Lithium battery fault diagnosis method based on support vector machine and K mean value

Abstract

The invention relates to a lithium battery fault diagnosis method based on a support vector machine and a K mean value. The lithium battery fault diagnosis method comprises the following steps of acquiring an original data set through a working condition of a battery, and selecting a battery fault symptom; preprocessing the data of the diagnosis variables, including normalization and PCA; sendingthe preprocessed data into Kmeans clusters, and screening a clustering result which is the same as an actual label as a fault sample set; and randomly splitting the fault sample set into a training set and a testing set, sending training samples into an SVM model for learning, outputting an SVM classification model, and sending testing samples into the SVM classification model for testing. Throughthe method disclosed by the invention, the fault diagnosis research is conducted on the battery system to realize identification of 4 health states; by considering the conditions that the generationof the battery fault is influenced by various factors, a specific reason of the fault generation is hard to determine, and the diagnosis of the battery fault has a certain difficulty, the invention provides a fault classification method based on the support vector machine and the K mean value.

Description

technical field [0001] The invention relates to the technical field of battery fault diagnosis, in particular to a lithium battery fault diagnosis method based on a support vector machine and K-means. Background technique [0002] Lithium batteries are the energy source for electric vehicles and aircraft, and are also the most prone to failure. The faults of lithium batteries mainly include abnormal temperature, overcharge, overdischarge, undervoltage, overvoltage, equalization failure, abnormal charge and discharge current, self-discharge, abnormal internal resistance, battery aging and abnormal voltage of each single battery. [0003] At present, there are two main fault diagnosis methods for lithium batteries: first, model-based fault diagnosis: since the battery is a nonlinear system that changes in real time and is affected by various parameter changes, it is important to establish an accurate fault diagnosis model Very difficult; second, data-driven fault diagnosis: d...

AI Technical Summary

Problems solved by technology

[0003] At present, there are two main fault diagnosis methods for lithium batteries: first, model-based fault diagnosis: since the battery is a nonlinear system that changes in real time and is affected by various parameter changes, it is important to establish an accurate fault diagnosis model Very difficult; second, data-driven fault diagnosis: data-driven fault diagnosis uses historical monitoring data information to detect faults, does not require an accurate battery model, and the algorithm is fast, but commonly used neural networks and expert systems require A large number of data samples are used for training, which cannot be satisfied by general complex systems

Images