In the "double carbon" era, in the process of the rapid development of the new energy industry, the failure to give early warning to the internal faults of the battery cell has caused a series of new energy industry safety problems. Through the unique online impedance detection technology of dnb1168, Datang NXP makes intelligent early warning for the stray cells in a module or the whole pack, strangles the hidden trouble in the early stage, and ensures the safety of the whole life cycle of the new energy system.

Outlier detection technology based on electrochemical impedance detection

Make the faulty cell invisible

Cell outliers

As a monitoring method based on standard statistical technology and rigorously demonstrated, the outlier detection algorithm can provide more dimensional detection means than the traditional scheme through the unique online impedance detection technology, and make intelligent early warning for the outliers in a module or the whole pack. Whether before the pack leaves the factory or when the pack is working, the outlier detection method of the cell can avoid the "fish out of the net" Accurately detect the abnormal state of the cell, so as to nip the out of control hidden danger in the early stage and ensure the safety of the whole life cycle of the new energy system.

Application flow chart of cell outlier detection technology in pack factory

Application flow chart of cell outlier detection technology in pack work

The cell outlier detection method is not limited by the current cell type, working temperature, charge / discharge current, state of charge (SOC), state of Health (soh) and other conditions. In the detection environment of the same module or the same pack, it is only necessary to calculate the impedance amplitude of a group of frequencies according to a certain interval time point, and analyze the impedance amplitude and median deviation of each frequency point, which can more efficiently and quickly locate the problem cell, find another way, and unlock the new skill of cell monitoring.

Figure 1: data performance on amplitude frequency diagram when the cell is in normal state

By calculating the impedance value of each cell and comparing it with the median, the impedance deviation degree of each cell can be obtained. If the deviation in the full frequency band is within a certain range (in actual operation, the appropriate deviation range can be selected according to the internal resistance of different cells), as shown in figure 1, ± 1% range, it can be considered that the cells are in the same state, which also shows that there is no outlier phenomenon, and the cell state is normal.

Figure 2: data performance on amplitude frequency diagram when the cell is out of group

However, when the deviation of one or more cells is too large, it indicates that one or more cells are in an inconsistent state, that is, there is an outlier phenomenon, which indicates that abnormal cell material, abnormal temperature, mechanical failure or loose link points have occurred, as shown in Figure 2.

The birth of battery management system can not only ensure the safety and reliability of energy storage devices, improve the power and energy efficiency of battery packs, but also play a vital role in the development of new energy vehicle industry. In practical applications, partners can use electrochemical impedance spectroscopy (EIS) to diagnose battery status. While improving the operational reliability of the system, it also reduces the operating cost of the system and provides economic and technical guarantee for partners. In the future, we will further elaborate the system theory and application method of outlier detection technology in combination with the actual application case analysis. Please look forward to it.

In the future, Datang NXP will adhere to innovation and always be committed to transforming chip technology innovation into a driving force for change in the new energy industry, injecting new vitality into the development of battery management chips.