WO2023068899A1 - 배터리 팩 내의 이상 징후 셀 검출 장치 및 방법 - Google Patents
배터리 팩 내의 이상 징후 셀 검출 장치 및 방법 Download PDFInfo
- Publication number
- WO2023068899A1 WO2023068899A1 PCT/KR2022/016209 KR2022016209W WO2023068899A1 WO 2023068899 A1 WO2023068899 A1 WO 2023068899A1 KR 2022016209 W KR2022016209 W KR 2022016209W WO 2023068899 A1 WO2023068899 A1 WO 2023068899A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- series data
- cell voltage
- cell
- cells
- voltage time
- Prior art date
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 107
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000007600 charging Methods 0.000 claims abstract description 289
- 238000013136 deep learning model Methods 0.000 claims abstract description 32
- 208000024891 symptom Diseases 0.000 claims description 120
- 230000006399 behavior Effects 0.000 claims description 42
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical group [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 claims description 29
- 229910052744 lithium Inorganic materials 0.000 claims description 29
- 238000004891 communication Methods 0.000 claims description 17
- 238000001556 precipitation Methods 0.000 claims description 16
- 238000005259 measurement Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims 5
- 238000009529 body temperature measurement Methods 0.000 claims 1
- 210000004027 cell Anatomy 0.000 description 389
- 230000006870 function Effects 0.000 description 7
- 230000007423 decrease Effects 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 230000001747 exhibiting effect Effects 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 4
- 230000002238 attenuated effect Effects 0.000 description 4
- 238000004880 explosion Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 229920000642 polymer Polymers 0.000 description 4
- 230000008961 swelling Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010280 constant potential charging Methods 0.000 description 2
- 230000008021 deposition Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 206010000369 Accident Diseases 0.000 description 1
- 208000019901 Anxiety disease Diseases 0.000 description 1
- 210000004460 N cell Anatomy 0.000 description 1
- JDZCKJOXGCMJGS-UHFFFAOYSA-N [Li].[S] Chemical compound [Li].[S] JDZCKJOXGCMJGS-UHFFFAOYSA-N 0.000 description 1
- 230000036506 anxiety Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007850 degeneration Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000003487 electrochemical reaction Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000005022 packaging material Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/165—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values
- G01R19/16566—Circuits and arrangements for comparing voltage or current with one or several thresholds and for indicating the result not covered by subgroups G01R19/16504, G01R19/16528, G01R19/16533
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/3644—Constructional arrangements
- G01R31/3646—Constructional arrangements for indicating electrical conditions or variables, e.g. visual or audible indicators
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
- G01R31/3842—Arrangements for monitoring battery or accumulator variables, e.g. SoC combining voltage and current measurements
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Definitions
- the present invention relates to an apparatus and method for detecting anomaly symptom cells in a battery pack, and more particularly, detects cells with abnormal symptoms by statistically analyzing the difference between measured values and predicted values for voltage behavior of cells in the process of charging a battery pack. It relates to a device and method capable of detecting in advance.
- rechargeable secondary batteries are widely used in various fields ranging from small high-tech electronic devices such as smart phones, laptop computers, and tablet PCs to electric vehicles and energy storage systems (ESS). .
- the present invention was conceived on the background of the prior art as described above, and provides a device and method capable of improving the safety of a battery pack by detecting in advance cells showing abnormal symptoms among cells in a battery pack in the process of charging the battery pack. It has a purpose.
- An apparatus for detecting anomaly symptom cells in a battery pack measures voltage, current, and temperature of first to Nth cells included in the battery pack while the battery pack is being charged. It includes a voltage measuring unit, a current measuring unit, a temperature measuring unit, and a control unit operatively coupled with the voltage measuring unit, current measuring unit, and temperature measuring unit.
- control unit may be configured to determine, as an error, a maximum difference between the second cell voltage time-series data and the predicted cell voltage time-series data for each charging period, for each of the first to Nth cells.
- control unit determines a first average and a first standard deviation for errors of the first to Nth cells for each charging section, and corresponds to a first standardized value for the errors of each of the first to Nth cells for each charging section. (error - first average)/first standard deviation" may be determined, and cells having the first normalized value greater than a first threshold may be detected as abnormal symptom cells in at least one charging period.
- the deep learning model is pre-learned using first cell voltage time-series data and second cell voltage time-series data measured in the first half and second half of each charging section for the first to m-th learning cells, respectively, It may be learned in advance to receive the first cell voltage time series data and output predicted cell voltage time series data with a minimized error with the second cell voltage time series data.
- control unit determines, as an error, a maximum difference between the second cell voltage time-series data and the predicted cell voltage time-series data for each charging period, for each of the first to Nth cells; determining "(error - first average)/first standard deviation" corresponding to a first standardized value for errors of each of the first to Nth cells for each charging section; It may be configured to detect a cell having a first normalization value greater than a first threshold value as an abnormal symptom cell in at least one charging period.
- the first average and the first standard deviation may be predetermined values in the learning process of the deep learning model.
- control unit determines a second average and a second standard deviation of errors of each of a plurality of cells included in the module for each of the first to p-th modules constituting the battery pack; determining "(error - second average)/second standard deviation" corresponding to a second standardized value for errors of each of the first to Nth cells for each charging section; Battery cells having the first normalized value greater than the first threshold value and the second normalized value greater than the second threshold value may be detected as abnormal symptom cells in at least one charging period.
- control unit determines a relative change behavior of the second cell voltage time-series data and the predicted cell voltage time-series data as the charging period is shifted for each of the first to Nth cells. It may be configured to monitor whether it corresponds to the change behavior pattern, and to finally determine the abnormal symptom type of the cell in which the predefined change behavior pattern is confirmed a reference number of times or more.
- the predefined change behavior pattern is such that the second cell voltage time-series data increases faster than the predicted cell voltage time-series data in an early charging period and the predicted cell voltage time-series data increases at a faster rate than the second cell voltage time-series data in a later charging period. It increases faster than the time series data, and the type of anomaly may be lithium precipitation in the negative electrode.
- An apparatus includes a recording storage medium in which data, predefined parameters, programs, or combinations thereof are stored; And it may further include a display.
- the control unit may record identification information on the detected symptom cell in the recording storage medium, output a message indicating that an abnormal symptom cell has been detected in the battery pack through the display, or communicate with an external device. It may be configured to transmit identification information of an anomaly symptom cell to the side.
- the technical problem according to the present invention can be achieved by a battery management system including a device for detecting an abnormal symptom cell in a battery pack described above.
- a method for detecting anomaly cell in a battery pack according to another aspect of the present invention for achieving the above technical problem is, while the battery pack is being charged according to a charging profile having a plurality of charging sections, in each charging section, the first to the th For each of the N cells, (a) acquiring first cell voltage time-series data in the first half of the charging period; (b) applying a deep learning model to the first cell voltage time series data to determine predicted cell voltage time series data in the second half of a charging period; (c) obtaining second cell voltage time-series data in the second half; and (d) determining an error between the second cell voltage time series data and the predicted cell voltage time series data. and (e) detecting a cell having a relatively larger error than other cells in at least one charging period as an abnormal symptom cell.
- a cell with abnormal symptoms can be easily detected by dividing a charging profile of a battery pack into a plurality of charging sections and statistically comparing and analyzing behaviors of measured voltage and predicted voltage for each charging section. Therefore, it is possible to prevent human accidents in advance by detecting abnormal signs directly related to fire or explosion accidents, in particular, serious signs such as lithium precipitation in the negative electrode at an early stage and warning the user.
