TW201816415A - Expandable modular battery capacity estimation system using a Coulomb Counting method to accumulate a charging/discharging electric quantity in a charging/discharging mode, and substituting an open-circuit voltage into a relational expression in a rest mode - Google Patents

Abstract

An expandable modular battery capacity estimation system is disclosed, which has a mother board, at least a daughter board, and a man-machine interface. The at least one daughter board acquires an output voltage of each battery unit of a battery set, a series current and an operation temperature of the battery set in order to carry out a monitoring procedure and a battery capacity estimation procedure. The battery capacity estimation procedure comprises using a Coulomb counting method to accumulate charging/discharging electric quantity in a charging/discharging mode, and substituting an open-circuit voltage of one of the battery units into a relational expression of the open-circuit voltage and the battery capacity percentage in a rest mode, so as to estimate the battery capacity.

Description

   Scalable modular battery capacity estimation system   

The invention relates to an expandable modular battery capacity estimation system, in particular to an expandable modular battery capacity estimation system suitable for a series lithium-ion battery pack.

With the development of technology, portable and wearable digital products are becoming more and more popular, and there are more and more electronic devices such as smart home appliances and electric vehicles. These smart electronic devices must rely on battery power. The secondary battery can be seen on the power supply side of most portable products. In addition to being widely used in portable electronic products, secondary batteries are also used in electrical energy storage for aviation, vehicles and renewable energy. On the system. According to the chemical composition of batteries, secondary batteries are currently mainly classified into lead-acid batteries, nickel-cadmium batteries (Ni-Cd), nickel-metal hydride batteries (Ni-MH), and lithium-ion batteries (Lithium- ion) and other four types, of which lithium-ion batteries are the mainstream of current secondary batteries, which have the advantages of high energy density, light weight, high average operating voltage, and no memory effect. However, the performance of a lithium-ion battery is affected by its own capacity. Overcharge and overdischarge may cause safety issues such as explosions. Therefore, the battery power and related information are very important. If the user can know the percentage of the battery capacity, he can effectively use the battery and determine the charging and discharging time to avoid the above situation.

To design a perfect battery capacity estimation system, it must include the following functions: battery status monitoring: monitor the voltage of each battery cell, the current of the battery pack, and the temperature of the battery pack to prevent battery damage and other conditions.

Battery capacity percentage and current load usable time: Estimate the battery capacity percentage and current load usable time to help users understand the current capacity of the battery.

Modularization: Since a single battery does not have the ability to support the operation of a large system, modularization enables multiple groups of batteries to be monitored by a single system uniformly.

High accuracy: The higher the accuracy of battery information sampling, the more accurately the battery usage can be monitored.

Communication interface: Since the management and protection of multiple groups of batteries are controlled by a single system at the same time, the communication interface is very important for the battery capacity estimation system.

In addition, for the capacity estimation of lithium-ion batteries, the common estimation methods in the literature include the open circuit voltage method, Coulomb integration method, internal resistance method, Kalman filter, and impedance tracking. Among them, a mathematical model based on temperature is proposed in the literature, and a three-dimensional table (open circuit voltage-capacity percentage-temperature) and an unscented Kalman filter (UKF) are used to achieve capacity estimation. Another literature proposes a mathematical model based on impedance, and uses an equivalent impedance model of second-order circuit components combined with a Fractional Kalman Filter (FKF) to estimate capacity; some researchers have also proposed capacity estimation based on voltage retention curves Method, and extended Kalman filter to modify its algorithm equivalent model. In addition, Texas Instruments proposed a patent for battery capacity estimation based on impedance tracking technology, and introduced its basic principle and update time of various data. The database created is based on the use of different types of batteries. There is also a literature that proposes an adaptively adjusted capacity estimation method based on extended Kalman filters and electrochemical models, which uses high-order parameter adjustment systems to reduce the effect of errors; another researcher has proposed the use of Coulomb integral method to estimate battery capacity. At the same time, the accuracy of the capacity estimation is corrected by using the voltage to the coulomb change trend line. In addition, an adaptive non-linear correction algorithm is used to compensate for the non-linear situation in the actual application of the battery, and the accuracy of the increased battery capacity estimation is also the previous Proposed technology.

Although conventionally there are various battery capacity estimation schemes as described above, there is still room for improvement in its performance.

