Battery thermal runaway eradication system, method and device based on controllable micro-impact reaction

Abstract

The invention provides a battery thermal runaway eradication system, method and device based on controllable micro-impact reaction, and the system comprises a micro-channel array which is disposed on the surface of a battery current collector; the plurality of micro-energy control units are arranged at node positions of the micro-channel array and are used for dredging the micro-channel array through a controllable micro-impact effect; the intelligent control module is connected with the micro-energy control unit and used for controlling triggering of the micro-energy control unit; and the external fire-fighting module is used for conveying an inhibitor into the battery after the micro-channel array is dredged. A micro-channel network is preset in the battery manufacturing stage, and an expressway directly reaching the thermal runaway origin is laid for a fire extinguishing agent. Through precise controllable micro-impact reaction of the micro-energy control unit, a channel is instantly dredged within millisecond time, so that an external efficient fire extinguishing agent can directly act on a reaction core, and the industrial problem that the external fire extinguishing agent cannot effectively permeate is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of batteries, and particularly relates to a battery thermal runaway eradication system, method and device based on controllable micro-impact reaction. BACKGROUND

[0002] With the rapid development of new energy vehicles, large-scale energy storage and other industries, the energy density and capacity of lithium ion batteries and next-generation solid-state batteries are continuously improving, and their thermal runaway safety risks have become the most core bottleneck restricting the development of the industry. Thermal runaway is a process of chain heat release reaction out of control in the battery, which originates from the inside of the battery (such as internal short circuit caused by separator rupture), and spreads rapidly within milliseconds to seconds, eventually causing disastrous consequences such as fire and explosion.

[0003] At present, the prevention and control schemes for battery thermal runaway in the industry mainly include "passive safety" and "active safety". Passive safety schemes, such as the use of phase change materials, heat insulation cotton, etc., mainly delay heat spread through heat absorption and heat insulation, but their heat management capability is limited and cannot cope with the huge energy released in the moment of severe thermal runaway. Active safety schemes are mainly of the "external intervention" type, i.e. deploying detection systems and fire extinguishing agents (such as perfluorohexanone, aerosol, water mist, etc.) outside the battery pack or module, and spraying fire extinguishing agents when thermal runaway signs are detected.

[0004] However, these existing technical solutions all have insurmountable essential defects, which are specifically manifested as follows:

[0005] "Space-time barrier" problem: the "fire source" of thermal runaway is deeply hidden in the dense porous electrode core of the battery, and the external fire extinguishing agent needs to go through a long process of "detection-decision-spraying-penetration". When the fire extinguishing agent reaches the outside of the battery, the internal thermal runaway reaction has entered an irreversible severe stage; more importantly, it is extremely difficult for the fire extinguishing agent to penetrate the hard metal shell (such as steel shell, aluminum shell) of the battery within the time window of seconds and further penetrate to the thermal runaway origin inside the core. This results in that external fire extinguishing can only temporarily suppress external fire, but cannot eradicate the continuous heat release reaction inside, thus causing a very high rekindling rate, which belongs to "treating the symptoms but not the root cause".

[0006] "Failure rate" of next-generation batteries: for all-solid-state batteries, the solid-state electrolyte layer is extremely dense, hard and has very low porosity. Once internal short circuit occurs, any external fire extinguishing means is almost completely unable to penetrate to the reaction core, lacking effective internal intervention approaches, which makes solid-state batteries face huge safety challenges.

[0007] Limitations of internal intervention schemes: Some existing internal safety schemes, such as coating the separator with heat-sensitive materials or designing fragile structures, often have unreliable responses or negatively impact the battery's electrochemical performance. For example, "internal puncture" schemes may damage the battery's structural integrity and introduce new short-circuit risks.

[0008] In published patent applications, such as Chinese Invention Patent Application Publication No. CN114705999A, a battery thermal runaway testing method, device, apparatus, and computer-readable storage medium are disclosed. The battery thermal runaway testing method is applied to a battery thermal runaway testing device, which includes a housing and a pressure relief valve disposed within the housing, used to connect the inner cavity of the housing to the outside. The battery thermal runaway testing method includes: acquiring the gas pressure inside the housing as the runaway gas pressure value when a battery placed inside the housing experiences thermal runaway; determining the venting rate of the pressure relief valve based on the runaway gas pressure value; and determining the gas generation rate during battery thermal runaway based on the venting rate. The above-mentioned battery thermal runaway testing method provides relatively accurate test results for the gas generation rate of a battery under thermal runaway conditions.

