KR102515853B1 - Abnormality diagnosis device, system and method based on battery cell unit - Google Patents

Abstract

Disclosed is an apparatus, system, and method for diagnosing abnormality based on a battery cell unit. The abnormal diagnosis apparatus of the present invention performs pre-processing for diagnosing an abnormality based on the measured data when receiving a plurality of measurement data obtained in a battery cell unit from the energy storage device and the communication unit communicating with the energy storage device, and the pre-processed measurement A control unit for modeling data and diagnosing an abnormality of each battery cell using the modeled result is included.

Description

Abnormality diagnosis device, system and method based on battery cell unit {Abnormality diagnosis device, system and method based on battery cell unit}

The present invention relates to a device for diagnosing an abnormality in a battery, and more particularly, an abnormality diagnosing device and system based on a battery cell unit that detects an abnormal state of a battery in an early stage by diagnosing an abnormality in a battery included in an energy storage device in a cell unit. and methods.

A conventional method for diagnosing an abnormality in a battery uses a predefined reference value or performs diagnosis only on a module basis.

For this reason, when a predefined reference value is used, it is difficult to reflect the individual characteristics of the battery and to identify problems such as temporary errors and changes in the dynamic characteristics of the battery. In addition, in the case of using the branch of the module unit, a mismatch may occur in the measurement of the module unit and the measurement of the cell unit, so it may appear that there is no abnormality in the battery externally, but there may be problems in individual cells inside the module. It is difficult to diagnose the possibility.

In particular, it is difficult to use a conventional statistical-based method or a supervised learning method because it is difficult to obtain data on an abnormal state in a facility such as an energy storage device.

Therefore, there is a need for research to clearly understand the state of the battery of the energy storage device.

Korean Registered Patent Publication No. 10-2148204 (2020.08.26.)

An object of the present invention is to provide a device, system, and method for diagnosing abnormality based on a battery cell unit that accurately diagnoses anomaly of a battery module included in an energy storage device.

In order to achieve the above object, an abnormality diagnosis apparatus based on a battery cell unit according to the present invention receives a plurality of measurement data obtained from a battery cell unit from a communication unit that communicates with an energy storage device and the energy storage device to measure the measurement and a control unit that performs preprocessing for diagnosing an abnormality based on the data, models the preprocessed measurement data, and diagnoses an abnormality for each battery cell using the modeled result.

In addition, the control unit, at least one of a statistical method including at least one of maximum value, minimum value, average, median value, standard deviation, variance, moving average, skewness and kurtosis, Fourier transformation and wavelet transformation A feature combination that can best represent the operation of the battery cell among a signal processing method including one and a dimensionality reduction method including at least one of principal component analysis and t-stochasic nearest neighbor (T-SNE) and pre-processing the measurement data with the detected combination.

In addition, the control unit performs modeling to compare the dynamic characteristics of each battery cell using the preprocessed result, and at least one battery cell among a plurality of battery cells prepares for another battery cell using the modeled diagnostic model. It is characterized in that a battery cell exhibiting different characteristics than a predetermined standard is detected, and the detected battery cell is diagnosed as an abnormal state.

In addition, the control unit calculates a data distribution in a predetermined specific period (Δt), calculates a density of data using the calculated distribution, and the different characteristics based on data grouped according to the calculated density. Characterized in that detecting a battery cell indicating.

In addition, the control unit performs modeling to compare the current dynamic characteristics of each battery cell with past dynamic characteristics using the preprocessed result, and uses the modeled diagnostic model to determine the current dynamic characteristics of at least one battery cell among a plurality of battery cells. It is characterized in that a battery cell having a different dynamic characteristic than a preset standard is detected in preparation for the dynamic characteristic in the past, and the detected battery cell is diagnosed as an abnormal state.

In addition, the control unit is characterized in that the past dynamic characteristics of each battery cell is modeled by learning the autoencoder.

In addition, the control unit calculates a residual for the past dynamic characteristics of each battery cell, and compares the current dynamic characteristics of each battery cell based on learning data representing a residual distribution for the calculated residual to indicate the different characteristics. Characterized in detecting a battery cell.

An abnormality diagnosis system according to the present invention diagnoses abnormalities in each battery cell using an energy storage device that stores electrical energy using a battery and a plurality of measurement data obtained from each battery cell included in the battery. Including an abnormality diagnosis device, wherein the abnormality diagnosis device diagnoses an abnormality based on the measurement data when a plurality of measurement data obtained in units of battery cells is received from a communication unit communicating with the energy storage device and the energy storage device. and a control unit for performing preprocessing for processing, modeling the preprocessed measurement data, and diagnosing abnormality of each battery cell using the modeled result.

