基于阻抗在线测量的锂离子电池过放电诱发内短路识别研究

The over-discharge behavior of lithium-ion batteries can induce internal short circuits, which may lead to thermal runaway. Early detection of internal short circuits is crucial for preventing catastrophic accidents. However, the consistency of the single cell is different, and obvious electrical and thermal characteristics are lacking at the initial stage of the internal short circuit caused by over-discharge. It is difficult to reliably realize fault warnings with conventional physical parameters such as voltage and temperature. The impedance can reflect the internal information of the battery and has a good indication of the fault state. More research is needed on the impedance-based online diagnosis of internal short circuits in lithium-ion batteries. Acknowledging the complexity associated with modelling and identifying electrochemical impedance spectroscopy (EIS) parameters, this paper proposes anonline identification method for lithium-ion battery internal short circuits under a frequency of 70 Hz induced by over-discharge. This method does not involve complex mathematical models or parameter identification, ensuring accurate identification of internal short circuits caused by over-discharge with fault warning capabilities. Firstly, the mechanism of internal short circuits induced by over-discharge in lithium-ion batteries was studied, and the dynamic impedance change characteristics during the internal short circuit were analyzed. Secondly, an online impedance measurement device for lithium-ion batteries was designed, and an experimental platform was built to simulate internal short circuits through overdischarge. EIS under different temperature environments and state of charge conditions was obtained, and the characteristic impedance frequency for internal short circuit fault identification was determined. The dynamic impedance was measured during normal charging/discharging and over-discharge processes of lithium-ion batteries. The dynamic impedance and its change rate were analyzed for internal short-circuit fault identification. Based on the dynamic impedance characteristics, an internal short-circuit online identification method was proposed. Finally, according to the irreversible capacity loss and self-discharge characteristics of the internal short circuit battery and battery disassembly, the reliability of the proposed identification method was verified. The experimental results show that the dynamic impedance of lithium-ion batteries during discharge exhibits a semi-sinusoidal characteristic change, which can provide an early warning for over-discharge by about 144 seconds in advance. The needle-like characteristic change of dynamic impedance can provide an early warning of internal short circuit faults by about 152 seconds. The significant rebound feature of dynamic impedance can be used as a sign of internal short circuit occurrence. In addition, the feature of impedance change rate helps realize fault identification and warning of lithium-ion batteries. Experimental verification shows that batteries with a significant rebound characteristic of dynamic impedance experience a sharp increase in irreversible and self-discharge capacity loss up on cessation of discharging. Moreover, after disassembling, the positive electrode has copper metal deposition, the negative electrode graphite layer is damaged, the copper foil is dissolved, and the separator is sparse and shows signs of damage, demonstrating the reliability of the proposed method.