Battery performance management system and method using an electric vehicle charging station

Abstract

This invention provides a battery performance management system and method using an electric vehicle charging station. [Solution] In the battery performance management system 10, the battery performance management server collects battery performance evaluation information, including battery identification information and cumulative operating characteristics information, electric vehicle identification information and cumulative operating characteristics information, and the latest battery charging characteristics information, from multiple charging stations via a network. Using an artificial intelligence model that has been pre-trained to output the degree of battery degradation upon receiving the battery performance evaluation information, the server determines the current degree of degradation corresponding to the collected battery performance evaluation information, and also determines the latest control factor corresponding to the current degree of degradation. The server then transmits the latest control factor to the charging station via the network so that the charging station can transmit the latest control factor to the electric vehicle's control system and update the control factor.

Description

[Technical Field]

[0001] The present invention relates to a battery performance management system and method, and more particularly to a system and method in which a remote server collects and stores battery performance evaluation information in a database while an electric vehicle is being charged at an electric vehicle charging station, determines the state of health (SOH) of the battery using an artificial intelligence model trained with big data, and updates the control factors used for battery charging and discharging control.

[0002] This application claims priority based on Korean Patent Application No. 10-2020-0035892 filed on 24 March 2020 and Korean Patent Application No. 10-2021-0037625 filed on 23 March 2021, and all content disclosed in the specifications and drawings of the said applications is incorporated into this application. [Background technology]

[0003] Batteries are rapidly expanding their applications not only in mobile devices such as cell phones, laptop computers, smartphones, and smartpads, but also in fields such as electric vehicles (EVs, HEVs, PHEVs) and large-capacity energy storage systems (ESS).

[0004] The rate at which electric vehicle batteries degrade varies depending on the driver's driving habits and operating environment. For example, batteries in electric vehicles that frequently accelerate rapidly, or those operating in mountainous regions, desert regions, or cold regions will degrade relatively quickly.

[0005] The degree of battery performance degradation can be quantified as a factor called SOH. SOH is a numerical value that shows the performance of a battery in the MOL (Middle of Life) state as a relative ratio to the performance of a battery in the BOL (Beginning of Life) state.

[0006] Indicators used to show battery performance include battery capacity and internal resistance. As the number of charge-discharge cycles increases, the battery capacity decreases and the internal resistance increases. Therefore, State of Health (SOH) can be quantified by the rate of decrease in battery capacity and the rate of increase in internal resistance.

[0007] A battery in the BOL (Battery-Only) state will show a State of Health (SOH) of 100%, while a battery in the MOL (Moisture-Only) state will show a SOH lower than 100%. If the SOH falls below a certain level, it indicates that the battery's performance has deteriorated beyond its limit, and the battery needs to be replaced.

[0008] To maximize battery degradation and extend its lifespan, the charge and discharge control logic must be set differently according to the degree of performance degradation. This requires a system that centrally monitors the performance changes of multiple batteries of the same model and efficiently updates the various control logics used for charging and discharging electric vehicle batteries. [Overview of the project] [Problems that the invention aims to solve]

[0009] The present invention was conceived against the background of the prior art described above, and aims to provide a battery performance management system and method that can accumulate and collect battery performance evaluation information from an electric vehicle charging station while the electric vehicle is being charged at the charging station, diagnose the battery performance (e.g., degree of degradation) based on the collected big data, and update the control factors used for battery charging and discharging control on a platform basis according to the diagnosed performance. [Means for solving the problem]

[0010] To achieve the above objectives, the battery performance management system using electric vehicle charging stations according to the present invention includes a battery performance management server that is communicatively connected to multiple charging stations located in a geographically dispersed area via a network, and a database connected to the battery performance management server that stores information on the degree of degeneration of electric vehicles.

[0011] Preferably, the battery performance management server may be configured to collect battery performance evaluation information from charging stations via the network, including battery identification information and cumulative operating characteristics information, electric vehicle identification information and cumulative operating characteristics information, and the latest battery charging characteristics information, and store this information in a database. It may also be configured to determine the current degree of degeneration corresponding to the collected battery performance evaluation information using an artificial intelligence model that has been pre-trained to output the degree of battery degeneration upon receiving the battery performance evaluation information. If the current degree of degeneration has increased by more than a certain threshold compared to the previous degree of degeneration, it may determine the latest control factor corresponding to the current degree of degeneration using correlation information between the control factor used to control battery operation and the degree of degeneration, and to transmit the latest control factor to the charging station via the network so that the charging station can transmit the latest control factor to the electric vehicle's control system and update the control factor.

[0012] According to one perspective, the cumulative operating characteristics information of a battery may include at least one selected from the group that includes cumulative operating time by voltage interval, cumulative operating time by current interval, and cumulative operating time by temperature interval.

[0013] On the other hand, cumulative information on the operating characteristics of electric vehicles may include at least one selected from the group that includes cumulative operating time by speed section, cumulative operating time by operating area, and cumulative operating time by humidity section.

[0014] Furthermore, from another perspective, the latest charging characteristics information may include at least one selected from the group that includes battery charge state, voltage, current, and temperature data measured or predicted at multiple points in time.

[0015] Ideally, the battery performance management server should be: Each time battery performance evaluation information is received from multiple electric vehicle charging stations, if it is determined that the latest charging characteristics information contains enough data to determine the current degree of battery degradation, the system can be configured to determine the current degree of battery degradation from the latest charging characteristics information, save the cumulative battery operating characteristics information, cumulative electric vehicle operating characteristics information, and the latest charging characteristics information to a database as training input data for an artificial intelligence model, and save the current degree of battery degradation to the database as training output data for the artificial intelligence model.

[0016] Ideally, the battery performance management server can iteratively train the artificial intelligence model whenever training input data and training output data exceeding a certain threshold are accumulated and stored in the database.

[0017] One perspective is that the battery performance management server matches the battery identification information and / or the electric vehicle identification information and / or the electric vehicle's operating area, stores the learning input data and learning output data in a database, and each time learning input data and learning output data exceeding a certain threshold are matched with the battery identification information and / or the electric vehicle identification information and / or the electric vehicle's operating area and stored cumulatively, the artificial intelligence model can be iteratively trained to correspond to the battery identification information and / or the electric vehicle identification information and / or the electric vehicle's operating area.

[0018] On the other hand, the battery performance management server may be configured to analyze battery performance evaluation information and determine the degree of battery degradation using an artificial intelligence model that has been trained to correspond to battery identification information and / or electric vehicle identification information and / or electric vehicle operating area.

[0019] In the present invention, the battery performance management server can receive and store in a database cycle-specific performance evaluation information, including cumulative operating characteristic information and latest charging characteristic information measured each time a charge-discharge cycle is performed on the battery, from a battery data provision server via a network, as well as cycle-specific degradation levels.

[0020] In this case, the battery performance management server may further include an auxiliary artificial intelligence model that has been trained to output a degree of degeneration from cumulative battery operating characteristics information and latest charging characteristics information, using cycle-by-cycle performance evaluation information and cycle-by-cycle degeneration information stored in a database.

[0021] Preferably, the battery performance management server may be configured to input cumulative battery operating characteristics information and latest charging characteristics information included in the battery performance evaluation information into the auxiliary artificial intelligence model to determine the degree of battery degradation when the artificial intelligence model has not completed training.

[0022] Furthermore, the battery performance management server may be configured to input cumulative battery operating characteristic information and the latest charging characteristic information included in the battery performance evaluation information into an auxiliary artificial intelligence model to determine the degree of battery degradation, and to determine the battery degradation degree as a weighted average of the degree of degradation (first value) determined by the artificial intelligence model and the degree of degradation (second value) determined by the auxiliary artificial intelligence model.

[0023] Preferably, the battery performance management server may be configured to increase the weight assigned to the degree of degeneration of the artificial intelligence model when calculating the weighted average as the amount of training the artificial intelligence model increases.

[0024] In one embodiment, the artificial intelligence model may be an artificial neural network.

[0025] In the present invention, the control factor may include (i) at least one selected from the magnitude of the charging current applied for each charging state interval, the upper limit of the charging voltage, the lower limit of the discharge voltage, the maximum charging current, the maximum discharge current, the minimum charging current, the minimum discharge current, the maximum temperature, the minimum temperature, the power map for each charging state, and the internal resistance map for each charging state; (ii) at least one selected from the upper limit of the pulse current duty cycle (ratio of pulse duration to pulse pause time), the lower limit of the pulse current duty cycle, the upper limit of the pulse current duration, the lower limit of the pulse current duration, the maximum value of the pulse current, and the minimum value of the pulse current; or (iii) at least one selected from the magnitude of the current in constant current charging (CC) mode, the cutoff voltage at which constant current charging (CC) mode ends, and the magnitude of the voltage in constant voltage charging (CV) mode.

