Method for monitoring internal temperature of single cell of power battery

Abstract

The invention provides a method for monitoring the internal temperature of a single cell of a power battery, and relates to the technical field of battery monitoring, a monitoring method and a cell assembly are included, the cell assembly comprises a cell column, the cell column is wrapped with a cell lamination, the inner side of the cell lamination is provided with a miniature thin film thermocouple sensor, and the miniature thin film thermocouple sensor is arranged in the cell column. Aiming at the problems of insufficient precision, limited environmental adaptability, low intelligent degree and lack of thermal runaway prediction capability in monitoring of the internal temperature of a single cell of a power battery, a micro thin film thermocouple sensor array is embedded, an intelligent data processing flow and a multi-source data fusion design are adopted, and the thermal runaway prediction capability of the single cell of the power battery is improved. And the monitoring precision, reliability and automation level are obviously improved. Firstly, high-precision non-destructive detection and environmental adaptability are realized. The miniature thin film thermocouple sensor array is embedded between the positive and negative plates and the diaphragm, is attached to the non-coating area of the positive plate, and is in direct contact with the active material to construct a three-dimensional distributed temperature monitoring network.

Description

Technical Field

[0001] This invention relates to the field of battery monitoring technology, specifically a method for monitoring the internal temperature of a single power battery cell. Background Technology

[0002] As a core energy component in electric vehicles, energy storage power stations, and other fields, the safety and performance of power batteries directly affect the reliability and lifespan of the system. During charging and discharging, power batteries generate a large amount of heat due to internal resistance, chemical side reactions, and electrochemical reactions, leading to uneven temperature distribution within the cell. Localized high-temperature areas may trigger thermal runaway, causing performance degradation, capacity loss, and even serious safety accidents such as fires or explosions. According to industry research, during the service life of power batteries, users' priority requirements for internal temperature monitoring are, in order of importance: high-precision measurement, real-time performance, reliability, environmental adaptability, and prediction of thermal runaway risk. However, actual testing mainly relies on surface-mount sensors, infrared thermal imaging, or limited-point temperature measurement within the casing, resulting in significant discrepancies between performance and actual needs. Although internal temperature monitoring is one of the core requirements for battery safety management, current technologies are limited by insufficient accuracy of testing equipment, significant environmental interference, low reliability, and a lack of predictive capabilities, making it difficult to meet users' expectations for efficient and intelligent temperature management.

[0003] Modern power batteries (such as lithium-ion batteries and solid-state batteries) face higher demands for internal temperature management under high-rate charging and discharging, extreme environments (such as high or low temperatures), and long-term service, especially requiring dynamic monitoring and preventative maintenance during their service life. However, existing cell internal temperature monitoring technologies face the following key technical challenges:

[0004] Insufficient detection accuracy and non-destructive nature: Traditional surface-mount sensors (such as negative temperature coefficient thermistors, NTC) can only measure the temperature of the battery cell casing, and cannot obtain temperature data of the internal active areas. Positional errors are as high as ±20mm, and temperature errors are ±1.5℃. Infrared thermal imaging (such as Fluke Ti400), although non-contact detection, has a spatial resolution of only 10mm and a temperature accuracy of ±2℃, making it difficult to detect high temperatures in small areas (<0.05m) inside the battery cell. 2 This method cannot meet the requirements for high precision. In addition, surface temperature measurement may cause data distortion due to the thermal resistance of the casing, lagging behind the occurrence of thermal runaway.

[0005] Limited environmental adaptability: Existing detection methods are significantly affected by environmental factors (such as high temperature, low temperature, and electromagnetic interference). For example, infrared thermal imaging methods have a temperature error of ±2.5℃ and a false alarm rate as high as 15% in high-temperature environments (>40℃) or low-temperature environments (<-10℃). Surface-mount sensors and in-house temperature measuring devices lack environmental calibration mechanisms, making it difficult to adapt to seasonal changes (winter and summer) or extreme operating conditions (such as high-rate charging and discharging), thus limiting the reliability and applicability of the detection.

[0006] Low level of intelligence and automation: Current testing equipment is mostly single-function modules (such as thermocouples and infrared thermal imagers), lacking a unified control system architecture. Data is isolated and processing relies on manual analysis. For example, infrared thermal imaging requires professionals to interpret thermal images, and temperature measurement inside the casing requires manual sensor placement, resulting in long testing times (30-60 minutes per cell) and making automation impossible. Existing systems lack real-time data fusion, intelligent diagnostics, and risk prediction functions, requiring frequent manual operation by users and leading to low maintenance efficiency.

[0007] Therefore, a method for monitoring the internal temperature of a single power battery cell is needed to solve the above problems. Summary of the Invention

[0008] Technical problems to be solved

[0009] To address the shortcomings of existing technologies, this invention provides a method for monitoring the internal temperature of a single power battery cell, thus solving the problems mentioned in the background section.

[0010] Technical solution

[0011] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring the internal temperature of a single power battery cell, comprising a monitoring method and a cell assembly, wherein the cell assembly includes a cell column, the cell column is wrapped with cell laminations, and a micro-thin-film thermocouple sensor is installed inside the cell laminations; the monitoring method is applied to monitor the cell assembly and includes the following steps:

[0012] SP1: During the cell stacking and winding process, a micro thin-film thermocouple sensor array is embedded between the positive and negative electrode sheets through a roll-to-roll process. A temperature measurement node is set for every 5 layers of electrodes to build a three-dimensional distributed temperature monitoring network.

[0013] SP2: The micro-film thermocouple sensor is encapsulated with a corrosion-resistant flexible polyimide-alumina composite film to ensure that the resistance change rate is less than 2% after immersion in electrolyte for 240 hours and that the electrical continuity is maintained when the electrode charge-discharge deformation rate is greater than 10%.

