A method for online fault detection of pouch battery sensor

Abstract

The invention discloses an on-line detection method for sensor faults of soft pack batteries, which measures and records battery input current and terminal voltage data, collects soft pack battery surface temperature data through thermocouple sensors, and does not know the information of system models and parameters without knowing the information of system models and parameters. , using the online learning method in the rolling time domain to train and verify the preset prediction model according to the principal element matrix, the input current vector and the terminal voltage data vector, and obtain the estimated value of the temperature data at the corresponding moment; according to the thermocouple temperature data measurement The fault detection residual is obtained from the estimated value of the temperature data at the corresponding time and the fault detection residual, and the detection result is obtained according to the fault detection residual, the preset residual evaluation function and the preset fault detection threshold. Using a few thermocouples and current and voltage sensors, the sensor fault detection of soft-pack lithium-ion batteries is realized without knowing the exact model.

Description

technical field [0001] The invention belongs to the technical field of fault detection, in particular to an online fault detection method for a pouch battery sensor. Background technique [0002] Compared with square and cylindrical batteries, soft-pack lithium-ion batteries have the characteristics of high energy density, high safety, high flexibility and low cost. They are the power source of new energy vehicles, and their operation safety is the focus of scientific research and industry. The problem. The battery management system uses various sensors (temperature, voltage, current, etc.) to monitor battery status information, evaluate and manage the operating status, which is crucial to the high performance, long life and safe operation of the battery. However, there are many types of these sensors, and in the event of a failure, the battery management system will not be able to accurately obtain the status information of the battery system, resulting in irreversible dam...

AI Technical Summary

Problems solved by technology

However, there are many types of these sensors. In the event of a failure, the battery management system will not be able to accurately obtain the status information of the battery system, resulting in irreversible damage to the battery and even potential safety hazards.

At present, the fault detection method based on the internal thermodynamic model of the battery needs to know the information of the system model and parameters in advance, but it is often difficult to obtain an accurate model and there are various interferences in practical applications, which restricts the application and promotion of such methods

Images