Method for improving battery state information using long-term data

Abstract

The present invention relates to a method for improving battery state information using long-term data, and relates to a method for providing improved battery state information using limited data including a long-term statistical distribution of battery state information according to mileage for a plurality of specific products identical to a specific product that runs on batteries and operation data of the specific product.

Description

How to improve battery health information using long-term data

[0001] The present invention relates to a method for improving battery status information using long-term data, and more particularly, to a method for providing improved battery status information using limited data including a long-term statistical distribution of battery status information based on mileage for a plurality of specific products identical to a specific product that runs on batteries and operation data of the specific product.

[0002] Batteries that can be used for a long time until their lifespan ends by repeatedly charging and discharging are installed in various products, providing convenience to users by allowing them to operate the products without commercial power.

[0003] In particular, as electric vehicles equipped with batteries and powered by electricity supplied from the batteries have become more widespread, public interest in battery management for electric vehicles has increased.

[0004] In other words, ensuring safe use of batteries over long periods of time ultimately leads to improved safety of electric vehicles.

[0005] Ultimately, in order to improve the safety of electric vehicles, it is very important to estimate and provide information indicating how much the current performance of the battery, such as the state of health (SOH) or remaining useful life (RUL), has degraded compared to the initial performance of the battery.

[0006] The types of battery degradation include storage degradation and cycle degradation.

[0007] The above storage degradation refers to degradation that occurs over time even when the battery is not performing any operation, and cycle degradation refers to degradation that occurs as the battery is charged and discharged.

[0008] Therefore, as electric vehicles drive, the batteries are charged or discharged, so cycle degradation is correlated with the driving distance of the electric vehicle, and the lifespan of the batteries installed in the electric vehicle is correlated with storage degradation.

[0009] Therefore, in order to more accurately estimate battery status information such as SOH and RUL, the driving distance of the electric vehicle and the battery's usage period must be reflected.

[0010] Typically, electric vehicles do not provide battery status information, instead providing limited data consisting of vehicle operating data. There is a need to estimate battery status using this limited data, and the accuracy of this estimated battery status needs to be improved by reflecting cycle and storage degradation.

[0011] Accordingly, the present invention proposes a method for providing improved battery status information by using long-term data composed of long-term statistical distributions of battery status information according to mileage for a specific product (e.g., an electric vehicle) and limited data including operating data of a specific product.

[0012] That is, the present invention proposes a method of estimating battery status information using limited data (operating data) provided by a product, and providing improved battery status information estimated using long-term data reflecting storage deterioration and cycle deterioration of the battery.

[0013] Next, we will briefly explain the prior art existing in the technical field of the present invention, and then describe the technical details that the present invention seeks to achieve differently from the prior art.

[0014] First, Korean Patent No. 2452548 (October 4, 2022) relates to a battery deterioration state estimation device, a system including the same, and a method thereof, and relates to a battery deterioration state estimation device, a system including the same, and a method thereof, which estimates the battery deterioration state by using a correction coefficient calculated from a difference value of a second open circuit obtained based on a change in battery state of charge (SOC) according to a first open circuit voltage and current accumulation calculated using voltage, temperature, and current.

[0015] That is, Korean Patent No. 2452548 estimates the deterioration status of a battery using direct information on the battery, such as voltage, temperature, and current, and does not estimate battery status information using limited data including operating data according to the operation of a product equipped with a battery.

[0016] On the other hand, the present invention estimates battery status information using the product's operating data, and provides improved battery status information by reflecting the storage deterioration and cycle deterioration of the battery using long-term data, which is a statistical distribution of battery status information according to the product's mileage. Korean Patent No. 2452548 does not describe, suggest, or imply any technical features of the present invention.

[0017] In addition, Korean Patent Publication No. 2023-0101789 (July 6, 2023) relates to a battery SOH estimation system, a parameter extraction system and method therefor, which receives and stores voltage and current values ​​of a battery according to a preset cycle, calculates a response function in the frequency domain for the stored voltage and current values, and a change rate of the voltage value and charge capacity of the battery to estimate the SOH of the battery, and relates to a parameter extraction system and method therefor.

[0018] That is, Korean Patent Publication No. 2023-0101789 receives voltage and current values, which are direct battery data for calculating SOH, and does not utilize the operating data and long-term data of the product proposed in the present invention. Furthermore, it does not describe a method for calculating and providing improved battery status information using the operating data and long-term data. Therefore, the two inventions differ significantly in their technical structure, purpose, and effects.

[0019] The present invention was created to solve the above problems, and its purpose is to provide a method for improving battery status information using long-term data, which estimates battery status information including SOH, RUL, or a combination thereof using operating data of a product equipped with a battery even when direct data measured for the battery is not provided, and improves the estimated battery status information using long-term data, which is a statistical distribution of battery status information according to mileage for multiple products operated by a specific battery.

[0020] In addition, the present invention aims to provide a method for estimating battery status information using limited data including discharge data and charge data included in driving data.