[0014] SP3: Acquires microvolt-level voltage signals from miniature thin-film thermocouple sensors through a high-precision analog-to-digital conversion module, applies Kalman filtering algorithm to eliminate electromagnetic interference, achieves a temperature resolution of 0.1℃, and a sampling frequency of not less than 10Hz;

[0015] SP4: Based on real-time data of cell charging and discharging current, internal resistance and Joule heat, a temperature-current coupling calibration model is constructed to dynamically correct the measured values ​​of the micro thin-film thermocouple sensor and eliminate the local temperature rise error caused by electrochemical reaction heat.

[0016] SP5: Based on real-time temperature data from a three-dimensional distributed temperature monitoring network, it generates a temperature field distribution model inside the battery cell, predicts the risk of thermal runaway, and outputs early warning signals.

[0017] SP6: Based on the temperature field distribution model, dynamically adjust the thermal management strategy of the power battery pack, optimize the operating parameters of the cooling system, and improve thermal management performance;

[0018] SP7: By comparing real-time temperature data with historical temperature data, it identifies the aging trend of battery cell components, dynamically adjusts the temperature-current coupling calibration model parameters, and extends the service life of battery cell components.

[0019] SP8: Through periodic self-testing procedures, it detects the electrical connectivity and packaging integrity of the miniature thin-film thermocouple sensor array, generates a health status report, and optimizes the accuracy of the input data for the temperature field distribution model.

[0020] Preferably, in SP1, the miniature thin-film thermocouple sensor array has a size of 0.1mm × 0.5mm, is attached to the uncoated area of ​​the positive electrode edge, and is in direct contact with the active material without interfering with ion transport.

[0021] Preferably, in SP3, the Kalman filter algorithm optimizes signal processing by adaptively adjusting weights and combining the working status of the battery cell assembly, and the temperature measurement error is controlled within ±0.05℃.

[0022] Preferably, in SP4, the temperature-current coupling calibration model is trained using historical charge and discharge data through machine learning algorithms, and the calibration parameters are updated in real time to adapt to the temperature change characteristics of different operating conditions of the battery cell assembly.

[0023] Preferably, in SP5, the temperature field distribution model adopts the finite element analysis method, which combines the geometric structure of the battery cell assembly and the thermal conductivity of the material to simulate the three-dimensional temperature distribution inside the battery cell and calculate the probability of thermal runaway.

[0024] Preferably, in SP6, the thermal management strategy optimization is output through the temperature field distribution model, dynamically adjusting the coolant flow rate and cooling fan speed of the cooling system to control the temperature gradient of the battery cell assembly within 2℃ / cm.

[0025] Preferably, in SP7, the aging trend identification of the battery cell assembly is achieved by analyzing the time series characteristics of temperature data and combining them with the number of charge-discharge cycles to determine the degree of aging and adjust the temperature-current coupling calibration model parameters in stages.

[0026] Preferably, the monitoring method includes step SP9: based on the multi-node data of the three-dimensional distributed temperature monitoring network, a multi-point collaborative verification algorithm is applied to eliminate abnormal temperature measurement node data and improve the reliability of temperature monitoring by the micro-thin film thermocouple sensor.

[0027] Preferably, in SP8, the periodic self-test program detects the resistance change rate of the micro-thin film thermocouple sensor array by applying a weak test current, determines the package integrity, and generates a health status report.

[0028] Beneficial effects

[0029] This invention provides a method for monitoring the internal temperature of a single power battery cell. It has the following beneficial effects:

[0030] This invention addresses the problems of insufficient accuracy, limited environmental adaptability, low level of intelligence, and lack of thermal runaway prediction capability in the internal temperature monitoring of single-cell power batteries. By embedding a micro-thin-film thermocouple sensor array, employing an intelligent data processing workflow, and integrating multi-source data, it significantly improves monitoring accuracy, reliability, and automation. Firstly, high-precision non-destructive testing and environmental adaptability are achieved. The micro-thin-film thermocouple sensor array is embedded between the positive and negative electrodes and the separator, attached to the uncoated area of ​​the positive electrode, and in direct contact with the active material, constructing a three-dimensional distributed temperature monitoring network with a temperature resolution of 0.1 degrees Celsius and an error controlled within ±0.05 degrees Celsius, completely eliminating the blind spots of traditional surface temperature measurement. Simultaneously, the polyimide-alumina composite film encapsulation ensures the sensor's resistance to electrolyte corrosion (resistance change rate less than 2% after 240 hours) and deformation adaptability (maintaining electrical continuity when the deformation rate is higher than 10%), adapting to complex environments such as high and low temperatures and electromagnetic interference.

[0031] This invention employs a Kalman filter algorithm combined with dynamic adjustment signal processing based on the cell's operating state, along with a temperature-current coupling calibration model trained by machine learning, to calibrate temperature data in real time, adapting to high-rate charging and discharging conditions. The three-dimensional temperature field distribution model generated by finite element analysis, combined with a multi-point collaborative verification algorithm, automatically eliminates abnormal data and predicts thermal runaway risks, reducing the early warning response time to less than 1 second, a several-fold improvement over traditional methods (30-60 seconds). The battery management system integrates thermal management strategies, dynamically adjusting coolant flow rate and fan speed, reducing single-cell monitoring time to less than 5 minutes, eliminating the need for manual intervention and significantly improving maintenance efficiency.

[0032] This invention identifies cell aging trends and dynamically adjusts calibration parameters by comparing real-time and historical temperature data, extending cell lifespan by approximately 10-15%. A periodic self-test program monitors sensor status, generates health status reports, optimizes data accuracy, and reduces maintenance costs. This method is applicable to various scenarios such as electric vehicles and energy storage power stations, providing an efficient and intelligent solution for power battery safety management. Attached Figure Description

[0033] Figure 1 This is a flowchart of the monitoring process of the present invention;

[0034] Figure 2 This is a diagram illustrating the framework of the present invention;

[0035] Figure 3 This is a partially enlarged view of the present invention;

[0036] Figure 4 This is a schematic diagram of the internal structure of the battery cell assembly of the present invention.

