high-value unattended terminals.

[0115] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for cascading detection of thermal runaway in lithium batteries based on multiple characteristic parameters, characterized in that, The detection process includes the following steps: sequentially determining whether the real-time voltage, real-time stress change rate, real-time hydrogen concentration, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate of the lithium battery exceed their respective thresholds; if the current characteristic parameter exceeds its corresponding threshold, the battery is stopped from use; if all characteristic parameters do not exceed their corresponding thresholds, the lithium battery is determined to be normal and can continue to be used.

2. The method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters according to claim 1, characterized in that, The real-time voltage value of the lithium battery is calculated as follows: In the formula, V real V represents the real-time voltage value; k is the sensitivity coefficient of the voltage sensor used to detect the voltage; V out is the output voltage value of the voltage sensor; b is the zero-point offset.

3. The method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters according to claim 1, characterized in that, The real-time stress change rate is calculated as follows: In the formula, dσ / dt represents the real-time stress change rate; E is the elastic modulus of the lithium battery material; Δt is the time change; ε1 and ε2 are the strain values ​​before and after Δt, respectively.

4. The method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters according to claim 3, characterized in that, The real-time stress calculation is as follows: In the formula, σ represents the real-time stress; R0 represents the initial resistance of the strain gauge in the strain sensor attached to the surface of the lithium battery; ΔR represents the change in resistance of the strain gauge after deformation; and K represents the sensitivity coefficient of the strain gauge.

5. The method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters according to claim 1, characterized in that, The real-time ultrasonic transmission amplitude is calculated as follows: In the formula, B represents the real-time ultrasonic transmission amplitude; U p The peak output voltage of the ultrasonic sensor when it receives the echo is represented by M; M represents the sensitivity of the ultrasonic sensor, which is the output voltage per unit sound pressure.

6. The method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters according to claim 1, characterized in that, The threshold corresponding to the real-time voltage value is V. cr The threshold corresponding to the real-time stress change rate is σ. cr The threshold corresponding to the real-time hydrogen concentration is C. cr The threshold corresponding to the real-time ultrasonic transmission amplitude is B. cr .

7. The method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters according to claim 1, characterized in that, When the real-time temperature T of the battery surface is within the set temperature range τ or the real-time temperature change rate dT / dt is greater than or equal to the set temperature change rate threshold ΔT', it is determined that the real-time temperature and real-time temperature change rate of the lithium battery exceed the corresponding threshold, and the lithium battery is in the thermal runaway stage.

8. A method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters in a tiered manner according to any one of claims 1-3, characterized in that, Before performing characteristic parameter detection, it is necessary to detect the overcharge or overheat state of the lithium battery. When the lithium battery is in an overheated state, it is determined in sequence whether the real-time stress change rate, real-time hydrogen concentration, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate of the lithium battery exceed the corresponding threshold. When the battery is in an overcharged state, it is determined in sequence whether the real-time voltage value, real-time stress change rate, real-time hydrogen concentration, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate of the lithium battery exceed the corresponding threshold.

9. The method for detecting thermal runaway of lithium batteries based on multiple characteristic parameters according to claim 8, characterized in that, Overcharging refers to applying a charging voltage to a lithium battery that exceeds its design cutoff voltage; overheating refers to the lithium battery's temperature exceeding its safe operating temperature range due to external heat sources or abnormal internal heating.

10. A system for cascade detection of lithium battery thermal runaway based on multiple characteristic parameters, the system being used to execute the method for cascade detection of lithium battery thermal runaway based on multiple characteristic parameters as described in claim 9, characterized in that, include: The data acquisition module is used to acquire various characteristic parameters of lithium batteries in real time. The data transmission module is electrically connected to the data acquisition module and is used to transmit various characteristic parameters; The data processing module, which communicates with the data transmission module, is used to receive and process various characteristic parameters. The data acquisition module includes: a voltage sensor for acquiring the real-time voltage value of the lithium battery; a strain sensor for acquiring strain signals on the surface of the lithium battery and calculating the stress change rate based on the strain signals; a hydrogen sensor for acquiring the real-time hydrogen concentration value of the environment in which the lithium battery is located; an ultrasonic sensor assembly, including an ultrasonic transmitter, a receiver, and a signal generator, for acquiring the ultrasonic transmission amplitude penetrating the lithium battery; and a temperature sensor for acquiring the real-time temperature value of the lithium battery surface and calculating the temperature change rate. The data processing module is configured to perform tiered early warning judgments. Its judgment logic is as follows: sequentially judging whether the real-time voltage value, real-time stress change rate, real-time hydrogen concentration value, real-time ultrasonic transmission amplitude, real-time temperature, and real-time temperature change rate of the lithium battery exceed the corresponding thresholds; if any judgment result is yes, an early warning signal of the corresponding level is generated and a battery shutdown command is triggered; if all judgment results are no, the battery status is determined to be normal.

Images