[0026] In another embodiment of the present invention, the battery performance management server may be configured to transmit the electric vehicle's operating distance and current degeneration rate, as well as electric vehicle identification information, to the insurance company's server, and the insurance company's server may be configured to calculate the insurance premium for the electric vehicle by referring to the electric vehicle identification information and referring to the electric vehicle's operating distance and current degeneration rate.

[0027] A battery performance management method using an electric vehicle charging station according to another aspect of the present invention to achieve the above objectives may include the steps of: collecting battery performance evaluation information from the charging station via a network while the electric vehicle is being charged at the charging station, including battery identification information and cumulative operating characteristics information, electric vehicle identification information and cumulative operating characteristics information, and the latest battery charging characteristics information, and storing it in a database; determining the current degree of degeneration corresponding to the collected battery performance evaluation information using an artificial intelligence model that has been pre-trained to output the degree of degeneration of the battery upon receiving the battery performance evaluation information; determining the latest control factor corresponding to the current degree of degeneration using correlation information between the control factor used to control the battery operation and the degree of degeneration, if the current degree of degeneration has increased to or above a standard value compared to the previous degree of degeneration; and transmitting the latest control factor to the charging station via a network so that the charging station can transmit the latest control factor to the electric vehicle's control system and update the control factor.

[0028] The above technical challenges can also be achieved by a computer device. The computer device may include a non-transitory memory device that stores multiple processor execution instructions, and a processor configured to execute multiple processor execution instructions. The processor may be configured to (a) receive battery performance evaluation information from a charging station via a network, including battery identification information and cumulative operating characteristics information, electric vehicle identification information and cumulative operating characteristics information, and the latest battery charging characteristics information, by executing the processor execution instructions, (b) train an artificial intelligence model to output the degree of battery degeneration from the battery performance evaluation information, (c) use the trained artificial intelligence model to determine the current degree of degeneration corresponding to the collected battery performance evaluation information, (d) read the previous degree of battery degeneration from a database, (e) if the current degree of degeneration has increased by more than a threshold compared to the previous degree of degeneration, determine the latest control factor corresponding to the current degree of degeneration using correlation information between the control factor used to control battery operation and the degree of degeneration, and (f) transmit the latest control factor to the charging station via the network. [Effects of the Invention]

[0029] According to the present invention, by using an artificial intelligence platform system based on big data linked with multiple charging stations, the performance of the battery can be evaluated with high reliability based on the operation history of the electric vehicle and the battery's operating history, and the control factors used for battery charging and discharging control can be optimized. This not only extends the battery's lifespan but also increases safety.

[0030] Providing electric vehicle users with a reliable battery performance management service not only helps guide battery replacement at the appropriate time, but also enhances the reliability of battery manufacturers.

[0031] By building a database of battery performance evaluation information that reflects the driving habits of electric vehicle users on a big data platform, it can be used as accurate data for calculating insurance premiums by automobile insurance companies.

[0032] The following drawings accompanying this specification illustrate preferred embodiments of the invention and, together with the detailed description of the invention, serve to further illustrate the technical idea of ​​the invention. Therefore, the invention should not be construed as being limited solely to what is shown in the drawings. [Brief explanation of the drawing]

[0033] [Figure 1] This is a block diagram showing the configuration of a battery performance management system using an electric vehicle charging station according to one embodiment of the present invention. [Figure 2] This graph illustrates frequency distribution data generated from cumulative operating characteristic information of an electric vehicle battery according to one embodiment of the present invention. [Figure 3] This graph illustrates frequency distribution data generated from cumulative operating characteristic information of an electric vehicle battery according to one embodiment of the present invention. [Figure 4] This graph illustrates frequency distribution data generated from cumulative operating characteristic information of an electric vehicle battery according to one embodiment of the present invention. [Figure 5] This graph illustrates frequency distribution data generated from cumulative operational characteristics information of an electric vehicle according to one embodiment of the present invention. [Figure 6] This graph illustrates frequency distribution data generated from cumulative operational characteristics information of an electric vehicle according to one embodiment of the present invention. [Figure 7] This graph illustrates frequency distribution data generated from cumulative operational characteristics information of an electric vehicle according to one embodiment of the present invention. [Figure 8] This figure illustrates the structure of an artificial neural network according to one embodiment of the present invention. [Figure 9] This figure illustrates the structure of an auxiliary artificial neural network according to one embodiment of the present invention. [Figure 10]This is a flowchart illustrating a method for managing battery performance using an electric vehicle charging station according to one embodiment of the present invention. [Figure 11] This is a flowchart illustrating a method for managing battery performance using an electric vehicle charging station according to one embodiment of the present invention. [Modes for carrying out the invention]

[0034] Preferred embodiments of the present invention will now be described in detail with reference to the attached drawings. Prior to this, terms and words used in this specification and in the claims should not be interpreted in a manner limited to their usual or dictionary meanings, but rather in a manner corresponding to the technical idea of ​​the present invention, in accordance with the principle that the inventor himself can appropriately define the concepts of terms in order to best describe the invention. Accordingly, it should be understood that the embodiments and configurations shown in the drawings described herein are merely the most preferred embodiments of the present invention and do not represent the entirety of the technical idea of ​​the present invention, and that there may be a variety of equivalents and modifications that can be substituted for them at the time of this application.

[0035] Figure 1 is a block diagram showing the configuration of a battery performance management system using an electric vehicle charging station according to one embodiment of the present invention.

[0036] Referring to Figure 1, the battery performance management system 10 according to an embodiment of the present invention is connected to multiple charging stations EVC k This includes a battery performance management server 11, where k is an index to indicate that there are multiple objects indicated by the reference numerals, and the charging station EVC k If there are 10,000 of these, then k can range from 1 to 10,000.

[0037] Ideally, the charging station EVC k The battery performance management server 11 can be connected to each other via the network 12 so that they can communicate with one another.

[0038] Network 12 is a charging station EVC k The type of device is not limited as long as it supports communication with the battery performance management server 11.

[0039] Network 12 includes wired networks, wireless networks, or a combination thereof. Wired networks include short-range or wide-area internet supporting the TCP / IP protocol. Wireless networks include base station-based wireless networks, satellite communication networks, short-range wireless networks such as Wi-Fi®, or a combination thereof.

[0040] Network 12 may include, for example, 2G (second generation) to 5G (fifth generation) networks, LTE (Long Term Evolution) networks, GSM (Registered Trademark) (Global System for Mobile Communication) networks, Code Division Multiple Access (CDC) networks, EVDO (Evolution-Data Optimization) networks, Public Land Mobile networks, and / or other networks.

[0041] Network 12 may include, in other examples, a Local Area Network (LAN), a Wireless Local Area Network (WLAN), a Wide Area Network, a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), an ad-hoc network, a managed IP network, a Virtual Private Network, an intranet, the Internet, an optical fiber infrastructure network, and / or combinations thereof, or other types of networks.

[0042] Charging station EVC k is a charging device provided domestically and / or overseas to charge the battery B n of the electric vehicle EV n n is an index for indicating that there are multiple objects indicated by the reference numerals. If the number of electric vehicles is one million, n is from 1 to one million. Charging station EVC k can be provided in domestic and / or overseas parking lots, gas stations, public institutions, buildings, apartments, condominiums, single-family houses, etc. Charging station EVC k can be coupled to the network 12 in a communicable manner with the battery performance management server 11.

[0043] Preferably, the electric vehicle EV n includes the battery B n and the control system 15. The control system 15 is a computer device that controls the charging and discharging operations of the battery B n and measures the voltage, current, and temperature of the battery B n during charging and discharging of the battery B n and records them in the storage means 15a. Also, the control system 15 is for the electric vehicle EV nIt can perform control operations of mechanical and / or electronic mechanisms involved in the operation of the vehicle.

[0044] The storage means 15a is a non-transitory memory device, which is a computer storage medium capable of recording and / or erasing and / or modifying and / or transmitting data. The storage means 15a is, for example, a flash memory (registered trademark), a hard disk, an SSD (Solid State Disk), or other types of data storage hardware.

[0045] Electric vehicles (EVs) n The control system 15 controls battery B n Battery B is being charged or discharged n The operating characteristics information can be collected and recorded in the storage means 15a. The operating characteristics information is stored in battery B n The control system 15 may include one or more selected from the voltage, current, and temperature of battery B. n Operating characteristics information for battery B n The charge state (SOC) and / or timestamp of the battery B can be recorded in storage means 15a. The control system 15 uses known ampere counting methods, OCV methods, extended Kalman filters, etc., in the industry to record the battery B n The charge state of battery B can be estimated. The control system 15 can estimate the charge state of battery B. n Battery B n It can be electrically coupled to voltage sensors, current sensors, and temperature sensors provided therein.

