JP6411538B2 - Prediction system, prediction program, prediction device - Google Patents

Description

  Embodiments described herein relate generally to a technique for estimating and predicting a deterioration state of a lithium ion secondary battery.

  In the prior art, when estimating and predicting the future deterioration state of a lithium ion secondary battery, the deterioration state of the battery is predicted based on the information of the usage history of the secondary battery itself currently used.

  The following documents are disclosed.

JP 2013-181875 A JP 2014-149280 A

  In the prior art, since the past usage history of the user is used, it is necessary to store and store the history data each time. In addition, secondary batteries that cannot store history data have a problem that deterioration cannot be predicted.

  Also, in the prior art, since the past usage history of the secondary battery currently in use is used, the prediction is extrapolated (using known data, a value outside the range of the known data is obtained). Will be used. For example, when the value suddenly drops on the time axis or when the value rises rapidly, it is difficult to predict a sudden change in value by extrapolation.

  The problem to be solved by the present invention is to provide a technique for improving the prediction accuracy of the future deterioration state of the lithium ion secondary battery currently in use.

The prediction system of the embodiment is a system that predicts the lifetime of a secondary battery based on internal state quantities that are characteristic data of a plurality of materials constituting a lithium ion secondary battery, and includes a storage unit and a control unit. Have. Storage unit storing the elapsed time and the internal state quantity from the start of use of the secondary battery composed of different individuals from the secondary battery to be predicted, the same product or the same positive electrode material and the same negative electrode material as a main material To do. The control unit acquires the elapsed time and the internal state quantity from the start of use of the secondary battery to be predicted to the diagnosis reference date, and stores information on an individual whose relationship between the elapsed time and the internal state quantity is identical or approximate The capacity maintenance rate of the individual information of the search results is calculated , the predicted deterioration rate after the lapse of a specific period from the diagnosis reference date of the secondary battery to be predicted is obtained from this capacity maintenance rate transition , and the life prediction result is obtained. calculate.

It is a figure which shows the system configuration example of embodiment. It is a figure which shows the hardware structural example of a prediction server. It is a figure which shows an example of the functional block of the system of embodiment. It is a flowchart which shows the operation example of the prediction server of embodiment. It is a figure which shows an example of the log hit by search. It is a figure which shows an example of the log on the other use conditions which carried out the search hit. It is a figure which shows an example of the ratio of the period from a diagnostic reference date to a prediction date when the use start date of a diagnostic object cell, the capacity maintenance rate in a diagnostic reference date, and the period from a use start date to a diagnostic reference date are set to 1. is there. It is a figure explaining the derivation method of a prediction degradation rate. It is the figure which illustrated the list of the prediction deterioration rates of the cell obtained by a search result. It is a figure which shows the example of a display of the prediction result of embodiment. It is a figure which shows the hardware structural example of the prediction apparatus of embodiment.

  In the embodiment, a system for predicting the lifetime of a lithium ion secondary battery mounted on an EV vehicle (EV: Electric Vehicle) will be described. The life prediction is performed by predicting how much the capacity maintenance rate decreases in this example.

  The system of the embodiment is performed using the internal state quantity of the secondary battery of the same product, which is different from the secondary battery to be predicted. In the embodiment, usage histories of secondary batteries that are separate from each other, such as secondary batteries of the same product that have been used in advance and secondary batteries of the same product that have reached the end of product life, are registered in a plurality of databases. . In the system of the embodiment, the current state of the prediction target is matched by searching from the usage history of the secondary battery registered in the database using the current internal state quantity of the prediction target secondary battery. Or, extract data of past products that are approximate. The system of the embodiment uses this extracted data to predict the lifetime of the secondary battery to be predicted.

  In the present embodiment, life prediction is performed using an index called an internal state quantity. Each cell (secondary battery) has a configuration including materials such as a positive electrode active material and a negative electrode active material. The internal state quantity is made up of characteristic data of each material constituting the cell. That is, the internal state quantity is characteristic data of each of a plurality of materials. For example, the capacity of the active material A (for example, lithium manganate) constituting the positive electrode, the capacity of the active material B (for example, lithium cobaltate) constituting the positive electrode, and the negative electrode It includes characteristic data such as capacity (for example, graphite and lithium titanate), charge amount of the active material A constituting the positive electrode, charge amount of the active material B constituting the positive electrode, charge amount of the negative electrode, and internal resistance value.