The apparatus may further include a user terminal configured to receive monitoring information for diagnosing an abnormality of each battery cell from the abnormality diagnosis device and output the received monitoring information.

An abnormality diagnosis method based on a battery cell unit according to the present invention includes the steps of receiving, by an abnormality diagnosis device, a plurality of measurement data acquired in a battery cell unit from the energy storage device; Preprocessing for diagnosing an abnormality, modeling the preprocessed measurement data, and diagnosing an abnormality in each battery cell using the modeled result by the abnormality diagnosis apparatus.

An abnormality diagnosis apparatus, system, and method based on a battery cell unit of the present invention obtains individual data of the battery cell, diagnoses anomalies in dynamic characteristics of individual battery cells, and diagnoses abnormalities in current dynamic characteristics and past dynamic characteristics of individual batteries, A multi-dimensional abnormality diagnosis of a battery cell included in a battery module may be performed.

Through this, the present invention can solve data inconsistency generated in conventional module unit measurement.

In addition, the present invention can perform efficient operation by increasing the productivity and reliability of the energy storage device and at the same time increasing the scalability of the system.

1 is a configuration diagram for explaining an anomaly diagnosis system according to an embodiment of the present invention.
2 is a block diagram for explaining an abnormal diagnosis apparatus according to an embodiment of the present invention.
3 is a block diagram for explaining a control unit according to an embodiment of the present invention.
4 is a diagram for explaining a process of diagnosing an abnormal state by analyzing dynamic characteristics of each battery cell according to an embodiment of the present invention.
5 is a diagram for explaining a process of diagnosing an abnormal state by analyzing current dynamic characteristics and past dynamic characteristics of each battery cell according to an embodiment of the present invention.
6 is a flowchart illustrating a method for diagnosing an abnormality according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. First, in adding reference numerals to the components of each drawing, it should be noted that the same components have the same numerals as much as possible even if they are displayed on different drawings. In addition, in describing the present invention, if it is determined that a detailed description of a related known configuration or function is obvious to those skilled in the art or may obscure the gist of the present invention, the detailed description will be omitted.

1 is a configuration diagram for explaining an anomaly diagnosis system according to an embodiment of the present invention.

Referring to FIG. 1 , an abnormality diagnosis system 300 diagnoses an abnormality of a battery. The abnormal diagnosis system 300 includes the abnormal diagnosis device 100 and the energy storage device 200 .

The abnormality diagnosis apparatus 100 may perform abnormality diagnosis using a plurality of measurement data acquired in units of battery cells 210 , 220 , and 230 of the battery module included in the energy storage device 200 . At this time, the abnormality diagnosis apparatus 100 can clearly determine which battery cell in the battery module has an abnormality by determining whether or not there is an abnormality using the dynamic characteristics of each battery cell.

The energy storage device 200 is a device that stores electrical energy. That is, the energy storage device 200 is a system for storing and managing energy to efficiently use energy, and is used in power plants, transmission and distribution facilities, homes, factories, and companies. To this end, the energy storage device 200 includes a battery module, and the battery module includes a plurality of battery cells 210, 220, and 230. For example, the battery module may include the first battery cell 210, the second battery cell 220 to the Nth battery cell 230 (N is a natural number greater than or equal to 3). At this time, the energy storage device 200 measures characteristic information in units of battery cells 210 , 220 , and 230 and transmits the measured measurement data to the abnormal diagnosis device 100 . Here, the characteristic information may be battery state information.

In addition, the abnormal diagnosis system 300 may further include a user terminal (not shown) although not disclosed in the drawing. The user terminal receives monitoring information obtained by monitoring the current state of the battery module through communication with the abnormal diagnosis device 100 . The user terminal may remotely monitor the battery module in real time using the received monitoring information. The user terminal may be a computing system such as a smart phone, a desktop computer, a laptop computer, a tablet PC, and a handheld PC.

