A method for diagnosing short-circuit faults in solid-state batteries based on internal sensing
By monitoring internal strain and electrochemical signals of the battery through embedded sensors, and combining filtering and differential analysis, the problem of difficulty in diagnosing internal short circuits in solid-state batteries by traditional systems is solved, realizing real-time and accurate fault diagnosis and safety response, and improving the safety and reliability of the battery system.
Patent Information
- Application Number
- CN202510996942.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-18
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2045-07-18
AI Technical Summary
Traditional battery management systems struggle to accurately diagnose early localized short-circuit faults within solid-state batteries in real time, posing a safety hazard.
By using embedded sensors to monitor the internal strain and electrochemical signals of the battery in real time, and combining moving average filtering, differential method and multi-level threshold judgment, the strain differential curve can be analyzed, diagnostic results can be generated and safety response strategies can be triggered.
It enables real-time and accurate diagnosis of internal short-circuit faults in solid-state batteries, improving the safety and reliability of battery systems and reducing economic losses caused by faults.
Smart Images

Figure CN120686129B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of solid-state battery technology, specifically relating to a method for diagnosing short-circuit faults in solid-state batteries based on internal sensing. Background Technology
[0002] Solid-state batteries, which use solid electrolytes instead of the flammable organic electrolytes in traditional lithium-ion batteries, are considered a next-generation energy storage technology with high safety and high energy density. However, during long-term charge-discharge cycles, the volume expansion and contraction of electrode materials in solid-state batteries can generate significant mechanical stress at the electrode-electrolyte interface. This stress can lead to cracks in the solid electrolyte or electrode materials, and the propagation of these cracks can further provide pathways for the growth of lithium dendrites, ultimately causing internal short circuits.
[0003] Traditional battery management systems (BMS) diagnose faults by monitoring external parameters such as voltage and temperature. However, for localized, early-stage short circuits occurring inside solid-state batteries, changes in external parameters often exhibit significant lag, making effective early warning difficult and posing a safety hazard. Therefore, developing a method for real-time and accurate diagnosis of internal short circuits in solid-state batteries is crucial. Summary of the Invention
[0004] In order to address the shortcomings of existing technologies in solid-state battery fault diagnosis, this invention aims to provide a solid-state battery short-circuit fault diagnosis method based on internal sensing. By directly monitoring key physical quantities inside the battery, short-circuit faults can be diagnosed, significantly improving the safety of solid-state battery systems.
[0005] To solve the technical problem, the technical solution of the present invention is as follows:
[0006] A method for diagnosing short-circuit faults in solid-state batteries based on internal sensing, the method comprising:
[0007] S1: Acquire real-time strain data inside the battery from embedded sensors, synchronously record electrochemical signals, and generate raw strain data sequences and electrochemical auxiliary information;
[0008] S2: Based on the original strain data sequence obtained in S1, the original data is filtered by moving average to remove noise introduced by electromagnetic interference and mechanical vibration, and a smooth strain-time curve is constructed to obtain a smoothed strain data sequence, which provides a clean signal for subsequent differential calculations.
[0009] S3: Input the smoothed strain data sequence, use the finite difference method to calculate the first derivative of strain with respect to time, thereby obtaining the strain differential curve, which is used to provide information about the dynamic changes of internal strain;
[0010] S4: Analyze the strain differential curve g(t), which is expressed by the following formula:
[0011]
[0012] Where ε(t) is the strain-time data sequence after smoothing and filtering in step S2, t is time, and g(t) is the first derivative of strain with respect to time, i.e. the strain differential curve;
[0013] During normal battery operation, when the accumulation of mechanical stress inside the battery causes instantaneous, minute cracks in the electrode or electrolyte materials, it leads to a rapid change in local strain. This abrupt change is manifested in g(t) as one or more characteristic peaks P. For each detected valid characteristic peak P, its key characteristic parameters are extracted, including peak amplitude and occurrence time; peak amplitude A p The vertex value of this characteristic peak is represented as:
[0014]
[0015] Among them, T p It is the time interval during which the characteristic peak lasts, and the time t during which it appears. p Peak amplitude A p The point in time is represented as:
[0016]
[0017] Each extracted set of feature parameters is combined into a peak data point P. data , in the form of (t p A p );
[0018] S5: Input the peak data P data And the preset multi-level diagnostic thresholds will extract the peak amplitude A. p The system compares and analyzes various thresholds to determine whether a fault alarm condition has been triggered, generates diagnostic results, and indicates the battery status and the corresponding alarm signal level.