[0037] Legend:

[0038] 1. Battery cell assembly; 2. Battery cell stack; 3. Miniature thin-film thermocouple sensor; 4. Battery cell column. Detailed Implementation

[0039] 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. Specific Implementation Example 1:

[0041] like Figures 1 to 4 As shown, this invention provides a method for monitoring the internal temperature of a single power battery cell, including a monitoring method and a cell assembly 1. The cell assembly 1 includes a cell post 4, the cell post 4 is wrapped with a cell stack 2, and a micro thin-film thermocouple sensor 3 is installed inside the cell stack 2. The monitoring method is applied to monitor the cell assembly 1 and includes the following steps:

[0042] SP1: During the winding process of cell stacking 2, a micro thin-film thermocouple sensor 3 array is embedded between the positive and negative electrode sheets through a roll-to-roll process. A temperature measurement node is set for every 5 layers of electrodes to build a three-dimensional distributed temperature monitoring network.

[0043] SP2: The micro film thermocouple sensor 3 is encapsulated in a corrosion-resistant and flexible manner using a polyimide-alumina composite film to ensure that the resistance change rate is less than 2% after immersion in electrolyte for 240 hours and that the electrical continuity is maintained when the electrode charge-discharge deformation rate is greater than 10%.

[0044] SP3: The microvolt-level voltage signal of the miniature thin-film thermocouple sensor 3 is acquired through a high-precision analog-to-digital conversion module, and electromagnetic interference is eliminated by applying a Kalman filter algorithm to achieve a temperature resolution of 0.1℃ and a sampling frequency of not less than 10Hz.

[0045] SP4: Based on real-time data of cell charging and discharging current, internal resistance and Joule heat, a temperature-current coupling calibration model is constructed to dynamically correct the measured value of the micro thin-film thermocouple sensor 3 and eliminate the local temperature rise error caused by electrochemical reaction heat.

[0046] SP5: Based on real-time temperature data from a three-dimensional distributed temperature monitoring network, it generates a temperature field distribution model inside the battery cell, predicts the risk of thermal runaway, and outputs early warning signals.

[0047] SP6: Based on the temperature field distribution model, dynamically adjust the thermal management strategy of the power battery pack, optimize the operating parameters of the cooling system, and improve thermal management performance;

[0048] SP7: By comparing real-time temperature data with historical temperature data, the aging trend of cell component 1 is identified, and the temperature-current coupling calibration model parameters are dynamically adjusted to extend the service life of cell component 1.

[0049] SP8: Through periodic self-testing procedures, it detects the electrical connectivity and packaging integrity of the miniature thin-film thermocouple sensor array 3, generates a health status report, and optimizes the accuracy of the input data for the temperature field distribution model.

[0050] In SP1, the array of miniature thin-film thermocouple sensors 3 has a size of 0.1mm × 0.5mm. It is attached to the uncoated area at the edge of the positive electrode and is in direct contact with the active material, without interfering with ion transport.

[0051] In SP3, the Kalman filter algorithm optimizes signal processing by adaptively adjusting weights and combining the working status of cell component 1, and the temperature measurement error is controlled within ±0.05℃.

[0052] In SP4, the temperature-current coupling calibration model is trained using machine learning algorithms based on historical charge and discharge data, and the calibration parameters are updated in real time to adapt to the temperature variation characteristics of different operating conditions of the battery cell assembly 1.

[0053] In SP5, the temperature field distribution model uses the finite element analysis method, combined with the geometry of the battery cell assembly 1 and the thermal conductivity of the material, to simulate the three-dimensional temperature distribution inside the battery cell and calculate the probability of thermal runaway.

[0054] In SP6, the thermal management strategy optimization is achieved by outputting the temperature field distribution model, which dynamically adjusts the coolant flow rate and cooling fan speed of the cooling system to control the temperature gradient of the battery cell assembly 1 within 2℃ / cm.

[0055] In SP7, the aging trend identification of cell component 1 is achieved by analyzing the time series characteristics of temperature data and combining the number of charge and discharge cycles to determine the degree of aging and adjust the temperature-current coupling calibration model parameters in stages.

[0056] The monitoring method includes step SP9: Based on the multi-node data of the three-dimensional distributed temperature monitoring network, a multi-point collaborative verification algorithm is applied to eliminate abnormal temperature measurement node data and improve the temperature monitoring reliability of the micro-thin film thermocouple sensor 3.

[0057] In SP8, the periodic self-test program detects the resistance change rate of the micro-thin film thermocouple sensor array 3 by applying a weak test current, judges the package integrity, and generates a health status report. Specific Implementation Example 2:

[0059] like Figures 1 to 4 As shown, the following supplements the content of Example 1:

[0060] This invention provides an operational mode for a method of monitoring the internal temperature of a single power battery cell. This method is applied to real-time, high-precision, and high-reliability temperature monitoring of battery cell assemblies, enabling early warning of thermal runaway and optimizing the thermal management performance of the power battery pack. The implementation steps and specific applications of this method are described in detail below.