An object of the present invention is to provide a modular battery capacity estimation system suitable for a series lithium-ion battery pack. Its functions include monitoring the voltage of each battery cell, the temperature of the battery pack, and the current of the battery pack, and estimating the percentage of battery capacity. And remaining capacity, and avoid overcharging, overdischarging, overvoltage, undervoltage, overcurrent or overtemperature of the battery.

In order to achieve the above purpose, an expandable modular battery capacity estimation system has been proposed, which has: a mother board with a first digital controller; at least one daughter board, each with at least a differential amplifier, a current A sensor, a temperature sensor, a second digital controller and a third digital controller, the at least one differential amplifier is used to capture the unit output of each battery cell of a series lithium-ion battery module Voltage, the current sensor is used to sense the series current flowing through the series lithium-ion battery module, the temperature sensor is used to sense the operating temperature of the series lithium-ion battery module, and the third digit The controller communicates with the second digital controller through a universal asynchronous transmission / reception transmission interface, and communicates with the first digital controller through a controller area network bus interface to communicate the second digital The information of the unit output voltage, the series current, and the operating temperature read by the controller is transmitted to the first digital controller; and a human-machine interface is transmitted from the first digital through an RS232 interface. The controller obtains the output voltage, the series current, and the operating temperature of each of the units to perform a monitoring procedure and a battery capacity estimation procedure. The battery capacity estimation procedure includes using a Coulomb integral in a charge / discharge mode. Method to accumulate charge / discharge power, and in a rest mode, substitute an open circuit voltage of the battery cell into a relation between the open circuit voltage and a battery capacity percentage to estimate a battery capacity.

In one embodiment, the rest mode further includes a step of correcting the open circuit voltage by a product of the series current and an internal resistance.

In one embodiment, the human-machine interface has the functions of issuing an over-voltage warning, an under-voltage warning, an over-current warning, and an over-temperature warning.

In order to enable your reviewers to further understand the structure, characteristics, and purpose of the present invention, drawings and detailed descriptions of the preferred embodiments are attached below.

100‧‧‧Motherboard

101‧‧‧The first digital controller

110‧‧‧ daughter board

120‧‧‧ human-machine interface

111‧‧‧ Differential Amplifier

112‧‧‧Current sensor

113‧‧‧Temperature sensor

114‧‧‧Second Digital Controller

115‧‧‧Third Digital Controller

200‧‧‧series lithium-ion battery module

FIG. 1 is a schematic diagram of an embodiment of an expandable modular battery capacity estimation system according to the present invention.

FIG. 2 is a schematic diagram of a battery current sampling circuit adopted in the present invention.

FIG. 3 is a schematic diagram of a voltage sampling circuit adopted in the present invention.

FIG. 4 is a schematic diagram of a battery temperature sampling circuit adopted in the present invention.

Figure 5 is a schematic diagram of RS-232 voltage level conversion.

Figure 6. Schematic diagram of differential signals for CAN Bus.

Figure 7. Schematic diagram of CAN Bus hardware circuit.

FIG. 8 is a flowchart of a sampling plate system program adopted by the present invention.

FIG. 9 is a flowchart of a CAN Bus communication daughter board system program adopted by the present invention.

FIG. 10 is a flowchart of a CAN Bus communication motherboard system program adopted by the present invention.

FIG. 11 illustrates the overall operation structure of a VSP potentiostat / constant current instrument manufactured by BioLogic Corporation and used EC-Lab software for experiments.

FIG. 12 is a relationship diagram between the open circuit voltage and the capacity percentage of the battery.

FIG. 13 is a fitting relationship diagram of 20% to 100% battery capacity and one of the open circuit voltages.

Figure 14 is a fitting relationship between 0% to 20% of battery capacity and one of the open circuit voltages.

FIG. 15 is a flowchart of a battery capacity estimation program adopted in the present invention.

FIG. 16 is a flowchart of estimating the remaining capacity of a battery adopted in the present invention.

FIG. 17 is a flow chart of a procedure for updating the internal resistance value of a battery adopted in the present invention.

FIG. 18 is a flowchart of a rest mode subroutine adopted by the present invention.

FIG. 19 is the authentication-battery pack information provided by the man-machine interface of the present invention.

FIG. 20 is a verification-battery unit information provided by the human-machine interface of the present invention.

FIG. 21 is an internal resistance and open-circuit voltmeter provided by the man-machine interface of the present invention.

FIG. 22 is a waveform diagram of a charging experiment according to the present invention.

FIG. 23 is a graph of the estimation result of the capacity percentage of a charging experiment according to the present invention.