- the present invention also captures not only lithium precipitation on the negative electrode but also voltage change behavior caused by swelling or micro-short circuits, thereby enabling effective countermeasures against other anomalies.
- FIG. 1 is a block diagram showing a schematic configuration of an apparatus for detecting abnormal symptom cells in a battery pack according to an embodiment of the present invention.
- FIG. 2 is a graph showing an example of a charging profile according to an embodiment of the present invention.
- FIG. 3 is a graph showing voltage change behavior of cells while a battery pack is being charged according to the charging profile illustrated in FIG. 2 .
- FIG 4 is a graph showing an example in which the first half and the second half of a charging section are specifically set in a charging profile according to an embodiment of the present invention.
- FIG. 5 is a graph showing an example of first and second cell voltage time-series data measured in each charging section while the first to m-th learning cells are charged according to a charging profile according to an embodiment of the present invention.
- V k,i (j) shows second cell voltage time-series data V k,i (j) measured in the second half of the charging section 2 of the charging profile for a specific cell in which abnormal signs are detected, and predicted cell voltage time-series data V predicted by a deep learning model. * It is a graph illustrating the changing aspect of "V k,i ( j)-V * k,i (j)" corresponding to k,i (j) and their difference.
- V k,i (j) is a second cell voltage time-series data V k,i (j) measured in the second half of the charging section 1 to 5 of the charging profile for a specific cell in which lithium is deposited on the negative electrode, and the predicted cell voltage predicted by the deep learning model. It is a graph illustrating the changing aspect of "V k ,i (j)-V * k, i (j)" corresponding to time series data V * k,i (j) and their difference.
- FIGS. 8 to 10 are flowcharts illustrating a flow of a method for detecting abnormal symptom cells in a battery pack according to an embodiment of the present invention.
- FIG. 1 is a block diagram showing a schematic configuration of an apparatus for detecting abnormal symptom cells in a battery pack according to an embodiment of the present invention.
- the battery pack 20 is being charged according to a charging profile having a plurality of charging sections by a charging device 30, the battery pack 20 It is a device that detects abnormal symptom cells within.
- An abnormal symptom cell refers to a cell exhibiting an abnormal voltage change behavior different from a normal cell.
- a lithium polymer cell when lithium is precipitated from the negative electrode, a micro short circuit occurs inside the cell, or a swelling phenomenon occurs, the behavior of voltage change during charging is different from that of a normal cell.
- the battery pack 20 includes first through p-th modules 21 .
- the first to p-th modules 21 may be connected in series and/or parallel to each other.
- the a-th module includes first through n a -th cells 22 .
- n a is the total number of cells included in the a-th module.
- the first to n a -th cells 22 may be connected in series and/or parallel to each other.
- the number of cells included in each module may be the same or different.
- the total number of cells included in the battery pack 20 is am.
- the total number of cells included in the battery pack 20 is defined as N, and all cells of the battery pack 20 may be referred to as first through Nth cells.
- the first to Nth cells may be pouch-type lithium polymer cells.
- the present invention is not limited by the type of cell or the type of packaging material. Accordingly, the present invention can be applied to other types of secondary battery cells such as lithium-sulfur batteries and sodium-tium batteries. In addition, the present invention can be applied to cells having structures such as cylindrical cells and prismatic cells.
- the charging device 30 is a device that applies a charging current to the battery pack 20 according to a charging profile having a plurality of charging sections.
- the charging device 30 may be a charging station.
- the charging device 30 may be a Power Converting System (PCS) installed between the power storage device and the power grid.
- PCS is a system that controls the charging and discharging of power storage devices.
- the charging profile is a protocol defining how to change the magnitude of the charging current supplied to the battery pack 20 according to time.
- FIG. 2 is a graph showing an example of a charging profile 40 according to an embodiment of the present invention.
- the charging profile 40 has a plurality of charging sections (1 to 7). Each charging section has a different magnitude and duration of charging current. Duration is a time period during which each charging period is maintained.
- charging profile 40 is a step charging profile.
- the magnitude of the charging current is gradually reduced until the voltage of the cells included in the battery pack 20 reaches the cut-off voltage, and then the magnitude of the charging current according to the constant-voltage charging mode. is regulated
- charging sections 1 to 5 are sections in which the magnitude of the charging current gradually decreases to the cut-off voltage
- charging section 6 is a section to which the constant-voltage charging mode is applied
- charging section 7 is a section in which the battery cell is fully charged. This is a period in which a small charging current close to 0 is applied until the charging current is substantially zero.
- the actual charging section is the charging section 1 to 6.
- FIG. 3 is a graph showing voltage change behavior of cells while the battery pack 20 is being charged according to the charging profile 40 illustrated in FIG. 2 .
- the cells are lithium polymer cells operating at 3.2V to 4.2V.
- the total number of cells is 100, and since the cells do not have the same deterioration degree, voltage change behaviors of the cells also show differences.
- the present invention is not limited by a specific change pattern of the charging profile 40 . Accordingly, the charging profile 40 of FIG. 2 is only an example. If a certain charging profile has a plurality of charging sections, it should be understood that the corresponding profile corresponds to the charging profile according to the present invention.
- the device 10 includes a voltage measuring unit 11, a current measuring unit 12, a temperature measuring unit 13, and a control unit 14.
- the voltage measuring unit 11 measures the voltages of the first to Nth cells included in the battery pack 20 at regular time intervals while the battery pack 20 is being charged according to the charging profile 40 having a plurality of charging sections. is measured, and the cell voltage measurement value is output to the control unit 14.
- the voltage measurement unit 11 may include a voltage measurement circuit known in the art, and since the voltage measurement circuit is widely known, a detailed description thereof will be omitted.
- the current measuring unit 12 measures the magnitude of the charging current at regular time intervals while the battery pack 20 is being charged according to the charging profile 40 having a plurality of charging sections, and converts the measured current value to the control unit 14. output as
- the current measuring unit 12 may be a Hall sensor or a sense resistor that outputs a voltage value corresponding to the magnitude of the current. A voltage value can be converted into a current value according to Ohm's law.
- the current measuring unit 12 may be installed on a line through which the charging current flows.
- current measurement values measured by the current measurement unit 12 correspond to cell current values of all cells included in the battery pack 20 .
- a current measuring unit 12 may be additionally installed at an appropriate point of a line through which a charging current flows to measure a cell current value.
- the temperature measuring unit 13 measures the temperature of the first to Nth cells at regular time intervals while the battery pack 20 is being charged according to the charging profile 40 having a plurality of charging sections, and obtains the measured cell temperature values. It is output to the control unit 14.
- the temperature measuring unit 13 may be a thermocouple or a temperature measuring device that outputs a voltage value corresponding to the temperature.
- a voltage value can be converted into a temperature value using a voltage-temperature conversion lookup table (function).
- the temperature measuring unit 13 may be attached to each of the first through p-th modules 21 .
- the temperature of each module may be regarded as the temperature of cells included in the module.
- the installation of the temperature measuring unit 13 in units of cells is not limited.
- the control unit 14 While the battery pack 20 is being charged according to the charging profile 40, the control unit 14 periodically receives cell voltage values of the first to Nth cells from the voltage measuring unit 11 in each charging section, and receives cell voltage values for each cell. Acquire first and second cell voltage time-series data.
- the cell voltage time series data is a set of cell voltage data continuously measured at a plurality of time points.
- the first cell voltage time-series data is a set of cell voltage data measured in the first half of each charging period.
- the second cell voltage time-series data is a set of cell voltage data measured in the second half of each charging section.
- the boundary between the first half and the second half of the charging section may be arbitrarily set.