[0061] In the manufacturing process of cell assembly 1, firstly, in the cell lamination 2 winding process, a micro-thin-film thermocouple sensor array 3 is embedded between the positive and negative electrode sheets using a roll-to-roll process. Specifically, a micro-thin-film thermocouple sensor 3 with dimensions of 0.1mm × 0.5mm is selected, with its thickness controlled at the micrometer level to minimize interference with the cell structure. The attachment process is completed during the electrode coating process, using precision automated equipment to accurately attach the sensor array to the uncoated area at the edge of the positive electrode sheet. The uncoated area refers to the region on the positive electrode sheet where no active material is coated, located in the blank foil portion at the electrode edge. High-precision bonding technology ensures tight contact between the sensor and the boundary of the positive electrode sheet's active material area. The contact method involves fixing the sensor to the positive electrode sheet surface using a thermally conductive adhesive. This adhesive has high thermal conductivity and electrical insulation, ensuring that the sensor can directly sense temperature changes in the active material area without interfering with lithium-ion transport between the electrode and the separator. A temperature measurement node is set every 5 electrode layers, with the node distribution following a three-dimensional grid design, forming a three-dimensional distributed temperature monitoring network covering the inside of the cell. The location of the temperature measurement nodes is pre-planned using computer-aided design software to ensure that the nodes are evenly distributed along the X, Y, and Z axes of the cell stack 2, covering the key heat-sensitive areas inside the cell assembly 1.

[0062] Subsequently, the miniature thin-film thermocouple sensor 3 underwent corrosion-resistant flexible encapsulation using a polyimide-alumina composite film as the encapsulation material. This composite film was prepared using vacuum deposition technology, uniformly depositing an alumina layer onto a polyimide substrate to form a composite film approximately 10 micrometers thick. The encapsulation process was carried out in a dust-free environment, using automated encapsulation equipment to cover the sensor surface, achieving a tight bond through a hot-pressing process. The encapsulated sensor maintained a resistance change rate of less than 2% after immersion in electrolyte for 240 hours, and its corrosion resistance was verified by periodically testing the resistance value. Simultaneously, the flexible nature of the polyimide material allows the sensor to adapt to the deformation caused by the volume expansion of the electrodes during charging and discharging. When the deformation rate exceeds 10%, the sensor still maintains electrical connectivity through flexible wires made of a highly ductile metal material, ensuring signal transmission stability under deformation.

[0063] Temperature signal acquisition is accomplished through a high-precision analog-to-digital converter module with 16-bit resolution, capable of capturing microvolt-level voltage signals generated by miniature thin-film thermocouple sensors. During acquisition, the signal is transmitted to the data processing unit, where a Kalman filter algorithm is applied to eliminate electromagnetic interference. The Kalman filter algorithm establishes a state-space model to estimate the true value of the sensor signal in real time, and adaptively adjusts the weights based on the operating state of the battery cell assembly (such as charging, discharging, or resting). Specifically, this is achieved by monitoring the charging and discharging current and voltage changes of the battery cell and dynamically adjusting the filter's gain parameters. For example, during high-rate charging and discharging, when electromagnetic interference is strong, the algorithm increases the filter weights to suppress noise; during low-rate charging or resting, the filter weights are reduced to improve signal response speed. Through this optimization, the temperature measurement error is controlled within ±0.05℃, and the sampling frequency is maintained above 10Hz, ensuring real-time performance.

[0064] Based on real-time data of charge / discharge current, internal resistance, and Joule heat of battery cell assembly 1, a temperature-current coupled calibration model is constructed to correct sensor measurements. The model is trained using machine learning algorithms (such as support vector regression) on historical charge / discharge data to extract the nonlinear relationship between current, internal resistance, and temperature changes. The training dataset includes current, internal resistance, and temperature data of the battery cell under different operating conditions (such as 0.5C, 1C, and 2C rates). After training, the model can update calibration parameters in real time. The specific calibration process is as follows: based on the real-time collected current and internal resistance data, the Joule heat (Q = I0) is calculated. 2 R), and combined with the empirical formula of electrochemical reaction heat, estimate the impact of local temperature rise on sensor measurement, and calibrate the original temperature data by outputting correction coefficients through the model to eliminate errors and adapt to the temperature change characteristics of battery cell assembly 1 under different operating conditions.

[0065] A temperature field distribution model of the internal structure of battery cell assembly 1 is generated using real-time temperature data from a three-dimensional distributed temperature monitoring network. The model employs finite element analysis (FEM), based on the geometry and thermal conductivity of the materials used in battery cell assembly 1. The precise dimensions of the battery cell stack are obtained through 3D scanning technology, and the thermal conductivity data is sourced from a material database, including the thermophysical parameters of the positive electrode, negative electrode, separator, and electrolyte. FEM divides the battery cell into multiple mesh elements, and, combined with temperature data from sensor nodes, calculates the three-dimensional temperature distribution within the cell. By analyzing the temperature distribution gradient and anomalous hotspots, the risk of thermal runaway is predicted. When the temperature in a certain area exceeds a preset threshold (e.g., 80°C) or the temperature rise rate exceeds 2°C / s, an early warning signal is generated and transmitted to the battery management system, triggering a protection mechanism.

[0066] Based on a temperature field distribution model, the thermal management strategy of the power battery pack is dynamically adjusted. The model outputs temperature gradient data inside the battery cell, and the thermal management control unit adjusts the operating parameters of the cooling system according to this data. For example, when a certain area of ​​battery cell assembly 1 is detected to be too hot, the coolant flow rate is increased to 1.5L / min or the cooling fan speed is increased to 3000rpm to reduce the temperature of the local high-temperature area and control the cell temperature gradient within 2℃ / cm to ensure temperature uniformity.

[0067] By comparing real-time temperature data with historical temperature data, the aging trend of cell assembly 1 is identified. Historical data is stored in the battery management system database, including temperature profiles for each charge-discharge cycle. The analysis method involves extracting time-series features of the temperature data, such as average temperature, peak temperature, and temperature rise rate, and combining this with the number of charge-discharge cycles to determine the degree of aging. Based on the aging degree classification (e.g., mild, moderate, severe aging), the parameters of the temperature-current coupling calibration model are dynamically adjusted, for example, by increasing the calibration coefficient to compensate for the increase in internal resistance caused by aging, thereby extending the cell's lifespan.