  For example, when only one piece of characteristic data is focused on and life prediction is performed using only that piece of characteristic data, factors other than the piece of characteristic data are not considered, so the accuracy of the prediction result is not high. In the present embodiment, since life prediction is performed using a plurality of different types of characteristic data, it is possible to improve the accuracy of search results.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating an example of a system configuration of the embodiment. The system 100 according to the embodiment includes a prediction server 1 and an EV vehicle 2, which transmit / receive data to / from each other via the Internet 3 (wide area network). The prediction server 1 is installed in a place connected via the Internet 3, that is, a place far from the EV vehicle 2.

  The prediction server 1 receives a current value, a voltage value, and a battery temperature of a secondary battery mounted on the EV vehicle 2 via the Internet 3, calculates a life prediction result, and transmits it to the EV vehicle 2. Although the prediction server 1 is described as a single computer in the embodiment, it may be a server group including a plurality of computers, or a cloud system including a large-capacity storage and a network device.

  The EV vehicle 2 is an electric vehicle that has a lithium ion secondary battery and an electric motor and travels by driving the motor with electric power supplied from the secondary battery. The EV vehicle 2 has a mechanism that enables communication with the prediction server 1 via the Internet 3. The EV vehicle 2 measures the current value, voltage value, and battery temperature of the secondary battery, and transmits these measurement results to the prediction server 1. This transmission is a conventional communication means.

  FIG. 2 is a diagram illustrating a hardware configuration example of the prediction server 1. The prediction server 1 is a computer in this embodiment, and includes a processor 11, a memory 12, a ROM (Read Only Memory) 13, an HDD (Hard Disk Drive) 14, an input device 15, an output device 16, and a network I / F (Interface). 17. These devices transmit and receive control signals and data to and from each other via the communication bus 18.

  The processor 11 is an arithmetic processing unit such as a CPU (Central Processing Unit), for example, and loads a program stored in the ROM 13 or the HDD 14 into the memory 12 and executes the calculation. Thereby, the processor 11 performs various processes according to the program and controls each hardware. The memory 12 is a volatile main storage device. The ROM 13 is a non-volatile storage device that stores permanently, and BIOS (Basic Input Output System) at the time of system startup is stored in advance. The HDD 14 is a nonvolatile auxiliary storage device that can be permanently stored, and data, programs, and operating systems used by the user are stored in the HDD 14.

  The input device 15 is a device that receives information input from the user to the prediction server 1 and is a keyboard, a mouse, or the like. The output device 16 is a device that notifies the user of the processing contents and processing results of the prediction server 1, and is a monitor, a printer, or the like. The input device 15 and the output device 16 are used when the administrator of the prediction server 1 performs maintenance, for example.

  The network I / F 17 is a unit responsible for communication with an external device of the prediction server 1 and includes a LAN (Local Area Network) board. Note that the information obtained from the EV vehicle 2 is obtained via the network I / F 17.

  FIG. 3 is a functional block diagram of the system according to the embodiment. First, the configuration of the EV vehicle 2 will be described. The EV vehicle 2 includes a battery pack 201, a control unit 210, and a display operation unit 204. The battery pack 201 is a module in which a plurality of modules formed by combining a plurality of cells are connected and accommodated in a case. In the embodiment, the cell is a lithium ion secondary battery.

  The control unit 210 is a unit that controls the battery pack 201 via a CAN (Controller Area Network), and includes a measuring instrument 202 and a measurement result DB 203. The measuring device 202 receives and measures monitoring information on the current value, voltage value, and temperature of each cell accommodated in the battery pack 201 from a sensor attached in the battery pack 201. These measurement values are accumulated in the measurement result DB 203 that is a storage unit, and then transmitted to the prediction server 1 in accordance with a user operation or periodically. In the case of an EV vehicle that does not have a storage unit, the measurement value obtained by the measuring device 202 is transmitted to the prediction server 1 as it is.