- the duration of the i-th charging section is T i
- the first half of the charging section is from the start of the charging section to the time 0.3T i
- the second half of the charging section is from the time 0.3T i to the time T i .
- the duration of the first half is 0.3T i
- the duration of the second half is 0.7T i .
- FIG 4 is a graph showing an example in which the first half and the second half of a charging section are specifically set in the charging profile 40 according to an embodiment of the present invention.
- the vertical solid line is the boundary between charging sections
- the dotted line is the boundary between the first half and the second half of the charging section.
- the durations of the first half and second half of the charging section 1 are 0.3T 1 and 0.7T 1 , respectively.
- the durations of the first half and the second half of the charging section 2 are 0.3T 2 and 0.7T 2 , respectively.
- the durations of the first half and the second half of the charging section 3 are 0.3T 3 and 0.7T 3 , respectively.
- the durations of the first half and the second half of the charging section 4 are 0.3T 4 and 0.7T 4 , respectively.
- the durations of the first half and the second half of the charging section 5 are 0.3T 5 and 0.7T 5 , respectively.
- the durations of the first half and the second half of the charging section 6 are 0.3T 6 and 0.7T 6 , respectively.
- the durations of the first half and the second half of the charging section 7 are 0.3T 7 and 0.7T 7 , respectively.
- the duration of the charging section may be different for each charging section.
- the duration of the first half and the second half of the charging section may be the same for the entire charging section or may be different for each charging section. That is, the boundary between the first half and the second half of the charging section may be set at any time within the duration of the charging section.
- the controller 14 acquires the first cell voltage time series in each charging section.
- a deep learning model can be applied to the data to determine the predicted cell voltage time series data in the second half of the charging period.
- the deep learning model measures the first to mth learning cells having the same specifications as the cells in the battery pack 20 in the first half of each charging period for each learning cell while being charged according to the charging profile 40 described above. It is a pre-learned model that receives first cell voltage time series data and outputs predicted cell voltage time series data having a minimum error with second cell voltage time series data measured in the second half of the corresponding charging section.
- FIG. 5 is a graph showing an example of first and second cell voltage time-series data measured in each charging section while the first to m-th learning cells are charged according to the charging profile 40 according to an embodiment of the present invention. .
- the graph at the top of FIG. 5 is an example of overlapping voltage curves for 100 of m learning cells, and the graph at the bottom of FIG. 5 is the first cell voltage time series measured in charging sections 1 to 7. Data and second cell voltage time-series data are shown as an example.
- thousands to tens of thousands of learning cells with different degrees of degeneration may be used to improve the accuracy and reliability of the model.
- the deep learning model receives the first cell voltage time series data (X) measured for the learning cell in the first half of the charging period 1 and the second cell measured for the learning cell in the second half of the charging period 1.
- the predicted cell voltage time series data having a minimized error with the voltage time series data Y may be pre-learned to be output.
- the first cell voltage time series data (X) measured for the learning cell in the first half of the charging period 2 is input, and the error with the second cell voltage time series data (Y) measured for the learning cell in the second half of the charging period 2 is It may be pre-learned to output minimized predicted cell voltage time-series data.
- the error with the second cell voltage time series data (Y) measured for the learning cell in the second half of the charging section 3 is It may be pre-learned to output minimized predicted cell voltage time-series data.
- the error with the second cell voltage time series data (Y) measured for the learning cell in the second half of the charging section 4 is It may be pre-learned to output minimized predicted cell voltage time-series data.
- the first cell voltage time-series data (X) measured for the learning cell in the first half of the charging section 5 is input, and the error with the second cell voltage time-series data (Y) measured for the learning cell in the second half of the charging section 5 is It may be pre-learned to output minimized predicted cell voltage time-series data.
- the error with the second cell voltage time series data (Y) measured for the learning cell in the second half of the charging section 6 is It may be pre-learned to output minimized predicted cell voltage time-series data.
- the error with the second cell voltage time series data (Y) measured for the learning cell in the second half of the charging section 7 is It may be pre-learned to output minimized predicted cell voltage time-series data.
- the data used for learning the deep learning model may further include first cell voltage time-series data measured in the first half of the charging section as well as time-series data on the cell current and cell temperature measured in the first half of the charging section.
- the deep learning model receives the first cell voltage time series data, cell current time series data, and cell temperature time series data measured in the first half of the charging section, and the error with the second cell voltage time series data measured in the second half of the charging section It can be learned to output minimized predicted cell voltage time-series data.
- any deep learning model based on an artificial neural network suitable for predicting the time-series behavior of cell voltage may be used without limitation.
- the artificial neural network may be a recurrent neural network (RNN), a convolution neural network (CNN), or the like.
- RNN recurrent neural network
- CNN convolution neural network
- the present invention is not limited by the type of artificial neural network.
- the control unit 14 obtains the second cell voltage time series data and the predicted cell voltage time series data in the second half of each charging section for each of the first to Nth cells in the battery pack 20, and then obtains the second cell voltage time series data ( It is possible to determine the error between the measured value) and the predicted cell voltage time-series data (predicted value).
- the control unit 14 also monitors an error between the second cell voltage time-series data determined in each charging section and the predicted cell voltage time-series data for each of the first to Nth cells, and determines that the error is relatively large in at least one charging section.
- a cell can be detected as an abnormal symptom cell.
- the control unit 14 uses Equation 1 below to determine the second cell voltage time-series data V k,1 (j) and the predicted cell voltage time-series data V k,1 * for each charging section.
- the error E k,1 between (j) can be calculated.
- E k,i max(
- E k,i Error of the k-th charging section for the i-th battery cell
- V k,i (j) Second cell voltage measured in the second half of the k-th charging section for the i-th battery cell
- V * k,i (j) predicted cell voltage predicted by the j-th battery cell in the second half of the k-th charging period for the i-th battery cell by the deep learning model
- Num measure The total number of cell voltage measurements in the second half of the k-th charging section or the total number of cell voltage predictions using a deep learning model in the second half of the k-th charging section
- the controller 14 also calculates errors E k,2 , E k,3 , E k,n between the second cell voltage time-series data and the predicted cell voltage time-series data for each charging section for the second to Nth cells. can be calculated
- V k,i (j) shows second cell voltage time-series data V k,i (j) measured in the second half of the charging period 2 of the charging profile 40 for a specific cell in which abnormal symptoms are detected, and the predicted cell voltage predicted by the deep learning model. It is a graph illustrating the changing aspect of "V k ,i (j)-V * k, i (j)" corresponding to time series data V * k,i (j) and their difference.
- the black solid line is V k,i (j), and the black dotted line is V * k,i (j).
- V k,i (j)-V * k,i (j) increases to a significant level in a cell with abnormal symptoms in a specific charging period. Therefore, the error E k,i can be used as a parameter for detecting abnormal symptom cells.
- the second cell voltage time-series data and the predicted cell voltage time-series data of cells without abnormal signs are shown as gray lines and are not well distinguished because they coincide with each other.
- the control unit 14 may also determine a first average Avr k and a first standard deviation ⁇ k for the errors E k,i of the first to Nth cells for each charging section using Equation 2 below.
- Avr k mean(E k,i ), 1 ⁇ i ⁇ N
- N the number of cells in the battery pack
- E k,i Error of the k-th charging section for the i-th battery cell
- the controller 14 may also determine a first standardized value Std_Value k,i for the error E k,i of each of the 1st to Nth cells using Equation 3 below for each charging section.
- the number of charging sections is Num charge
- the number of first standardized values Std_Value k,i for each cell is Num charge .
- the control unit 14 may also detect cells having a first normalized value Std_Value k,i greater than a preset first threshold in at least one charging period among the first to Nth cells of the battery pack 20 as abnormal symptom cells. .