FIG. 24 is a capacity percentage estimation error graph of a charging experiment according to the present invention.

FIG. 25 is a waveform diagram of a discharge experiment according to the present invention.

FIG. 26 is a graph showing a result of capacity percentage estimation in a discharge experiment according to the present invention.

FIG. 27 is a capacity percentage estimation error graph of a discharge experiment according to the present invention.

Please refer to FIG. 1, which is a schematic diagram of an embodiment of an expandable modular battery capacity estimation system according to the present invention. As shown in FIG. 1, the expandable modular battery capacity estimation system includes a motherboard 100, at least one daughter board 110, and a human-machine interface 120.

The motherboard 100 has a first digital controller 101, and the first digital controller 101 can be implemented by a microprocessor.

Each of the at least one daughter board 110 has at least one differential amplifier 111, a current sensor 112, a temperature sensor 113, a second digital controller 114, and a third digital controller 115. The at least one differential amplifier 111 is used to capture a unit output voltage (V ₁ , V ₂ , V ₃ , V ₄ …) of each battery cell of a series lithium-ion battery module 200. The current sensor 112 Is used to sense the series current flowing through the series lithium-ion battery module 200, the temperature sensor 113 is used to sense the operating temperature of the series lithium-ion battery module 200, and the third digital controller 115 is Communicate with the second digital controller 114 through a UART (Universal Asynchronous Transmit / Receive Transmission) interface, and communicate with the first digital controller 101 through a CAN Bus (controller area network bus) interface, The information of the unit output voltage, the series current, and the operating temperature read by the second digital controller 114 is transmitted to the first digital controller 101. In addition, the second digital controller 114 and the third digital controller 115 can both be implemented by a microprocessor or integrated in the same microprocessor.

The human-machine interface 120 obtains the output voltage of each of the units, the series current, and the operating temperature from the first digital controller 101 through an RS232 interface to perform a monitoring procedure and a battery capacity estimation procedure, wherein the battery The capacity estimation procedure includes using a Coulomb integration method to accumulate charge / discharge power in a charge / discharge mode, and substituting an open circuit voltage of the battery cell into one of the open circuit voltage and a battery capacity percentage in a rest mode. The relationship is used to estimate a battery capacity. In addition, the rest mode may further include the step of using the product of the series current and the internal resistance of a battery to correct the open-circuit voltage to improve the estimation accuracy, and the human-machine interface may further have an over-voltage warning, an under-voltage warning, and an over-voltage warning. Flow warning and over temperature warning function.

The principle of the present invention will be explained below:

Hardware architecture of capacity estimation system:

The expandable battery capacity estimation system of the present invention samples battery voltage, temperature, and current respectively, and transmits them to a computer through a communication interface to display and calculate the battery capacity through a human-machine interface, and simultaneously monitors the battery for overcharging and overcharging. In the case of overheating, overcurrent, overtemperature, etc., a warning signal will be issued if there is any abnormality.

Design specifications and sampling circuit:

The sampling circuit includes battery voltage, battery pack current, and battery pack temperature sampling. The parameters of the sampling circuit designed by the present invention are as follows: Test battery specifications: The test battery selected by the present invention is the NCR18650B lithium-ion battery introduced by PANASONIC and provided by PANASONIC. The battery specification table is shown in Table 1.

Table 1. NCR18650B specifications     

Various data measurement ranges:

Number of batteries: 4 strings of batteries as a module

Measurable battery voltage range: 0V ~ 5V

Measurable battery current range: 0A ~ 10A

Measurable battery temperature range: 0 ℃ ~ 55 ℃

protection mechanism:

Overcharge voltage protection: 4.25V

Over-discharge voltage protection: 2.8V

Overcharge current protection: 3.4A

Over-discharge current protection: 6.8A

Over temperature protection: 55 ℃

Current sampling circuit:

FIG. 2 is a schematic diagram of a battery current sampling circuit adopted in the present invention. The present invention uses a HX-10P Hall element produced by LEM to measure current. The HX-XXP current sensor introduced by LEM company can convert the two-way rated current flowing into a voltage signal of plus or minus 4 volts. Taking the HX-10P used in the present invention as an example, it is assumed that the flowing current is 4 amps in the positive direction. The voltage value converted by the Hall element is 1.6 volts. Because the battery has two possibilities of charging and discharging, the value converted by the Hall element will also have positive and negative values. Therefore, the voltage value converted by the Hall element must be first added by the adder and 4 volts. The voltage signal The range becomes 0 to 8 volts, and then the maximum value of the voltage signal is reduced to the full value of the analog / digital converter. In this way, there is no need to adjust the circuit depending on the charge and discharge conditions, and no negative voltage will appear.