[0019] Furthermore, after step S5, the method further includes:
[0020] Input the diagnostic results and the current operating status of the battery, trigger the corresponding system control strategy based on the diagnostic results, issue control commands, generate fault reports, and provide safety measure suggestions to ensure battery safety and realize a diagnosis-response closed loop.
[0021] Furthermore, in step S1, the internal sensor is a fiber Bragg grating sensor or a thin-film pressure sensor.
[0022] Furthermore, in step S1, when a fiber Bragg grating (FBG) sensor is used, the center wavelength shift of the FBG is monitored in real time by a high-precision fiber Bragg grating demodulator, and it is precisely demodulated into strain data inside the battery according to the pre-calibrated strain sensitivity coefficient, thereby providing a high-quality raw signal for subsequent differential calculations and fault diagnosis.
[0023] Furthermore, in step S4, the characteristic peak of the strain differential curve originates from the sudden change in local strain caused by instantaneous cracking of the internal material of the battery, and its peak amplitude characterizes the severity of the short circuit caused by the cracking event.
[0024] Furthermore, in step S5, the method for constructing the multi-level diagnostic threshold includes:
[0025] S501: Multiple solid-state battery samples were subjected to charge-discharge cycle tests at different charge rates, and their voltage and internal strain were monitored simultaneously to confirm the occurrence of short-circuit faults.
[0026] S502: Monitor and screen out cyclic data where the voltage curve shows an abnormal drop, calculate the corresponding strain differential curve, and extract the characteristic peaks related to the voltage anomaly as the basis for short circuit diagnosis.
[0027] S503: For initial fault samples, perform statistical distribution analysis on their strain differential characteristic peaks, and select the lower limit of their confidence interval as the first-level threshold.
[0028] S504: For severe fault samples, perform statistical distribution analysis on their strain differential characteristic peaks and select the lower limit of their confidence interval as the secondary threshold.
[0029] Compared with the prior art, the advantages of the present invention are as follows:
[0030] By monitoring the internal strain changes of solid-state batteries in real time, potential short-circuit faults can be detected promptly, significantly improving the safety of the battery system. Furthermore, the combination of differential curve analysis and multi-level threshold judgment makes the fault diagnosis process more accurate, helping to formulate reasonable safety response strategies and reduce economic losses caused by faults. Compared with traditional detection methods, this method not only provides detailed fault evolution information but also enables online monitoring, significantly improving the safety and reliability of solid-state batteries in various application scenarios. Attached Figure Description
[0031] Figure 1 The main flowchart of a solid-state battery short-circuit fault diagnosis method based on internal sensing according to the present invention. Detailed Implementation
[0032] The specific implementation of the present invention is described below with reference to embodiments:
[0033] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to complement the content disclosed in the specification for those skilled in the art to understand and read, and are not intended to limit the conditions under which the present invention can be implemented. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.
[0034] Furthermore, the terms such as "upper," "lower," "left," "right," "middle," and "one" used in this specification are merely for clarity of description and are not intended to limit the scope of the invention. Any changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.
[0035] Example 1:
[0036] Reference Figure 1 The present invention proposes a method for diagnosing short-circuit faults in solid-state batteries based on internal sensing, comprising the following steps:
[0037] S1. In the solid-state battery manufacturing process, internal sensors such as fiber optic sensors and thin-film pressure sensors are embedded in key locations such as the positive electrode, negative electrode, or solid electrolyte layer to monitor internal strain in situ and simultaneously acquire external electrochemical signals.