- the first standardized value Std_Value k,i is somewhat different from the average error value (Avr k ) when the error E k,i of the i-th cell determined in the k-th charging section is based on the standard deviation ( ⁇ k ) of the error. It is a factor indicating whether
- the first threshold may be set to 3 or more, more preferably 4 or more, and more preferably 4.5 or more.
- the first average Avr k and the first standard deviation ⁇ k of each charging section may be determined in advance in the learning process of the deep learning model. That is, after the learning of the deep learning model is completed, the first to m-th learning cells are charged according to the charging profile 40 having a plurality of charging intervals, and the first cell voltage time-series data, the second cell voltage time-series data, and prediction After collecting the cell voltage time series data, the error E k,i of each charging section can be determined for each learning cell.
- the average and standard deviation of the error E k,i calculated using Equation 2 for each charging section may be preset as the first average Avr k and the first standard deviation ⁇ k , respectively.
- control unit 14 may determine the second average Avr k, a and the second standard deviation ⁇ k,a of the error E k ,i in units of modules using Equation 4 below.
- a second average Avr k,a and a second standard deviation ⁇ k,a are determined for each filling interval.
- Avr k,a mean(E k,i@a ), 1 ⁇ i ⁇ n a , 1 ⁇ k ⁇ Num charge
- ⁇ k,a std(E k,i ), 1 ⁇ i ⁇ n a , 1 ⁇ k ⁇ Num charge
- n a total number of cells included in the a-th module
- E k,i@a Error of the k-th charging section for the i-th cell in the a-th module
- the control unit 14 may also determine a second standardized value Std_Value * k,i within the module for the error E k,i of each of the 1st to Nth cells using Equation 5 for each charging section.
- the number of second standardized values Std_Value * k,i in the module for each cell is Num charge .
- Std_Value * k,i (E k,i - Avr k,a )/ ⁇ k,a , 1 ⁇ i ⁇ N, 1 ⁇ a ⁇ p, 1 ⁇ k ⁇ Num charge
- N the total number of cells
- Avr k,a Second average defined according to the module in which the cell is included
- ⁇ k,a Second standard deviation defined according to the module in which the cell is included
- the controller 14 also classifies cells in which the first normalized value Std_Value k,i is greater than the first threshold value and the second normalized value Std_Value * k,i greater than the second threshold value in the module in at least one charging section as abnormal symptom cells. can be configured to detect
- the second standardized value Std_Value * k,i is the average value within the module of error ( Avr k, when the error E k,i of the cell determined in the k-th charging section is based on the standard deviation within the module of error ( ⁇ k,a ). ,a ) is a factor indicating how far away it is from.
- the second threshold may be smaller than the first threshold.
- the second threshold may be preferably set to 3.0 or less, more preferably 2.5 or less.
- controller 14 may additionally perform logic for detecting the type of anomaly of the battery cell.
- the controller 14 determines the relative change behavior of the second cell voltage time-series data of the first to Nth cells measured in the second half of each charging section and the predicted cell voltage time-series data predicted in the second half of each charging section. As this is shifted, it is possible to monitor whether a predefined change behavior pattern for each symptom type is displayed. In addition, the controller 14 may identify the corresponding cell as an abnormal symptom cell and determine the abnormal symptom type when a predefined change behavior is detected in the same cell more than a reference number of times while a plurality of charging cycles are in progress.
- second cell voltage time series data increases faster than predicted cell voltage time series data in a charging period at an early stage of charging. This is because the potential of the negative electrode increases at the beginning of charging when lithium is deposited on the negative electrode. Since the cell voltage corresponds to the difference between the anode potential and the cathode potential, the change slope of the cell voltage increases as the cathode potential increases. As a result, the second cell voltage increases faster than the predicted cell voltage in the initial charging period. Since the predicted cell voltage is the voltage predicted by the deep learning model, it shows a voltage change behavior close to that of a normal cell at the beginning of charging, so the increase in cell voltage is not steep.
- the predicted cell voltage time-series data increases faster than the second cell voltage time-series data.
- the potential of the negative electrode gradually decreases as charging progresses. This is because when lithium is precipitated from the negative electrode, the amount of lithium participating in the electrochemical reaction decreases and the decrease in potential of the negative electrode is attenuated.
- the decrease in potential of the cathode is attenuated, the increase in cell voltage is also attenuated correspondingly.
- the predicted cell voltage increases faster than the second cell voltage in the latter charging period. Since the predicted cell voltage is the voltage predicted by the deep learning model, it shows a voltage change behavior close to that of a normal cell even in the latter half of the charge, so the decrease in potential of the negative electrode is not attenuated.
- V k,i (j) is a second cell voltage time-series data V k,i (j) measured in the second half of the charging section 1 to 5 of the charging profile 40 for a specific cell in which lithium is actually deposited on the negative electrode, predicted by a deep learning model It is a graph illustrating changes in “V k , i ( j)-V * k,i (j)” corresponding to the predicted cell voltage time series data V * k,i (j) and their difference. In each graph, the horizontal axis is time (seconds) and the vertical axis is voltage (milli-volt).
- the black solid line is V k,i (j), and the black dotted line is V * k,i (j).
- the second cell voltage time-series data increases faster than the predicted cell voltage time-series data in the initial period 1 of charging.
- the predicted cell voltage time-series data increases faster than the second cell voltage time-series data.
- the second cell voltage time-series data and the predicted cell voltage time-series data of cells without abnormal symptoms are indicated by gray lines and are not well distinguished because they coincide with each other.
- control unit 14 increases the second cell voltage time-series data faster than the predicted cell voltage time-series data in the first charging period among the first to Nth cells, and the predicted cell voltage time-series data in the later charging period. If a cell exhibiting a behavior of increasing faster than the second cell voltage time-series data is identified, it can be identified as having an abnormal sign of lithium precipitation on the anode of the corresponding cell.
- the control unit 14 while charging the battery pack 20 according to the charging profile 40 including a plurality of charging sections, the control unit 14 generates second cell voltage time-series data in an initial charging section among the first to Nth cells. If a cell that increases faster than the predicted cell voltage time series data and shows a behavior in which the predicted cell voltage time series data increases faster than the second cell voltage time series data in the latter charging period is identified, an anomaly symptom count is counted for the corresponding cell each time. can be incremented by 1.
- the control unit 14 may also finally determine that there is an abnormal symptom in which lithium is deposited on the negative electrode of the cell when the abnormal symptom count is equal to or greater than the reference number.
- the device 10 may further include a recording storage medium 15 and a display 16 storing data, predefined parameters, programs, or a combination thereof.
- control unit 14 may record identification information of the abnormal symptom cell and/or information on the abnormal symptom type together with a time stamp in the recording storage medium 15 .
- the identification information of the symptom cell includes a model code of the battery pack 20, a module code including the symptom cell, a production lot number of the symptom cell, or a combination thereof.
- Information on the type of abnormality may include a diagnostic code indicating lithium precipitation of the anode.
- the controller 14 may also be configured to output a message indicating that an abnormal symptom cell is detected in the battery pack 20 through the display 16 when detecting an abnormal symptom cell according to the above-described embodiment.
- the display 16 When the battery pack 20 is mounted in the electric vehicle, the display 16 may be an instrument panel of the electric vehicle or an integrated vehicle control display. In another example, when the battery pack 20 is mounted in the power storage device, the display 16 may be a display included in the integrated control computer of the power storage device. However, the present invention is not limited by the type of display.
- the device 10 may further include a communication interface 17.