Meanwhile, the abnormality diagnosis system 300 may establish a communication network 350 between the abnormality diagnosis device 100, the energy storage device 200, and the user terminal so that they can communicate with each other. The communication network 350 may be composed of a backbone network and a subscriber network. The backbone network may be composed of one or a plurality of integrated networks among an X.25 network, a Frame Relay network, an ATM network, a Multi Protocol Label Switching (MPLS) network, and a Generalized Multi Protocol Label Switching (GMPLS) network. Subscriber networks include FTTH (Fiber To The Home), ADSL (Asymmetric Digital Subscriber Line), cable network, zigbee, Bluetooth, and wireless LAN (IEEE 802.11b, IEEE 802.11a, IEEE 802.11g, IEEE 802.11n ), Wireless Hart (ISO/IEC62591-1), ISA100.11a (ISO/IEC 62734), COAP (Constrained Application Protocol), MQTT (Multi-Client Publish/Subscribe Messaging), WIBro (Wireless Broadband), Wimax, 3G, It may be High Speed Downlink Packet Access (HSDPA), 4G and 5G. In some embodiments, the communication network 350 may be an internet network or a mobile communication network. In addition, the communication network 350 may include all other well-known wireless communication or wired communication methods to be developed in the future.

2 is a block diagram for explaining an abnormal diagnosis apparatus according to an embodiment of the present invention.

Referring to FIGS. 1 and 2 , the abnormal diagnosis apparatus 100 includes a communication unit 10 and a control unit 30 , and may further include an output unit 50 and a storage unit 70 .

The communication unit 10 communicates with the energy storage device 200 . The communication unit 10 receives a plurality of measurement data obtained for each cell unit of the battery cell of the battery module from the energy storage device 200 . The communication unit 10 transmits monitoring information on the battery cells of the battery module to the user terminal.

The control unit 30 performs overall control of the abnormal diagnosis apparatus 100 . The control unit 30 performs preprocessing for diagnosing an abnormality based on a plurality of measurement data received from the communication unit 10, and models the preprocessed measurement data. The control unit 30 diagnoses an abnormality in each battery cell using a diagnosis model, which is a modeled result. In this case, the control unit 30 may perform modeling for diagnosis of abnormality using two methods. In detail, the control unit 30 may perform modeling to compare dynamic characteristics of each battery cell or modeling to compare current dynamic characteristics of each battery cell with past dynamic characteristics. Through this, the control unit 30 performs modeling suitable for the driving situation of the abnormality diagnosis apparatus 100 to support accurate diagnosis of abnormality for each battery cell. In addition, the control unit 30 may generate monitoring information for diagnosing abnormality of each battery cell.

The output unit 50 outputs measurement data received from the communication unit 10 and outputs monitoring information generated from the control unit 30 . The output unit 50 may include a display that outputs a screen, a speaker that outputs sound, and the like. Here, the display includes a liquid crystal display (LCD), a thin film transistor-liquid crystal display (TFT LCD), an organic light-emitting diode (OLED), a flexible display, It may include at least one of 3D displays.

The storage unit 70 stores a program or algorithm for driving the abnormal diagnosis apparatus 100 . The storage unit 70 stores measurement data received from the communication unit 10 . The storage unit 70 stores the diagnostic model modeled by the control unit 30 and generated monitoring information. The storage unit 70 is a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, It may include at least one storage medium of a magnetic disk and an optical disk.

Figure 3 is a block diagram for explaining a control unit according to an embodiment of the present invention, Figure 4 is a diagram for explaining a process of diagnosing an abnormal state by analyzing the dynamic characteristics of each battery cell according to an embodiment of the present invention, Figure 5 is a diagram for explaining a process of diagnosing an abnormal state by analyzing current dynamic characteristics and past dynamic characteristics of each battery cell according to an embodiment of the present invention.

Referring to FIGS. 2 to 5 , the control unit 30 includes a data collection unit 31 , a modeling unit 33 and an anomaly diagnosis unit 35 .

The data collection unit 31 collects a plurality of pieces of measurement information acquired from a battery cell unit included in the energy storage device 200 . Here, the measurement information is sensing data for each battery cell and may include voltage, current, temperature, internal resistance, and the like. The data collection unit 31 may store a plurality of collected measurement information in the storage unit 70 .