Voltage sampling circuit:

One voltage sampling circuit adopted in the present invention is shown in FIG. 3. After the voltage of each battery cell enters the differential amplifier and is converted into a voltage signal (V ₁ ~ V ₄ ), it enters the analog / digital converter inside the MCU for conversion. The battery pack of the present invention is connected in series. If the battery cells are simply connected to the analog / digital converter, the short circuit between the batteries will be caused by the common ground of the negative terminals of the battery cells, and the battery may be damaged and there may be an explosion. Therefore, the present invention designs a differential amplifier between the analog / digital converter and the battery pack to avoid the above-mentioned danger.

Temperature sampling circuit:

In the present invention, an AD590 temperature sensor introduced by Analog Devices is used for temperature sampling. The temperature sensing range is -55 ° C to 150 ° C. The output current has a linear relationship with the temperature. Each 1 ° K rise increases the current by 1 μA. FIG. 4 is a schematic diagram of a battery temperature sampling circuit adopted in the present invention. As shown in Figure 4, the AD590 itself is a current source, so connecting the AD590 negative terminal to a resistor can convert its current into a corresponding voltage signal. In order to avoid loading effects when entering the analog / digital converter, the AD590 A first-level voltage follower is added to the output terminal, so that the wrong voltage signal generated by the AD590 due to the load effect can be avoided.

Communication interface circuit:

The battery capacity estimation system needs to deal with the update and transmission of a large amount of data, so the choice of communication interface is very important. As can be seen from the system architecture diagram, the communication methods used by the present invention according to different applications include CAN Bus and RS-232 (UART). The reason for the selection is as follows: under different conditions, the battery capacity estimation system must increase or decrease the number of communication daughter boards due to different battery numbers, and the data must be continuously updated. Therefore, it is necessary to choose a bus type and an interface that can communicate instantly. CAN Bus is a bus type. Its nodes are expandable, the communication distance is long, and it is universal and safe. Therefore, the present invention uses CAN Bus as the communication interface between the mother board and the daughter board.

Between the sampling board and the CAN Bus communication sub-board, because of the point-to-point transmission and the short communication distance, the present invention chooses to use a simple UART (Universal Asynchronous Transmit / Receive Transmission) interface as the communication interface between the sampling board and the sub-board.

The user computer generally has a serial communication connector (COM PORT). Therefore, the present invention selects RS-232 as the communication interface between the computer man-machine interface and the CAN Bus communication motherboard.

The RS-232 communication protocol uses a full-duplex working mode. In the line configuration, transmission and reception are performed by independent lines. In addition to the ground reference line, at least 3 lines are required for transmission. Figure 5 is a schematic diagram of RS-232 voltage level conversion. In the figure, the UART logic 0 and logic 1 voltage levels of the MCU (microcontroller) are 0V and 3.3V respectively; and the logic voltage level of the RS-232 terminal is ± 3V ~ Between 15V, RS-232 uses negative logic definition. Before the data enters the computer, it will pass through a first-level reverse circuit. Therefore, the MCU's UART must pass through the chip that converts the voltage level before transmitting the data to the computer. The data complies with RS-232 specifications.

The transmission bit of the controller area network bus (CAN Bus) has only 0 (recessive) and 1 (dominant, dominate) status, which mainly depends on two lines: CAN_H (high potential signal) and CAN_L (low Potential signal). In the ISO 11898 standard, when CAN_H and CAN_L are equal to 2.5V, the bit value is set to 1; when CAN_H rises to 3.5V and CAN_L drops to 1.5V, the bit value is set to 0, as shown in Figure 6. In the CAN Bus circuit design, the specific specification must add a 120Ω terminal resistor between CAN_H and CAN_L at the start and end points of its network. Its function is to detect errors. If 60Ω is measured at both ends, it is normal. If 120Ω is measured at both ends, it means that one end of the main line is open. In communication, each node sending CAN data packets must be sent to the CAN bus through the CAN transceiver. The present invention selects the MCP2551 CAN transceiver produced by Microchip, which complies with the ISO 11898 specification. The signal is transmitted using differential voltage. The CAN Bus hardware circuit of the present invention is shown in FIG. 7, where the MCU's C1TX and C1RX are connected to the TXD and RXD ends of the CAN transceiver, and after conversion, they are connected to the bus CAN_H and CAN_L, so basically CAN Bus The communication only needs two copper stranded wires as the main line of the bus, and other nodes can communicate normally as long as they are connected to the main line.