[0009] The existing testing methods described above still have significant limitations in the early identification and warning of thermal runaway: they typically rely on threshold judgments based on a single parameter (such as air pressure or temperature), failing to effectively integrate multi-dimensional and multi-physical field signals (such as voltage drops, changes in gas composition, and internal stress fluctuations) for collaborative analysis and early diagnosis. This detection mechanism based on a single criterion is insufficiently sensitive to latent thermal runaway behaviors in their initial stages or with atypical evolutionary paths, making it difficult to provide accurate early warnings before the thermal runaway chain reaction enters an irreversible phase. Consequently, it cannot provide sufficient response time for active safety systems, limiting the effective intervention of prevention and control measures.

[0010] In view of the above-mentioned technical problems in the existing technology, the present invention provides a battery thermal runaway eradication system, method and device based on controllable micro-impact reaction. Summary of the Invention

[0011] This invention proposes a battery thermal runaway eradication system, method, and apparatus based on controllable micro-impact reaction.

[0012] The present invention adopts the following technical solution:

[0013] The battery thermal runaway eradication system based on controllable micro-impact response includes:

[0014] Microchannel array, disposed on the surface of the battery current collector;

[0015] Multiple micro-energy control units are located at the node positions of the microchannel array, and are used to clear the microchannel array through controllable micro-impact reactions;

[0016] An intelligent control module, connected to the micro-energy control unit, is used to control the triggering of the micro-energy control unit;

[0017] An external fire suppression module is used to deliver inhibitors into the battery after the microchannel array has been cleared.

[0018] Furthermore, the microchannel array is a groove-shaped channel formed on the surface of the current collector by laser etching or precision imprinting, wherein the depth of the groove-shaped channel is 10–20 μm and the width is 30–50 μm.

[0019] Furthermore, the micro-energy control unit is a multilayer thin film structure made by MEMS process using nanoscale aluminum powder and nanoscale copper oxide powder, including an aluminum layer and a copper oxide layer.

[0020] Furthermore, the molar ratio of aluminum to copper oxide in the multilayer thin film structure is 2:3, and the particle size of the aluminum and copper oxide nanoparticles is 80–100 nm.

[0021] Furthermore, the surface of the micro-energy control unit is provided with a parylene protective film.

[0022] Furthermore, the intelligent control module is connected to each micro-energy control unit via a circuit, and triggers the micro-energy control unit by applying an electrical pulse. The voltage of the electrical pulse is not less than 5V and the duration is on the order of microseconds.

[0023] Furthermore, the intelligent control module employs a multi-parameter fusion algorithm for thermal runaway detection. The algorithm includes calculating a synergy index between the change rates of different parameters. The parameters include at least two of the following: voltage change rate, temperature change rate, internal pressure change rate, hydrogen concentration, and carbon monoxide concentration. The synergy index includes the Pearson correlation coefficient between the voltage change rate and the temperature change rate.

[0024] Furthermore, the intelligent control module is integrated into the battery management system and thermal runaway identification is achieved through a support vector machine classifier.

[0025] This invention also provides a method for eliminating battery thermal runaway based on controllable micro-impact reaction, comprising the following steps:

[0026] The micro-energy control unit set inside the battery is triggered by an electrical pulse to cause a controllable micro-impact reaction and clear the microchannel array embedded in the surface of the current collector.

[0027] Fire suppression inhibitors are delivered into the battery through a cleared microchannel array.

[0028] The present invention also provides a battery thermal runaway eradication device based on controllable micro-impact reaction, comprising:

[0029] The battery cell has the microchannel array and micro-energy control unit pre-embedded on its current collector surface;

[0030] The intelligent control module is integrated with the battery management system and is used to activate the micro-energy control unit and external fire suppression module when a risk of thermal runaway is identified.