- the control unit 14 may transmit identification information about the symptom cell and/or information about the symptom type to an external device through the communication interface 17 .
- the communication interface 17 supports wired or wireless communication.
- the communication interface 17 may support data transmission/reception by CAN (Controller Area Network), Daisy Chain, RS-232, and the like.
- the communication interface 17 may support data transmission and reception through short-range wireless communication such as Wi-Fi, Bluetooth, and ZigBee.
- the communication interface 17 can support wide-area data transmission and reception through wired and wireless Internet, base station communication, and satellite communication.
- the external device may be the charging device 30 .
- the external device may be a cloud server that collects state information of the battery pack 20 .
- the external device may be a diagnostic device for checking performance of the battery pack 20 .
- the controller 14 selectively uses a processor known in the art, an application-specific integrated circuit (ASIC), other chipsets, logic circuits, registers, communication modems, data processing devices, etc. to execute various control logics.
- ASIC application-specific integrated circuit
- the type of the recording medium 15 is not particularly limited as long as it is a medium capable of recording and erasing information.
- the recording medium 15 may be a hard disk, RAM, ROM, EEPROM, register or flash memory.
- the recording medium 15 stores a program including control logic executed by the control unit 14 and/or data generated when the control logic is executed and predefined lookup tables, functions, parameters, chemical/physical/ Electrical constants, etc. may be stored and/or updated and/or erased and/or transmitted.
- At least one or more control logics of the control unit 14 are combined, and the combined control logics may be written in a computer readable code system and recorded in the recording medium 15 .
- the code system may be distributed and stored and executed on computers connected through a network.
- functional programs, codes and code segments for implementing the combined control logics can be easily inferred by programmers skilled in the art to which the present invention belongs.
- Device 10 may be included in a battery management system or a battery diagnostic system.
- the battery management system is a system that controls overall operation of the battery pack 20 .
- Such a battery management system may be an integrated control system included in a load device on which the battery pack 20 is mounted, such as an electric vehicle or a power storage device.
- the device 10 according to the present invention may be included as part of other devices or systems other than a battery management system or a battery diagnosis system, if necessary.
- the charging device 30 applies a charging current to the battery pack 20 according to the charging profile 40 including a plurality of charging sections.
- the charging device 30 may be a charging station of an electric vehicle or a PCS of a power storage device.
- Application of the charging current may be initiated at the request of the control unit 14 . That is, the controller 14 may recognize that the charging cable is connected to the high potential line and the low potential line of the battery pack 20 and request the charging device 30 to start charging. Alternatively, application of the charging current may be automatically initiated when the charging device 30 is connected to the battery pack 20 .
- step S20 the control unit 14 measures the voltage of the first to Nth cells in the battery pack 20 in each charging section while the battery pack 20 is being charged according to the charging profile 40. ), the cell voltage value is periodically input, and the first and second cell voltage time-series data are obtained.
- step S30 the control unit 140 separates the first and second cell voltage time-series data obtained through the voltage measurement unit 11 in each charging period, and separates each charging period for the first to Nth cells.
- a pre-learned deep learning model is applied to determine the predicted cell voltage time series data in the second half of the charging period.
- step S40 the control unit 14 calculates an error E k,i between the second cell voltage time-series data and the predicted cell voltage time-series data in each charging section using Equation 1 for the first to Nth cells. do.
- the number of charging sections is Num charge
- the total number of errors E k,i is Num charge *N.
- step S50 the control unit 14 determines a first average Avr k and a first standard deviation ⁇ k for the errors E k,i of the first to Nth cells for each charging section by using Equation 2.
- step S60 the control unit 14 determines a first standardized value Std_Value k,i for the error E k,i of each of the 1st to Nth cells by using Equation 3 for each charging section.
- the number of charging sections is Num charge
- the total number of first standardized values Std_Value k,i is Num charge *N.
- control unit 14 may detect a cell having a first normalized value Std_Value k,i greater than a preset first threshold as an abnormal symptom cell in at least one charging section.
- step S70, step S80 or step S90 may be selectively performed.
- the control unit 14 may record identification information of the abnormal symptom cell and/or information on the abnormal symptom type together with a time stamp in the recording storage medium 15.
- the identification information of the symptom cell includes a model code of the battery pack 20, a module code to which the symptom cell belongs, a production lot number of the symptom cell, or a combination thereof.
- Information on the type of abnormality may include a diagnostic code indicating lithium precipitation of the anode.
- step S90 when detecting an abnormal symptom cell, the controller 14 may output a message indicating that an abnormal symptom cell has been detected in the battery pack 20 through the display 16 .
- the message When the message is output, the user can stop using the battery pack 20 and replace the battery pack 20 or request a detailed inspection of the battery pack 20 at a service center or after service center.
- step S50 can be omitted.
- the first average Avr k and the first standard deviation ⁇ k may be determined in advance during the learning process of the deep learning model. That is, after the learning of the deep learning model is completed, the error E k,i of each charging section may be determined for each learning cell while charging the first to mth learning cells according to the charging profile 40 .
- the average and standard deviation of the error E k,i may be calculated using Equation 2 for each charging section, and the calculated values may be preset as the first average Avr k and the first standard deviation ⁇ k , respectively.
- the first average Avr k and the second standard deviation ⁇ k set in the learning process of the deep learning model may be pre-stored in the recording medium 15 and referred to by the control unit 14 when step S60 proceeds.
- steps after step S60 can be modified as shown in FIG. 9 .
- step S100 the control unit 14 can determine the second average Avr k,a and the second standard deviation ⁇ k ,a of the error E k,i in each charging section in units of modules using Equation 4. there is.
- step S110 the control unit 14 determines a second standardized value Std_Value * k,i within the module for the error E k,i of each of the first to Nth cells for each charging section by using Equation 5. do.
- the number of charging sections is Num charge
- the number of second standardized values Std_Value * k,i in the module for each cell is Num charge .
- step S120 the control unit 14 determines that the first standardized value Std_Value k,i in at least one charging section is greater than the first threshold and at the same time the second standardized value Std_Value * k,i within the module is greater than the second threshold. Large cells can be detected as abnormal symptom cells.
- the second threshold may be smaller than the first threshold.
- the first threshold may be 3.0 or more, preferably 4.0 or more, and more preferably 4.5 or more.
- the second threshold may be preferably set to 3.0 or less, more preferably 2.5 or less.
- step S120, step S80 or step S90 may be performed substantially the same as in the foregoing embodiment.
- the controller 14 when the controller 14 detects an abnormal symptom cell in the battery pack 20, it may record identification information of the abnormal symptom cell and/or information about the abnormal symptom type together with a time stamp in the recording storage medium 15. there is.
- the identification information of the symptom cell includes a model code of the battery pack 20, a module code to which the symptom cell belongs, a production lot number of the symptom cell, or a combination thereof.
- Information on the type of abnormality may include a diagnostic code indicating lithium precipitation of the anode.
- the controller 14 may also output, through the display 16 , a message indicating that an abnormal symptom cell has been detected in the battery pack 20 when detecting an abnormal symptom cell in the battery pack 20 .
- the method for detecting an anomaly symptom cell in a battery pack according to the present invention may further include steps of identifying an anomaly symptom type of a battery cell.
- step S130 the control unit 14 uses the second cell voltage time-series data of the first to Nth cells measured in the second half of each charging section and the predicted cell voltage predicted in the second half of each charging section. It is possible to monitor whether the relative change behavior of the time-series data exhibits a predefined change behavior pattern for each type of anomaly as the charging period is shifted.