The firmware architecture of the battery capacity estimation system:

The present invention selects different microcontrollers according to different communication interfaces. In the CAN Bus communication board part, the dsPIC33FJ64GP802 microcontroller produced by Microchip is used as the core controller, which mainly plays the role of data communication; and the sampling board selects dsPIC33FJ16GS502 microcontroller. Controller, which is mainly responsible for functions such as communication with CAN Bus communication MCU and sampling data calculation.

System program flow:

According to the different control conditions, it can be divided into three system programs: sampling of the daughter board, CAN Bus communication program, and CAN Bus communication program of the mother board.

Sampling system program: The sampling circuit uses dsPIC33FJ16GS502 as the main controller. The analog / digital converter has 10-bit resolution and the sampling frequency is set to 200kHz. Timer2 is set to interrupt every 25 milliseconds. It is used as a conversion trigger source for analog / digital converters. Channels 0 to 5 are temperature, current, and battery 1 to battery 4 voltages. After sampling and conversion, they are averaged 64 times and reused. The UART is sent to dsPIC33FJ64GP802. The program flow chart is shown in Figure 8.

CAN Bus communication sub-board system program: The CAN Bus communication sub-board uses dsPIC33FJ64GP802 as the main controller. Figure 9 shows the CAN Bus communication sub-board system program flow chart. After the MCU receives the battery information transmitted by the sampling controller through the UART, it stores it in In the DMA RAM, waiting for the RTR signal transmitted by the CAN Bus motherboard. Once the RTR signal is received, the controller will determine the mask and filter of the CAN Bus, and then pair the ID to check whether the RTR signal matches the distance of the daughter board. Remote control signal, if it is true, the battery information is transmitted to the CAN Bus communication motherboard through DMA RAM.

CAN Bus communication motherboard system program: The CAN Bus communication motherboard also uses dsPIC33FJ64GP802 as the main controller. Figure 10 shows the CAN Bus communication motherboard system program flow chart. Timer1 is set to interrupt once every 0.4 seconds. After accumulating 2 seconds, it depends on the different daughter boards. ID, the corresponding RTR signal is transmitted. After receiving the RTR signal, the daughter board will return the corresponding battery information. At this time, the mother board will receive it into the register through DMA. After confirming that all the battery information of the daughter board is received, the motherboard will use RS-232 communication to send the battery information to the computer and display it on the LabVIEW HMI.

Capacity estimation method:

In the capacity estimation method of the present invention, a modified open-circuit voltage method and a Coulomb Counting Method are used. In practice, the Mathscript module in LabVIEW is selected. First, the table building conditions and program flow of the capacity estimation method are explained.

Relationship between open circuit voltage, battery internal resistance and capacity percentage:

In the battery charge and discharge experiments of the present invention, VSP potentiostat / constant current instruments produced by BioLogic are used and EC-Lab software is used for experiments. The overall operation structure is shown in FIG. 11. EC-Lab can perform constant current charge / discharge, constant voltage charge, and constant potential AC impedance measurement. The battery test part of the present invention includes the relationship between the open circuit voltage and the capacity percentage and the battery DC internal resistance and the capacity percentage. The open circuit voltage and capacity percentage relationship is used for the resting state of the capacity estimation method. This method defines the charge and discharge current below 0.05. Below C is the resting state, and in the resting state, the measured voltage can be directly used to find the capacity percentage. The relationship between the DC internal resistance and the capacity percentage is used in two parts: When the first part is used in the resting state, because a small current (<0.05C) may still exist in the resting state set by this method, the internal resistance is the correction factor here. , The measured voltage will be corrected to an approximate open-circuit voltage and then the corresponding capacity percentage will be used; the second part is used to estimate the remaining capacity. This method uses the current load current as the basis for voltage simulation until the battery voltage reaches the cut-off voltage, and The internal resistance is used to correct the actual voltage.

The relationship between the open circuit voltage and the capacity percentage: charge at a constant current of 0.5C to 4.2V, and then charge at a constant voltage of 4.2V until the charging current is less than 20mA (full battery), and then discharge at 0.5% discharge current per 1% of capacity After that, take another 1.5 hours to take the open circuit voltage, and the relationship curve obtained is shown in Figure 12.