[0031] Compared with the prior art, the superior effects of the present invention are as follows:

[0032] 1. The battery thermal runaway eradication system, method, and apparatus based on controllable micro-impact reaction described in this invention pre-installs a microchannel network during the battery manufacturing stage, paving a "highway" for the extinguishing agent to directly reach the origin of thermal runaway. Through the precise and controllable micro-impact reaction of the micro-energy unit (MECU), the channels are instantly cleared within milliseconds, allowing the external high-efficiency extinguishing agent to directly act on the reaction core. This solves the industry problem of the inability of external extinguishing agents to effectively penetrate, achieving "annihilation" extinguishing from the inside out, thereby completely preventing reignition.

[0033] 2. The battery thermal runaway eradication system, method, and device based on controllable micro-impact reaction described in this invention deeply integrates the early warning and triggering mechanism into the battery management system (BMS). Based on a multi-parameter fusion machine learning algorithm, it can identify and locate thermal runaway in the earliest stage (T1 stage). The entire response time from early warning to MECU triggering and unblocking can be controlled at the millisecond level, far exceeding the second-level response of any external fire suppression system, and can nip the thermal runaway chain reaction in the bud before it fully erupts.

[0034] 3. The battery thermal runaway eradication system, method, and apparatus based on controllable micro-impact reaction described in this invention employs an electrically triggered MECU with a trigger voltage far exceeding the battery's operating voltage, fundamentally eliminating accidental activation caused by normal battery voltage fluctuations or erroneous signals from the BMS. Simultaneously, the MECU and microchannels are fully encapsulated using materials such as parylene, ensuring compatibility and long-term stability with the electrolyte throughout the entire battery lifespan, without any negative impact on the battery's electrochemical performance.

[0035] 4. The battery thermal runaway eradication system, method and device based on controllable micro-impact reaction described in this invention, through nanomaterial formulation and MEMS process design, precisely controls the reaction of the MECU to a "controllable micro-impact reaction" mode. Its energy is concentrated on clearing blockages at the entrance of the channel or slightly softening the solid electrolyte interface, rather than causing a destructive explosion. While effectively opening the path, it protects the overall structural integrity of the battery to the greatest extent. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the battery thermal runaway eradication system based on controllable micro-impact reaction of the present invention;

[0037] Figure 2 This is a schematic diagram of the current collector structure integrating a microchannel array and a MECU of the present invention;

[0038] Figure 3 This is a schematic diagram of the micro-MEMS structure of the MECU unit of the present invention;

[0039] Figure 4 This is a schematic diagram comparing the voltage and temperature-time curves of the acupuncture experiment in Embodiment 2 of the present invention;

[0040] Figure 5 This is a schematic diagram of the workflow of the intelligent control module of the present invention;

[0041] Figure 6 This is a schematic diagram of the working process of the battery thermal runaway eradication system based on controllable micro-impact reaction of the present invention.

[0042] The figure shows: 1 - current collector, 2 - microchannel array, 3 - micro-energy control unit, 4 - parylene encapsulation layer, 5 - electrode slurry layer, 6 - seed layer, 7 - Al / CuO alternating layer, 8 - Al layer, 9 - CuO layer. Detailed Implementation

[0043] To better understand the above-mentioned objectives, features and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features in the embodiments can be combined with each other.

[0044] like Figures 1-6 As shown, the present invention provides a battery thermal runaway eradication system based on controllable micro-impact reaction, comprising:

[0045] 2. Microchannel array; 3. Several micro-energy control units; 4. Intelligent control module; 5. External fire protection module.

[0046] like Figure 1 As shown, the microchannel array 2 and the micro-energy control unit 3 are located inside a single battery cell. The intelligent control module is integrated into the battery management system, and the external fire protection module is located outside the battery module. The intelligent control module controls the micro-energy control unit 3 through a trigger circuit and directly controls the external fire protection module.

[0047] The microchannel array 2 is formed on the surface of the current collector 1 of the battery by laser etching or precision imprinting. The current collector 1 is a negative electrode copper foil or a positive electrode aluminum foil. The typical cross-section of the microchannel is rectangular or trapezoidal, with a depth of 10 to 20 micrometers and a width of 30 to 50 micrometers. The microchannel array 2 covers the entire active area of ​​the current collector 1 in a mesh-like form. Under normal battery operation, the microchannels are tightly bonded by the electrode slurry layer 5 and the parylene encapsulation layer 4, and are in a physically sealed state.