  The display operation unit 204 is a control panel that receives a user operation for controlling the EV vehicle 2 and displays a vehicle state and the like. The control panel of the car navigation system may be used.

  Next, the configuration of the prediction server 1 will be described. The prediction server 1 includes a past battery state DB 101, an internal state amount calculation unit 102, a deterioration transition DB 104, a search unit 105, a prediction calculation unit 107, and a prediction result DB 108. Temporary data includes the current internal state quantity 103 and the coincidence log 106. It should be noted that the internal state quantity calculation unit 102, the search unit 105, and the prediction calculation unit 107 are different from each other when the processor 11 shown in FIG. It is a functional part realized in cooperation with the hardware resource of. The past battery state DB 101, deterioration transition DB 104, and prediction result DB 108 are realized by a data management mechanism of a relational database management system (RDBMS) that has been introduced in advance, but are realized by other means for data storage. May be.

  The past battery state DB 101 stores current, voltage, and temperature measurement results obtained from the same product of an individual different from the secondary battery to be predicted (battery pack 201 in this example). That is, the past battery state DB 101 stores the measurement results collected in the past for secondary batteries that are separate from the secondary battery to be predicted and are the same product. As indicated by the broken line arrows in FIG. 3, the current value, voltage value, and temperature measurement value currently transmitted from the EV vehicle 2 are registered in the past battery state DB 101, and the battery pack mounted on the other vehicle. When the lifetime is predicted, the measurement value may be used as past data.

  The same product refers to, for example, a car that is sold with the same name equipped with a cell composed of the same positive electrode material and the same negative electrode material as a main material, or a cell composed of the same positive electrode material and the same negative electrode material as a main material. A stationary storage battery system sold under the same name. In addition, battery modules and battery packs for cars equipped with cells sold under the same name may be the same product, or battery modules and battery packs for stationary applications equipped with cells sold under the same name. May be the same product.

  The internal state quantity calculation unit 102 acquires the measurement result of the prediction target secondary battery transmitted from the EV vehicle 2 and calculates the internal state quantity. This calculation result is shown as the current internal state quantity 103 in FIG. The internal state quantity is various characteristic data of each material used in the secondary battery, which is calculated by a conventional technique, for example, the battery performance estimation method disclosed in Patent Document 2. On the other hand, the internal state quantity calculation unit 102 obtains accumulated data accumulated in the past battery state DB 101 and calculates the internal state quantity. This calculation result is accumulated in the deterioration transition DB 104.

  The deterioration transition DB 104 is an internal state quantity of the same product of an individual different from the secondary battery to be predicted, such as a secondary battery of the same product used in advance or a secondary battery of the same product whose product life has expired. And the performance of the battery, for example, the transition of the capacity maintenance rate, are accumulated in correspondence (hereinafter referred to as a log). At this time, logs having different use conditions (temperature, SOC range, charge / discharge current, and their patterns) are accumulated, but details of use conditions are not necessarily accumulated.

  The prediction result DB 108 permanently stores the prediction results obtained by the prediction calculation unit 107 and accumulates data.

  The search unit 105, the prediction calculation unit 107, and the match log 106 will be described with reference to the flowchart of FIG.

  The search unit 105 refers to the deterioration transition DB 104 based on the internal state quantity of the secondary battery to be diagnosed, and searches for one or more secondary batteries having the same or similar internal state quantity in the deterioration transition DB 104. (S001). For example, the deterioration transition DB 104 is searched using the elapsed time from the start of use of the battery pack 201 to the present and the current internal state quantity, and the relationship between the elapsed time and the internal state quantity is identical or similar. Extract logs. For the similarity determination, for example, the closest value is used, or one or a plurality of values within a certain range are used. The search unit may be any of a pack unit, a module unit, and a cell unit. In this example, the search unit is described as a cell unit.