[0068] A periodic self-test procedure is used to check the electrical connectivity and package integrity of the miniature thin-film thermocouple sensor array. The self-test procedure starts after every 100 charge-discharge cycles, measuring the resistance change rate of the sensor array by applying a weak test current (e.g., 10 μA). If the resistance change rate exceeds 2%, it indicates a possible micro-crack in the package, generates a health status report, marks abnormal sensor nodes, and optimizes the input data accuracy of the temperature field distribution model.

[0069] Based on multi-node data from a three-dimensional distributed temperature monitoring network, a multi-point collaborative verification algorithm is applied to improve monitoring reliability. The algorithm compares temperature data from adjacent monitoring nodes, identifies outliers (such as node data with deviations exceeding 3℃), and removes abnormal data to ensure the accuracy of the temperature field distribution model. Specific Implementation Example 3:

[0071] like Figures 1 to 4 As shown, based on the above method for monitoring the internal temperature of a single power battery cell, the following use cases are provided for different application scenarios to demonstrate the implementation and effect of this method in actual operation.

[0072] Case 1: Electric vehicle driving at high speed:

[0073] In high-speed driving scenarios of electric vehicles, the power battery pack needs to support high-rate discharge, causing a rapid rise in internal temperature of the battery cells and posing a risk of thermal runaway. When applying this monitoring method, during the battery cell manufacturing process, a miniature thin-film thermocouple sensor array measuring 0.1mm × 0.5mm is embedded between the positive and negative electrode sheets using a roll-to-roll process. Temperature measurement nodes are set every five electrode layers, forming a three-dimensional distributed temperature monitoring network. The sensor is attached to the uncoated area at the edge of the positive electrode sheet, using thermally conductive adhesive to ensure tight contact with the active material boundary, ensuring that temperature changes are sensed without interfering with ion transport. The sensor is encapsulated using a polyimide-alumina composite film, which is resistant to electrolyte corrosion and can accommodate electrode deformation of over 10%.

[0074] When the vehicle is traveling at a high speed of 120 km / h, the battery cell discharges at a 2C rate, generating significant Joule heat. Temperature signals are acquired via a high-precision analog-to-digital converter. A Kalman filter algorithm, combined with the high-rate discharge state of the battery cell, increases the filter weight to suppress electromagnetic interference, maintaining a temperature resolution of 0.1℃, a sampling frequency of 10Hz, and an error controlled within ±0.05℃. A temperature-current coupling calibration model calculates Joule heat and calibrates sensor measurements based on real-time current and internal resistance data, eliminating errors caused by electrochemical reaction heat. Real-time temperature data generates a temperature field distribution model, simulating the three-dimensional temperature distribution inside the battery cell through finite element analysis. When the temperature in a certain area of ​​the battery cell reaches 78℃, approaching the 80℃ threshold, the system outputs an early warning signal, notifying the battery management system to reduce the discharge rate.

[0075] Meanwhile, the temperature field distribution model showed a local temperature gradient of 3℃ / cm. The thermal management control unit dynamically adjusted the cooling system, increasing the coolant flow rate to 1.8L / min and the cooling fan speed to 3500rpm, reducing the temperature gradient to below 2℃ / cm within 2 minutes to prevent thermal runaway. During long-term operation, by comparing real-time and historical temperature data, the aging trend of the battery cells was identified, and the calibration model parameters were adjusted to extend the cell life. A periodic self-test program checks the sensor array resistance change every 100 charge-discharge cycles, generating a health status report to ensure data accuracy. A multi-point collaborative verification algorithm eliminates abnormal node data with deviations exceeding 3℃, ensuring monitoring reliability.

[0076] Case 2: Frequent Start-Stop Scenario of Electric Buses

[0077] Electric buses frequently start and stop in urban traffic conditions, causing the battery cells to undergo multiple short-duration high-rate charge and discharge cycles, resulting in significant temperature fluctuations. This method constructs a three-dimensional temperature monitoring network by embedding a micro-thin-film thermocouple sensor array during the battery cell stacking and winding process, with temperature measurement nodes set every five electrode layers. The sensors are attached to the uncoated area of ​​the positive electrode, in contact with the active material, and the encapsulation material ensures corrosion resistance and flexibility. During bus operation, the battery cells switch between 1C and 1.5C rates. The temperature signal acquisition module captures microvolt-level signals at a frequency of 10Hz. The Kalman filter algorithm dynamically adjusts the weights according to the start-stop status, enhancing noise immunity during high-rate discharge and improving response speed during standby, while maintaining an error of ±0.05℃.

[0078] The temperature-current coupling calibration model utilizes machine learning algorithms to update parameters in real time based on historical charge-discharge data, adapting to the temperature variations caused by frequent start-stop cycles. The temperature field distribution model simulates temperature distribution through finite element analysis, combined with cell geometry and thermal conductivity. When the temperature rise rate at a certain node reaches 2.1℃ / s, the system generates an early warning signal, triggering the battery management system to limit charging power. Based on the model output, the thermal management strategy adjusts the coolant flow rate to 1.2L / min and the fan speed to 2500rpm, controlling the temperature gradient within 2℃ / cm. Periodic self-checks detect the integrity of the sensor packaging, eliminating abnormal node data to ensure model accuracy. During long-term operation, the temperature time series characteristics and cycle count are analyzed to identify mild aging, and calibration parameters are adjusted accordingly to extend cell lifespan.