[0048] The specific placement of the microchannel array 2 on the surface of the current collector 1 can be optimized according to the battery system to improve intervention efficiency: For liquid electrolyte batteries, it is preferable to place it on the surface of the negative electrode current collector 1 (copper foil) because the ductility of copper foil is better than that of the positive electrode aluminum foil. After laser etching to form microchannels, it can better maintain mechanical integrity. At the same time, the side reactions in the early stage of thermal runaway are mostly concentrated at the negative electrode interface. Placing microchannels here is conducive to achieving rapid response. For solid-state batteries, given that the solid electrolyte layer is dense and hard, the microchannels and the integrated micro-energy control unit 3 are adjusted to the interface between the positive electrode current collector 1 and the solid electrolyte, so that the energy generated by the controllable micro-impact reaction can directly act on the dense electrolyte layer that needs to be opened, and efficiently form a penetration crack network.

[0049] The core material of the Micro Energy Control Unit 3 (MECU) is a metastable composite energetic material based on aluminum and copper oxide. This material is stable at room temperature (metastable) and requires specific electrical pulse energy to trigger. Its reaction formula is 2Al + 3CuO → Al₂O₃ + 3Cu + Heat. By controlling the nanoparticle size (80-100nm), ratio (Al:CuO≈2:3mol), and density, its reaction mode is precisely regulated to a rapid "controllable micro-impact reaction," rather than slow combustion or violent explosion. The "controllable micro-impact reaction" is a "localized, microsecond-level, controllable instantaneous energy release behavior." The characteristics of the "controllable micro-impact reaction" are: concentrated energy release (acting on ~1mm particles). 2The micro-energy control unit 3 is a multilayer thin-film structure fabricated using microelectromechanical systems (MEMS) technology. It consists of alternating Al layers 8 and CuO layers 9. The thickness of each layer is matched to the nanoparticle size, controlled between 50 and 200 nanometers, ensuring a large contact area and extremely short diffusion distance between reactants, thus achieving rapid and concentrated release of reaction energy. The total thickness is designed according to the required reaction energy, typically between 1 and 10 micrometers, to ensure integration onto the surface of the current collector 1 without affecting the battery's electrochemical performance. The thickness and number of layers are precisely calculated and controlled based on the stoichiometric ratio of the aluminothermic reaction (2Al + 3CuO) to ensure complete reaction. The interfaces between layers are steep to ensure a clear reaction initiation point and controllable propagation speed. A sublayer 6, made of nickel or gold and with a thickness of 10 to 20 nanometers, is placed between the current collector 1 and the first reactive film. Its function is twofold: firstly, to enhance interfacial adhesion and prevent detachment during cycling; secondly, it acts as a catalyst layer, effectively lowering the reaction initiation energy barrier and ensuring rapid and uniform reaction propagation upon electrical pulse triggering. When the number of film layers is large (e.g., more than 5 layers), a bridging layer, made of alumina or silicon dioxide and with a thickness of 5 to 10 nanometers, is introduced between the alternately stacked Al layers 8 and CuO layers 9. This layer alleviates internal stress caused by the multilayer structure, preventing film cracking, while ensuring its thickness is insufficient to hinder the overall propagation of the reaction. The MECU is not a thermosensitive material. It requires a specific energy supply (≥5V) from an external high-voltage pulse triggering circuit to activate. This voltage is much higher than the battery operating voltage, fundamentally eliminating accidental ignition caused by battery voltage fluctuations or BMS malfunctions, ensuring extremely high safety.

[0050] like Figure 2 As shown, the micro-energy control unit 3 is located at the nodes of the microchannel array 2. The nodes are the intersections of the grid of the microchannel array 2. The node distance is optimized according to the controllable micro-impact reaction energy propagation range of the micro-energy control unit 3, typically 3 mm to 5 mm, to ensure that the thermal runaway area can be effectively covered. Each micro-energy control unit 3 is approximately 1 mm by 1 mm in size. The entrance of the node, i.e., the convergence point of 3 to 4 microchannels, is precisely covered by conductive adhesive to ensure that multiple microchannels can be unblocked simultaneously after triggering.