  The prediction server 1 calculates the capacity maintenance rate from the difference of the searched internal state quantity with respect to the internal state quantity at the start of use in the searched and extracted cell. The capacity maintenance ratio is a capacity ratio at the time of measurement with respect to the initial capacity. FIG. 5 is a diagram showing a transition graph (log) of the capacity maintenance rate of the cell that has been searched for among the cells stored in the deterioration transition DB 104. P1 indicated by a circle indicates the hit value of the current internal state quantity 103. FIG. 6 is a transition graph of the capacity maintenance rate for the log under another use condition of the cell that has been searched for, and the circle indicates the hit value of the current internal state quantity 103. This transition graph corresponds to the matching log 106.

  The initial capacity is a voltage defined as 0% of a cell predetermined at the time of product completion or product purchase (for example, discharge end voltage) until reaching an open circuit voltage (for example, end-of-charge voltage) defined as 100%. Various values such as total charge capacity [Ah] when charged can be defined and used.

  Returning to the flowchart of FIG. Next, the prediction calculation unit 107 calculates a prediction deterioration rate based on the obtained matching log 106 (S002). Calculation of the predicted deterioration rate is performed in the following order.

  The prediction calculation unit 107 calculates the ratio between the period from the use start date of the diagnosis target cell to the diagnosis reference date (data acquisition date) and the period from the diagnosis reference date to the prediction destination. FIG. 7 is a diagram showing the change over time of the capacity maintenance rate of the battery pack 201. The use start date in FIG. 7 is the date when the battery pack 201 is used, and the diagnosis reference date is the current date. Here, the prediction calculation unit 107 sets the period (15 months) from the use start date to the diagnosis reference date to 1, and the ratio of the period from the diagnosis reference date to the prediction date (3 months in this example) to this. calculate. In this example, since the period from the use start date to the diagnosis reference date is 15 months, the ratio is 0.2.

  Next, the prediction calculation unit 107 applies the ratio calculated above to the matching log 106 and extracts the deterioration rate at the prediction destination. FIG. 8 is the log shown in FIG. The prediction calculation unit 107 obtains the capacity maintenance rate of the value 0.2 (P2) ahead of the search hit value (P1), and calculates the difference between the capacity maintenance rates of P1 and P2. Here, it is derived that the value has decreased by 0.01, and this is assumed to be the predicted deterioration rate. The prediction calculation unit 107 extracts the prediction deterioration rate by the above method for all the matching logs.

  In the above description, the predicted deterioration rate with the predicted date 3 months after the diagnosis reference date is extracted. However, the predicted calculation unit 107 is not limited to this, but the same applies to other predicted dates such as 1 month and 2 months later. The predicted deterioration rate is calculated by the method. FIG. 9 shows a list of predicted deterioration rates of the searched cells after one month, two months, and three months. In the example of FIG. 9, the logs of the cells 1 to 8 are searched and extracted, and the predicted deterioration rate after 1 to 3 months is calculated for the cells 1 to 8. The list shown in FIG. 9 corresponds to the predicted deterioration rate 301 shown in the flowchart of FIG.

  Returning to the flowchart of FIG. Next, the prediction calculation unit 107 calculates the distribution of the predicted deterioration rate (S003). The prediction calculation unit 107 calculates the distribution of the predicted deterioration rate for each prediction date and determines the final prediction range. In this example, the prediction range is obtained from the average value ± standard deviation (μ ± σ) of the predicted deterioration rate. The life prediction result calculated in this way is data of this final prediction range, and is accumulated and stored in the prediction result DB 108.

  The life prediction result calculated according to the flowchart of FIG. 4 is transmitted to the EV vehicle 2, and the display operation unit 204 of the EV vehicle 2 displays each data of the life prediction result as a graph. FIG. 10 is a display example at this time. As shown in FIG. 10, the user can select cell units, module units, pack units, and, in the case of cell units, select cell identification information. The predicted date can also be selected, and the display operation unit 204 performs redisplay according to the setting selected by the user. In this way, the user can confirm future life predictions.