[0021] In addition, the present invention provides a method for calculating statistical information including a median, an average, a maximum, a minimum, or a combination thereof from a statistical distribution of battery status information of long-term data corresponding to the mileage of a specific product, calculating a weight for the degree of reflection of the median or average of the calculated statistical information and the estimated battery status information, and improving the estimated battery status information according to the weight.

[0022] In addition, the present invention estimates battery status information of a specific product using statistical information and mileage included in the operation data of the specific product, calculates weights for the degree of reflection of the battery status information estimated using the mileage and the battery status information estimated using the operation data, and provides a method for improving the battery status information estimated using the operation data according to the weights.

[0023] In addition, the present invention aims to provide a method for estimating the weight using battery status information estimated by mileage using full data on the battery of a product equipped with the same battery as a specific product.

[0024] A method for improving battery status information using long-term data according to one embodiment of the present invention includes a step of receiving operating data indicating a current operating status of a specific product that runs on a battery, a step of estimating first battery status information using the received operating data, and a step of improving battery status information using the estimated first battery status information and long-term data, wherein the long-term data is characterized in that it is a statistical distribution of battery status information according to mileage for a plurality of products that run on a specific battery.

[0025] In addition, the product is an electric vehicle, and the driving data includes discharge data including speed, external temperature, internal temperature, heating and cooling set temperature, or a combination thereof when discharging the electric vehicle, charging data including charging rate, internal temperature, external temperature, or a combination thereof when charging the electric vehicle, mileage data including driving distance of the electric vehicle, usage period of the battery of the electric vehicle, or a combination thereof, and the battery status information is characterized in that it includes SOH (state of health), RUL (remaining useful lifetime), or a combination thereof.

[0026] In addition, the above long-term data is characterized by being collected by country, region, season, or a combination thereof.

[0027] In addition, the battery status information improvement step includes a statistical information calculation step for calculating statistical information including a median, an average, a maximum, a minimum, or a combination thereof from a statistical distribution of battery status information corresponding to the mileage of a specific electric vehicle in the long-term data, and a weight calculation step for calculating a weight for the degree of reflection of the median or average value among the calculated statistical information and the estimated first status information of the battery, and the first status information of the battery is improved by reflecting the average or the median value of the calculated statistical information and the estimated first status information of the battery according to the calculated weight, and the result of improving the first status information of the battery is characterized in that it is limited to the maximum and minimum values ​​of the calculated statistical information.

[0028] In addition, the weight calculation step is characterized in that the weight calculation step extracts the battery status information corresponding to the mileage of the specific electric vehicle from a mapping table that maps the battery status information estimated by mileage using full data including voltage, current, temperature, charge, or a combination thereof for the corresponding battery of an electric vehicle equipped with the same battery as the specific electric vehicle, calculates a first error obtained by subtracting the first status information of the estimated battery from the extracted status information of the battery, and a second error obtained by subtracting the median or average value among the calculated statistical information from the extracted status information of the battery, and calculates the weight by calculating the ratio of the result of adding the first error and the second error to the first error (= first error / (first error + second error)).

[0029] In addition, the method for improving the battery status information further includes a battery second status information estimation step for estimating the second status information of the battery by using statistical information including a median, an average, a maximum, a minimum, or a combination thereof calculated from a statistical distribution of battery status information corresponding to the mileage of the specific electric vehicle in the long-term data, and the battery second status information estimation step is characterized in that the mileage and statistical information of the specific electric vehicle are input into a learning model generated by learning each learning data labeling the status information of the battery for each data set composed of the mileage and statistical information, thereby estimating the second status information of the corresponding battery.

[0030] In addition, the battery status information improvement step includes a weight calculation step of calculating weights for the degree of reflection of the first status information of the estimated battery and the second status information of the estimated battery, and the first status information of the battery is improved by reflecting the first status information of the estimated battery and the second status information of the estimated battery according to the calculated weights, and the result of improving the first status information of the battery is characterized in that it is limited to the maximum and minimum values ​​of the calculated statistical information.

[0031] In addition, the weight calculation step is characterized in that the weight calculation step extracts the battery status information corresponding to the mileage of the specific electric vehicle from a mapping table that maps the battery status information estimated by mileage using full data including voltage, current, temperature, charge, or a combination thereof for the corresponding battery of an electric vehicle equipped with the same battery as the specific electric vehicle, calculates a third error obtained by subtracting the estimated first status information of the battery from the extracted status information of the battery, and a fourth error obtained by subtracting the estimated second status information of the battery from the extracted status information of the battery, and calculates the ratio of the result of adding the third error and the fourth error to the third error (= third error / (third error + fourth error)) to calculate the weight.

[0032] In addition, a device for improving battery status information using long-term data according to one embodiment of the present invention is characterized by including a memory for storing a program code implementing a method for improving battery status information using the long-term data, and a processor configured to load and execute the program code stored in the memory.