[0051] The micro-energy control unit 3 and the microchannel array 2 are encapsulated by a parylene encapsulation layer 4 through vacuum phase deposition, forming a 5-10 micrometer-thick protective film that is insulating, moisture-proof, and resistant to electrolyte corrosion. The encapsulation covers the entire surface of the micro-energy control unit 3 and the inner wall of the microchannels, extending from the microchannel inlet node to the microchannel end.

[0052] The parylene encapsulation layer 4 possesses excellent insulation, moisture resistance, and electrolyte corrosion resistance properties. It is disposed on the entire integrated structure surface of the microchannel and micro-energy control unit 3, and its encapsulation range completely covers the integrated structure: First, the parylene encapsulation layer 4 covers the entire outer surface of the micro-energy control unit 3 with a thickness of 5-10 μm, including its top, sides, and edge areas in contact with the current collector 1; Second, it also covers the inner wall surface of the microchannel, forming a thin film without blocking the channel groove space, thereby assisting the electrode slurry layer 5 in sealing the microchannel and isolating it from the electrolyte under normal conditions; In addition, the boundary of the parylene encapsulation layer 4 extends from the entrance node (each node corresponds to a micro-energy control unit 3) which is the intersection of the microchannel array 2 to the area near the core of the electrode active material at the end of the microchannel, forming a continuous and complete protection system. The entrance node is where 3-4 microchannels converge (in a "+" or "Y" shape). The micro-energy control unit 3 is fixed directly above the node by conductive adhesive and directly covers the entrance end of the microchannel.

[0053] In one specific embodiment, the micro-MEMS structure of the micro-energy control unit 3 is as follows: Figure 3 As shown, from the outside to the inside, there are parylene encapsulation layer 4, Al / CuO alternating layer 7, seed layer 6, and current collector 1. The Al / CuO alternating layer 7 consists of alternating CuO layer 9 and Al layer 8, with the outermost layer being CuO layer 9 and the innermost layer being Al layer 8.

[0054] The integrated structure of the micro-energy control unit 3, the microchannel array 2, and the parylene encapsulation layer 4 can be extended to the battery module level. For example, a functional coating with microchannels and MECUs can be prepared on the surface of the metal fireproof separator of the battery module (the coating thickness can be adapted to the module design, such as 200 μm) to achieve secondary barrier against heat spread.

[0055] The intelligent control module, based on the battery management system platform, integrates multi-parameter sensor information for early warning of thermal runaway. It is connected to the conductive contacts of the micro-energy control unit 3 via a trigger circuit. This trigger circuit consists of independent wires (50-100μm diameter silver-plated copper wire) arranged along the inner wall of the battery casing, completely isolated from the charging and discharging circuit of the current collector to ensure an independent and stable trigger signal. The intelligent control module, as shown... Figure 5The system comprises a data acquisition layer, a feature extraction layer, and a decision layer. The data acquisition layer monitors voltage, voltage change rate, temperature, temperature change rate, internal pressure, hydrogen concentration, carbon monoxide concentration, and ultrasonic signals in real time, with a sampling frequency of at least 100 Hz. The feature extraction layer extracts key feature vectors from the raw data, including the voltage's decrease slope within a 1-millisecond time window, the temperature's increase rate within a 3-second time window, the acceleration of gas concentration changes, and a synergy index for multiple parameter changes. The synergy index is a set of features used to measure the spatiotemporal anomalous correlation between different parameter pairs, not a single formula. Its core principle is that thermal runaway is a chain reaction involving multi-physics coupling. When this reaction occurs, parameters that were originally independent or weakly correlated will exhibit strong and specific synergistic anomalous patterns. The synergy index includes calculating the Pearson correlation coefficient between the voltage change rate and the temperature change rate within a 500-millisecond sliding time window. The formula is: the synergy index equals the covariance of the voltage change rate and the temperature change rate within the window divided by the product of their standard deviations. Other synergistic indicators include the correlation coefficients between voltage and pressure change rates, and temperature and hydrogen concentration change rates, forming a subset of synergistic features. The decision layer uses a support vector machine classifier for thermal runaway identification, trained on a large amount of battery abuse test data. The kernel function maps the feature vectors to a high-dimensional space to find the optimal classification hyperplane. A thermal runaway probability value is output, triggering an alert when the probability value is greater than 0.95. The localization algorithm monitors abnormal voltage change patterns at each sampling point, combines this with a heat propagation model for three-dimensional spatial localization, outputs a list of micro-energy control units (3) to be activated, and activates these units via a trigger circuit. The trigger circuit is a dedicated LC pulse discharge circuit that generates high-energy electrical pulses with a voltage greater than or equal to 5 volts and a duration in the microsecond range.