[0046] The control system 15 controls electric vehicles (EVs). n The operational characteristics information can be recorded in the storage means 15a. The operational characteristics information is for electric vehicles (EVs). n Speed, electric vehicle (EV) n It includes at least one selected from the group consisting of the operating area and humidity. Preferably, the control system 15 is for electric vehicles (EVs). nThe operation characteristics information of the electric vehicle (EV) can be recorded in the storage means 15a along with a timestamp. The control system 15 collects and stores operation characteristics information. n It can be electrically coupled with a speed sensor, GPS sensor, and humidity sensor.

[0047] EVC Charging Station k is an electric vehicle (EV) n Using the charging port of an electric vehicle (EV) n Battery B n Charge battery B n While charging is in progress, battery performance evaluation information is collected and transmitted to the battery performance management server 11. k This is from the battery performance management server 11 to battery B n Electric vehicles (EVs) are subject to various control factors used during charge and discharge control. n This can be transmitted to the control system 15. n The control system 15 controls battery B n The control factor used for charge and discharge control can be updated. This will be discussed later.

[0048] Preferably, the battery performance management system 10 may include a large-capacity database 16 connected to the battery performance management server 11.

[0049] According to one source, the battery performance management server 11 is for electric vehicles (EVs). n EVC charging station k While being charged, the charging station EVC is connected to network 12. k From, electric vehicles (EVs) n Cumulative information on the operational characteristics and battery B n Battery performance evaluation information, including cumulative operating characteristics information and latest charging characteristics information, can be collected and stored in the performance evaluation information storage unit 16a of the database 16.

[0050] Ideally, battery B nThe cumulative operating characteristics information may include at least one selected from the group that includes cumulative operating time by voltage interval, cumulative operating time by current interval, and cumulative operating time by temperature interval.

[0051] Ideally, electric vehicles (EVs) n The cumulative operational characteristics information may include at least one selected from the group that includes cumulative operational time by speed section, cumulative operational time by operating area, and cumulative operational time by humidity section.

[0052] Ideally, the latest charging characteristics information should be for battery B n Battery B measured or predicted while charging n The operating characteristics information may include at least one selected from the group that includes the battery charge state, voltage, current, and temperature measured or predicted at multiple points in time.

[0053] EVC Charging Station k is an electric vehicle (EV) n While charging is taking place, electric vehicle EV n It can communicate with the control system 15 to exchange information and / or data. For example, communication is performed by a data communication line included in the charging cable. Alternatively, communication is performed by the charging station EVC k and electric vehicles (EVs) n This is done via wireless communication between them. For this purpose, the charging station EVC k and electric vehicles (EVs) n This may include short-range wireless communication devices.

[0054] EVC Charging Station k is an electric vehicle (EV) n Information and / or data collected from can be transmitted to the battery performance management server 11 via the network 12 using a predefined communication protocol.

[0055] Battery performance management server 11 is for electric vehicles (EVs) n EVC charging station k While being charged, the charging station EVCk From, electric vehicles (EVs) n Identification information and battery B n Identification information and battery B n Cumulative information on operating characteristics, electric vehicles (EVs) n Battery performance evaluation information, including cumulative operational characteristics information and latest charging characteristics information, can be received and stored in the performance evaluation information storage unit 16a of the database 16.

[0056] Here, electric vehicles (EVs) n The identification information could be the vehicle model code, and battery B n The identification information is battery B n This could be the model code.

[0057] Ideally, the charging station EVC k is an electric vehicle (EV) n Identification information and battery B n Identification information and battery B n Cumulative information on operating characteristics, electric vehicles (EVs) n Battery performance evaluation information including cumulative operational characteristics information and latest charging characteristics information, and electric vehicle (EV) n While charging is taking place, electric vehicle EV n The control system 15 can receive and transmit the received information and / or data to the battery performance management server 11 via the network 12.

[0058] According to one source, the battery performance management server 11 is connected to the charging station EVC k Electric vehicles (EVs) transmitted from n After analyzing the cumulative information of the operating characteristics and generating frequency distribution data for each operating characteristic, electric vehicle EV n Identification information and / or Battery B n The identification information can be matched and recorded in the learning data storage unit 16b of the database 16.

[0059] According to one perspective, in frequency distribution data related to cumulative operating characteristics information, the variable is voltage, current, or temperature, and the frequency is the battery B for each variable.n It can be the cumulative operation time.

[0060] FIG. 2 is a graph showing an example of frequency distribution data related to voltage among the cumulative operation characteristic information of the battery B n FIG. 3 is a graph showing an example of frequency distribution data related to current among the cumulative operation characteristic information of the battery B n and FIG. 4 is a graph showing an example of frequency distribution data related to temperature among the cumulative operation characteristic information of the battery B n .

[0061] Referring to FIGS. 2 to 4, the frequency distribution data can provide the cumulative operation time of the battery B n for each voltage interval, the cumulative operation time of the battery B n for each current interval, and the cumulative operation time of the battery B n for each temperature interval while the electric vehicle EV n is running. The frequency distribution data shows the operation history of the electric vehicle EV n and can be used for the battery performance management server 11 to train the artificial intelligence model. This will be described later.

[0062] On the other hand, the battery performance management server 11 analyzes the cumulative operation characteristic information of the electric vehicle EV k transmitted from the charging station EVC n to generate frequency distribution data for each operation characteristic, and then matches it with the identification information of the electric vehicle EV n and / or the identification information of the battery B n and records it in the learning data storage unit 16b of the database 16.

[0063] In the frequency distribution data for the operation characteristics, the variable is the speed of the electric vehicle EV n , the operation area of the electric vehicle EV n or the humidity of the area where the electric vehicle EV n is running, and the frequency can be the cumulative operation time of the electric vehicle EV n in each variable.

[0064] FIG. 5 is a graph showing an example of frequency distribution data related to speed among the cumulative operation characteristic information of the electric vehicle EV n and FIG. 6 is a graph showing an example of frequency distribution data related to the operation area among the cumulative operation characteristic information of the electric vehicle EV n and FIG. 7 is a graph showing an example of frequency distribution data related to the humidity of the area where the electric vehicle EV n operates among the cumulative operation characteristic information of the electric vehicle EV n and FIG. 7 is a graph showing an example of frequency distribution data related to the humidity of the area where the electric vehicle EV n operates.

[0065] Referring to FIGS. 5 to 7, the frequency distribution data can provide information related to the cumulative operation time by speed interval, the cumulative operation time by operation area, and the cumulative operation time by humidity interval of the electric vehicle EV n during operation. The area can be an administrative area in the country and / or overseas. In one example, the area can be a city, but is not limited thereto. The frequency distribution data can be used for the battery performance management server 11 to train an artificial intelligence model. This will be described later. n

[0066] On the other hand, the battery performance management server 11 can record the latest charging characteristic information of the electric vehicle EV k transmitted from the charging station EVC n in the performance evaluation information storage unit 16a of the database 16.

[0067] Preferably, the latest charging characteristic information includes at least one operation characteristic data selected from the group including SOC, voltage, current, and temperature measured or predicted at multiple time points while the battery B n of the electric vehicle EV n is being charged at the charging station EVC k .

[0068] The operation characteristic data measured at each measurement time point is a 4-dimensional vector, SOC k , I k , V k , T kThis can be shown as follows: k is an index related to the measurement time of the operating characteristics. If the number of measurements is n, then k is a natural number from 1 to n, and the number of data points included in the latest charging characteristics information is n.

[0069] When predetermined conditions are met, the battery performance management server 11 uses the operating characteristics data included in the latest charging characteristics information to determine the battery B n Determine the degree of degeneration, and the degree of degeneration is applied to electric vehicles (EVs). n Identification information and / or Battery B n This information, along with its identification, can be recorded in the learning data storage unit 16b of the database 16.

[0070] For example, the battery performance management server 11 determines whether the latest charging characteristic information was collected within a pre-set estimated voltage interval for the degree of degradation. To this end, the battery performance management server 11 checks the voltage data V included in the latest charging characteristic information. k The distribution of can be examined. If the result of this determination is "yes", the battery performance management server 11 can determine the change in charge capacity by integrating the current data measured in the degeneration estimation voltage interval, and determine the ratio of the change in charge capacity to the reference change in charge capacity as the degree of degeneration. The reference change in charge capacity is the B of a battery in a BOL state. n The change in charging capacity shown while charging in the degeneration estimation voltage interval is, and the reference change in charging capacity is, battery B n For each model, the parameters can be pre-recorded in the parameter storage unit 16c of the database 16.