The modeling unit 33 first performs preprocessing for diagnosing an abnormality based on the measurement data collected from the data collection unit 31 before modeling. The modeling unit 33 is a statistical method including at least one of maximum, minimum, average, median, standard deviation, variance, moving average, skewness, and kurtosis, Fourier transformation, and wavelet transformation. A feature combination that can best represent the operation of a battery cell among a signal processing method including at least one and a dimensionality reduction method including at least one of principal component analysis and t-stochasic nearest neighbor (T-SNE) detection, and the measurement data may be pre-processed with the detected combination. The modeling unit 33 creates a diagnostic model by performing modeling based on the preprocessed measurement information. At this time, the modeling unit 33 may perform modeling using two methods. First, the modeling unit 33 creates a diagnosis model by performing modeling to compare the dynamic characteristics of each battery cell based on the preprocessed result. The second modeling unit 33 creates a diagnostic model by performing modeling to compare current dynamic characteristics of each battery cell with past dynamic characteristics based on the preprocessed result. Through this, the modeling unit 33 can create an optimal diagnosis model by performing modeling suitable for each situation.

The abnormality diagnosis unit 35 diagnoses abnormality of each battery cell using the generated diagnostic model and generates monitoring information for the abnormality diagnosis. The abnormal diagnosis unit 35 may output the generated monitoring information or transmit it to a user terminal.

As shown in FIG. 4 , when a battery cell abnormality diagnosis is performed using a diagnostic model generated through modeling that compares dynamic characteristics of each battery cell, the abnormality diagnosis unit 35 uses the modeled diagnostic model to diagnose a plurality of battery cells. At least one battery cell among the battery cells of the battery cell exhibiting a different characteristic than a predetermined standard in preparation for other battery cells is detected. The abnormality diagnosis unit 35 diagnoses the detected battery cell as an abnormal state. At this time, the abnormal diagnosis unit 35 may determine the criterion for determining the abnormality using data for a specific period (Δt) rather than a signal at a single time. That is, the abnormal diagnosis unit 35 may compare general dynamic characteristics compared to a single time period by comparing the operations of a plurality of battery cells during a specific period of time. In general, when comparing the dynamic characteristics of battery cells for a specific period of time, when battery cells with the same specifications are used, the dynamic characteristics of the battery cells appear similar, whereas a battery cell in an abnormal state has different dynamic characteristics from a battery cell in a normal state. By using these characteristics, the abnormal diagnosis unit 35 can diagnose the abnormal state of the battery cell. In detail, the anomaly diagnosis unit 35 calculates a data distribution in a predetermined specific period, and calculates a data density using the calculated distribution. After detecting battery cells exhibiting different characteristics based on data grouped according to the calculated density, the abnormality diagnosis unit 35 may diagnose that the detected battery cells are in an abnormal state. Here, the abnormal diagnosis unit 35 may use Density Based Spatial Clustering of Applications with Noise (DBSCAN) as a technique for grouping data according to density, but is not limited thereto. Density-based clustering is a technique that classifies n data samples into the same cluster when they are in a circle with a radius of ε. n and ε are hyperparameters, which are values that need to be tuned, and can be optimized to improve performance. In addition, data belonging to the same cluster is called core data, and data not belonging to the same cluster is called noise data. Therefore, in the present invention, it can be confirmed that noise data can be matched with abnormal data, and that battery cells belonging to the abnormal data exhibit different dynamic characteristics from other battery cells for a specific period of time.

As shown in FIG. 5 , in the case of diagnosing an abnormality of a battery cell using a diagnostic model generated through modeling that compares the current dynamic characteristics of each battery cell with the past dynamic characteristics, the abnormality diagnosis unit 35 is a modeled diagnostic model. A battery cell exhibiting a different characteristic than a preset standard in comparison with past dynamic characteristics of at least one battery cell among a plurality of battery cells is detected by using . The abnormality diagnosis unit 35 diagnoses the detected battery cell as an abnormal state. At this time, the abnormality diagnosis unit 35 may determine using data for a specific period rather than a signal at a single time. In detail, the abnormality diagnosis unit 35 models the past dynamic characteristics of each battery cell by learning the autoencoder, and determines the current state of the battery cell through the modeled diagnosis model. The anomaly diagnosis unit 35 calculates a residual for the past dynamic characteristics of each battery cell, and compares the current dynamic characteristics of each battery cell based on learning data representing a residual distribution for the calculated residual to indicate different characteristics. After detecting the battery cell, it may be diagnosed that the detected battery cell is in an abnormal state. At this time, the criterion for determination may use the largest value among the residual values, but is not limited thereto and may be flexibly set by the user. On the other hand, the autoencoder learns the reconstruction neural network together with the reduced neural network, and tries to decode a representation that is as close as possible to the original input from the reduced representation. That is, the autoencoder can build a neural network that reflects the characteristics of conventional data through encoding, and calculate the difference from conventional data through decoding.