[0056] The entire triggering circuit system adopts a centrally branched tree topology. The central control and pulse generation layer is located at the battery pack level, containing the battery management system main controller and the central high-voltage pulse generation circuit. The zoned triggering and control layer is located inside the battery module and is responsible for managing the triggering circuits of multiple cells. The terminal execution network is built into a single cell, with metallized traces fabricated on the current collector 1 using thin-film deposition and photolithography processes, and connected to the micro-energy control unit 3. The layers are connected via low-voltage control buses and high-voltage pulse transmission lines, and electrical isolation is achieved using optocouplers or magnetic couplers. The triggering network busbars inside the cell are connected to flexible circuit boards or insulated wires using micro-welding technology, leading out from the cell tabs and connecting to the module-level triggering harness.

[0057] Example 1: Calculating the thermal runaway probability P

[0058] Step 1: Data Acquisition and Preprocessing

[0059] The system continuously monitors the voltage V(t) and temperature T(t) of each cell at a sampling frequency of 100Hz, and can optionally monitor parameters such as internal pressure P(t) and hydrogen concentration H2(t).

[0060] All parameter data are recorded in real time and stored in a buffer for subsequent time window analysis.

[0061] Step 2: Calculation of Synergy Indicators

[0062] The synergy index is calculated using a sliding time window method, with a fixed window length of 500 milliseconds (i.e., backtracking from the current analysis time t to t-0.5s). For each parameter pair, the Pearson correlation coefficient is calculated using the following sub-steps:

[0063] 2.1 Core Synergy Indicators (Required)

[0064] The core indicator is the Pearson correlation coefficient between the rate of change of voltage (dV / dt) and the rate of change of temperature (dT / dt), which is calculated as follows:

[0065] Within the time window, the first derivatives of the voltage series V(t) and temperature series T(t) are calculated respectively, yielding the dV / dt and dT / dt arrays. The derivative calculations employ numerical difference methods, such as the central difference method.

[0066] The Pearson correlation coefficient S_index(t) for these two derivative arrays is calculated using the following formula:

[0067] S_index(t)=ρ_{dV / dT}=[Σ((dV_i / dt-μ_{dV})*(dT_i / dt-μ_{dT}))] / [n*σ_{dV}*σ_{dT}]

[0068] Where dV_i / dt and dT_i / dt are the rates of change of voltage and temperature at each sampling point within the window;

[0069] μ_{dV} and μ_{dT} are the average values ​​of the rate of change of voltage and the rate of change of temperature within the window;

[0070] σ_{dV} and σ_{dT} are the standard deviations of the rate of change of voltage and the rate of change of temperature within the window;

[0071] n: Number of sampling points within the window (n = 50 at 100Hz sampling).

[0072] Physical meaning: S_index(t)∈[-1,1]. When S_index(t) is consistently below the negative threshold (e.g., -0.7), it indicates a strong negative correlation between voltage drop and temperature rise, which is an early characteristic of thermal runaway.

[0073] 2.2 Other synergy metrics (optional, used for cross-validation)

[0074] The system simultaneously calculates the synergy of other parameter pairs to form a multidimensional feature subset:

[0075] Voltage change rate (dV / dt) and pressure change rate (dP / dt): A strong negative correlation is expected, reflecting the pressure increase caused by heat generated by internal short circuit.

[0076] Temperature change rate (dT / dt) and hydrogen concentration change rate (dH2 / dt): expected to be strongly positively correlated, reflecting that temperature rise accelerates electrolyte decomposition to produce hydrogen.

[0077] Pressure change rate (dP / dt) and hydrogen concentration change rate (dH2 / dt): expected to be strongly positively correlated, reflecting that gas production directly leads to pressure change.

[0078] The synergy of each parameter pair is calculated using the Pearson correlation coefficient formula described above, yielding corresponding indices such as ρ_dVdP, ρ_dTdH2, and ρ_dPdH2.

