EP-4738151-A1 - DATA ANALYSIS DEVICE AND OPERATING METHOD THEREOF

EP4738151A1EP 4738151 A1EP4738151 A1EP 4738151A1EP-4738151-A1

Abstract

A data analysis apparatus according to an embodiment disclosed in this document may include an information acquisition unit configured to acquire a plurality of data sets, each containing time-series data for a plurality of parameters, and a controller configured to extract features of each data set based on the correlation between the plurality of parameters, extract patterns of each data set based on the similarity between the extracted features, and classify the plurality of data sets into a plurality of clusters based on the patterns.

Inventors

PARK, Hyung Oak
CHOI, JEE SOON
PARK, KYU TAE
KIM, HA YEON

Assignees

LG Energy Solution, Ltd.

Dates

Publication Date: 20260506
Application Date: 20240806

Claims (15)

A data analysis apparatus comprising: an information acquisition unit configured to acquire a plurality of data sets, each containing time-series data for a plurality of parameters; and a controller configured to extract features of each data set based on the correlation between the plurality of parameters, extract patterns of each data set based on the similarity between the extracted features, and classify the plurality of data sets into a plurality of clusters based on the patterns.
The data analysis apparatus of claim 1, wherein the controller extracts the features as a matrix representing the correlation between the plurality of parameters.
The data analysis apparatus of claim 2, wherein the correlation between the plurality of parameters is represented by the Pearson correlation coefficient for each parameter pair.
The data analysis apparatus of claim 1, wherein the controller calculates the correlation coefficients between the features of the plurality of data sets and generates a correlation coefficient matrix composed of the correlation coefficients.
The data analysis apparatus of claim 4, wherein the controller applies a classification algorithm to the correlation coefficient matrix to extract the patterns.
The data analysis apparatus of claim 5, wherein the classification algorithm is a K-means algorithm.
The data analysis apparatus of claim 6, wherein the controller extracts the patterns based on the results of applying the classification algorithm to the correlation coefficient matrix while varying a K value.
The data analysis apparatus of claim 1, wherein the controller analyzes the characteristics of each of the plurality of clusters.
The data analysis apparatus of claim 8, wherein the controller analyzes the anomaly occurrence history related to the data sets included in each of the plurality of clusters to estimate the parameters associated with the anomaly occurrence history.
The data analysis apparatus of claim 1, wherein the controller extracts the features by setting a window interval for each of the plurality of data sets.
An operating method of a data analysis apparatus, the method comprising: acquiring a plurality of data sets, each containing time-series data for a plurality of parameters; extracting features of each data set based on the correlation between the plurality of parameters; extracting patterns of each data set based on the similarity between the extracted features; and classifying the plurality of data sets into a plurality of clusters based on the patterns.
The method of claim 11, wherein the extracting of the features of each data set comprises extracting the features as a matrix representing the correlation between the plurality of parameters.
The method of claim 11, wherein the extracting of the patterns of each data set comprises: generating a correlation coefficient matrix composed of the correlation coefficients calculated between the features of the plurality of data sets; and applying a classification algorithm to the correlation coefficient matrix to extract the patterns.
The method of claim 13, wherein the classification algorithm is a K-means algorithm, and the extracting of the patterns by applying the classification algorithm comprises extracting the patterns based on the results of applying the classification algorithm while varying a K value to the correlation coefficient matrix.
The method of claim 11, further comprising estimating the parameters associated with the anomaly occurrence history by analyzing the anomaly occurrence history related to the data sets included in each of the plurality of clusters.

Description

TECHNICAL FIELD CROSS REFERENCE TO RELATED APPLICATION This application claims priority to Korean Patent Application No. 10-2023-0106325, filed August 14, 2023, the entire contents of which is incorporated herein for all purposes by this reference. Technical Field The embodiments disclosed in this document relate to a data analysis apparatus and operating method thereof. BACKGROUND ART Recently, research and development on secondary batteries have been actively conducted. Here, the secondary battery is a rechargeable battery and includes all of the conventional Ni/Cd batteries, Ni/MH batteries, and recent lithium ion batteries. Among the secondary batteries, Lithium ion batteries have an advantage of much higher energy density than the conventional Ni/Cd batteries and Ni/MH batteries. In addition, Lithium-ion batteries can be manufactured small and lightweight enough to be used as power sources for mobile devices and recently expand their usage range to power sources for electric vehicles, drawing attention as a next-generation energy storage medium. Typically, batteries are monitored and controlled by a Battery Management System (BMS). The data collected from the battery management system includes data on a wide variety of factors, and the quantity of data is so vast that it has been difficult to analyze the cause of anomalies even when they occur in the battery. DISCLOSURE TECHNICAL PROBLEM It is an object of the embodiments disclosed in this document to provide a data analysis apparatus and operating method thereof capable of effectively identifying the cause of anomalies in a battery by classifying battery data. The technical objects of the embodiments disclosed in this document are not limited to the aforesaid, and other objects not described herein with be clearly understood by those skilled in the art from the descriptions below. TECHNICAL SOLUTION A data analysis apparatus according to an embodiment disclosed in this document may include an information acquisition unit configured to acquire a plurality of data sets, each containing time-series data for a plurality of parameters, and a controller configured to extract features of each data set based on the correlation between the plurality of parameters, extract patterns of each data set based on the similarity between the extracted features, and classify the plurality of data sets into a plurality of clusters based on the patterns. According to an embodiment, the controller may extract the features as a matrix representing the correlation between the plurality of parameters. According to an embodiment, the correlation between the plurality of parameters may be represented by the Pearson correlation coefficient for each parameter pair. According to an embodiment, the controller may calculate the correlation coefficients between the features of the plurality of data sets and generate a correlation coefficient matrix composed of the correlation coefficients. According to an embodiment, the controller may apply a classification algorithm to the correlation coefficient matrix to extract the patterns. According to an embodiment, the classification algorithm may be a K-means algorithm. According to an embodiment, the controller may extract the patterns based on the results of applying the classification algorithm to the correlation matrix while varying a K value. According to an embodiment, the controller may analyze the characteristics of each of the plurality of clusters. According to an embodiment, the controller may analyze the anomaly occurrence history related to the data sets included in each of the plurality of clusters to estimate the parameters associated with the anomaly occurrence history. According to an embodiment, the controller may extract the features by setting a window interval for each of the plurality of data sets. According to an embodiment disclosed in this document, an operating method of a data analysis apparatus may include acquiring a plurality of data sets, each containing time-series data for a plurality of parameters, extracting features of each data set based on the correlation between the plurality of parameters, extracting patterns of each data set based on the similarity between the extracted features, and classifying the plurality of data sets into a plurality of clusters based on the patterns. According to an embodiment, the extracting of the features of each data set may include extracting the features as a matrix representing the correlation between the plurality of parameters. According to an embodiment, the extracting of the patterns of each data set may include generating a correlation coefficient matrix composed of the correlation coefficients calculated between the features of the plurality of data sets, and applying a classification algorithm to the correlation coefficient matrix to extract the patterns. According to an embodiment, the classification algorithm may be a K-means algorithm, and the extracting of the