6 is a flowchart illustrating a method for diagnosing an abnormality according to an embodiment of the present invention.

1 and 6, the method for diagnosing abnormalities includes acquiring individual data of battery cells, diagnosing anomaly in dynamic characteristics of individual battery cells, and diagnosing abnormalities in current dynamic characteristics and past dynamic characteristics of individual batteries, thereby A multi-dimensional abnormality diagnosis of a battery cell may be performed. Through this, the abnormality diagnosis method can resolve data inconsistency generated in the conventional module-based measurement, increase productivity and reliability of the energy storage device 200, and at the same time increase system scalability to perform efficient operation. .

In step S110 , the abnormality diagnosis apparatus 100 receives a plurality of measurement data acquired in units of battery cells from the energy storage device 200 . Here, the measured data is sensing data for each battery cell and may include voltage, current, temperature, internal resistance, and the like. The abnormal diagnosis apparatus 100 may store the received measurement information.

In step S120, the abnormal diagnosis apparatus 100 performs pre-processing for diagnosing an abnormality based on the measurement data, and models the pre-processed measurement data. The abnormal diagnosis apparatus 100 may perform modeling to compare dynamic characteristics of each battery cell or modeling to compare current dynamic characteristics of each battery cell with past dynamic characteristics. Through this, the device 100 for diagnosing abnormalities may support accurate diagnosis of abnormalities for each battery cell by performing modeling suitable for a driving situation of the device.

In step S130, the abnormal diagnosis apparatus 100 diagnoses an abnormality in each battery cell using the modeled result. The abnormality diagnosis apparatus 100 diagnoses a battery cell exhibiting different dynamic characteristics among a plurality of battery cells as an abnormal state, or diagnoses a battery cell exhibiting different characteristics between a current dynamic characteristic and a past dynamic characteristic among a plurality of battery cells as an abnormal state. can do. In addition, the abnormality diagnosis apparatus 100 may generate monitoring information for abnormality diagnosis for each battery cell, output the generated monitoring information, or transmit the generated monitoring information to a user terminal so that the user terminal can output the monitoring information.

As described above, the present invention discloses only a configuration in which the abnormal diagnosis device 100 diagnoses an abnormality in a battery cell of a battery module included in the energy storage device 200, but is not limited thereto, and the energy storage device 200 ), it can be applied to all systems using battery modules.

The method according to an embodiment of the present invention may be provided in the form of a computer readable medium suitable for storing computer program instructions and data. Such a computer-readable recording medium may include program commands, data files, data structures, etc. alone or in combination, and includes all types of recording devices storing data that can be read by a computer system. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROMs (Compact Disk Read Only Memory) and DVDs (Digital Video Disks). Optical media), magneto-optical media such as floptical disks, and program instructions such as ROM (Read Only Memory), RAM (RAM, Random Access Memory), flash memory, etc. and a hardware device specially configured to do so. In addition, the computer-readable recording medium is distributed in computer systems connected through a network, so that computer-readable codes can be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the technical field to which the present invention belongs.

Although the above has been described and illustrated in relation to preferred embodiments for illustrating the technical idea of the present invention, the present invention is not limited to the configuration and operation as shown and described in this way, without departing from the scope of the technical idea. It will be readily apparent to those skilled in the art that many changes and modifications can be made to the present invention. Accordingly, all such appropriate changes and modifications and equivalents should be regarded as falling within the scope of the present invention.

10: Ministry of Communications
30: control unit
31: data collection unit
33: modeling unit
35: abnormal diagnosis unit
50: output unit
70: storage unit
100: abnormal diagnosis device
200: energy storage device
210: first battery cell
220: second battery cell
230: Nth battery cell
300: abnormal diagnosis system
350: communication network

Claims (10)