[0079] Case 3: Low-rate operation scenario of energy storage power station:

[0080] In energy storage power stations, battery cells operate at a low rate of 0.5C, where temperature changes are relatively slow, but long-term stable monitoring is required to ensure safety. A miniature thin-film thermocouple sensor array is embedded in the battery cell stack, with temperature measurement nodes set every five electrode layers, attached to the uncoated area of ​​the positive electrode, and encapsulated for corrosion resistance and flexibility. Temperature signal acquisition operates at a frequency of 10Hz. A Kalman filter algorithm reduces the filtering weight at low rates to improve signal response speed, with errors controlled within ±0.05℃. A temperature-current coupling calibration model is trained based on low-rate data, calibrating the temperature in real time to eliminate the thermal effects of minute electrochemical reactions.

[0081] A temperature field distribution model simulates the temperature distribution of the battery cells. When the temperature in a certain area reaches 75℃, an early warning signal is output to prompt maintenance personnel to check. The thermal management strategy adjusts the coolant flow rate to 0.8L / min to maintain a temperature gradient below 2℃ / cm. Periodic self-checks are performed every 100 cycles, detecting the sensor resistance change rate and generating a health status report. A multi-point collaborative verification algorithm eliminates abnormal data to ensure data reliability. By comparing historical temperature data, the aging trend of the battery cells is identified, and calibration parameters are dynamically adjusted to extend their service life, making it suitable for long-term stable operation of energy storage power stations.

[0082] Case 4: Extreme Low Temperature Charging Scenarios

[0083] In environments as low as -10℃, rapid charging of electric vehicles may lead to abnormal localized temperature rises in the battery cells. This method constructs a three-dimensional monitoring network by embedding a micro-thin-film thermocouple sensor array. The sensors are attached to the uncoated area of ​​the positive electrode and encapsulated to resist low-temperature electrolyte corrosion. During charging, the battery cell operates at a 1C rate, and the temperature signal acquisition module captures data at a frequency of 10Hz. The Kalman filter algorithm, combined with the low-temperature charging state, optimizes the filtering parameters, controlling the error within ±0.05℃. The temperature-current coupling calibration model is trained based on low-temperature data, calibrating the temperature values ​​to adapt to low-temperature characteristics.

[0084] The temperature field distribution model detected a localized cell temperature of 65°C at a rate of 1.8°C / s, generating an early warning signal. The battery management system then reduced the charging current to 0.8C. The thermal management strategy increased the coolant flow rate to 1.5L / min, controlling the temperature gradient to within 2°C / cm. A self-test program checked sensor performance, eliminating abnormal data to ensure model accuracy. Temperature data and charging cycles were analyzed to identify aging trends, adjust calibration parameters, and ensure safe low-temperature charging.

[0085] The above examples demonstrate the flexible application of this method in different scenarios, ensuring cell safety and performance through high-precision monitoring, dynamic calibration, and thermal management optimization. Specific Implementation Example 4:

[0087] like Figures 1 to 4 As shown, based on the above content, the following experimental data is provided:

[0088] Comparison of monitoring methods:

[0089]

[0090] Performance improvement comparison:

[0091]

[0092]

[0093] Cost vs. Feasibility Comparison

[0094]

[0095] Comparison of improvements to technical defects:

[0096]

[0097] Based on the comparison in the above four aspects, this patented solution has the following significant advantages over traditional solutions:

[0098] 1. Enhanced safety: Internal temperature monitoring transforms thermal runaway warning time from "post-event response" to "pre-event prevention," avoiding battery pack-level thermal diffusion;

[0099] 2. Precise monitoring: The internal temperature monitoring error is < ±0.5℃, enabling monitoring of the internal temperature field of the battery cell;

[0100] 3. High reliability: Utilizing encapsulation technology, it boasts high resistance to corrosion and interference;

[0101] 4. Expanded application scenarios: Supports mainstream high-rate charge and discharge conditions in the industry, which can broaden the application boundaries of batteries. Specific Implementation Example 5:

[0103] like Figures 1 to 4 As shown, the following are supplementary implementation details of the monitoring methods and devices, including improved hardware parameters, algorithm implementation, anomaly handling, maintenance procedures, and compliance specifications.

[0104] Hardware Parameters: The hardware system of this monitoring method consists of the following core components: 1. Miniature Thin-Film Thermocouple Sensor Array: Utilizing K-type thin-film thermocouples, measuring 0.1 mm × 0.5 mm with a thickness of 5 micrometers, the temperature measurement range is from -40°C to 200°C. Each degree Celsius temperature change generates a 41 microvolt voltage signal, with a response time of less than 10 milliseconds. A temperature measurement node is set for every 5 electrode layers, with 20 to 50 nodes per cell assembly, spaced 5 mm apart. The sensor encapsulation uses a polyimide-alumina composite film with a thickness of 10 micrometers, capable of maintaining a resistance change rate of less than 2% after immersion in electrolyte for 240 hours, and maintaining electrical connection even when the electrode deformation rate exceeds 10%. 2. High-Precision Analog-to-Digital Conversion Module: Model ADS1256, featuring 16-bit resolution, a sampling frequency of 10 times per second, an input voltage range of 0 to 5 millivolts, a noise level of less than 1 microvolt, and support for 4-channel input, allowing simultaneous acquisition of multiple sensor signals. 3. Data Processing Unit: Employs an STM32F429 microcontroller with a 180 MHz clock speed, 512 KB of memory, and 2 MB of storage. It runs signal processing and analysis algorithms and communicates with the battery management system via a CAN bus. 4. Thermally Conductive Adhesive: With a thermal conductivity of 2.5 Kelvin per meter, an electrical insulation strength greater than 10 kV per millimeter, and a thickness of 20 micrometers, it is used to fix the sensor to the uncoated area of ​​the positive electrode. 5. Flexible Conductor: Utilizes 0.05 mm diameter nickel-based alloy conductors with a ductility exceeding 15%. Connects the sensor to the analog-to-digital converter module, ensuring stable signal transmission under deformation. 6. Battery Management System: Integrates a thermal management control unit equipped with a quad-core Cortex-A7 processor. It performs temperature distribution analysis and thermal management control, controlling the coolant pump (flow rate range 0.5 to 2 liters per minute) and the fan (speed range 1000 to 4000 rpm). The entire hardware system consumes less than 5 watts, operates at a voltage of 3.3 to 5 volts, is compatible with ambient temperatures ranging from -40 degrees Celsius to 85 degrees Celsius, and has an IP67 protection rating, making it suitable for electric vehicles, energy storage power stations, and other scenarios.