The relationship between battery DC internal resistance and capacity percentage: The DC impedance part uses HIOKI 3561 battery internal resistance tester produced by HIOKI. The DC impedance table is shown in Table 2.

Capacity estimation:

The capacity estimation method of the present invention is a modified open-circuit voltage method. At the beginning, the measured terminal voltage is corrected to obtain an approximate open-circuit voltage, which is then sent to a curve fitting function to obtain the initial capacity percentage. The battery capacity is 0% to The 20% fit uses a sin aggregate function model, and the battery capacity of 20% to 100% uses a polynomial function fit. Formula (1) and formula (2) are the battery capacity of 20% to 100% and 0% to 20, respectively. % Applied mathematical model, where V _OC is the open circuit voltage and SOC is the percentage of battery capacity; Table 3 and Table 4 are fitting coefficients, and the fitting results are shown in Figures 13 and 14. By mistake! No reference source found. It can be seen that there is still an error when using curve fitting to approximate the percentage of real capacity, so first use curve fitting to find a similar percentage of capacity within 0% to 100%, and then use interpolation to correct the error.

SOC ( V _OC ) = a × V _OC ² + b × V _OC + c (1)

SOC ( V _OC ) = a 1 × sin ( b 1 × V _OC + c 1) + a 2 × sin ( b 2 × V _OC + c 2) + a 3 × sin ( b 3 × V _OC + c 3) (2)

The capacity estimation of the present invention is set to be executed once every two seconds. The flowchart is shown in FIG. 15. At the beginning, the program will first determine whether it is the first time to enter the algorithm. If it is true, it will build the table, initialize the parameters, and calculate the initial capacity. Then the current will be used as the mode judgment criterion. If it is the charge mode or the discharge mode, the coulomb integration will be performed. The remaining capacity estimation is based on the current load current to determine how much capacity the user can provide if the discharge current is discharged to the cut-off voltage. The influencing factors include the discharge current, the internal resistance of the battery, and the set cut-off voltage. Figure 16 is a flow chart of the remaining capacity estimation. At the beginning of the program, the current capacity percentage is used as the initial capacity percentage, and then every 2% is used as a stage simulation. Each stage will determine whether the cut-off voltage has been reached. If not, proceed to In the next stage, if it is true, the simulated capacity percentage at this moment is regarded as the end capacity percentage, and then the difference between the initial and end capacity percentages is multiplied by the total power to obtain the current remaining capacity.

The update of the internal resistance is mainly to correct the situation that the internal resistance of the battery rises due to aging or other factors. It can also avoid errors in initializing the meter and actual use. Its action timing is the end of discharge. FIG. 17 is a flowchart of a procedure for updating the internal resistance. The internal resistance meter of this estimation method is established with a step of 5% between 20% and 100% of the capacity, and a step of 2% between 0% and 20%, so 0% is deducted when the internal resistance is updated. There will be a total of 25 points at 100%. At the beginning of updating the internal resistance program, it will determine whether the battery is currently at these 25 points. If true, then use the current measured voltage minus the current open circuit voltage corresponding to the current capacity percentage divided by the current The current, that is, the corrected internal resistance, is written into the internal resistance versus capacity percentage table to complete the update of the internal resistance program.

If the load current is less than ± 0.05C, it enters the rest mode subroutine. The flowchart is shown in Figure 18. At the beginning of the program, it will first determine whether to enter the rest mode from the discharge mode. If it is true, calculate the remaining capacity and capacity percentage, and estimate the time that the battery can be used in the current state of use. Because the battery must change from charging or discharging to a resting state, it must wait for the battery chemical reaction to make the measured voltage approach the open circuit voltage. Therefore, the invention is designed to use the rest mode for 30 minutes and then use it every 100 seconds. The voltage is converted into the initial capacity at this stage, and the converted initial capacity is obtained by substituting the measured voltage into the fitting function to find the approximate initial capacity, and then correcting by interpolation to obtain the true initial capacity.