[0071] In another example, the battery performance management server 11 analyzes the latest charging characteristics information and determines the battery B within a pre-set degeneration estimation voltage range. n The battery is charged, and it is determined whether multiple voltage data have been measured under variable charging current conditions. For this purpose, the battery performance management server 11 uses the voltage data V included in the latest charging characteristics information. k and current data I kThe distribution of can be examined. If the result of this determination is "yes", the battery performance management server 11 performs linear regression analysis on multiple current and voltage data measured within a preset degeneration estimation voltage interval from the latest charging characteristic information and calculates the average value of |dV / dI| for battery B n The internal resistance value is determined, and the ratio of the reference internal resistance value to the internal resistance value is set to Battery B n This can be determined as the degree of degeneration. In this embodiment, the charging station EVC k Battery B n While the battery B is being charged within a pre-set degeneration-estimated voltage interval, the AC charging current and / or different amplitude charging pulses are applied to battery B. n It can be applied to the battery. Then, under variable charging current conditions, multiple voltage data can be measured. The reference internal resistance value is the battery B in the BOL state. n This is the internal resistance value, and the reference internal resistance value is for Battery Model B n Each parameter can be pre-recorded in the parameter storage unit 16c of the database 16.

[0072] The battery performance management server 11 uses an artificial intelligence model to manage the charging station EVC k Battery B transmitted from n Cumulative information on operating characteristics, electric vehicles (EVs) n Battery performance evaluation information including cumulative operational characteristics information and latest charging characteristics information for Battery B n This can determine the degree of degeneration.

[0073] In this invention, the degree of degeneration calculated from the latest charging characteristics information constitutes part of the big data used to train the artificial intelligence model. Therefore, the determination of the degree of degeneration for achieving the technical objectives of this invention is determined substantially by an artificial intelligence model trained on big data.

[0074] The reason is that the degree of degradation calculated from the latest charging characteristics information has the limitation that it can only be determined when certain conditions are added, and battery B nThis is because the degree of degeneration determined by an artificial intelligence model trained on big data is more accurate and reliable, as the past usage history is not adequately considered.

[0075] Preferably, the artificial intelligence model is a software algorithm coded in a programming language and may be an artificial neural network. However, the present invention is not limited thereto.

[0076] Figure 8 shows the structure of an artificial neural network 100 according to one embodiment of the present invention.

[0077] Referring to Figure 8, the artificial neural network 100 includes an input layer 101, multiple hidden layers 102, and an output layer 103. The input layer 101, the multiple hidden layers 102, and the output layer 103 each contain multiple nodes.

[0078] When the battery performance management server 11 trains the artificial neural network 100, or when it uses the artificial neural network 100 to analyze battery B n When determining the degree of degeneration, the charging station EVC is input to the input layer 101. k Battery B collected from n Frequency distribution data generated from cumulative operating characteristics information, electric vehicle (EV) n Frequency distribution data generated from cumulative operational characteristics information and data included in the latest charging characteristics information can be input.

[0079] The cumulative operating characteristic information input (assigned) to the node of the input layer 101 may include a first cumulative time value for each voltage interval and / or a second cumulative time value for each current interval and / or a third cumulative time value for each temperature interval. The first to third cumulative time values ​​are ratios based on the total usable time corresponding to the guaranteed lifespan of battery Bn, and it is desirable to normalize them. For example, if the cumulative time value in a particular voltage interval is 1,000 hours and the total usable time is 20,000 hours, the normalized cumulative time value is 1 / 20 (0.05).

[0080] The number of first cumulative time values ​​may correspond to the number of voltage intervals, the number of second cumulative time values ​​may correspond to the number of current intervals, and the number of third cumulative time values ​​may correspond to the number of temperature intervals. For example, if there are 5 voltage intervals, 9 current intervals, and 10 temperature intervals, the number of first to third cumulative time values ​​will be 5, 9, and 10, respectively.

[0081] Preferably, the input layer 101 may include a number of nodes corresponding to the number of first cumulative time values ​​and / or the number of second cumulative time values ​​and / or the number of third cumulative time values.

[0082] The cumulative operational characteristics information input (assigned) to the nodes of the input layer 101 may include a fourth cumulative time value per speed section and / or a fifth cumulative time value per operating area and / or a sixth cumulative time value per humidity section. It is desirable to normalize the fourth to sixth cumulative time values ​​as a percentage based on the total usable time corresponding to the guaranteed lifespan of battery Bn. For example, if the cumulative time value in a particular speed section is 2,000 hours and the total usable time is 20,000 hours, the normalized cumulative time value is 1 / 10 (0.1).

[0083] The number of cumulative time values ​​in the fourth category corresponds to the number of speed intervals, and the number of cumulative time values ​​in the fifth category corresponds to the number of electric vehicles (EVs). n The number of cumulative time values ​​in the 6th cumulative time value corresponds to the number of areas in which the train operates, and the number of humidity intervals can correspond to the number of humidity intervals. For example, if there are 8 speed intervals, 20 operating areas, and 6 temperature intervals, then the number of cumulative time values ​​in the 4th to 6th cumulative time values ​​will be 8, 20, and 6, respectively.

[0084] Preferably, the input layer 101 may include a number of nodes corresponding to the number of fourth cumulative time values ​​and / or the number of fifth cumulative time values ​​and / or the number of sixth cumulative time values.

[0085] The latest charge identification information input (assigned) to the node of input layer 101 may include voltage data and temperature data. Battery B nSince both voltage and temperature are measured for each SOC, 100 nodes may be allocated for voltage data input, and another 100 nodes may be allocated for temperature data input.

[0086] Here, 100 is the number of nodes corresponding to the SOC from 1% to 100%, assuming the SOC changes by 1% increments from 0% to 100%. If the SOC range is 31-50%, then battery B n Once the voltage and temperature are measured, voltage data may be input to 20 nodes corresponding to 31-50%, and temperature data may be input to another 20 nodes corresponding to 31-50%. Then, the nodes corresponding to the 1-30% and 51-100% SOC ranges may not receive voltage and temperature data and may be assigned a value of 0.

[0087] On the other hand, voltage and temperature data measured at a SOC that includes decimal values ​​can be converted to voltage and temperature data from a nearby SOC without decimals using interpolation or extrapolation. In some cases, temperature data may be excluded from the input data to reduce the computational load on the artificial neural network during training. In this case, the input layer 101 may not include nodes to which temperature data is input.

[0088] Output layer 103 is battery B n It may include a node that outputs information on the degree of degeneration. As shown in Figure 8, if the artificial neural network 100 is designed based on a stochastic model, the output layer 103 is battery B n It may include multiple nodes for outputting the probability distribution of the degree of degeneration.

[0089] For example, if an artificial neural network 100 is designed to determine a degree of degeneration from 71% to 100% in 1% increments, the output layer 103 may contain a total of 30 nodes. In this case, the degree of degeneration corresponding to the node that outputs the highest probability value among the 30 nodes is Battery B nThis can be determined as the degree of degeneration. For example, if the probability of outputting from the 10th node is highest, then battery B n The degree of degeneration can be determined as 80%. It is obvious to those skilled in the art that the number of nodes may be further increased to improve the accuracy of the degree of degeneration.

[0090] Alternatively, if the artificial neural network 100 is designed based on a deterministic model, the output layer 103 will be battery B n It may include at least one node for directly outputting the degree of degeneration.

[0091] The number of hidden layers 102 interposed between the input layer 101 and the output layer 103, and the number of nodes included in each hidden layer 102, can be appropriately selected considering the learning computational load of the artificial neural network 100 and the accuracy and reliability of the artificial neural network 100.

[0092] In artificial neural networks 100, the sigmoid function may be used as the activation function. Alternatively, a variety of activation functions known in this field may be used, such as the SiLU (Sigmoid Linear Unit) function, ReLU (Rectified Linear Unit) function, softplus function, ELU (Exponential Linear Unit) function, and SQLU (Square Linear Unit) function.

[0093] In the artificial neural network 100, the initial values ​​of the interconnection weights and biases between nodes can be set randomly. Furthermore, the interconnection weights and biases can be optimized during the learning process of the artificial neural network.

[0094] In one embodiment, the artificial neural network may be trained by a backpropagation algorithm. Furthermore, the connected weights and biases may be optimized by an optimizer while the artificial neural network is being trained.

[0095] In one example, the SGD (Stochastic Gradient Descent) algorithm could be used as the optimizer. Alternatively, algorithms such as NAG (Nesterov Accelerated Gradient), Momentum, Nadam, Adagrad, RMSProp, Adadelta, and Adam could be used.

[0096] The battery performance management server 11 can periodically and iteratively train the artificial neural network 100 using the training data stored in the training data storage unit 16b of the database 16.

[0097] For this purpose, the battery performance management server 11 uses the method described above to manage multiple charging stations EVC k Many electric vehicles (EVs) n While the device is being charged, it collects learning data and accumulates and records it in the learning data storage unit 16b of the database 16.

[0098] The training data consists of training input data and training output data. The training input data is from electric vehicles (EVs). n Frequency distribution data generated from cumulative operational characteristics information, Battery B n The learning output data may include frequency distribution data generated from cumulative operating characteristic information and data included in the latest charging characteristic information. n This includes the degree of degeneration. The training data is from electric vehicles (EVs). n EVC charging station k It is obtained while being charged.

