![](https://allindiaev.com/wp-content/uploads/2024/05/WhatsApp-Image-2024-05-14-at-5.18.38-PM-1024x512.jpeg)
Part 4: Experimental results
In lengthy and high-precision magnetic field measurement experiments, the inherent fluctuations of the Earth’s magnetic field and the electromagnetic radiation emitted by laboratory equipment pose significant challenges to data accuracy.
To ascertain the authenticity and precision of experimental data and assess the impact of background magnetic fields on results, we initiated an investigation. Initially, the fluxgate sensor was positioned within the laboratory’s geomagnetic environment and subjected to stable measurements over an uninterrupted period of 8 h, meticulously documenting the outcomes.
As illustrated in Fig. 13, our findings revealed that while the background magnetic field exhibited fluctuations of approximately 10.55nT throughout the measurement period, these fluctuations were substantially smaller in magnitude compared to the targeted magnetic field range pertinent to the experiment.
This revelation underscores our ability to accurately capture experimental phenomena amidst the intricate backdrop of the geomagnetic field, thus facilitating the derivation of scientifically reliable conclusions. Consequently, we posit that the influence of background magnetic fields on experimental outcomes is negligible, obviating the necessity for conducting experiments within a magnetic shield cylinder. This discovery not only streamlines experimental setups and procedures, enhancing efficiency, but also confirms the viability of utilizing magnetic detection methods to assess battery status within the geomagnetic field.
Figure 13
![](https://allindiaev.com/wp-content/uploads/2024/05/image-13.png)
Fluctuations of the ambient magnetic field (continuous measurement for 8 h).
During the experiment, we employed the constant current discharge method to assess the battery pack, maintaining a consistent discharge current of 0.5A for a duration of 1 h. Figure 14 depicts the curves detailing the current, voltage, capacity, and energy of the tested battery pack over the course of charging and discharging.
As illustrated, under the constant current discharge condition, the battery pack consistently maintains a current value of 0.5A, showcasing excellent current stability. Notably, the voltage of the battery pack remains nearly unchanged, hovering around 4.8 V at both the onset and conclusion of the discharge, indicating remarkable voltage stability throughout the discharge process. Given the parallel connection of batteries in this experiment, the output voltage remains steady even if individual batteries encounter issues.
Furthermore, owing to the constant current discharge mode employed, the capacity and energy of the battery pack exhibit a linear correlation with time. This discovery furnishes a crucial experimental foundation for comprehending the performance dynamics of battery packs during constant current discharge, thereby offering valuable guidance for researchers seeking to delve deeper into battery performance optimization.
Figure 14
![](https://allindiaev.com/wp-content/uploads/2024/05/image-14.png)
The results of battery tests: (a) The current; (b) The voltage; (c) The capacity; (d) The energy.
The magnetic field distribution image of the healthy battery pack during 0.5A discharge is depicted in Fig. 15, with each individual battery being marked with a corresponding number. Within the image, four black dotted boxes delineate the four battery slots within the box. Examination of the magnetic field distribution image reveals a degree of unevenness in the induced magnetic field distribution of the battery pack, attributable to the background field presence and the compounded effect of magnetic fields.
Notably, as each individual battery undergoes discharge, current accumulation becomes apparent at positive and negative positions, leading to a discernible uneven distribution of magnetic fields in these areas. Additionally, the upper section of the image exhibits a more uniform magnetic field distribution compared to the lower portion. This disparity can be attributed to the upper part corresponding to the power bank’s monitor, where internal currents are smaller and more uniform, thus resulting in a more homogenous induced magnetic field distribution.
Figure 15
![](https://allindiaev.com/wp-content/uploads/2024/05/image-15.png)
Magnetic field distribution when the healthy battery pack is discharged at 0.5A.
The magnetic field of the power bank is too high, to the extent that it would cover the failed positions. To observe clearer experimental phenomena, we measured the magnetic field distribution, B2, of the faulty battery pack and compared it with the magnetic field value, B1, at the corresponding positions within a non-faulty battery pack to assess the magnetic anomaly ΔB.