[0038] The steps involved in sensor implantation and battery encapsulation include:
[0039] S101. The sensor portion is pre-packaged, and corrosion-resistant and insulating materials are used to protect the sensor and its leads to prevent damage during battery manufacturing.
[0040] S102. Implant a sensor at the center of the battery core or cell stack to ensure that the sensor is in close contact with the cell material and to ensure the accuracy of strain measurement.
[0041] S103. The core or cell stack with the sensor is encapsulated into the battery casing.
[0042] S104. The battery is sealed using methods such as vacuum sealing, while ensuring the airtightness of the sensor lead-out position and the airtightness of the battery.
[0043] To ensure the accuracy of monitoring and the reliability of data, a fiber Bragg grating (FBG) sensor is preferably used in this embodiment. A high-precision FBG demodulator monitors the center wavelength shift of the FBG in real time and, based on a pre-calibrated strain sensitivity coefficient, precisely demodulates it into strain data within the battery, thereby providing high-quality raw signals for subsequent differential calculations and fault diagnosis.
[0044] S2. Preprocess the raw strain monitoring data, and obtain the differential curve characterizing the strain change rate through smoothing filtering and differential calculation;
[0045] To accurately capture the subtle strain response caused by changes in internal state, in this embodiment, in S1, the acquired raw strain data is first smoothed using a moving average filtering method to effectively eliminate or reduce random noise introduced by factors such as electromagnetic interference and mechanical vibration. Subsequently, based on the smoothed strain time series, its first derivative with respect to time is calculated to generate a strain differential curve, which reflects the rate of strain change.
[0046] Methods for obtaining strain differential curves include:
[0047] S11. Set the sliding window size and perform a moving average calculation on the original strain data sequence;
[0048] S12. Connect the filtered data points sequentially to reconstruct a smooth strain-time curve;
[0049] S13. The first derivative of each point on the smoothed strain curve is approximately calculated using the finite difference method.
[0050] S14. Based on the calculated first derivative value, plot the curve of strain differential as a function of time.
[0051] In the data processing process, in this embodiment, in S11, the sliding window setting is dynamically adjusted according to the actual data noise level and signal change frequency.
[0052] S3. When a crack occurs inside the battery, the sudden change in strain causes a characteristic peak to appear on the differential curve. The amplitude of this peak is compared with a preset diagnostic threshold.
[0053] S4. If the peak amplitude exceeds the preset first-level threshold, it is determined that an early short circuit has occurred and a warning of the attention level is issued; if the peak amplitude continues to increase and exceeds the second-level threshold, it is determined that a short circuit is developing and a warning of the danger level is issued, and intervention measures such as limiting the charging and discharging current are recommended.
[0054] This invention enables the diagnosis of internal short-circuit faults in solid-state batteries, providing key technical support for the development of safer and more reliable next-generation energy storage systems.
[0055] Example 2:
[0056] The specific steps in this solid-state battery short-circuit fault diagnosis method are as follows:
[0057] (1) Embed an FBG sensor into a solid-state battery;
[0058] (2) The battery was subjected to cyclic charge-discharge tests at 0.1C, 0.3C and 1C rates, and internal strain data was collected by a fiber optic demodulator at a sampling frequency of 1 second.
[0059] (3) The program performs moving average filtering on the internal strain data in real time and calculates its time differential curve;
[0060] (4) When the peak amplitude of the strain differential curve is below the first-level threshold, it is judged as normal;
[0061] (5) When the peak amplitude of the strain differential curve is above the first-level threshold, a first-level alarm is triggered;
[0062] (6) When the peak amplitude of the strain differential curve is above the secondary threshold, a secondary alarm is triggered;
[0063] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.
[0064] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.