[0099] Ideally, the training data should be from electric vehicles (EVs). n Identification information and / or Battery B n The identification information can be matched and recorded in the learning information storage unit 16b of the database 16. Therefore, the learning data storage unit 16b contains the same model battery B nThe same model electric vehicle (EV) equipped with it. n A large amount of training data collected from the charging station EVC can be recorded. Additionally, the training data can be stored in the charging station EVC. k Because it is collected continuously, the amount can gradually increase.

[0100] Preferably, the battery performance management server 11 reduces the computational load for training the artificial neural network 100 through distributed processing of data and improves the reliability of the output predicted by the artificial neural network 100, in order to improve the performance of electric vehicles (EVs). n Model and / or Battery B n An artificial neural network can be trained individually for each model.

[0101] In other words, when the battery performance management server 11 periodically trains the artificial neural network 100, it uses the training data stored in the training data storage unit 16b to select data for electric vehicles (EVs). n Model and / or Battery B n The model extracts only the same training data and the relevant electric vehicle (EV) n Model and / or Battery B n An artificial neural network 100 dedicated to the model can be independently trained. Furthermore, the battery performance management server 11 can manage electric vehicles (EVs). n Model and / or Battery B n For this model, if the amount of newly collected training data increases beyond a certain threshold, training of the corresponding artificial neural network 100 can be resumed to further improve the accuracy of the artificial neural network 100.

[0102] Meanwhile, electric vehicles (EVs) n If the frequency distribution data (see Figure 6) generated from cumulative operating time information by operating region within the cumulative operating characteristics information has too many variables, an artificial neural network 100 can be individually trained for each wide-area region formed by grouping multiple regions.

[0103] For example, electric vehicles (EVs) nThere are a total of 100 models, including electric vehicles (EVs). n Battery B installed in n There are a total of 10 models, including electric vehicles (EVs). n Let's assume that the number of cities served by the service, including both domestic and international cities, totals 1000. In this case, the battery performance management server 11 can group the cities according to predetermined criteria and train a number of artificial neural networks corresponding to a total of "100 × 10 × (number of regional groupings)". For example, city grouping may be done on a country-by-country basis. In another example, grouping may be done on a predetermined number of adjacent cities within the same country.

[0104] In this case, when the battery performance management server 11 trains the artificial neural network 100, it selects from the training data stored in the training data storage unit 16b that is used for electric vehicles (EVs). n Model and / or Battery B n Only training data where the models are identical and the variable (city) of the frequency distribution data for the operating area is the same is extracted, and the operating area and / or electric vehicle EV n Model and / or Battery B n An artificial neural network 100 dedicated to the model can be independently trained. Furthermore, the battery performance management server 11 can monitor the operating area and / or electric vehicle (EV). n Model and / or Battery B n If the amount of new training data increases beyond a certain threshold, the model can restart training the corresponding artificial neural network 100, further improving its accuracy.

[0105] In this invention, the artificial intelligence model is not limited to an artificial neural network. Therefore, in addition to artificial neural networks, Gaussian process models and the like can also be used. (Electric vehicle EV) n Cumulative operational characteristics information and / or battery B nWhen learning the correlation between cumulative operating characteristic information and / or the latest charging characteristic data and the degree of degeneration, support vector machines (SVMs), k-nearest neighbor algorithms, and naive Bayes classifiers can be used. If there are reliability issues with the degree of degeneration information used for learning, k-means clustering can be used as an auxiliary means to obtain the degree of degeneration information.

[0106] On the other hand, the battery performance management server 11 may be equipped with an auxiliary artificial neural network that has been trained using cumulative operating characteristic information for each cycle and the latest charging characteristic information for each cycle, provided by the battery manufacturer.

[0107] Figure 9 illustrates the structure of an auxiliary artificial neural network 100' according to one embodiment of the present invention.

[0108] Referring to Figure 9, the auxiliary artificial neural network 100' includes an input layer 101', multiple hidden layers 102', and an output layer 103'. The auxiliary artificial neural network 100' includes an input layer 101' for electric vehicles (EVs). n Except for the absence of a node to which data corresponding to the cumulative operational characteristics information is input, it is substantially identical to the artificial neural network 100 shown in Figure 8.

[0109] The auxiliary artificial neural network 100' will use battery B if the artificial neural network 100 has not been sufficiently trained. n This can be used to determine the degree of degeneration.

[0110] The battery performance management server 11 can be connected to the battery data provision server 17 via the network 12 in a way that allows it to communicate with the battery data provision server 17 in order to collect data used for training the auxiliary artificial neural network 100'.

[0111] Preferably, the battery data provision server 17 may be located within the battery manufacturing company. The battery data provision server 17 is for electric vehicles (EVs). n Battery B installed in n Cumulative information on cycle-specific operating characteristics, latest charge characteristics information for each cycle, and cycle-specific battery B obtained from charge-discharge cycle experiments. n The degree of degeneration of battery B n This information, along with its identification details, can be transmitted to the battery performance management server 11 via the network 12.

[0112] The charge-discharge cycle experiment was conducted using a charge-discharge simulator with battery B n This refers to an experiment in which charging and discharging are repeated a predetermined number of times under various charge and discharge conditions. Charge-discharge cycle experiments are performed on battery B. n This is an essential experiment that battery manufacturers must conduct before commercialization. The charge and discharge conditions are for electric vehicles (EVs). n It is desirable to follow the diverse operating conditions (mountain driving, rough road driving, urban driving, highway driving, etc.) and climatic conditions (temperature, humidity, etc.).

[0113] The charge / discharge simulator is an automated experimental setup that combines a control computer, a charge / discharge device, and a temperature / humidity control chamber. Each time a charge cycle is performed, the charge / discharge simulator generates cumulative operating characteristic information by accumulating the cumulative operating time for each voltage interval and / or the cumulative operating time for each current interval and / or the cumulative operating time for each temperature interval. It can also measure or predict the State of Charge (SOC) and / or voltage and / or current and / or temperature during charging to generate the latest charging characteristic information, which can then be recorded in a storage device.

[0114] Furthermore, the charge / discharge simulator, once each charging cycle is complete, uses the point of charge completion as a reference point for battery B n The degree of degeneration can be determined. The degree of degeneration can be calculated from the change in charging capacity determined by the ampere count method over a predetermined charging voltage interval, or from the internal resistance of the battery obtained by linear regression analysis of voltage and current data measured over a predetermined charging voltage interval, as described above.

[0115] The battery data provision server 17 may include a database 18 that stores data obtained from charge-discharge cycle experiments. n For each charge-discharge cycle, the cumulative operating characteristics information for each cycle, the latest charging characteristics information for each cycle, and the degree of degradation for each cycle are used for battery B n The data can be matched with the identification information and stored in the database 18. The data stored in the database 18 can be transmitted from the charge / discharge simulator via the network 12.

[0116] The battery data provision server 17 periodically provides auxiliary learning data, including cumulative information on cycle-specific operating characteristics, latest charge characteristics information for each cycle, and the degree of degeneration for each cycle, stored in the database 18, to Battery B n This information, along with its identification details, can be transmitted to the battery performance management server 11 via the network 12. The number of auxiliary training data points corresponds to the number of charge-discharge cycle experiments performed. For example, if 200 charge-discharge cycle experiments are performed on a particular model of battery, the number of auxiliary training data points will be 200.

[0117] The battery performance management server 11 receives auxiliary learning data transmitted from the battery data provision server 17 to battery B n The identification information can be matched and recorded in the learning data storage unit 16b of the database 16.

[0118] Preferably, among the auxiliary learning data, information relating to the cumulative operating time by voltage interval and / or the cumulative operating time by current interval and / or the cumulative operating time by temperature interval included in the cumulative operating characteristics information may be converted into frequency distribution data and stored in the learning data storage unit 16b of the database 16.

[0119] After the auxiliary training data is stored in the database 16, the battery performance management server 11 can use the auxiliary training data to train an auxiliary artificial neural network 100' for each battery model.

[0120] The structure of the auxiliary artificial neural network 100' is similar to the structure of artificial neural network 100 shown in Figure 8. The difference is that it is used in electric vehicles (EVs). n This means that the node into which the frequency distribution data generated from the cumulative information of the operational characteristics is input is deactivated. However, the learning method and remaining features of the auxiliary artificial neural network 100' are substantially the same as those described above.

[0121] The battery performance management server 11 uses an auxiliary artificial neural network 100' trained on auxiliary training data transmitted from the battery data provision server 17, and multiple charging stations EVC k An artificial neural network 100, trained on data transmitted from, and the charging station EVC, are used in a complementary manner. k From electric vehicles (EVs) n After charging, battery B n The degree of degeneration is determined, and based on the determined degree of degeneration, battery B n The control factor used for charge and discharge control in electric vehicles (EVs) n This can be provided to the control system 15.