  In the above-described embodiment, a mode in which a server (computer) is used as one form of a unit that performs prediction calculation has been described. This is an implementation example for simultaneously processing requests from a plurality of EV vehicles and for collecting and collecting a plurality of data on a large scale. On the other hand, you may provide the device for a user to obtain the prediction result of an object battery easily on the spot. The prediction device 3 shown in FIG. 11 is a stationary or easy-to-carry device, and includes a microprocessor 311, an ASIC 312 (ASIC: Application specific integrated circuit), a memory 313, a nonvolatile auxiliary storage device 314, a display unit 315, an operation A portion 316 and an input / output terminal 317. The microprocessor 311 comprehensively controls hardware in the apparatus. The ASIC 312 is an integrated circuit for performing a specific process at high speed, and performs a part or all of the processes described above at high speed. The input / output terminal 317 is a terminal for directly connecting a cable to the control unit 210 of the EV vehicle 2, but may be a LAN connection or the like. The memory 313, the auxiliary storage device 314, the display unit 315, and the operation unit 316 employ conventional units. The display unit 315 and the operation unit 316 may be a touch panel display.

  Each function unit of the internal state quantity calculation unit 102, the search unit 105, and the prediction calculation unit 107 illustrated in FIG. 3 is realized by software control in the microprocessor 311 or ASIC 312. In order to reduce the size, the prediction device 3 may have a configuration without the past battery state DB 101 and the prediction result DB 108. In this case, data registration in the deterioration transition DB 104 is performed in advance and stored in the auxiliary storage device 315. The life prediction result calculated according to the flowchart of FIG. 4 is displayed as a graph on the display unit 315 as it is.

  In the present embodiment, the battery pack of the EV vehicle has been described as a prediction target. That is, any aspect may be used as long as it is a lithium ion secondary battery.

  In the prior art, the future prediction is performed by extrapolating using the past data of the secondary battery itself to be predicted, but the prediction result is extrapolated until an appropriate prediction model is constructed. It is difficult to get. On the other hand, in this embodiment, as shown in FIG. 5, FIG. 6, and FIG. 8, interpolation is performed using past data of the same product (determining the range of each section of the data string based on the existing data string). To predict. Thus, since the lifetime prediction of the target secondary battery is performed based on the past actual value of the same product, it is possible to predict a sudden change in value that is difficult to predict by extrapolation.

  Since the prediction is performed based on the past data of the same product in this way, if the forecast date is outside the range of the past data of the same product, the display operation unit 204 displays a message indicating that it is out of the scope of application. May be.

  Like this embodiment, lifetime prediction can be performed using the past data of the internal state quantity of the secondary battery of the same product of the individual different from the secondary battery of prediction object. Thereby, lifetime prediction can be performed without requiring information such as the usage history of the prediction target cell itself.

  In the present embodiment, an internal state quantity, which is composed of characteristic data of each material constituting the cell, is employed as an index. That is, by using the internal state quantity, it is possible to perform life prediction using a plurality of and many kinds of characteristic data. As in this embodiment, by using various types of characteristic data, the characteristic data A matches or approximates, the characteristic data B matches or approximates, and the characteristic data C matches or approximates when searching. In addition, the search condition can be narrowed down and the search result can be obtained with higher accuracy than the case of using one characteristic data or a small number of characteristic data.

  The term “life prediction” also includes lithium ion battery deterioration prediction, performance prediction, and performance estimation. The term of life prediction includes a concept for predicting and estimating a future battery state such as a state prediction and a deterioration state prediction of a lithium ion battery.

  In each embodiment, the description has been given in the case where the function for implementing the embodiment of each embodiment is recorded in advance in the apparatus. However, the present invention is not limited to this, and the same function may be downloaded from the network to the apparatus. Those having these functions stored in a recording medium may be installed in the apparatus. The recording medium may be any form as long as the recording medium can store the program and can be read by the apparatus, such as a CD-ROM. In addition, the function obtained by installing or downloading in advance may be realized in cooperation with an OS (operating system) inside the apparatus.

  In addition, although the term “cell” is used in the present embodiment, it may be referred to as a battery, a single battery, a secondary battery, a storage battery, or a battery cell. The battery module and the battery pack may each include a control board. Moreover, although this embodiment has been described using a battery mounted on an EV vehicle, it may be replaced with a stationary storage battery system.

  According to the embodiment, the prediction accuracy of the future deterioration state of the lithium ion battery can be improved.