[0033] As described above, the present invention provides improved battery status information using long-term data consisting of long-term statistical distributions of battery status information according to mileage for multiple products and operation data of a specific product, thereby enabling efficient management of the battery and stable use of the product.

[0034] FIG. 1 is a diagram illustrating a method for improving battery status information using long-term data according to one embodiment of the present invention.

[0035] FIG. 2 is a diagram illustrating long-term data according to one embodiment of the present invention.

[0036] FIG. 3 is a drawing illustrating in detail a method for improving battery status information according to one embodiment of the present invention.

[0037] FIG. 4 is a drawing illustrating in detail a method for improving status information of a battery according to another embodiment of the present invention.

[0038] FIG. 5 is a drawing illustrating in detail a method for improving status information of a battery according to another embodiment of the present invention.

[0039] FIG. 6 is a diagram illustrating a learning model for estimating battery status information according to one embodiment of the present invention.

[0040] FIG. 7 is a block diagram showing the configuration of a battery status information improvement device using long-term data according to one embodiment of the present invention.

[0041] FIG. 8 is a flowchart illustrating a procedure for providing improved status information of a battery according to one embodiment of the present invention.

[0042] FIG. 9 is a flowchart illustrating a procedure for providing improved status information of a battery according to another embodiment of the present invention.

[0043] [Description of symbols] 100: Battery status information enhancement device; 110: Driving data receiving unit; 120: Long-term data receiving unit; 130: Battery status information estimation unit; 131: Battery first status information estimation unit; 132: Battery second status information estimation unit; 140: Battery status information enhancement unit; 141: Statistical information calculation unit; 142: Weight calculation unit; 150: Battery status information providing unit; 200: Driving data providing device; 300: Long-term data providing server; 400: Database.

[0044] Hereinafter, with reference to the attached drawings, a preferred embodiment of a method for improving battery status information using long-term data of the present invention will be described in detail. The same reference numerals in each drawing represent the same components. In addition, specific structural and functional descriptions of embodiments of the present invention are merely illustrative for the purpose of explaining embodiments according to the present invention, and unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by a person of ordinary skill in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning they have in the context of the related technology, and it is preferable not to interpret them in an ideal or excessively formal sense unless explicitly defined in this specification.

[0045] FIG. 1 is a diagram illustrating a method for improving battery status information using long-term data according to one embodiment of the present invention.

[0046] As illustrated in FIG. 1, a battery status information enhancement device (100) using long-term data according to an embodiment of the present invention (hereinafter referred to as a battery status information enhancement device) is configured to receive operation data according to the use of the product (10) from the product (10), collect long-term data from a long-term data provision server (300), and provide enhanced battery status information to the product (10).

[0047] Here, the product (10) refers to various battery-powered devices, including electric vehicles. The present invention will be described in detail below using electric vehicles as an example.

[0048] Additionally, the product (10) is configured to provide driving data to a battery status information improvement device (100) through a driving data provision device (200).

[0049] The above driving data provision device (200) may be an OBD (on board diagnostics) of an electric vehicle (10).

[0050] The above operation data is configured to include discharge data when discharging the product (10), charge data when charging, mileage (mileage data) according to the operation of the product (10), or a combination thereof.

[0051] That is, the above driving data is configured to include discharge data including speed, external temperature, internal temperature, heating and cooling set temperature of the electric vehicle (10) or a combination thereof when discharging the electric vehicle (10); charging data including charging rate of the electric vehicle (10), charging rate of the electric vehicle (10), internal temperature, external temperature or a combination thereof when charging the electric vehicle (10); mileage data including driving distance of the electric vehicle (10), usage period of the battery mounted on the electric vehicle (10) or a combination thereof; or a combination thereof.

[0052] The long-term data provision server (300) may be provided by the manufacturer server of the electric vehicle (10) and may be configured to provide the long-term data to the driving data provision device (200). It is preferable that such long-term data be provided by the long-term data provision server (300). However, the present invention is not limited thereto and may be generated, stored, and managed in advance.

[0053] The above long-term data refers to the statistical distribution of battery status information according to mileage for multiple products (10) (electric vehicles) that run on a specific battery, and can be collected by country, region, season, or a combination thereof.

[0054] The above battery status information includes state of health (SOH), remaining useful lifetime (RUL), or a combination thereof. However, it is not limited thereto and may be composed of various indicators indicating the battery status according to deterioration, such as state of power (SOP), state of energy (SOE), etc.

[0055] The above battery status information improvement device (100) estimates the status information (hereinafter referred to as first status information of the battery) of the battery mounted on the specific electric vehicle (10) using the discharge data or charging data of the driving data received from the specific electric vehicle (10).

[0056] That is, the battery status information enhancement device (100) is configured to estimate the first status information of the battery from the discharge data received when the battery is discharged according to the operation of a specific electric vehicle (10), or to estimate the first status information of the battery from the charge data received when the battery is charged according to the charging of a specific electric vehicle (10).