[0122] The following refers to the charging station EVC, with reference to Figures 10 and 11. K From electric vehicles (EVs) n During the process of collecting training data for the artificial intelligence model while the battery is being charged, n The process of determining the degree of degeneration, and the determined degree of degeneration, which determines battery B n This document details the process by which the control factors used for charge and discharge control are updated.

[0123] Referring to Figure 10, in step S10, the battery performance management server 11 controls the charging station EVC k electric vehicles (EVs) n During or after charging, the charging station EVC k From, Battery B nIdentification information and electric vehicle (EV) n Identification information and electric vehicle (EV) n Cumulative information on the operational characteristics, battery B n The battery performance evaluation information, including cumulative operating characteristics information and the latest charging characteristics information, is received. In step S10, the battery performance management server 11 may record the battery performance evaluation information transmitted via the network 12 in the performance evaluation information storage unit 16a of the database 16.

[0124] In stage S20, the battery performance management server 11 receives the voltage data V included in the latest charging characteristics information. k and / or current data I k Refer to this to determine if the conditions for calculating the degree of degeneration are met.

[0125] For example, the conditions under which the degree of degeneration can be calculated are: voltage data V k Battery B n This can be met when the battery B is charged. In another example, the condition for calculating the degree of degeneration is that within a predetermined degeneration estimation voltage interval, battery B n Multiple voltage data V under conditions where the battery is charged and the charging current is variable. k This can be true when it is measured.

[0126] If the decision in stage S20 is "yes," proceed to stage S30; if the decision in stage S20 is "no," proceed to stage S60.

[0127] In stage S30, the battery performance management server 11 receives the voltage data V included in the latest charging characteristics information. k and / or current data I k Using battery B n The degree of degeneration is determined. The method for determining the degree of degeneration was described above. After stage S30, proceed to stage S40.

[0128] At stage S40, the battery performance management server 11 reports battery B n Frequency distribution data related to voltage and / or current and / or temperature is generated from the cumulative operating characteristics information of electric vehicles (EVs).n Frequency distribution data related to speed and / or operating area and / or humidity is generated from the cumulative information of the operating characteristics. After step S40, proceed to step S50.

[0129] In stage S50, the battery performance management server 11 receives frequency distribution data generated from cumulative operating characteristics information, frequency distribution data generated from cumulative operational characteristics information, the latest charging characteristics information, and battery B determined in stage S30. n The degree of degeneration of battery B n Identification information and / or electric vehicle (EV) information n The identification information is matched and recorded in the learning information storage unit 16b of the database 16. Here, the frequency distribution data generated from the cumulative operating characteristics information, the frequency distribution data generated from the cumulative operation characteristics information, and the latest charging characteristics information become learning input data, and battery B n The degree of degeneration becomes the training output data. After stage S50, proceed to stage S60.

[0130] At stage S60, the battery performance management server 11 reports battery B n Identification information and / or electric vehicle (EV) information n Refer to the identification information of Battery B n Models and / or electric vehicles (EVs) n Determine if a trained artificial neural network 100 corresponding to the model is available.

[0131] For example, battery B n The model is BBB001, and it is an electric vehicle (EV). n Let's assume the model is EV001. In this case, the battery performance management server 11 will determine that an EV001 model electric vehicle equipped with a BBB001 model battery is at the charging station EVC k The system determines whether there is an artificial neural network 100 trained using data exceeding a certain threshold collected during the charging process. The threshold could be, for example, several hundred to several thousand.

[0132] In step S60, the battery performance management server 11 refers to the frequency distribution data by operating area generated in step S40, and the battery B n Models and / or electric vehicles (EVs) n The models are the same, and the operating areas are the same. n It is possible to determine whether an artificial neural network 100 trained on data collected from exists.

[0133] For example, battery B n The model is BBB001, and it is an electric vehicle (EV). n The model is EV001, and it is an electric vehicle (EV). n Assume that the regional variable of the frequency distribution data generated from the operating area is a city within South Korea. In this case, the battery performance management server 11 will determine that an EV001 model electric vehicle equipped with a BBB001 model battery is located at a charging station EVC in South Korea. k The system determines whether an artificial neural network 100, trained using data exceeding a certain threshold collected during the charging process, is ready. The threshold can range from several hundred to several thousand.

[0134] If the decision in stage S60 is "yes," proceed to stage S70.

[0135] In stage S70, the battery performance management server 11 sends the input layer 101 of the artificial neural network 100 to the electric vehicle (EV) n Frequency distribution data generated from cumulative operational characteristics information, and battery B n Frequency distribution data generated from cumulative operating characteristics information, and battery B n Voltage data V included in the latest charging characteristics information k And, temperature data T k And, input. Since the artificial neural network 100 is trained with training data above a certain threshold, when data is input from the input layer 101, the output layer 103 will input battery B n The degree of degeneration is output. Then, the battery performance management server 11 outputs the battery B via the artificial neural network 100. nThe current degree of degeneration can be determined. After step S70, proceed to step S80 in Figure 11.

[0136] On the other hand, if the judgment in stage S60 is "no", then in stage S70', the battery performance management server 11 will send a message to the input layer 101' of the auxiliary artificial neural network 100' regarding battery B n Frequency distribution data generated from cumulative operating characteristics information (see Figures 2-5), and battery B n Voltage data V included in the latest charging characteristics information k And, temperature data T k Enter the following, Battery B n The current degree of degeneration can be determined. The auxiliary artificial neural network 100' uses battery B provided by the battery data server 17. n This is an artificial neural network pre-trained using charge-discharge cycle experimental data, and the training method is as described above.

[0137] At stage S70 or stage S70', Battery B n Once the current degree of degeneration is determined, the process proceeds to stage S80 in Figure 11.

[0138] In stage S80, the battery performance management server 11 stores the degree of degeneration determined by the artificial neural network 100 or auxiliary artificial neural network 100' in the degeneration degree information storage unit 16d of the database 16 for electric vehicles (EVs). n Identification information and / or Battery B n The identification information is matched and saved along with the timestamp. After step S80, proceed to step S90.

[0139] At stage S90, the battery performance management server 11 stores the battery B recorded in the degeneration information storage unit 16d of the database 16. n The degree of degeneration in the past is compared with the current degree of degeneration to determine whether the current degree of degeneration has increased above the baseline value.

[0140] A reference value is a value defined in advance, and Battery B nThis is to determine whether or not to perform updated logic on the various control factors used in the charge-discharge control process. For example, the baseline value could be 3-5%.

[0141] For example, the control factor may be at least one selected from the following: the magnitude of the charging current applied to each charging state interval, the upper limit of the charging voltage, the lower limit of the discharging voltage, the maximum charging current, the maximum discharging current, the minimum charging current, the minimum discharging current, the maximum temperature, the minimum temperature, the power map for each charging state, and the internal resistance map for each charging state.

[0142] In another example, the control factor is battery B n This is used when pulse charging and discharging is performed and may include at least one selected from the following: an upper limit of the pulse current duty cycle (the ratio of pulse duration to pulse pause time), a lower limit of the pulse current duty cycle, an upper limit of the pulse current duration, a lower limit of the pulse current duration, a maximum pulse current, and a minimum pulse current.

[0143] In yet another example, the control factor is battery B n This is used when step charging is performed and may include the magnitude of the charging current applied to each charging state interval.

[0144] In yet another example, the control factor is battery B n This is used when charging in CC / CV mode and may include at least one selected from the following: the magnitude of the current in constant current charging (CC) mode, the cutoff voltage at which constant current charging (CC) mode ends, and the magnitude of the voltage in constant voltage charging (CV) mode.

[0145] If the judgment at stage S90 is "yes," proceed to stage S100.

[0146] In stage S100, the battery performance management server 11 refers to the control factor storage unit 16e of the database 16 and checks the battery B nThe latest control factor corresponding to the current degree of degeneration is read, and the battery performance evaluation results, including the current degree of degeneration and the latest control factor, are transmitted via network 12 to the charging station EVC. k It will be transmitted to [the specified location].

[0147] The control factor storage unit 16e is battery B n It includes lookup tables that define control factor information for each degree of degeneration. The lookup table is for battery B n Identification information and / or electric vehicle (EV) information n It is recorded in conjunction with the identification information. This allows the control factor to be determined by battery B n Models and / or electric vehicles (EVs) n It is preferable to read from the lookup table corresponding to the model. After step S100, proceed to step S110.

[0148] At stage S110, the charging station EVC k Battery B n After receiving the battery performance evaluation results, including the current degree of degradation and the corresponding latest control factors, via network 12, the results are transmitted via the charging cable's communication line or short-range wireless communication to the electric vehicle (EV). n The signal is transmitted to the control system 15. After step S110, the process proceeds to step S120.