[0079] Step 3: Feature Subset Construction

[0080] Combine all the synergy metrics calculated in step 2 into a synergy feature subset (Synergy_Feature_Subset):

[0081] Synergy_Feature_Subset=[ρ_dVdT,ρ_dVdP,ρ_dTdH2,ρ_dPdH2,...]

[0082] This subset is combined with the independent features of each parameter (such as voltage drop slope and temperature rise rate) to form the final feature vector, which serves as the input to the machine learning model.

[0083] Step 4: Machine Learning Classification and Decision Making

[0084] The feature vectors are trained and inferred using a Support Vector Machine (SVM) classifier. The SVM model is trained on historical data to identify cooperative anomalous patterns in thermal runaway.

[0085] Decision logic: When the synergy index (such as ρ_dVdT) is consistently abnormal (e.g., <-0.7) and combined with other features (such as voltage drop), the SVM outputs a high-risk classification.

[0086] Step 5: Probability Output

[0087] The Platt scaling method is used to convert the SVM output into probability values ​​P to quantify the risk of thermal runaway. Specifically:

[0088] Train a logistic regression model and map the SVM decision value f to a probability:

[0089] Parameters A and B are determined through cross-validation to ensure probability calibration.

[0090] When the probability P exceeds a threshold (e.g., 0.95), the system triggers a thermal runaway warning.

[0091] Example 2: Used in high-energy-density ternary lithium-ion pouch batteries

[0092] Based on a 100 Ah NCM811 / graphite pouch cell with an energy density of no less than 280 Wh / kg, a mesh channel structure with a depth of approximately 15 μm and a width of approximately 40 μm was formed on the copper foil surface of the negative electrode current collector using ultraviolet laser etching. Micro-energy control units (3) with dimensions of 1 mm × 1 mm were arranged at each mesh node. All processes were completed in a dry environment, and finally, encapsulation was performed using parylene vacuum phase deposition, with an encapsulation layer thickness of 8 μm. During testing, the battery was charged to 100% SOC and then subjected to a nail penetration test. The results are as follows... Figure 4 As shown, the battery management system identifies and locates the fault point within 1.2ms after detecting a sudden voltage drop, triggers the three nearest micro-energy control units 3, successfully clears the corresponding microchannels and injects perfluorohexanone, keeping the battery's maximum temperature below 380℃ and preventing deflagration.

[0093] Example 3: Used in oxide solid-state batteries

[0094] Based on a 2 Ah NCM / lithium zirconate lanthanum oxide / lithium metal solid-state battery, a micro-energy control unit 3 is integrated at the interface between the positive electrode current collector 1 and the lithium zirconate lanthanum oxide solid electrolyte layer. The controllable micro-impact reaction energy setting is appropriately increased, making it slightly higher than that in Example 1. After an internal short circuit is triggered by overcharging, the system successfully activates the micro-energy control unit 3, inducing the generation of a micron-scale crack network within the electrolyte layer. This allows perfluorohexanone to penetrate along the cracks into the short-circuit region, effectively suppressing the thermal runaway reaction.

[0095] Example 4: For large-scale lithium iron phosphate energy storage modules

[0096] The energy storage module, based on 280 Ah lithium iron phosphate cells, employs a current collector 1 with microchannels and a micro-energy control unit 3 during the cell manufacturing stage. Simultaneously, a protective coating with similar functionality, approximately 200 μm thick, is applied to the surface of the module's metal fireproof partition. In a simulated single-cell thermal runaway experiment, the internal system of the cell responds first, followed by a secondary triggering of the partition surface system, effectively delaying the time for heat to spread to adjacent cells to over 20 minutes.

[0097] Working principle and process as follows Figure 6As shown:

[0098] Step 1: Early warning of thermal runaway

[0099] Warning: The battery management system has detected a T1 level thermal runaway warning through a multi-parameter fusion algorithm.

[0100] Location and Decision: The system accurately locates the faulty battery cell and calculates the number of micro-energy control units 3 that need to be activated.

[0101] Step 2: High-voltage pulse triggers MECU

[0102] Trigger: The intelligent control module activates the trigger circuit to apply an electrical pulse to the target micro-energy control unit 3.