The resulting magnetic field distribution image is presented in Fig. 16. Experimental findings reveal that in the failure of the battery at position 1, a magnetic field anomaly ranging from approximately 2000–3000 nT manifests near the fault location, forming a circular magnetic field distribution image. The maximum magnetic anomaly value recorded is approximately 2934 nT, with its location corresponding to the number of the battery’s position, indicated by the blue triangle in Fig. 16a. Similarly, as depicted in Fig. 16b, when the battery at position 2 fails, a distribution pattern akin to that in Fig. 16a emerges near the fault location, with the magnetic field anomaly peaking at around 6735 nT.
Figure 16
![](https://allindiaev.com/wp-content/uploads/2024/05/image-16.png)
Magnetic field distribution when the battery pack is discharged at 0.5A: (a) The battery at position 1 fails. (b) The battery at position 2 fails.
In conclusion, the location of a failed battery within the power bank can be determined by identifying the location of the magnetic anomaly. This approach circumvents the need for monitoring battery pack capacity anomalies or reconstructing internal current images to pinpoint faults, thus enabling precise identification of failed battery locations.
Similarly, this method can also detect defective batteries when the internal current of the battery pack increases abnormally. If some batteries within the pack experience an internal short circuit leading to an abnormal rise in current, this deviation in current triggers an abnormal increase in the magnetic field surrounding the battery pack.
Consequently, this method can detect the abnormal magnetic field resulting from the abnormal current and concurrently identify the faulty battery. The abnormal current caused by an internal short circuit often induces an abnormal rise in battery temperature. When the temperature of a lithium-ion battery surpasses 90 °C, the interface film of the solid electrolyte begins to decompose, potentially triggering a chain reaction leading to thermal runaway. Hence, it is imperative to promptly identify and investigate battery faults.
However, during normal battery operation, temperature also rises, which may impede the detection of abnormal temperature, resulting in a delay in identifying anomalies. As outlined in Sect. 3.4.1, the magnetic field measurement method exhibits a faster response time compared to temperature measurement, enabling more timely detection and localization of battery pack faults.
Conclusion
This paper establishes a coupled 3D multiphysics model for the lithium-ion battery pouch cell by integrating electrochemical, magnetic field, and thermal models. Numerical simulations are conducted to investigate the distribution of physical fields surrounding the cell. Based on this model, the paper calculates and analyzes the physical field distribution under conditions of internal short circuits and the presence of cracks.
The results indicate that under internal short circuit conditions, there is a significant abnormal increase in the magnetic field near the short circuit point, with the magnetic field anomaly on the battery surface approximately 5 μT higher than in other regions. The primary sources of heat generation in the battery, stemming from lithium dendrites, are the positive and negative electrodes. The larger the radius of the lithium dendrites, the greater the maximum magnetic field value in the system. In the case of a cell’s crack, there is a noticeable abnormal increase in the magnetic field at both ends of the crack.
Whether in the scenario of internal short circuits or the presence of the crack, the number and location of faults can be determined by the abnormal increase in the magnetic field. The key advantage of this method is the non-destructive detection of the battery’s health state, avoiding material deformation caused by disassembly and similar methods.
Furthermore, since the temperature response is delayed compared to the magnetic field, utilizing magnetic field detection allows for early identification of short circuit issues, facilitating timely intervention to prevent accidents resulting from short circuits. Additionally, a magnetic field scanning method is proposed for model verification. The results demonstrate that the magnetic field image can accurately reflect the location of faulty battery cells and facilitate their selection and localization.
In future research, enhancing the accuracy of the model will be a primary challenge and focus. During battery faults, the abnormal magnetic field generated within the fault area exhibits large magnetic field gradients. To improve the model’s precision, further refinement of the grid is required to obtain more realistic computational results. Simultaneously, integrating this model with an aging model in future studies can facilitate the prediction of the battery’s lifespan. The large size of existing fluxgate sensors limits the measurement accuracy of small size batteries.
In the future, high-resolution devices such as optical fiber magnetic field sensors can be introduced to obtain finer magnetic field distribution images, providing accurate references for battery production, detection and screening. In addition, future research can utilize small magnetic sensor arrays to achieve multi-point real-time measurements and accurately reconstruct magnetic field images. These research outcomes will contribute to the exploration of the mechanisms behind battery faults, providing a theoretical foundation for non-destructive battery testing. Additionally, they offer reference methods for the safety assessment and health evaluation of commercial batteries.