Claims
1. A method for diagnosing short-circuit faults in solid-state batteries based on internal sensing, characterized in that, The method includes: S1: Acquire real-time strain data of the solid-state battery from embedded sensors, synchronously record electrochemical signals, and generate raw strain data sequences and electrochemical auxiliary information; S2: Based on the original strain data sequence obtained in S1, the original data is filtered by moving average to remove noise introduced by electromagnetic interference and mechanical vibration, and a smooth strain-time curve is constructed to obtain a smoothed strain data sequence, which provides a clean signal for subsequent differential calculations. S3: Input the smoothed strain data sequence, use the finite difference method to calculate the first derivative of strain with respect to time, thereby obtaining the strain differential curve, which is used to provide information about the dynamic changes of internal strain; S4: Analyze the strain differential curve. ,in: , in, The strain-time data sequence is after smoothing and filtering in step S2. For time, The first derivative of strain with respect to time; During normal battery operation, when the accumulation of mechanical stress inside the solid-state battery causes instantaneous, minute cracks in the electrode or electrolyte materials, it leads to a rapid change in local strain. This abrupt change in local strain is reflected in the strain differential curve. The above is manifested as one or more characteristic peaks ; For each detected valid feature peak Extracting peak amplitude and time of occurrence Peak amplitude The peak value and occurrence time of the corresponding characteristic peak. The peak amplitude is defined as the time point in time, and each set of peak amplitudes and occurrence times is combined to form peak data points. Its form is ; The characteristic peak of the strain differential curve originates from the sudden change in local strain caused by instantaneous cracking of the material inside the solid-state battery. The peak amplitude of the characteristic peak characterizes the severity of the short circuit caused by the cracking. S5: Input the peak data points And preset multi-level diagnostic thresholds, extracting peak amplitude The results are compared with various diagnostic thresholds to determine whether a fault alarm condition has been triggered, and diagnostic results indicating the solid-state battery status and the corresponding alarm signal level are generated. The method for constructing the multi-level diagnostic thresholds includes: S501: Multiple solid-state battery samples were subjected to charge-discharge cycle tests at different charge rates, and their voltage and internal strain were monitored simultaneously to confirm the occurrence of short-circuit faults. S502: Monitor and screen out cyclic data where the voltage curve shows an abnormal drop, calculate the corresponding strain differential curve, and extract the characteristic peaks related to the voltage anomaly as the basis for diagnosing short-circuit faults. S503: For initial fault samples, perform statistical distribution analysis on their strain differential characteristic peaks, and select the lower limit of their confidence interval as the first-level diagnostic threshold. S504: For severe fault samples, perform statistical distribution analysis on their strain differential characteristic peaks and select the lower limit of their confidence interval as the secondary diagnostic threshold.
2. The method for diagnosing short-circuit faults in solid-state batteries based on internal sensing according to claim 1, characterized in that, After step S5, the method further includes: Input the diagnostic results and the current operating status of the battery, trigger the corresponding system control strategy based on the diagnostic results, issue control commands, generate fault reports, and provide safety measure suggestions to ensure battery safety and realize a diagnosis-response closed loop.
3. The method for diagnosing short-circuit faults in solid-state batteries based on internal sensing according to claim 1, characterized in that, In step S1, the embedded sensor is a fiber Bragg grating (FBG) sensor or a thin-film pressure sensor.
4. The method for diagnosing short-circuit faults in solid-state batteries based on internal sensing according to claim 3, characterized in that, In step S1, when a fiber Bragg grating (FBG) sensor is used, the center wavelength shift of the FBG sensor is monitored in real time by a high-precision fiber Bragg grating demodulator, and the sensor is precisely demodulated into strain data inside the battery according to the pre-calibrated strain sensitivity coefficient, thereby providing a high-quality raw signal for subsequent differential calculations and fault diagnosis.
Citation Information
Patent Citations
Battery safety state monitoring method and device, electronic equipment and storage medium
CN118198537A
Battery lithium precipitation detection method and device
CN119535256A
Optical fiber embedded lithium battery for researching electrode phase change and manufacturing method and application thereof
CN119674293A