Monitoring interface:

The present invention uses LabVIEW graphical system design software developed by National Instruments to write a monitoring interface to display battery parameters and waveforms, and simultaneously output to an Excel file to record changes in parameters such as voltage, current, temperature, and capacity percentage. The display section of the human-machine interface is divided into three pages, which are the battery pack status, the status of each battery cell, and the internal resistance and open circuit voltage versus capacity percentage table of each battery cell. Battery pack status includes information such as 1 battery pack voltage, 1 battery pack temperature, 1 battery pack current value and waveform, battery pack percentage and remaining capacity, and current load usage time. The display information of the status of each battery cell includes the value and waveform of the voltage of the four battery cells, the open circuit voltage and the remaining capacity after the calculation of each battery cell. The relationship table between internal resistance and open circuit voltage and capacity percentage of each battery cell is recorded on the last page. The capacity estimation method will change the battery internal resistance through different conditions. Therefore, the system displays the relationship table between internal resistance and capacity percentage on the man-machine interface. For users to observe the change of battery internal resistance. The system is set to access the battery and capacity estimation information every 2 seconds. The battery information includes real-time date and time, 4 battery cell voltages, 1 battery voltage, 1 battery current, and 1 battery temperature. The rule of law records real-time date and time, 4 battery cell capacity percentages and remaining capacity, and 1 battery cell capacity percentages and remaining capacity. The two storage paths can be selected from the battery pack status tab.

Figure 19 is the battery pack information page, which shows the battery pack information including total voltage and current values and waveforms, battery temperature, cumulative charge and discharge capacity, total capacity percentage, remaining capacity and current load usable time, and battery status. The bottom left corner is RS -232 input port setting, from the top half to the bottom, battery and capacity algorithm information storage path selection, warning sign display and parameter correction, the storage file of the present invention is mainly comma-separated files of Excel; warning sign Divided into five types, each battery unit over-voltage and under-voltage warning, battery pack over-temperature warning, battery pack over-current warning and battery pack over-current warning; parameter correction function is to adjust and adapt to different battery use conditions.

FIG. 20 is an information page of each battery cell. The displayed information of each battery cell includes the value and waveform of each battery cell voltage, the percentage and remaining capacity of each battery cell capacity, and the open circuit voltage of each battery cell after modification. Figure 21 is the page of internal resistance and open circuit voltmeter. The display information includes the internal resistance meter of 0% to 20% and 20% to 100% of each battery cell and the open circuit voltmeter of 0% to 20% and 20% to 100%. During the execution of the algorithm, the internal resistance is updated according to the battery usage and aging conditions, and the updated value will be displayed on this page immediately.

Experimental results and verification:

Verification accuracy of lithium ion battery parameter estimation:

The sampling data of this estimation system are battery pack temperature, battery pack current, and battery cell voltage. The analog / digital converters used in the present invention are all analog / digital converters inside dsPIC33FJ16GS502, and the resolution is 10 bits. The control group used the GDM-8342 digital electricity meter produced by Guwei Company and the thermal imaging device Ti 105 produced by Fluke Company as instruments for measuring current, voltage and temperature. GDM-8342 has a voltage accuracy of 0.002% and a current accuracy of 0.002%; the temperature accuracy of Ti 105 is 2%. Table 5 shows the voltage and temperature measurement results, and Table 6 shows the battery pack current measurement results. The charge and discharge currents are measured at different intervals from 0.1C to 1C. It can be seen from the table that the voltage accuracy can reach 99.94%. In addition to the larger current range error (2.4% ~ 4.8%), the accuracy can reach below 99.78%, and the temperature accuracy can reach 96.99%.

Table 6. Current measurement results     

Lithium-ion battery capacity estimation accuracy verification:

The invention designs different charge and discharge waveforms (including rest mode) to verify the accuracy of the capacity estimation system under different charge and discharge conditions. Figure 22 shows the charging experiment waveforms. Tables 7 and 8 show the capacity percentage and cumulative capacity test results and error table of the charging experiment. The error value is the actual measured value minus the theoretical value. Figures 23 and 24 are the capacity percentage estimates of the charging experiment. Measurement and error graph.

According to the test chart of the charging experiment, the capacity percentage of the battery to be tested should be 15% at the beginning, but the actual measurement result falls between 16.3% and 16.8%. The reason may be that the battery under test has passed after discharging to 15%. One day of standing time results in an error; as far as the Coulomb integral method is concerned, the error in the part of continuous current variation is as large as -83.92C (-0.69%), but the error in the constant current charging part can be controlled to less than -26C (- 0.21%) or less.