[0105] Algorithm Implementation: This method involves four algorithms: Kalman filtering, temperature-current coupling calibration, temperature field distribution analysis, and multi-point collaborative verification. The computational logic is described below. 1. The Kalman filtering algorithm is used to eliminate electromagnetic interference. The input is the sensor voltage signal acquired by the analog-to-digital converter module, and the output is the filtered temperature value. The algorithm establishes a mathematical model to predict the current temperature and corrects it based on the actual measured value. During prediction, it assumes the current temperature is close to the previous temperature and adds a small amount of random fluctuation; during correction, it dynamically adjusts the weight of the prediction result based on the actual voltage signal. At high charging / discharging rates (e.g., current exceeding 100 amps), electromagnetic interference is strong, so the algorithm increases the confidence level of the predicted value to suppress noise; at low rates or in a static state (current below 10 amps), it increases the confidence level of the measured value to improve response speed, ultimately controlling the output temperature error to ±0.05 degrees Celsius. 2. The temperature-current coupling calibration algorithm calibrates the sensor temperature. The input is the filtered temperature, real-time current, and internal resistance, and the output is the calibrated temperature value. The algorithm first calculates Joule heat (current squared multiplied by internal resistance) based on current and internal resistance, then combines empirical data on electrochemical reaction heat to estimate the impact of heat on temperature measurement. Using a support vector regression model, trained on historical charge / discharge data (including current, internal resistance, and temperature at different rates such as 0.5C, 1C, and 2C), it learns the relationship between current, internal resistance, and temperature deviation. In real-time operation, the model calculates a temperature correction value based on the current current and internal resistance, subtracts the error caused by heat, and outputs a calibrated temperature to adapt to different operating conditions. 3. The temperature field distribution algorithm generates a three-dimensional temperature distribution inside the battery cell. Inputs include the calibrated temperature of each node, the cell's geometric dimensions (80 mm long, 50 mm wide, 10 mm thick), and the thermal conductivity of the materials (positive electrode, negative electrode, separator, etc.). The algorithm divides the cell into 1000 grid cells, analyzes heat conduction within the cell, and calculates the temperature value of each grid point based on the sensor node temperatures. When the temperature in a certain area exceeds 80 degrees Celsius or the temperature rise rate exceeds 2 degrees Celsius per second, an early warning signal is output to the battery management system. 4. The multi-point collaborative verification algorithm verifies the reliability of the temperature of each node. The input is the calibration temperature of all nodes. The temperature difference between adjacent nodes is compared. If the difference exceeds 3 degrees Celsius, it is marked as abnormal and removed. The temperature distribution is then recalculated until there are no abnormal nodes.

[0106] Anomaly Handling: The following strategies are implemented to address potential anomalies: 1. Sensor Failure: If the self-test program detects a resistance change rate exceeding 2% at a node, it is marked as a failure. The battery management system reconstructs the temperature distribution using temperature data from adjacent nodes through interpolation, with the error controlled within 0.2 degrees Celsius. If more than 20% of nodes fail, the system issues a maintenance alarm, prompting replacement of the cell assembly. 2. Analog-to-Digital Conversion Module Signal Loss: If there is no signal after 5 consecutive samplings, the system restarts the module. If there is still no signal, historical temperature data is used to predict the current temperature for no more than 10 seconds, with an error of less than 1 degree Celsius. A fault log is also recorded. 3. Algorithm Processing Anomalies: If abnormal input data (e.g., current exceeding 1000 amps) causes an algorithm error, the system switches to a backup linear calibration method. The temperature deviation is simply estimated based on the current and internal resistance, and the coolant flow rate is limited to no more than 2 liters per minute, and the fan speed to no more than 4000 revolutions per minute. 4. Thermal runaway warning false alarm: If the temperature returns to normal (below 75 degrees Celsius and the rate of increase is less than 1 degree Celsius per second) within 10 seconds after the warning is triggered, the system cancels the warning, records it as a false alarm, and adjusts the warning threshold. 5. Communication interruption: If communication between the battery management system and the thermal management control unit is interrupted for more than 1 second, the control unit enters default mode, setting the coolant flow rate to 1 liter per minute and the fan speed to 2000 revolutions per minute, until communication is restored.

[0107] Maintenance Process: The maintenance process ensures long-term stable system operation. 1. Regular Self-Check: Every 100 charge-discharge cycles, a self-check program is run, applying a 10 microamp test current to detect the sensor resistance change rate, generating a health status report, and marking abnormal nodes. 2. Data Backup: Temperature, current, and internal resistance data are backed up to the cloud monthly via the battery management system, with a storage capacity of at least one year for aging trend analysis. 3. Sensor Replacement: If the health status report indicates a sensor failure, technicians disassemble the cell assembly in a cleanroom environment, remove the failed sensor using specialized tools, install a new sensor, and repackage it, testing for a resistance change rate below 2%. 4. System Calibration: The analog-to-digital conversion module and data processing unit are calibrated every 6 months. Sensor output is verified using a standard temperature source (accuracy ±0.01 degrees Celsius), and calibration parameters are updated. 5. Cooling System Check: The coolant pump and fan operation status are checked every 3 months to ensure flow rate and speed are within specified ranges, and fan dust is cleaned. 6. Fault Recording and Analysis: All abnormal events are recorded in the battery management system log. Technicians analyze the logs quarterly to optimize algorithm parameters and warning thresholds.