[0149] Stage S120, electric vehicle (EV) n The control system 15 refers to the latest control factor included in the battery performance evaluation results, and controls the battery B n The control system 15 updates the previous control factor used to control the charging and discharging of the charging station EVC. k After charging is complete, Battery B n Battery B n The charging and discharging of the device can be safely controlled.

[0150] On the other hand, if the judgment at stage S90 is "no", the process proceeds to stage S130.

[0151] At stage S130, the battery performance management server 11 controls battery B n Since the current degree of degeneration did not increase beyond the baseline value, a message indicating that updating the control factor is unnecessary, along with the battery performance evaluation results including the current degree of degeneration, were sent to the charging station EVC via network 12. k The data is transmitted to [location]. After step S130, the process proceeds to step S140.

[0152] At stage S140, the charging station EVC k When the battery performance evaluation results are received, the electric vehicle (EV) is connected via a charging cable or short-range wireless communication. n The battery performance evaluation results are transmitted to the control system 15. After step S140, the process proceeds to step S150.

[0153] Stage S150, electric vehicle (EV) n The control system 15 confirms a message from the battery performance evaluation results indicating that updating the previous control factor is unnecessary, and then the battery B n The control factors used to control the charging and discharging of the device are maintained as they are.

[0154] Although not shown in the diagram, the battery performance management server 11 also monitors battery B even after the artificial neural network 100 has finished learning. n When determining the degree of degeneration, an auxiliary artificial neural network 100' can be used as an aid.

[0155] In other words, the battery performance management server 11 uses the artificial neural network 100 in stage S70 to analyze battery B n After determining the degree of degeneration (first value), step S70' is performed further using the auxiliary artificial neural network 100' to process battery B n The degree of degeneration (second value) can be determined. Then, the weighted average of the first and second values ​​is calculated for battery B nThis can be determined as the degree of degeneration. In this case, the weight assigned to the first value can be gradually increased compared to the weight assigned to the second value as the amount of training data used to train the artificial neural network 100 increases.

[0156] For example, the weight assigned to the first value can be determined as the ratio of the data used to train the artificial neural network 100 to the total amount of data used to train the artificial neural network 100 and the auxiliary artificial neural network 100'.

[0157] According to this modified embodiment, the more the artificial neural network 100 is trained, the more battery B n The degree of degeneration (first value) converges to the degree of degeneration determined by artificial neural network 100. Conversely, if the amount of training data used to train artificial neural network 100 is small, battery B n The degree of degeneration converges to a degree of degeneration (second value) determined by the auxiliary artificial neural network 100.

[0158] In the present invention, electric vehicle EV n The control system 15 is for the charging station EVC k Battery B included in the battery performance evaluation results transmitted from n The current degree of degeneration of electric vehicles (EVs) n An integrated control display panel can be provided to the driver via a graphical user interface. Preferably, the graphical user interface may include numbers and / or graphic gauges indicating the degree of degeneration.

[0159] On the other hand, the battery performance management system 10 according to the embodiment of the present invention described above may further include an insurance company server 19 that is communicably connected to the battery performance management server 11 via a network 12.

[0160] In this case, the battery performance management server 11 controls the electric vehicle (EV). nDegree of degeneration, total operating distance, and electric vehicles (EVs) n The identification information may be configured to be transmitted to the insurance company server 19. n The total distance traveled is by electric vehicles (EVs). n EVC charging station k While being charged, the charging station EVC k via electric vehicles (EVs) n It can be received from the control system 15.

[0161] Insurance company server 19 is electric vehicle EV n Refer to the identification information of the relevant electric vehicle (EV) n The insurance premiums related to this can be configured to be calculated by referring to information on the degree of degeneration. In other words, the insurance company server 19 is battery B n The higher the degree of degeneration, the more electric vehicles (EVs) n Increase the depreciation rate for electric vehicles (EVs) n The price can be calculated. Also, the insurance company server 19 can calculate the price of an electric vehicle (EV). n Battery B n If the degree of deterioration is greater than average, it may be judged that the driver's driving habits are poor, and the risk rate for automobile accidents may be increased, thereby raising the insurance premium.

[0162] The insurance premium calculated by the insurance company server 19 is stored in the insurance company server 19's database (not shown), and then for electric vehicles (EVs). n It is self-evident that this can be referenced during the insurance renewal process.

[0163] A battery performance management method using an electric vehicle charging station according to an embodiment of the present invention can be coded with a plurality of processor execution instructions and then stored in a non-transitory memory device (11a in Figure 1) provided in the battery performance management server 11. The processor execution instructions can be used to instruct a processor (11b in Figure 1) provided in the battery performance management server 11 to execute at least a portion of the steps described above. Alternatively, a hardware logic circuit can be provided in the battery performance management server 11 to perform at least a portion of the steps described above instead of the processor execution instructions. The hardware logic circuit may be an ASIC (Application-Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array). However, it will be obvious to those skilled in the art that the steps of the embodiment can be executed not only by specific processor execution instructions, specific hardware circuits or combinations thereof, but also by other known software, hardware circuits or combinations thereof.

[0164] According to the present invention, by using an artificial intelligence platform system on a big data base linked with multiple charging stations, the performance of the battery can be evaluated with high reliability based on the operation history of the electric vehicle and the battery's operating history, and the control factors used for battery charging and discharging control can be optimized. This not only extends the battery's lifespan but also increases safety.

[0165] Providing electric vehicle users with a reliable battery performance management service not only helps guide battery replacement at the appropriate time, but also enhances the reliability of battery manufacturers.

[0166] By building a database of battery performance evaluation information that reflects the driving habits of electric vehicle users on a big data platform, it can be used as accurate data for calculating insurance premiums by automobile insurance companies.

[0167] In describing the various embodiments of the present invention, components named "~server" should be understood as functionally distinct elements rather than physically distinct elements. Therefore, each component may be selectively integrated with other components, or each component may be divided into sub-components for efficient execution of control logic. However, it will be obvious to those skilled in the art that if functional identity is maintained even after integration or division of components, the integrated or divided components should also be interpreted as falling within the scope of the present invention.

[0168] Although the present invention has been described above with reference to limited embodiments and drawings, it goes without saying that the present invention is not limited thereto, and that various modifications and variations are possible within the equivalent scope of the technical concept and claims of the present invention by persons with ordinary skill in the art to which the present invention pertains.

Claims

1. A communication interface, Includes at least one processor, The at least one processor is Battery performance evaluation information, including at least one of the following: identification information of a battery installed in a vehicle, cumulative information on the operating characteristics of a battery installed in a vehicle, identification information of a vehicle, cumulative information on the operating characteristics of a vehicle, or charging characteristics information of a battery installed in a vehicle, is collected via the communication interface. The degree of battery degradation corresponding to the collected battery performance evaluation information is determined. Using correlation information between the control factor used to control battery operation and the degree of degeneration, the control factor corresponding to the determined degree of degeneration is determined. A battery performance management system configured to transmit signals for controlling the charging of the battery according to the control factor to the vehicle or charging station via the communication interface.

2. A communication interface, Includes at least one processor, The at least one processor is Battery performance evaluation information, including at least one of the following: battery identification information, cumulative battery operating characteristics information, vehicle identification information, cumulative vehicle operation characteristics information, or battery charging characteristics information, is collected via the communication interface. The degree of battery degradation corresponding to the collected battery performance evaluation information is determined. Using correlation information between the control factor used to control the operation of a battery installed in an automobile and the degree of degeneration, the control factor corresponding to the determined degree of degeneration is determined. A battery performance management system configured to transmit signals for controlling the charging of the battery according to the control factor to the vehicle or charging station via the communication interface.

3. A communication interface, Includes at least one processor, The at least one processor is Battery performance evaluation information, including at least one of the following: battery identification information, cumulative battery operating characteristics information, vehicle identification information, cumulative vehicle operation characteristics information, or battery charging characteristics information, is collected via the communication interface. The degree of battery degradation corresponding to the collected battery performance evaluation information is determined. Using correlation information between the control factor used to control battery operation and the degree of degeneration, the control factor corresponding to the determined degree of degeneration is determined. The system is configured to transmit a signal for controlling the charging of the battery according to the control factor to the vehicle or charging station via the communication interface. The aforementioned battery performance evaluation information includes vehicle identification information and cumulative information on the vehicle's operating characteristics, in a battery performance management system.

4. The at least one processor is A battery performance management system according to any one of claims 1 to 3, wherein if the determined degree of degeneration increases by more than a standard value compared to the previous degree of degeneration, the system is configured to determine a control factor corresponding to the determined degree of degeneration using correlation information between the control factor used to control the battery operation and the degree of degeneration.

5. The at least one processor is The system is configured to use an artificial intelligence model to determine the degree of battery degradation corresponding to the collected battery performance evaluation information. The artificial intelligence model, A battery performance management system according to any one of claims 1 to 4, wherein the model is trained to take performance evaluation information of multiple automobiles as input data and output the degree of degeneration of multiple batteries included in the multiple automobiles.