Figure 25 shows the discharge experiment waveforms. Tables 9 and 10 show the capacity percentage and cumulative capacity test results and error tables for the discharge experiments. Figures 26 and 27 show the discharge experiment capacity percentage estimation results and error charts. From the experimental chart of the discharge experiment, it can be seen that the part with the larger error appears in the rest mode, and the maximum error at the end of the discharge reaches 2.42%, while the part of the Coulomb integral method falls between 8.8C (0.07%) to -16.3C (- 0.13%).

Verification of Lithium-ion battery remaining capacity estimation accuracy:

In addition to estimating the capacity percentage of the battery, the capacity estimation method of the present invention also calculates the remaining capacity of the battery based on the current load current. The invention is designed to test the discharge current of 0.5C and 0.6C after the battery is fully charged, and the capacity estimation method calculates the remaining capacity and the use time after entering the discharge mode, and then continues to discharge the battery at this discharge current until Some batteries in the battery pack are lower than the cut-off voltage (the present invention is set to 2.8V), and the entire discharge process is compared with the calculation result of the capacity estimation method to verify the accuracy of the remaining capacity estimation. Table 11 and Table 12 show the remaining capacity verification results using 0.5C and 0.6C as the load current. Because the capacity estimation method has the function of updating the internal resistance, this verification experiment will compare the use of the original internal resistance with the updated internal resistance. Estimated results. As can be seen from the table, the results of the two groups of experiments using the original internal resistance are significantly different from the actual use situation. The estimated use time is about 8 minutes different from the actual discharge time, and the estimated use time and actual discharge time of the updated internal resistance are used. The error can be reduced to within 2 minutes.

in conclusion:

The lithium-ion battery capacity estimation system of the present invention uses a 10-bit analog / digital converter inside the MCU with peripheral circuits to sample battery parameters. The system of the present invention can achieve 99.94% accuracy in voltage sampling, and the current sampling is in addition to small current. , The accuracy can reach 99.78%, and the temperature sampling accuracy can reach 96.99%. In terms of capacity estimation, although the open-circuit voltage method will have an error of 2% to 3% in the initial capacity estimation, in the next rest period, the error caused by the open-circuit voltage method will be reduced to 1.5%; and the Coulomb integral method will cause The error is between 0.13% and 0.96%. After the use time estimated from the remaining capacity is corrected by the updated internal resistance, the error from the actual use time is less than two minutes. The human-machine interface plan uses page display to make it humane and easy to read. The human-machine interface of the present invention can also automatically detect battery abnormalities and issue warnings, and record the battery and capacity estimation information in Excel files. Facilitate user inspection and archiving.

The disclosure of the present invention is a preferred embodiment, and any change or modification that is partly derived from the technical idea of the present invention and easily inferred by those skilled in the art will not depart from the scope of patent rights of the present invention.

To sum up, the present invention, regardless of the purpose, means and effect, is showing its technical characteristics that are quite different from the conventional ones, and its first invention is practical, and it also meets the patent requirements of the invention. Pray for granting patents at an early date.

Claims (3)

    An expandable and modular battery capacity estimation system includes: a mother board with a first digital controller; at least one daughter board, each with at least one differential amplifier, a current sensor, and a temperature sensor A second digital controller and a third digital controller, the at least one differential amplifier is used to capture the unit output voltage of each battery cell of a series lithium-ion battery module, and the current sensor system is The temperature sensor is used to sense the series current flowing through the series lithium-ion battery module. The temperature sensor is used to sense the operating temperature of the series lithium-ion battery module. The third digital controller is a The synchronous transmitting and receiving transmission interface communicates with the second digital controller, and communicates with the first digital controller through a controller area network bus interface to communicate with each of the second digital controllers. The information of the unit output voltage, the series current and the operating temperature are transmitted to the first digital controller; and a human-machine interface is obtained by the first digital controller through an RS232 interface. Voltage, the series current, and the operating temperature to perform a monitoring procedure and a battery capacity estimation procedure, wherein the battery capacity estimation procedure includes using a Coulomb integration method to accumulate charge / discharge power in a charge / discharge mode, and In a rest mode, an open circuit voltage of the battery cell is substituted into a relation between the open circuit voltage and a percentage of a battery capacity to estimate a battery capacity.         The scalable modular battery capacity estimation system described in item 1 of the scope of the patent application, wherein the rest mode further includes a step of correcting the open circuit voltage by a product of the series current and an internal resistance of a battery.         The expandable modular battery capacity estimation system described in the first item of the patent application scope, wherein the man-machine interface has the functions of issuing an over-voltage warning, an under-voltage warning, an over-current warning, and an over-temperature warning.