- control unit 14 increases the second cell voltage time-series data faster than the predicted cell voltage time-series data in the first charging period among the first to Nth cells, and the predicted cell voltage time-series data in the later charging period. Whether a battery cell exhibiting a behavior pattern that increases faster than the second cell voltage time-series data exists may be monitored.
- step S140 the control unit 14 determines whether the relative change behavior of the second cell voltage time-series data and the predicted cell voltage time-series data corresponds to a change behavior pattern predefined according to the abnormal symptom type.
- step S150 proceeds. Conversely, if the result of step S140 is NO, the process returns to step S130.
- step S140 If the result of step S140 is YES, the control unit 140 identifies the abnormal symptom type corresponding to the predefined change behavior pattern in step S150.
- the control unit 14 increases the second cell voltage time-series data faster than the predicted cell voltage time-series data in the first charging period among the first to Nth cells, and the predicted cell voltage time-series data in the later charging period.
- a battery cell exhibiting a behavior pattern that increases faster than the second cell voltage time series data it can be identified as having an abnormal sign of lithium precipitation on the negative electrode of the corresponding cell.
- step S160 the control unit 14 increases the abnormal symptom count by 1 for the corresponding cell in which the abnormal symptom type (eg, negative electrode lithium deposition) is identified.
- the abnormal symptom type eg, negative electrode lithium deposition
- step S170 the control unit 14 determines whether the abnormal symptom count exceeds the reference number.
- step S180 proceeds. Conversely, if the determination in S170 is NO, the process returns to step S130.
- step S170 determines the abnormal symptom type (eg, negative electrode lithium deposition) for the cell whose abnormal symptom count exceeds the reference number in step S180.
- the abnormal symptom type eg, negative electrode lithium deposition
- step S80 or step S90 may be performed as in the above-described embodiment.
- the controller 14 records the identification information of the abnormal symptom cell and information about the abnormal symptom type together with a time stamp. It can be recorded on the storage medium 15.
- the identification information of the symptom cell includes a model code of the battery pack 20, a module code to which the symptom cell belongs, a production lot number of the symptom cell, or a combination thereof.
- Information about the type of abnormality may include a diagnostic code indicating lithium precipitation of the anode.
- the control unit 14 When the abnormal symptom cell is identified in the battery pack 20 and the abnormal symptom type is finally determined, the control unit 14 also sends a message indicating that the abnormal symptom cell has been detected in the battery pack 20 together with information about the abnormal symptom type. It can be output through the display 16.
- control unit 14 may transmit identification information about the abnormal cell and/or information about the abnormal symptom type to an external device through the communication interface 17 .
- the communication interface 17 may support wired communication or wireless communication.
- the external device may be the charging device 30 .
- the external device may be a cloud server that collects state information of the battery pack 20 .
- the external device may be a diagnostic device for checking performance of the battery pack 20 .
- the abnormal symptom type may include other abnormal symptom types such as cell swelling and internal short circuit in addition to lithium precipitation of the negative electrode.
- each abnormal symptom type it is possible to easily confirm through experiment what relative change behavior pattern the second cell voltage time-series data and the predicted cell voltage time-series data show according to the shift of the charging section. It is self-evident to those skilled in the art.
- the present invention it is possible to easily detect cells with abnormal symptoms by dividing the charging profile of the battery pack 20 into a plurality of charging sections and statistically comparing and analyzing behaviors of measured voltage and predicted voltage for each charging section. . Therefore, it is possible to prevent human accidents in advance by detecting abnormal signs directly related to fire or explosion accidents, in particular, serious signs such as lithium precipitation in the negative electrode at an early stage and warning the user.
- the present invention also captures not only lithium precipitation on the negative electrode but also voltage change behavior caused by swelling or micro-short circuits, so that other symptoms can be effectively dealt with.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Secondary Cells (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
Description
Claims (17)
- 배터리 팩이 충전되는 동안 배터리 팩에 포함된 제1 내지 제N셀에 대한 전압, 전류 및 온도를 측정하는 전압 측정부, 전류 측정부 및 온도 측정부 및상기 전압 측정부, 전류 측정부 및 온도 측정부와 동작 가능하게 결합된 제어부를 포함하고,상기 제어부는, 배터리 팩이 복수의 충전 구간을 가진 충전 프로파일에 따라 충전되는 동안, 각 충전구간에서 제1 내지 제N셀의 각각에 대해, 충전 구간의 전반부에서 상기 전압 측정부를 통해 제1셀전압 시계열 데이터를 취득하고; 상기 제1셀전압 시계열 데이터에 딥러닝 모델을 적용하여, 충전 구간의 후반부에서 예측 셀전압 시계열 데이터를 결정하고; 상기 후반부에서 상기 전압 측정부를 통해 제2셀전압 시계열 데이터를 취득하고, 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터의 오차를 결정하고,상기 제어부는, 적어도 하나 이상의 충전구간에서 다른 셀들보다 오차가 상대적으로 큰 셀을 이상 징후 셀로 검출하도록 구성된, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제1항에 있어서,상기 제어부는, 제1 내지 제N셀 각각에 대해, 충전구간 별로 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터 간의 최대 차이를 오차로 결정하고; 충전구간 별로 제1 내지 제N셀의 오차에 대한 제1평균과 제1표준편차를 결정하고; 충전구간별로 제1 내지 제N셀 각각의 오차에 대한 제1표준화 값에 해당하는 "(오차 - 제1평균)/제1표준편차"를 결정하고, 적어도 하나 이상의 충전구간에서 상기 제1표준화 값이 제1임계치보다 큰 셀을 이상 징후 셀로 검출하도록 구성된, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제1항에 있어서,상기 딥러닝 모델은, 제1 내지 제m학습 셀들에 대해 각 충전구간의 전반부 및 후반부에서 각각 측정된 제1셀전압 시계열 데이터 및 제2셀전압 시계열 데이터를 이용하여 미리 학습된 것으로서, 제1셀전압 시계열 데이터를 입력 받아 제2셀전압 시계열 데이터와의 오차가 최소화된 예측 셀전압 시계열 데이터를 출력하도록 미리 학습된 것인, 임을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제3항에 있어서,상기 제어부는, 제1 내지 제N셀 각각에 대해, 충전구간 별로 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터 간의 최대 차이를 오차로 결정하고; 충전구간별로 제1 내지 제N셀 각각의 오차에 대한 제1표준화 값에 해당하는 "(오차 - 제1평균)/제1표준편차"를 결정하고, 적어도 하나 이상의 충전구간에서 상기 제1표준화 값이 제1임계치보다 큰 셀을 이상 징후 셀로 검출하도록 구성되고,상기 제1평균 및 상기 제1표준편차는 상기 딥러닝 모델의 학습 과정에서 미리 결정된 값인, 임을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제2항 또는 제4항에 있어서,상기 배터리 팩은 직렬 및/또는 병렬로 연결된 제1 내지 제p모듈을 포함하고,상기 제어부는, 제1 내지 제p모듈 각각에 대해, 모듈 내에 포함된 복수 셀 각각의 오차에 대한 제2평균과 제2표준편차를 결정하고; 충전구간별로 제1 내지 제N셀 각각의 오차에 대한 제2표준화 값에 해당하는 "(오차 - 제2평균)/제2표준편차"를 결정하고, 적어도 하나 이상의 충전구간에서 상기 제1표준화 값이 제1임계치보다 크고 상기 제2표준화 값이 제2임계치보다 큰 배터리 셀을 이상 징후 셀로 검출하도록 구성된, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제1항에 있어서,상기 제어부는, 상기 제1 내지 제N셀 각각에 대해,충전구간이 쉬프트됨에 따라 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터의 상대적 변화 거동이 이상 징후 유형 별로 미리 정의된 변화 거동 패턴에 대응되는지 모니터하고, 미리 정의된 변화 거동 패턴이 기준 횟수 이상 확인된 셀의 이상 징후 유형을 최종 결정하도록 구성된, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제6항에 있어서,상기 미리 정의된 변화 거동 패턴은, 초반 충전구간에서 상기 제2셀전압 시계열 데이터가 상기 예측 셀전압 시계열 데이터보다 빠르게 증가하고 후반 충전구간에서는 상기 예측 셀전압 시계열 데이터가 상기 제2셀전압 시계열 데이터보다 빠르게 증가하는 것이고,상기 이상 징후 유형은 음극에서의 리튬 석출인, 임을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제1항에 있어서,데이터, 사전 정의된 파라미터, 프로그램 또는 이들의 조합이 저장되는 기록저장매체; 및 디스플레이를 더 포함하고,상기 제어부는, 상기 검출된 이상 징후 셀에 관한 식별정보를 상기 기록저장매체에 기록하거나, 또는 배터리 팩 내에 이상 징후 셀이 검출되었음을 나타내는 메시지를 상기 디스플레이를 통해 출력하거나, 또는 통신을 통해 외부 디바이스 측으로 이상 징후 셀의 식별정보를 전송하도록 구성된, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 장치.