[0103] Step 3: Rapid unblocking of microchannels

[0104] Activation path: The micro-energy control unit 3 is triggered to generate a controllable micro-impact reaction, producing a shock wave and localized high temperature. For liquid electrolyte batteries, the shock wave vaporizes the electrolyte blocking the microchannel inlet and causes the parylene encapsulation layer 4 to separate instantaneously from the current collector 1; for solid-state batteries, the controllable micro-impact reaction generates a micron-scale crack network in the solid electrolyte.

[0105] Step 4: Precise penetration of extinguishing agent to extinguish the fire

[0106] Empowering Extinguishing: The external fire extinguishing system is activated, and the extinguishing agent enters from the battery tabs and penetrates into the core area of ​​thermal runaway through the cleared microchannel network.

[0107] A method for eliminating battery thermal runaway based on controllable micro-impact response includes the following steps:

[0108] The micro-energy control unit 3, which is set inside the battery, is triggered by an electrical pulse to cause a controllable micro-impact reaction and clear the microchannel array 2 pre-embedded on the surface of the current collector 1.

[0109] Fire suppression inhibitors are delivered into the battery through the cleared microchannel array 2.

[0110] A battery thermal runaway eradication device based on controllable micro-impact response, comprising:

[0111] The battery cell has the microchannel array 2 and micro-energy control unit 3 pre-embedded on the surface of its current collector 1;

[0112] The intelligent control module is integrated with the battery management system and is used to activate the micro-energy control unit 3 and the external fire suppression module when a risk of thermal runaway is identified.

[0113] This invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims.

Claims

1. A battery thermal runaway eradication system based on controllable micro-impact reaction, characterized in that, include: Microchannel array, disposed on the surface of the battery current collector; Multiple micro-energy control units are located at the node positions of the microchannel array, and are used to clear the microchannel array through controllable micro-impact reactions; An intelligent control module, connected to the micro-energy control unit, is used to control the triggering of the micro-energy control unit; An external fire suppression module is used to deliver inhibitors into the battery after the microchannel array has been cleared.

2. The system according to claim 1, characterized in that, The microchannel array is a groove-shaped channel formed on the surface of the current collector by laser etching or precision imprinting. The groove-shaped channel has a depth of 10–20 μm and a width of 30–50 μm.

3. The system according to claim 1, characterized in that, The micro-energy control unit is a multilayer thin film structure made by MEMS process using nanoscale aluminum powder and nanoscale copper oxide powder, including an aluminum layer and a copper oxide layer.

4. The system according to claim 3, characterized in that, The molar ratio of aluminum to copper oxide in the multilayer thin film structure is 2:3, and the particle size of the aluminum and copper oxide nanoparticles is 80–100 nm.

5. The system according to claim 1, characterized in that, The surface of the micro-energy control unit is provided with a parylene protective film.

6. The system according to claim 1, characterized in that, The intelligent control module is connected to each micro-energy control unit through a trigger circuit, and triggers the micro-energy control unit by applying an electrical pulse. The voltage of the electrical pulse is not less than 5V and the duration is on the order of microseconds.

7. The system according to claim 1, characterized in that, The intelligent control module uses a multi-parameter fusion algorithm to determine thermal runaway. The algorithm includes calculating the synergy index between the change rates of different parameters. The parameters include at least two of the following: voltage change rate, temperature change rate, internal pressure change rate, hydrogen concentration, and carbon monoxide concentration. The synergy index includes the Pearson correlation coefficient between the voltage change rate and the temperature change rate.

8. The system according to claim 1, characterized in that, The intelligent control module is integrated into the battery management system and thermal runaway identification is achieved through a support vector machine classifier.

9. A method for eliminating battery thermal runaway based on a controllable micro-impact reaction according to the system described in any one of claims 1-8, characterized in that, Includes the following steps: The micro-energy control unit set inside the battery is triggered by an electrical pulse to cause a controllable micro-impact reaction and clear the microchannel array embedded in the surface of the current collector. Fire suppression inhibitors are delivered into the battery through a cleared microchannel array.

10. A battery thermal runaway eradication device based on a controllable micro-impact reaction according to any one of claims 1-8, characterized in that, include: The battery cell has the microchannel array and micro-energy control unit pre-embedded on its current collector surface; The intelligent control module is integrated with the battery management system and is used to activate the micro-energy control unit and external fire suppression module when a risk of thermal runaway is identified.

Images