[0057] Meanwhile, to estimate battery status information during battery discharge, the amount of charge supplied to the electric vehicle (passed charge) must be calculated. To calculate this amount of charge, current (discharge current) is required. However, the discharge data of the present invention is not data directly measured on the battery (voltage, current, charge, temperature, etc.).

[0058] Accordingly, the battery status information improvement device (100) is configured to estimate the current value of the battery when discharging by using the discharge data of the received driving data, estimate the amount of charge provided to the electric vehicle (10) by using the estimated current value, and estimate the first status information of the battery by using the estimated amount of charge.

[0059] At this time, the battery status information improvement device (100) learns learning data labeled with current values ​​for each discharge data, inputs the discharge data of the received driving data into a current estimation learning model to estimate the current value, and estimates the amount of charge provided to the electric vehicle (10) during discharge using the estimated current value, and then calculates the ratio of the previously estimated amount of charge to the estimated amount of charge, thereby estimating the status information of the battery during discharge.

[0060] In addition, the battery state information enhancement device (100) estimates the first state information of the battery by using charging data when charging an electric vehicle (10), divides the charging section of the battery into a plurality of detailed charging sections according to the charging data, calculates the amount of supply charge supplied to the battery for a predetermined time during which the state of charge (SOC) changes, estimates the reference supply charge amount supplied to the battery when the battery is new, and calculates the ratio of the estimated reference supply charge amount to the calculated amount of supply charge, thereby estimating the first state information of the battery when charging. The reference supply charge amount may be configured to be estimated by inputting the received charging data into a learning model for estimating the reference supply charge amount, which is generated by learning learning data in which the reference supply charge amount is labeled for each charging data collected for each detailed charging section when the battery is new.

[0061] Meanwhile, there are no limitations on the method for estimating the first state information of the battery. That is, the present invention can be performed through various methods that can estimate the first state information of the battery using limited data, such as discharge data and charging data according to the use of the electric vehicle (10), without using data directly measured from the battery.

[0062] The above battery status information improvement device (100) uses the estimated first status information of the battery and long-term data to improve the estimated first status information of the battery (i.e., improved status information of the battery) and provides it to the electric vehicle (10), thereby enabling the driver to efficiently manage the battery and stably operate the electric vehicle (10).

[0063] The database (400) illustrated in FIG. 1 is configured to store and manage various data or information for providing improved status information of the battery, including received driving data, collected long-term data, and each learning model for estimating status information of the battery.

[0064] Improving the status information of the above battery will be explained with reference to FIGS. 3 to 5.

[0065] FIG. 2 is a diagram illustrating long-term data according to one embodiment of the present invention.

[0066] As illustrated in FIG. 2, long-term data according to one embodiment of the present invention refers to a statistical distribution of battery status information based on mileage (i.e., mileage data) for multiple electric vehicles (10) powered by a specific battery. Here, mileage includes the driving distance of the electric vehicle (10), the battery's lifespan, or a combination thereof.

[0067] The above statistical distribution is expressed as the number of electric vehicles according to battery status information, and can be expressed as a normal distribution.

[0068] This long-term data is collected by product, battery, country, region, season, or a combination thereof, and is used to improve the battery's condition information by reflecting the battery's cycle degradation and storage degradation in the battery's first condition information.

[0069] The long-term data illustrated in Fig. 2 illustrates, as an example, the statistical distribution of battery status information according to the battery's usage period. The long-term data of the present invention may be composed of a statistical distribution of battery status information according to mileage, including the driving distance of an electric vehicle (10), the battery's usage period, or a combination thereof.

[0070] FIG. 3 is a drawing illustrating in detail a method for improving battery status information according to one embodiment of the present invention.

[0071] As illustrated in FIG. 3, a battery status information enhancement device (100) according to one embodiment of the present invention receives driving data indicating the current driving status of a specific electric vehicle from a driving data provision device (200) and collects long-term data from a long-term data provision server (300). Here, the driving data includes discharge data when the electric vehicle (10) is discharged, and charging data when it is charged.

[0072] In addition, the battery status information enhancement device (100) estimates the first status information of the battery using the discharge data or charge data of the received driving data. The estimation of the first status information of the battery is performed using a learning model, which has been described with reference to FIG. 1 and is therefore omitted here.

[0073] In addition, the battery status information improvement device (100) performs statistical analysis on long-term data corresponding to the mileage of the received driving data to produce statistical information including an average value (mean), a median value (mid), a minimum value (min), a maximum value (max) or a combination thereof of the battery status information. That is, the battery status information improvement device (100) produces the statistical information from a statistical distribution of battery status information corresponding to the mileage including the driving distance of the electric vehicle, the battery's usage period or a combination thereof of the received driving data from the collected long-term data.

[0074] At this time, it is desirable to use long-term data corresponding to the country, region and current season where the electric vehicle (10) is located.

[0075] Next, the battery status information improvement device (100) calculates the median or average value among the calculated statistical information and the weight (W) for the degree of reflection of the first status information of the estimated battery in order to calculate the improved status information of the battery.