6. The battery performance management system according to any one of claims 1 to 5, wherein the cumulative operating characteristics information of the battery includes at least one selected from the group including cumulative operating time by voltage interval, cumulative operating time by current interval, and cumulative operating time by temperature interval.

7. The battery performance management system according to any one of claims 1 to 6, wherein the cumulative operating characteristics information of the vehicle includes at least one selected from the group including cumulative operating time by speed section, cumulative operating time by operating area, and cumulative operating time by humidity section.

8. The battery performance management system according to any one of claims 1 to 7, wherein the charging characteristic information includes at least one selected from the group including battery charge state, voltage, current, and temperature data measured or predicted at multiple points in time.

9. The system further includes a database connected to at least one of the aforementioned processors, The at least one processor is The battery performance management system according to claim 5, wherein each time battery performance evaluation information is received from the charging station, if it is determined that the charging characteristics information contains enough data to determine the degree of battery degradation, the degree of battery degradation is determined from the charging characteristics information, the battery performance evaluation information is stored in the database as learning input data for the artificial intelligence model, and the determined degree of degradation is stored in the database as learning output data for the artificial intelligence model.

10. The at least one processor is The battery performance management system according to claim 9, wherein the artificial intelligence model is repeatedly trained each time that learning input data and learning output data exceeding a certain threshold are accumulated and stored in the database.

11. The at least one processor is The learning input data and the learning output data are stored in the database, matching them with at least one of the battery identification information, the vehicle identification information, and the vehicle's operating area. The battery performance management system according to claim 9 or 10, wherein each time the matching learning input data and learning output data exceeding a standard value are accumulated and stored, the artificial intelligence model is iteratively trained to correspond to at least one of the battery identification information, the vehicle identification information, and the vehicle's operating area.

12. The at least one processor is The battery performance management system according to claim 11, configured to analyze the battery performance evaluation information and determine the degree of battery degradation using an artificial intelligence model that has been trained to correspond to at least one of the battery identification information, the vehicle identification information, and the operating area of ​​the vehicle.

13. The at least one processor is The communication interface receives from the battery data provision server cycle-specific performance evaluation information, including cumulative operating characteristic information and charging characteristic information measured each time a charge-discharge cycle is performed on the battery, and cycle-specific degradation levels. The battery performance management system according to claim 5, further training an auxiliary artificial intelligence model to output a degree of degeneration from cumulative battery operating characteristic information and charging characteristic information using the cycle-specific performance evaluation information and the cycle-specific degree of degeneration.

14. The at least one processor is The battery performance management system according to claim 13, configured to input the battery performance evaluation information into the auxiliary artificial intelligence model to determine the degree of battery degradation.

15. The at least one processor is The battery performance evaluation information is input into the auxiliary artificial intelligence model to determine the degree of battery degradation. The battery performance management system according to claim 13 or 14, configured to determine the degree of degeneration of the battery as a weighted average of the degree of degeneration determined by the artificial intelligence model and the degree of degeneration determined by the auxiliary artificial intelligence model.

16. The battery performance management system according to claim 15, wherein the at least one processor is configured to increase the weight assigned to the degree of degeneration of the artificial intelligence model when calculating the weighted average as the amount of learning of the artificial intelligence model increases.

17. The battery performance management system according to any one of claims 13 to 16, wherein the artificial intelligence model or the auxiliary artificial intelligence model is an artificial neural network.

18. The battery performance management system according to claim 5, wherein the artificial intelligence model is an artificial neural network.

19. The aforementioned control factor is At least one selected from the following for each charging state interval: the magnitude of the charging current applied, the upper limit of the charging voltage, the lower limit of the discharging voltage, the maximum charging current, the maximum discharging current, the minimum charging current, the minimum discharging current, the maximum temperature, the minimum temperature, the power map for each charging state, and the internal resistance map for each charging state. At least one selected from the upper limit of the pulse current duty cycle (the ratio of pulse duration to pulse pause time), the lower limit of the pulse current duty cycle, the upper limit of the pulse current duration, the lower limit of the pulse current duration, the maximum value of the pulse current, and the minimum value of the pulse current, or A battery performance management system according to any one of claims 1 to 18, comprising at least one selected from the magnitude of the current in constant current charging (CC) mode, the cutoff voltage at which the constant current charging (CC) mode ends, and the magnitude of the voltage in constant voltage charging (CV) mode.

20. The at least one processor is configured to transmit the vehicle's mileage, the degree of battery degradation, and vehicle identification information to the insurance company's server. The battery performance management system according to any one of claims 1 to 19, wherein the insurance company's server is configured to refer to the vehicle identification information and calculate the insurance premium for the vehicle in question by referring to the vehicle's mileage and the degree of battery degradation.

21. A step of collecting battery performance evaluation information which includes at least one of the following: identification information of a battery installed in a vehicle, cumulative information on the operating characteristics of a battery installed in a vehicle, identification information of a vehicle, cumulative information on the operating characteristics of a vehicle, or charging characteristics information of a battery installed in a vehicle. The steps include determining the degree of degradation corresponding to the collected battery performance evaluation information, A step of determining a control factor corresponding to the determined degree of degeneration using correlation information between the control factor used to control battery operation and the degree of degeneration, A battery performance management method comprising the step of transmitting a signal to the automobile or charging station for controlling the charging of the battery according to the control factor.

22. A step of collecting battery performance evaluation information, which includes at least one of the following: battery identification information, cumulative battery operating characteristics information, vehicle identification information, cumulative vehicle operation characteristics information, or battery charging characteristics information. The steps include determining the degree of degradation corresponding to the collected battery performance evaluation information, A step of determining a control factor corresponding to the determined degree of degeneration using correlation information between the control factor used to control the operation of a battery installed in an automobile and the degree of degeneration, A battery performance management method comprising the step of transmitting a signal to the automobile or charging station for controlling the charging of the battery according to the control factor.

23. A step of collecting battery performance evaluation information, which includes at least one of the following: battery identification information, cumulative battery operating characteristics information, vehicle identification information, cumulative vehicle operation characteristics information, or battery charging characteristics information. The steps include determining the degree of degradation corresponding to the collected battery performance evaluation information, A step of determining a control factor corresponding to the determined degree of degeneration using correlation information between the control factor used to control battery operation and the degree of degeneration, The step includes transmitting a signal to the vehicle or charging station for controlling the charging of the battery according to the control factor, The battery performance evaluation information includes vehicle identification information and cumulative information on vehicle operation characteristics, as a battery performance management method.

24. A non-temporary memory device that stores multiple processor execution instructions, A processor configured to execute the plurality of processor execution instructions, including The processor executes the plurality of processor execution instructions, Battery performance evaluation information is collected via a network, including at least one of the following: battery identification information installed in a vehicle, cumulative operating characteristics information of a battery installed in a vehicle, vehicle identification information, cumulative operating characteristics information of a vehicle, or charging characteristics information of a battery installed in a vehicle. The current degree of degradation corresponding to the collected battery performance evaluation information is determined, Using correlation information between the control factor used to control battery operation and the degree of degeneration, the control factor corresponding to the determined degree of degeneration is determined. A computer device configured to transmit signals to the automobile or charging station for controlling the charging of the battery according to the control factor.

25. A non-temporary memory device that stores multiple processor execution instructions, A processor configured to execute the plurality of processor execution instructions, including The processor executes the plurality of processor execution instructions, Battery performance evaluation information is collected via the network, including at least one of the following: battery identification information, cumulative battery operating characteristics information, vehicle identification information, cumulative vehicle operation characteristics information, or battery charging characteristics information. The current degree of degradation corresponding to the collected battery performance evaluation information is determined, Using correlation information between the control factor used to control the operation of a battery installed in an automobile and the degree of degeneration, the control factor corresponding to the determined degree of degeneration is determined. A computer device configured to transmit signals to the automobile or charging station for controlling the charging of the battery according to the control factor.

26. A non-temporary memory device that stores multiple processor execution instructions, A processor configured to execute the plurality of processor execution instructions, including The processor executes the plurality of processor execution instructions, Battery performance evaluation information is collected via the network, including at least one of the following: battery identification information, cumulative battery operating characteristics information, vehicle identification information, cumulative vehicle operation characteristics information, or battery charging characteristics information. The current degree of degradation corresponding to the collected battery performance evaluation information is determined, Using correlation information between the control factor used to control battery operation and the degree of degeneration, the control factor corresponding to the determined degree of degeneration is determined. The system is configured to transmit a signal to the vehicle or charging station for controlling the charging of the battery according to the control factor, The aforementioned battery performance evaluation information includes vehicle identification information and cumulative vehicle operation characteristics information, and is provided by a computer device.

Images