- 제1항에 따른 배터리 팩 내의 이상 징후 셀 검출 장치를 포함하는 배터리 관리 시스템.
- 배터리 팩이 복수의 충전구간을 가진 충전 프로파일에 따라 충전되는 동안, 각 충전구간에서 제1 내지 제N셀의 각각에 대해,(a) 충전 구간의 전반부에서 제1셀전압 시계열 데이터를 취득하는 단계;(b) 상기 제1셀전압 시계열 데이터에 딥러닝 모델을 적용하여, 충전 구간의 후반부에서 예측 셀전압 시계열 데이터를 결정하는 단계;(c) 상기 후반부에서 제2셀전압 시계열 데이터를 취득하는 단계; 및(d) 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터의 오차를 결정하는 단계; 및(e) 적어도 하나 이상의 충전구간에서 다른 셀들보다 오차가 상대적으로 큰 셀을 이상 징후 셀로 검출하는 단계;를 포함하는, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
- 제10항에 있어서,상기 (d) 단계는, 상기 제1 내지 제N셀 각각에 대해, 충전구간별로 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터 간의 최대 차이를 오차로 결정하는 단계이고,상기 (e) 단계는,(e1) 충전구간별로 제1 내지 제N셀의 오차에 대한 제1평균과 제1표준편차를 결정하는 단계;(e2) 충전구간별로 제1 내지 제N셀 각각의 오차에 대한 제1표준화 값에 해당하는 "(오차 - 제1평균)/제1표준편차"를 결정하는 단계; 및(e3) 적어도 하나 이상의 충전구간에서 상기 제1표준화 값이 제1임계치보다 큰 셀을 이상 징후 셀로 검출하는 단계;를 포함하는, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
- 제10항에 있어서,상기 딥러닝 모델은, 제1 내지 제m학습 셀들에 대해 각 충전구간의 전반부 및 후반부에서 각각 측정된 제1셀전압 시계열 데이터 및 제2셀전압 시계열 데이터를 이용하여 미리 학습된 것으로서, 제1셀전압 시계열 데이터를 입력 받아 제2셀전압 시계열 데이터와의 오차가 최소화된 예측 셀전압 시계열 데이터를 출력하도록 미리 학습된 것인, 임을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
- 제12항에 있어서,상기 (d) 단계는, 제1 내지 제N셀 각각에 대해, 충전구간별로 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터 간의 최대 차이를 오차로 결정하는 단계이고,상기 (e) 단계는,(e1) 충전구간별로 제1 내지 제N셀 각각의 오차에 대한 제1표준화 값에 해당하는 "(오차 - 제1평균)/제1표준편차"를 결정하는 단계; 및(e2) 적어도 하나의 충전구간에서 상기 제1표준화 값이 제1임계치보다 큰 셀을 이상 징후 셀로 검출하는 단계를 포함하고,상기 제1평균 및 상기 제1표준편차는 상기 딥러닝 모델의 학습 과정에서 미리 결정된 값인, 임을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
- 제11항 또는 제13항에 있어서,상기 배터리 팩은 직렬 및/또는 병렬로 연결된 제1 내지 제p모듈을 포함하고,제1 내지 제p모듈 각각에 대해, 모듈 내에 포함된 복수 셀 각각의 오차에 대한 제2평균과 제2표준편차를 결정하는 단계;충전구간별로 제1 내지 제N셀 각각의 오차에 대한 제2표준화 값에 해당하는 "(오차 - 제2평균)/제2표준편차"를 결정하는 단계; 및상기 제1표준화 값이 제1임계치보다 크고 상기 제2표준화 값이 제2임계치보다 큰 셀을 이상 징후 셀로 검출하는 단계;를 더 포함하는, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
- 제10항에 있어서,상기 제1 내지 제N셀 각각에 대해, 충전구간이 쉬프트됨에 따라 상기 제2셀전압 시계열 데이터와 상기 예측 셀전압 시계열 데이터의 상대적 변화 거동이 이상 징후 유형 별로 미리 정의된 변화 거동 패턴에 대응되는지 모니터하는 단계; 및미리 정의된 변화 거동 패턴이 기준 횟수 이상 확인된 셀의 이상 징후 유형을 최종 결정하는 단계;를 더 포함하는, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
- 제15항에 있어서,상기 미리 정의된 변화 거동 패턴은 초반 충전구간에서 상기 제2셀전압 시계열 데이터가 상기 예측 셀전압 시계열 데이터보다 빠르게 증가하고 후반 충전구간에서는 상기 예측 셀전압 시계열 데이터가 상기 제2셀전압 시계열 데이터보다 빠르게 증가하는 것이고,상기 이상 징후 유형은 음극에서의 리튬 석출인, 임을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
- 제10항에 있어서,상기 검출된 이상 징후 셀에 관한 식별정보를 기록매체에 기록하는 단계;배터리 팩 내에 이상 징후 셀이 검출되었음을 나타내는 메시지를 디스플레이를 통해 출력하는 단계; 또는통신을 통해 외부 디바이스 측으로 이상 징후 셀의 식별정보를 전송하는 단계;를 더 포함하는, 것을 특징으로 하는 배터리 팩 내의 이상 징후 셀 검출 방법.