[0108] Compliance Statement: This method complies with the following standards and regulations: 1. International Electrotechnical Commission Standard (IEC 62660-3): Meets the safety performance requirements of power batteries; sensor packaging is resistant to electrolyte corrosion; system protection level is IP67; adaptable to extreme environments. 2. International Organization for Standardization Standard (ISO 26262): Functional safety reaches ASIL-C level; anomaly handling mechanism ensures system reliability; early warning signals reduce the risk of thermal runaway. 3. United Nations Economic Commission for Europe Regulation (UN ECE R100): Meets the safety requirements of electric vehicle batteries; hardware system passes vibration, shock, and temperature cycling tests. 4. Chinese National Standard (GB / T 31467.3): Meets the testing specifications for lithium-ion power batteries; temperature measurement accuracy and thermal management performance meet standards. 5. Environmental Protection Requirements: Sensors and packaging materials are lead-free and mercury-free, comply with the EU RoHS Directive; waste battery cells and components are recycled according to hazardous waste disposal regulations. This method has been verified by a third-party testing agency and obtained relevant certifications, ensuring compliant application in the electric vehicle and energy storage fields.

[0109] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a reference structure" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0110] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring the internal temperature of a power battery cell, comprising a monitoring method and a cell assembly (1), characterized in that: The electric core assembly (1) contains an electric core column (4), the outside of the electric core column (4) is wrapped with an electric core stack (2), a micro thin film thermocouple sensor (3) is installed inside the electric core stack (2), the monitoring method is a method applied to monitoring the electric core assembly (1), comprising the following steps: SP1: during the winding process of the electric core stack (2), embed the micro thin film thermocouple sensor (3) array between the positive electrode sheet and the negative electrode sheet through a roll-to-roll process, set a temperature measurement node every 5 layers of electrodes, and construct a three-dimensional distributed temperature monitoring network; SP2: adopt a polyimide-aluminum oxide composite film to flexibly encapsulate the micro thin film thermocouple sensor (3) to resist corrosion, ensure that the resistance change rate is less than 2% after 240 hours of electrolyte immersion, and maintain electrical connectivity when the electrode charge-discharge deformation rate is higher than 10%; SP3: collect the micro-volt level voltage signal of the micro thin film thermocouple sensor (3) through a high-precision analog-to-digital conversion module, apply Kalman filtering algorithm to eliminate electromagnetic interference, achieve temperature resolution of 0.1℃, and sampling frequency higher than 10Hz; SP4: based on real-time data of electric core charging and discharging current, internal resistance and Joule heat, construct a temperature-current coupling calibration model, dynamically correct the measurement value of the micro thin film thermocouple sensor (3), and eliminate local temperature rise error caused by electrochemical reaction heat; SP5: generate an electric core internal temperature field distribution model according to real-time temperature data of the three-dimensional distributed temperature monitoring network, predict thermal runaway risk, and output early warning signals; SP6: based on the temperature field distribution model, dynamically adjust the thermal management strategy of the power battery pack, optimize the cooling system operating parameters, and improve the thermal management performance; SP7: compare real-time temperature data with historical temperature data to identify the aging trend of the electric core assembly (1), dynamically adjust the temperature-current coupling calibration model parameters, and prolong the service life of the electric core assembly (1); SP8: through periodic self-checking program, detect the electrical connectivity and packaging integrity of the micro thin film thermocouple sensor (3) array, generate a health status report, and optimize the temperature field distribution model input data precision.

2. The method of claim 1, wherein: In the SP1, the micro thin film thermocouple sensor (3) array has a size of 0.1mmx0.5mm, is attached to the edge of the positive electrode sheet in the non-coated area, and directly contacts with the active material without interfering with ion transmission.

3. The method of claim 1, wherein: In the SP3, the Kalman filtering algorithm adjusts the weight adaptively, optimizes signal processing in combination with the working state of the electric core assembly (1), and controls the temperature measurement error within ±0.05℃.

4. The method of claim 1, wherein: In the SP4, the temperature-current coupling calibration model trains historical charging and discharging data through a machine learning algorithm, updates the calibration parameters in real time, and adapts to the temperature variation characteristics of the electric core assembly (1) under different working conditions.

5. The method of claim 1, wherein: In the SP5, the temperature field distribution model adopts a finite element analysis method, simulates the three-dimensional temperature distribution inside the electric core assembly (1) in combination with the geometric structure and material thermal conductivity of the electric core assembly (1), and calculates the thermal runaway occurrence probability.

6. The method of claim 1, wherein: In the SP6, the thermal management strategy is optimized by a temperature field distribution model, and the flow rate of the cooling liquid and the rotation speed of the cooling fan are dynamically adjusted to control the temperature gradient of the battery cell assembly (1) to be within 2 ℃ / cm.

7. The method of claim 1, wherein: In the SP7, the aging trend of the battery cell assembly (1) is identified by analyzing the time series characteristics of the temperature data, combining the number of charge and discharge cycles, determining the aging degree, and adjusting the temperature-current coupling calibration model parameters in stages.

8. The method of claim 1, wherein: The monitoring method includes the step SP9: based on the multi-node data of the three-dimensional distributed temperature monitoring network, a multi-point cooperative verification algorithm is applied to eliminate abnormal temperature measurement node data and improve the temperature monitoring reliability of the micro thin film thermocouple sensor (3).

9. The method of claim 1, wherein: In the SP8, the periodic self-checking program detects the resistance change rate of the micro thin film thermocouple sensor (3) array by applying a weak test current, judges the packaging integrity, and generates a health status report.

Images