  In addition, although embodiment of this invention was described, the said embodiment is shown as an example and is not intending limiting the range of invention. The novel embodiment can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

Claims (10)

A prediction system for predicting the lifetime of a secondary battery based on internal state quantities that are characteristic data of a plurality of materials constituting a lithium ion secondary battery,
Different individuals from the secondary battery to be predicted, a storage unit for storing the elapsed time and the internal state quantity from the start of use of the secondary battery composed of the same product or the same positive electrode material and the same negative electrode material as a main material ,
The elapsed time from the start of use of the secondary battery to be predicted to the diagnosis reference date and the internal state quantity are acquired, and information on the individual whose relationship between the elapsed time and the internal state quantity is identical or approximate is retrieved from the storage unit. , Calculate the capacity maintenance rate of the individual information of the search result, calculate the predicted deterioration rate after the lapse of a specific period from the diagnosis reference date of the secondary battery to be predicted from this capacity maintenance rate transition , and calculate the life prediction result A control unit,
A prediction system characterized by comprising: The prediction system according to claim 1, wherein the control unit interpolates an internal state quantity of a search result to calculate a life prediction result of the secondary battery to be predicted . The prediction system according to claim 1, wherein the internal state quantity is a plurality of characteristic data of a material constituting the lithium ion secondary battery. The storage unit, and characterized by storing one or both of the preceding and the internal state of the secondary battery of the same product became internal state or service life of the rechargeable battery of the same product being used The prediction system according to claim 1 . A control unit that measures and transmits the current, voltage, and battery temperature of the secondary battery to be predicted; and
An interface unit that receives measured values of the current, voltage, and battery temperature of the secondary battery to be predicted, and transmits the life prediction result;
A display unit for receiving the life prediction result and displaying the result in a graph;
The control unit calculates and obtains an internal state quantity based on a measurement value received by the interface unit, and controls transmission of the calculated lifetime prediction result of the predicted secondary battery via the interface unit. The prediction system according to claim 1, wherein: The prediction system according to claim 5, wherein the storage unit, the interface unit, and the control unit are installed at positions connected from the control unit via a wide area network. A program for causing a computer to execute a process of predicting the lifetime of a secondary battery based on internal state quantities that are characteristic data of a plurality of materials constituting a lithium ion secondary battery,
Obtain the elapsed time and internal state quantity from the start of use of the secondary battery to be predicted to the diagnosis reference date ,
Information about an individual in which the relationship between the elapsed time and the internal state quantity is identical or approximate is a secondary composed mainly of the same product or the same positive electrode material and negative electrode material of an individual different from the secondary battery to be predicted Search from the storage unit that has accumulated the elapsed time from the start of use of the battery and the internal state quantity,
The capacity maintenance rate of the individual information of the search results is calculated, and the predicted deterioration rate after the lapse of a specific period from the diagnosis reference date of the prediction target secondary battery is obtained from the transition of the capacity maintenance rate , and the life prediction result is calculated. A prediction program for causing a computer to execute processing. The prediction program according to claim 7 , wherein the computer executes a process of interpolating an internal state quantity of a search result and calculating a life prediction result of the prediction target secondary battery. A prediction device for predicting the lifetime of a secondary battery based on internal state quantities that are characteristic data of a plurality of materials constituting a lithium ion secondary battery,
A storage unit for storing different individuals from the secondary battery to be predicted, the elapsed time and the internal state quantity of the same product or the same positive electrode material and negative electrode material from the start of use of the secondary battery configured as a main material,
A first terminal for inputting measured values of the current, voltage, and battery temperature of the secondary battery to be predicted;
Based on the measurements received by the first terminal, the acquired elapsed time to the diagnostic reference date from with the use start calculating the internal state of the secondary battery to be predicted, and the elapsed time and the internal state quantity The storage unit is searched for information on individuals that have the same or similar relationship, and the capacity maintenance rate of the individual information in the search results is calculated . From this capacity maintenance rate transition , the diagnostic reference date of the secondary battery to be predicted is calculated. A control unit that calculates the predicted deterioration rate after a specific period from
The prediction apparatus characterized by having. The said control part interpolates the internal state quantity of a search result, and calculates the lifetime prediction result of the said secondary battery of prediction object, The prediction apparatus of Claim 9 characterized by the above-mentioned .

Images