[0076] At this time, the battery status information enhancement device (100) calculates the weight using a mapping table that maps battery status information estimated by mileage using full data including voltage, current, temperature, charge, or a combination thereof for the corresponding battery of an electric vehicle equipped with the same battery as a specific electric vehicle (10). At this time, since the battery status information is estimated using data directly measured for the battery, it is more accurate than the first status information of the battery estimated using limited data.

[0077] That is, the battery status information improvement device (100) extracts battery status information corresponding to the mileage of the driving data received from the mapping table for calculating the weight and calculates the weight.

[0078] In addition, the battery status information improvement device (100) calculates a first error by subtracting the first estimated battery status information from the battery status information extracted from the mapping table, and a second error by subtracting the median or average value among the calculated statistical information from the extracted battery status information.

[0079] In addition, the battery status information improvement device (100) calculates a weight by calculating the ratio of the result of adding the first error and the second error to the first error (= first error / (first error + second error)).

[0080] In addition, the battery status information improvement device (100) improves the first status information of the battery by calculating improved status information of the battery using the median or average value of statistical information, the estimated first status information of the battery, and the calculated weight according to the following [Mathematical Formula 1], thereby providing the improved first status information of the battery to the electric vehicle (10).

[0081] [Mathematical Formula 1]

[0082] F1 = A x (1 - W) + B x W

[0083] Here, F1 represents improved battery status information, A represents the median or average value of battery status information, which is statistical information calculated by performing statistical analysis on long-term data, W represents a weight, and B represents the first status information of the battery.

[0084] Additionally, the status information of the improved battery is limited to the maximum and minimum values ​​of the statistical information. That is, if the status information of the improved battery exceeds the maximum value, it is set to the maximum value, and if it is below the minimum value, it is set to the minimum value and provided to the electric vehicle (10).

[0085] FIG. 4 is a drawing illustrating in detail a method for improving battery status information according to one embodiment of the present invention.

[0086] Figure 4 illustrates a process for improving battery status information when there is no mapping table for calculating weights or when weights cannot be calculated.

[0087] At this time, the battery status information improvement device (100) is configured to provide the improved battery status information to the electric vehicle (10) by limiting the first status information of the battery estimated by the maximum and minimum values ​​of the statistical information.

[0088] FIG. 5 is a drawing illustrating in detail a method for improving status information of a battery according to another embodiment of the present invention.

[0089] Figure 5 illustrates a method for providing improved battery status information by estimating first and second status information of a battery using operating data and statistical information.

[0090] As illustrated in FIG. 5, a battery status information improvement device (100) according to one embodiment of the present invention receives driving data of a specific electric vehicle (10) from a driving data provision device (200) and collects long-term data from a long-term data provision server (300).

[0091] The above-described battery status information enhancement device (100) estimates the first status information of the battery using the discharge data or charge data of the received driving data, and performs statistical analysis on long-term data corresponding to the mileage of the received driving data to produce statistical information on the battery status information. The production of the statistical information is performed using the same method as that used to produce the statistical information described with reference to FIG. 3. As described above, the estimation of the first status information of the battery utilizes a learning model.

[0092] Additionally, the battery status information improvement device (100) estimates the second status information of the battery using the mileage of the received driving data and the calculated statistical information.

[0093] Estimating the second state information of the battery utilizes a battery state information estimation learning model for estimating the second state information of the battery. That is, the battery state information enhancement device (100) inputs the mileage of the received driving data and the calculated statistical information into the battery state information estimation learning model to estimate the second state information of the battery. The battery state information estimation learning model will be described in detail with reference to FIG. 6.

[0094] Thereafter, the battery status information improvement device (100) calculates a weight (W) for the degree of reflection of the first status information of the battery and the second status information of the battery in order to calculate improved status information of the battery.

[0095] At this time, the battery status information enhancement device (100) extracts battery status information corresponding to the mileage of the driving data received from the mapping table for calculating the weight and calculates the weight. The mapping table has been described with reference to FIG. 3, and is therefore omitted here.

[0096] In addition, the battery status information improvement device (100) calculates a third error obtained by subtracting the first state information of the estimated battery from the state information of the battery extracted from the mapping table, and a fourth error obtained by subtracting the median or average value among the calculated statistical information from the extracted state information of the battery, and calculates a weight by calculating the ratio of the result of adding the third error and the fourth error to the third error (= third error / (third error + fourth error)).

[0097] In addition, the battery status information improvement device (100) improves the first status information of the battery by calculating improved status information of the battery using the first status information of the battery estimated according to the following [Mathematical Formula 2], the second status information of the battery, and the calculated weight, thereby providing the improved first status information of the battery to the electric vehicle (10).

[0098] [Equation 2]

[0099] F2 = C x (1 - W) + D x W

[0100] Here, F2 represents enhanced battery status information, C represents the second status information of the estimated battery, W represents the weight, and D represents the first status information of the estimated battery.

[0101] Additionally, the improved battery status information is limited to the maximum and minimum values ​​of the statistical information. This limitation is performed in the same manner as in Fig. 3.

[0102] That is, as described with reference to FIGS. 3 and 5, the battery status information enhancement device (100) of the present invention can provide more precise battery status information by variably setting weights according to the operation of the electric vehicle (10). However, the weights may also be set to fixed values ​​in advance.

[0103] FIG. 6 is a diagram illustrating a learning model for estimating battery status information according to one embodiment of the present invention.

[0104] As illustrated in FIG. 6, a learning model for estimating battery status information according to one embodiment of the present invention can be created in a separate learning server (not shown).

[0105] The above learning server creates learning data by labeling battery status information for each data set composed of mileage including the driving distance of an electric vehicle, the battery's usage period, or a combination thereof, and statistical information according to each mileage.

[0106] Additionally, the learning server uses each of the above-mentioned learning data to create a learning model for estimating battery status information using a pre-prepared DNN (deep neural network).

[0107] The above DNN is configured to include an input layer including a plurality of input nodes, a hidden layer including a plurality of hidden nodes, and an output layer including an output node.

[0108] Each input node of the input layer and each hidden node of the hidden layer, and each hidden node of the hidden layer and each output node of the output layer are each connected through a link having a predetermined weight.

[0109] The above input layer receives the mileage and statistical information of the learning data, respectively.

[0110] The output node of the above output layer is configured to output the learning result (battery status information) according to the input, and since the learning server already knows the result (battery status information) according to the input input to the input layer during the learning process, it updates (adjusts) the weight of the link to reduce the error between the learning result output during the learning process and the actual current value.

[0111] The above weights refer to learning parameters, and the above learning is performed by updating the weights through the back propagation method, which allows the error to be reduced by back propagating the error to the DNN.

[0112] At this point, the DNN, which has completed learning by studying all training data, becomes the learning model for estimating battery condition information. The inputs to this learning model are the mileage of driving data received from an actual electric vehicle (10) and statistical information derived from statistical analysis of long-term data based on the mileage.

[0113] FIG. 7 is a block diagram showing the configuration of a battery status information improvement device using long-term data according to one embodiment of the present invention.

[0114] As illustrated in FIG. 7, a battery status information improvement device (100) according to one embodiment of the present invention is configured to include a driving data receiving unit (110), a long-term data collection unit (120), a battery status information estimation unit (130), a battery status information improvement unit (140), and a battery status information providing unit (150).

[0115] The above driving data receiving unit (110) receives driving data indicating the current driving status of the electric vehicle (10) from the driving data providing device (200).

[0116] The above driving data includes discharge data, charging data, mileage data including the driving distance of the electric vehicle (10), the usage period of the battery, or a combination thereof.

[0117] Here, the driving distance of the electric vehicle (10) refers to the driving distance of the electric vehicle (10) according to the life of the battery. In other words, it refers to the driving distance of the electric vehicle (10) after the battery is installed, not the total driving distance of the electric vehicle (10).

[0118] The above long-term data collection unit (120) collects long-term data from the long-term data provision server (300). The long-term data is collected every time it is updated.

[0119] The above battery status information estimation unit (130) is configured to include a battery first status information estimation unit (131) that estimates the first status information of the battery and a battery second status information estimation unit (132) that estimates the second status information of the battery.

[0120] The above battery first status information estimation unit (131) estimates the first status information of the battery using discharge data or charge data of the collected driving data.

[0121] The above-mentioned battery second status information estimation unit (132) is performed when it is desired to provide improved battery status information using statistical information.

[0122] The above battery status information improvement unit (140) is configured to include a statistical information calculation unit (141) and a weight calculation unit (142).

[0123] The above statistical information generating unit (141) generates statistical information from the statistical distribution of battery status information corresponding to the daily log of operation data received from the collected long-term data.

[0124] The above battery status information enhancement unit (140) calculates improved battery status information by using the median or average value of the calculated statistical information and the estimated first status information of the battery, and calculates a weight by using the median or average value of the statistical information and the estimated first status information of the battery through the weight calculation unit (142). Thereafter, the battery status information enhancement unit (140) improves the first battery status information by using the median or average value of the statistical information, the first status information of the battery, and the calculated weight. That is, the battery status information enhancement unit (140) calculates improved battery status information by using the median or average value of the statistical information, the first status information of the battery, and the calculated weight.

[0125] The improvement of the first state information of the battery by using the median or average value of the statistical information calculated above and the estimated first state information of the battery has been described with reference to Fig. 3, so it is omitted here.

[0126] Meanwhile, the battery second status information estimation unit (132) operates when it is desired to provide improved status information of the battery by using the battery's first status information and statistical information.

[0127] The above-described battery second status information estimation unit (132) estimates the second status information of the battery using the mileage of the received driving data and the calculated statistical information. The estimation of the second status information of the battery has been described with reference to FIGS. 5 and 6, and thus will be omitted here.

[0128] In addition, when the battery status information improvement unit (140) wants to calculate improved battery status information using the estimated first status information and the second status information of the battery, it calculates weights using the first status information and the second status information of the battery through the weight calculation unit (142). Thereafter, the battery status information improvement unit (140) improves the first battery status information using the first status information and the second status information of the battery.

[0129] The process of calculating improved battery status information using the first status information of the battery and the second status information of the battery has been described with reference to FIG. 5, so it is omitted here.

[0130] The above battery status information providing unit (160) provides improved battery status information to the electric vehicle (10).

[0131] FIG. 8 is a flowchart illustrating a procedure for providing improved status information of a battery according to one embodiment of the present invention.

[0132] As illustrated in FIG. 8, a procedure for providing improved status information of a battery according to one embodiment of the present invention includes a battery status information improvement device (100) performing a driving data receiving step of receiving driving data and a long-term data collecting step of collecting long-term data (S110).

[0133] The above driving data receiving step receives driving data indicating the current driving status of the electric vehicle (10), and the long-term data collecting step can collect the long-term data whenever the long-term data is updated.

[0134] Next, the battery status information enhancement device (100) performs a battery first status information estimation step (S120) of estimating the first status information of the battery using discharge data or charge data of the collected driving data, and performs a battery status information enhancement step using the long-term data and the first status information of the battery.

[0135] The above battery status information improvement step first performs a statistical information calculation step of calculating statistical information on battery status information from a statistical distribution of battery status information corresponding to mileage of driving data in long-term data (S130).

[0136] As described above, the above statistical information includes an average value, a median value, a maximum value, a minimum value, or a combination thereof for the battery status information.

[0137] Next, the battery status information improvement step performs a weight calculation step that calculates a weight for the median or average value of the calculated statistical information and the degree of reflection of the first status information of the estimated battery (S140).

[0138] As described above, the above weight is calculated using a mapping table that maps the battery status information estimated by mileage using full data on the battery.

[0139] Next, the battery status information improvement step performs the first battery improved status information calculation step that improves the first status information of the battery by reflecting the median or average value of the statistical information and the estimated first status information of the battery according to the calculated weight (S150).

[0140] That is, the first improvement step is to calculate improved status information of the battery by using the calculated weights, the median or average value of the calculated statistical information, and the estimated first status information of the battery.

[0141] Meanwhile, the battery health information enhancement step limits the improved battery health information to the maximum and minimum statistical values. That is, if the improved battery health information exceeds the maximum statistical value, it is set to the maximum value; if it falls below the minimum value, it is set to the minimum value.

[0142] Next, the battery status information improvement device (100) performs a battery status information provision step of providing the improved status information of the calculated battery to the electric vehicle (10) (S160).

[0143] FIG. 9 is a flowchart illustrating a procedure for providing improved status information of a battery according to another embodiment of the present invention.

[0144] As illustrated in FIG. 9, a procedure for providing improved status information of a battery according to another embodiment of the present invention first performs a battery status information improvement device (100) performing a driving data reception step of receiving driving data and a long-term data collection step of collecting long-term data (S210).

[0145] Next, the battery status information improvement device (100) performs a battery first status information estimation step (S220) for estimating the first status information of the battery using discharge data or charge data of the collected driving data, and performs a statistical information calculation step for calculating statistical information on the status information of the battery from a statistical distribution of the status information of the battery corresponding to the mileage of the driving data in the long-term data (S230).

[0146] The above driving data receiving step, long-term data collection step, and statistical information producing step are identical to the driving data receiving step, long-term data collection step, and statistical information producing step described with reference to FIG. 8.

[0147] Next, the battery status information enhancement device (100) performs a battery second status information estimation step of estimating the second status information of the battery by using the mileage of the received driving data and the calculated statistical information (S240).

[0148] Estimating the second state information of the above battery uses a learning model for estimating battery state information, and is omitted here as it has been described with reference to FIGS. 5 and 6.

[0149] Next, the battery status information improvement device (100) performs a battery status information improvement step.

[0150] The above battery condition information enhancement step first performs a weight calculation step (S250) that calculates weights for the degree of reflection of the estimated first and second condition information of the battery. The weights are calculated using a mapping table that maps the estimated battery condition information by mileage using full battery data.

[0151] Next, the battery status information enhancement step performs a second battery enhanced status information calculation step that enhances the first status information of the battery by reflecting the first status information of the battery and the second status information of the battery according to the calculated weights (S260).

[0152] In other words, the second battery enhanced status information calculation step calculates enhanced battery status information using the calculated weights and the estimated first and second battery status information. Meanwhile, the battery status information improvement step limits the enhanced battery status information to the maximum and minimum statistical values.

[0153] Next, the battery status information improvement device (100) performs a battery status information provision step of providing the improved status information of the calculated battery to the electric vehicle (10) (S270).

[0154] In addition, a battery management device (100) according to one embodiment of the present invention is configured to include a memory that stores a program code that implements a method for improving battery status information using the long-term data, as described with reference to FIGS. 7 and 8, and a processor configured to load and execute the program code stored in the memory.

[0155] As described above, the present invention has the effect of enabling efficient battery management and stable use of the product by calculating and providing improved battery status information using battery status information estimated with a limited battery and long-term data.

[0156] In addition, although the preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and various modifications can be implemented by a person having ordinary skill in the art to which the invention pertains without departing from the gist of the present invention claimed in the claims, and such modifications should not be understood individually from the technical idea or prospect of the present invention.

[0157] As described above, the present invention has industrial applicability because it provides improved battery status information using long-term data consisting of long-term statistical distributions of battery status information according to mileage for multiple products and operation data of a specific product, thereby enabling efficient management of the battery and stable use of the product.

Claims

1. A driving data receiving step for receiving driving data indicating the current driving status of a specific battery-powered product; A battery first state information estimation step for estimating the first state information of the battery using the received operating data; and A battery status information enhancement step for enhancing the first status information of the battery by using the first status information and long-term data of the battery estimated above; A method for improving battery status information using long-term data, characterized in that the above long-term data is a statistical distribution of battery status information according to mileage for multiple products running on a specific battery.

2. In claim 1, The above product is an electric vehicle, The above driving data includes discharge data including speed, external temperature, internal temperature, heating and cooling set temperature or a combination thereof when discharging the electric vehicle, charging data including charging rate, internal temperature, external temperature or a combination thereof when charging the electric vehicle, mileage data including driving distance of the electric vehicle, usage period of the battery of the electric vehicle or a combination thereof, or a combination thereof. A method for improving battery status information using long-term data, characterized in that the battery status information includes SOH (state of health), RUL (remaining useful lifetime), or a combination thereof.

3. In claim 1, The above long-term data is, A method for improving battery status information using long-term data, characterized in that the data is collected by country, region, season, or a combination thereof.

4. In claim 2, The above battery status information improvement step is: A statistical information generation step for generating statistical information including a median, average, maximum, minimum, or a combination thereof from a statistical distribution of battery status information corresponding to the mileage of a specific electric vehicle in the long-term data; and It includes a weight calculation step for calculating a weight for the median or average value among the statistical information calculated above and the degree of reflection of the first state information of the battery estimated above; The first state information of the battery is improved by reflecting the average or median of the statistical information calculated according to the weight calculated above and the first state information of the battery estimated above. A method for improving battery status information using long-term data, characterized in that the result of improving the first status information of the battery is limited to the maximum and minimum values ​​of the calculated statistical information.

5. In claim 4, The above weight calculation step is, Extracting battery status information corresponding to the mileage of the specific electric vehicle from a mapping table that maps battery status information estimated by mileage using full data including voltage, current, temperature, charge, or a combination thereof for the corresponding battery of an electric vehicle equipped with the same battery as the specific electric vehicle, A first error is calculated by subtracting the first state information of the estimated battery from the state information of the extracted battery, and a second error is calculated by subtracting the median or average value among the calculated statistical information from the state information of the extracted battery. A method for improving battery status information using long-term data, characterized in that the weight is calculated by calculating the ratio of the result of adding the first error and the second error to the first error (= first error / (first error + second error)).

6. In claim 2, The above battery status information improvement method is: It further includes a battery second state information estimation step for estimating second state information of the battery by using statistical information including a median, average, maximum, minimum or a combination thereof calculated from a statistical distribution of battery state information corresponding to the mileage of a specific electric vehicle in the long-term data; The above battery second state information estimation step is: A method for improving battery status information using long-term data, characterized in that the mileage and statistical information of a specific electric vehicle are input into a learning model created by learning each learning data labeled with battery status information for each data set composed of the mileage and statistical information, thereby estimating the second status information of the battery.

7. In claim 6, The above battery status information improvement step is: A weight calculation step for calculating weights for the degree of reflection of the first state information of the estimated battery and the second state information of the estimated battery; The first state information of the battery is improved by reflecting the first state information of the battery and the second state information of the battery estimated according to the calculated weights. A method for improving battery status information using long-term data, characterized in that the result of improving the first status information of the battery is limited to the maximum and minimum values ​​of the calculated statistical information.

8. In claim 7, The above weight calculation step is, Extracting battery status information corresponding to the mileage of the specific electric vehicle from a mapping table that maps battery status information estimated by mileage using full data including voltage, current, temperature, charge, or a combination thereof for the corresponding battery of an electric vehicle equipped with the same battery as the specific electric vehicle, A third error is calculated by subtracting the first state information of the estimated battery from the state information of the extracted battery, and a fourth error is calculated by subtracting the second state information of the estimated battery from the state information of the extracted battery. A method for improving battery status information using long-term data, characterized in that the weight is calculated by calculating the ratio of the result of adding the third error and the fourth error to the third error (= third error / (third error + fourth error)).

9. A memory storing a program code that implements a method for improving battery status information using the long-term data according to any one of claims 1 to 8; and A device for improving battery status information using long-term data, characterized in that it includes a processor configured to load and execute program code stored in the above memory.

Images