a communication unit that communicates with the energy storage device; and
Upon receiving a plurality of measurement data including voltage, current, temperature, and internal resistance in units of battery cells from the energy storage device, preprocessing is performed to diagnose an abnormality based on the measurement data, and modeling of the preprocessed measurement data is performed. And a control unit for diagnosing an abnormality for each battery cell using the modeled result;
The control unit,
A signal comprising at least one of a statistical method comprising at least one of maximum, minimum, mean, median, standard deviation, variance, moving average, skewness and kurtosis, Fourier transformation and wavelet transformation A processing method, and a dimensionality reduction method including at least one of principal component analysis and t-stochasic nearest neighbor (T-SNE).
Preprocessing the measurement data using at least one method of, detecting a feature combination that can best represent the operation of the battery cell,
An abnormality diagnosis device based on a battery cell unit, characterized in that modeling for comparing the dynamic characteristics of each battery cell or modeling for comparing the current dynamic characteristics of each battery cell with past dynamic characteristics using the preprocessed result. delete According to claim 1,
The control unit,
A battery cell in which at least one battery cell among a plurality of battery cells exhibits characteristics different from those of a predetermined reference by using a modeled diagnostic model that compares the dynamic characteristics of each battery cell using the preprocessed result. An abnormality diagnosis device based on a battery cell unit, characterized in that for detecting and diagnosing the detected battery cell as an abnormal state. According to claim 3,
The control unit,
A data distribution in a predetermined specific period (Δt) is calculated, a density of data is calculated using the calculated distribution, and battery cells exhibiting the different characteristics are based on data grouped according to the calculated density. An abnormality diagnosis device based on a battery cell unit, characterized in that for detecting. According to claim 1,
The control unit,
Using the preprocessed result, the current dynamic characteristics of at least one battery cell among a plurality of battery cells using a modeled diagnosis model that compares the current dynamic characteristics of each battery cell with the past dynamic characteristics is a preset standard in preparation for the past dynamic characteristics. An abnormality diagnosis device based on a battery cell unit, characterized in that for detecting a battery cell exhibiting different characteristics and diagnosing the detected battery cell as an abnormal state. According to claim 5,
The control unit,
Building a neural network that reflects the past dynamic characteristics of each battery cell by learning the past dynamic characteristics of each battery cell with an autoencoder, and modeling that calculates residuals for the past dynamic characteristics of each battery cell An anomaly diagnosis device based on a characterized battery cell unit. According to claim 6,
The control unit,
An abnormality diagnosis device based on a battery cell unit, characterized in that for detecting battery cells exhibiting the different characteristics by comparing current dynamic characteristics of each battery cell based on learning data representing a residual distribution for the calculated residual. An energy storage device that stores electrical energy using a battery; and
An abnormality diagnosis device for diagnosing an abnormality of each battery cell using a plurality of measurement data including voltage, current, temperature, and internal resistance of each battery cell included in the battery; including,
The abnormal diagnosis device,
a communication unit communicating with the energy storage device; and
When receiving the plurality of measurement data obtained in units of battery cells from the energy storage device, preprocessing is performed to diagnose an abnormality based on the measurement data, modeling of the preprocessed measurement data is performed, and the modeling result is used. And a control unit for diagnosing an abnormality for each battery cell;
The control unit,
A signal comprising at least one of a statistical method comprising at least one of maximum, minimum, mean, median, standard deviation, variance, moving average, skewness and kurtosis, Fourier transformation and wavelet transformation A processing method, and a dimensionality reduction method including at least one of principal component analysis and t-stochasic nearest neighbor (T-SNE).
Preprocessing the measurement data using at least one method of, detecting a feature combination that can best represent the operation of the battery cell,
The abnormal diagnosis system, characterized in that modeling for comparing the dynamic characteristics of each battery cell or modeling for comparing the current dynamic characteristics of each battery cell with past dynamic characteristics using the preprocessed result. According to claim 8,
a user terminal that receives monitoring information for diagnosis of abnormality of each battery cell from the abnormality diagnosis device and outputs the received monitoring information;
Abnormal diagnosis system further comprising a. receiving, by an abnormality diagnosis device, a plurality of measurement data including voltage, current, temperature, and internal resistance of each battery cell from the energy storage device;
performing pre-processing for diagnosing an abnormality based on the measurement data and modeling the pre-processed measurement data by the abnormality diagnosis apparatus; and
Including; diagnosing, by the abnormal diagnosis apparatus, an abnormality with respect to each battery cell using the modeled result;
The step of modeling the preprocessed measurement data,
A signal comprising at least one of a statistical method comprising at least one of maximum, minimum, mean, median, standard deviation, variance, moving average, skewness and kurtosis, Fourier transformation and wavelet transformation A processing method, and a dimensionality reduction method including at least one of principal component analysis and t-stochasic nearest neighbor (T-SNE).
Preprocessing the measurement data using at least one method of, detecting a feature combination that can best represent the operation of the battery cell,
A method for diagnosing an abnormality based on a battery cell unit, characterized in that modeling for comparing the dynamic characteristics of each battery cell or modeling for comparing the current dynamic characteristics of each battery cell with past dynamic characteristics using the preprocessed result.

Images