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2023538070A JP2024500893A (ja) | 2021-10-22 | 2022-10-21 | バッテリーパック内の異常兆候セルの検出装置及び方法 |
EP22884120.1A EP4321884A1 (en) | 2021-10-22 | 2022-10-21 | Apparatus of detecting abnormal cell in battery pack and method thereof |
CN202280013690.7A CN116829966A (zh) | 2021-10-22 | 2022-10-21 | 检测电池组中的异常电芯的设备及其方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2021-0142202 | 2021-10-22 | ||
KR1020210142202A KR20230057894A (ko) | 2021-10-22 | 2021-10-22 | 배터리 팩 내의 이상 징후 셀 검출 장치 및 방법 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023068899A1 true WO2023068899A1 (ko) | 2023-04-27 |
Family
ID=86059437
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2022/016209 WO2023068899A1 (ko) | 2021-10-22 | 2022-10-21 | 배터리 팩 내의 이상 징후 셀 검출 장치 및 방법 |
Country Status (5)
Country | Link |
---|---|
EP (1) | EP4321884A1 (ko) |
JP (1) | JP2024500893A (ko) |
KR (1) | KR20230057894A (ko) |
CN (1) | CN116829966A (ko) |
WO (1) | WO2023068899A1 (ko) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116466241A (zh) * | 2023-05-06 | 2023-07-21 | 重庆标能瑞源储能技术研究院有限公司 | 一种单体电池热失控定位方法 |
CN117310543A (zh) * | 2023-11-29 | 2023-12-29 | 中国华能集团清洁能源技术研究院有限公司 | 电池异常诊断方法及装置 |
CN117406007A (zh) * | 2023-12-14 | 2024-01-16 | 山东佰运科技发展有限公司 | 一种充电桩充电数据检测方法及系统 |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117578664B (zh) * | 2023-11-28 | 2024-06-11 | 浙江智格科技有限公司 | 基于智能物联网的换电柜安全控制方法及系统 |
CN117691722B (zh) * | 2024-02-01 | 2024-04-09 | 中山市福瑞特科技产业有限公司 | 一种锂电池充电安全监测预警方法及系统 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019521491A (ja) * | 2016-09-26 | 2019-07-25 | エルジー・ケム・リミテッド | 人工知能型燃料電池システム |
KR20200119383A (ko) * | 2019-03-26 | 2020-10-20 | 서강대학교산학협력단 | 인공 지능에 기반하여 배터리의 상태를 추정하는 장치 및 방법 |
WO2020240324A1 (ja) * | 2019-05-24 | 2020-12-03 | 株式会社半導体エネルギー研究所 | 二次電池の内部抵抗の推定方法及び二次電池の異常検知システム |
KR20210016154A (ko) * | 2019-07-31 | 2021-02-15 | 주식회사 에스제이 테크 | 머신러닝을 이용한 배터리 진단 방법 |
KR20210116801A (ko) * | 2020-03-16 | 2021-09-28 | 주식회사 로보볼트 | 신경망 기반의 배터리 잔존 수명 예측 방법 및 장치 |
KR20210142202A (ko) | 2019-06-12 | 2021-11-24 | 제지앙 둔안 아트피셜 인바이런먼트 컴퍼니 리미티드 | 전자 팽창 밸브 |
-
2021
- 2021-10-22 KR KR1020210142202A patent/KR20230057894A/ko active Search and Examination
-
2022
- 2022-10-21 JP JP2023538070A patent/JP2024500893A/ja active Pending
- 2022-10-21 CN CN202280013690.7A patent/CN116829966A/zh active Pending
- 2022-10-21 EP EP22884120.1A patent/EP4321884A1/en active Pending
- 2022-10-21 WO PCT/KR2022/016209 patent/WO2023068899A1/ko active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019521491A (ja) * | 2016-09-26 | 2019-07-25 | エルジー・ケム・リミテッド | 人工知能型燃料電池システム |
KR20200119383A (ko) * | 2019-03-26 | 2020-10-20 | 서강대학교산학협력단 | 인공 지능에 기반하여 배터리의 상태를 추정하는 장치 및 방법 |
WO2020240324A1 (ja) * | 2019-05-24 | 2020-12-03 | 株式会社半導体エネルギー研究所 | 二次電池の内部抵抗の推定方法及び二次電池の異常検知システム |
KR20210142202A (ko) | 2019-06-12 | 2021-11-24 | 제지앙 둔안 아트피셜 인바이런먼트 컴퍼니 리미티드 | 전자 팽창 밸브 |
KR20210016154A (ko) * | 2019-07-31 | 2021-02-15 | 주식회사 에스제이 테크 | 머신러닝을 이용한 배터리 진단 방법 |
KR20210116801A (ko) * | 2020-03-16 | 2021-09-28 | 주식회사 로보볼트 | 신경망 기반의 배터리 잔존 수명 예측 방법 및 장치 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116466241A (zh) * | 2023-05-06 | 2023-07-21 | 重庆标能瑞源储能技术研究院有限公司 | 一种单体电池热失控定位方法 |
CN116466241B (zh) * | 2023-05-06 | 2024-03-26 | 重庆标能瑞源储能技术研究院有限公司 | 一种单体电池热失控定位方法 |
CN117310543A (zh) * | 2023-11-29 | 2023-12-29 | 中国华能集团清洁能源技术研究院有限公司 | 电池异常诊断方法及装置 |
CN117406007A (zh) * | 2023-12-14 | 2024-01-16 | 山东佰运科技发展有限公司 | 一种充电桩充电数据检测方法及系统 |
CN117406007B (zh) * | 2023-12-14 | 2024-02-13 | 山东佰运科技发展有限公司 | 一种充电桩充电数据检测方法及系统 |
Also Published As
Publication number | Publication date |
---|---|
KR20230057894A (ko) | 2023-05-02 |
EP4321884A1 (en) | 2024-02-14 |
CN116829966A (zh) | 2023-09-29 |
JP2024500893A (ja) | 2024-01-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2023068899A1 (ko) | 배터리 팩 내의 이상 징후 셀 검출 장치 및 방법 | |
WO2021049800A1 (ko) | 배터리 이상 상태를 진단하기 위한 방법, 이를 위한 전자 장치 및 저장 매체 | |
WO2010016647A1 (en) | Apparatus and method for estimating state of health of battery based on battery voltage variation pattern | |
WO2017043890A1 (ko) | 이차 전지의 못 관통 시험 장치 및 방법 | |
WO2018038383A1 (ko) | 배터리 셀의 성능 테스트 장치 및 방법 | |
WO2016052900A1 (ko) | 이차 전지의 방전 출력 추정 방법 및 장치 | |
WO2022103185A1 (ko) | 전지의 용량 측정 장치 및 방법, 및 상기 장치를 포함하는 전지 제어 시스템 | |
WO2018194225A1 (ko) | 배터리 모니터링 및 보호 시스템 | |
WO2022114871A1 (ko) | 배터리 진단 장치, 배터리 진단 방법, 배터리 팩 및 자동차 | |
WO2020153637A1 (ko) | 배터리 관리 장치, 배터리 관리 방법 및 배터리 팩 | |
WO2021066555A1 (ko) | 배터리 진단 시스템 및 방법 | |
WO2022098096A1 (ko) | 배터리 진단 장치 및 방법 | |
WO2020262789A1 (ko) | 이상 배터리 셀 검출 방법 | |
WO2022145822A1 (ko) | 배터리 관리 장치 및 방법 | |
WO2022154354A1 (ko) | 배터리 시스템 진단 장치 및 방법 | |
WO2021080219A1 (ko) | 배터리 퇴화도 진단 장치 및 방법 | |
WO2022149917A1 (ko) | 배터리 관리 장치 및 방법 | |
WO2015056830A1 (ko) | 전동장치용 배터리의 전력모니터링 시스템 | |
WO2022215962A1 (ko) | 배터리 진단 장치 및 방법 | |
WO2022080837A1 (ko) | 배터리 진단 장치 및 방법 | |
WO2021230642A1 (ko) | 배터리를 진단하기 위한 장치 및 그 방법 | |
WO2022071776A1 (ko) | 배터리 진단 장치, 방법 및 시스템 | |
WO2023224414A1 (ko) | 배터리 진단 장치, 배터리 진단 방법, 배터리 팩 및 자동차 | |
WO2022250390A1 (ko) | 배터리 모니터링 장치 및 방법 | |
WO2023033480A1 (ko) | 배터리 진단 시스템 및 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22884120 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023538070 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202280013690.7 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022884120 Country of ref document: EP Ref document number: 22884120.1 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2022884120 Country of ref document: EP Effective date: 20231108 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |