A silicon carbide diode-based lithium battery management and protection system

Abstract

The application discloses a lithium battery management and protection system based on silicon carbide diode, by using the single conduction, photosensitive and other characteristics of the silicon carbide diode, several silicon carbide diodes are connected to the traditional lithium battery management system, so that the system realizes the functions of ESD protection, filtering, spark detection and the like. The ESD protection and the spark detection can effectively monitor abnormal phenomena such as high voltage and battery spark, and timely protect against the harm to the system, and meanwhile, the filtering function can effectively convert the AC component-containing electrical signal of the system into DC. The algorithm for estimating the SOC of the lithium battery by using the temperature correction extended Kalman filter in the system can reach convergence after 600 iterations, and the SOC estimation accuracy after iteration convergence reaches 0.5% or more, and the accuracy is improved by 0.2% compared with the method without using parameter correction.

Description

TECHNICAL FIELD

[0001] The application relates to a lithium battery management and protection system based on a silicon carbide diode, and belongs to the technical fields of lithium batteries and semiconductors. BACKGROUND

[0002] Silicon carbide (SiC) avalanche photodiodes (APDs) are high-performance photodetectors that utilize avalanche multiplication to enhance the detection sensitivity of optical signals. Compared to traditional silicon-based APDs, SiC APDs have a wider bandgap and higher thermal stability, allowing them to maintain excellent performance in extreme environments. The working principle of SiC APDs is based on the avalanche multiplication effect, where carriers gain enough energy in the depletion region under high reverse bias, generating more electron-hole pairs through collision ionization, thereby achieving signal amplification. This process can significantly improve the gain of the detector, making it advantageous in low-light signal detection. The high thermal conductivity and high breakdown voltage characteristics of SiC material enable SiC APDs to operate stably at high temperatures and high voltages, which is particularly important in aerospace, military, and industrial detection fields. In addition, SiC APDs also have fast response time and low dark current, making them widely used in high-speed optical communication and photon counting fields. With the advancement of SiC material preparation technology and the optimization of device design, the performance of SiC APDs has been significantly improved. Currently, SiC APDs have been applied to signal amplification and detection in optical communication systems, as well as high-precision photodetection in biomedical imaging, environmental monitoring, and space exploration fields.

[0003] Lithium battery management is a key technology to ensure the safe, efficient, and long-life operation of battery systems. With the widespread application of lithium batteries in portable electronic devices, electric vehicles, and energy storage systems, the importance of battery management systems (BMS) has become increasingly prominent. The main functions of BMS include battery state monitoring, charge and discharge control, thermal management, equalization control, and fault diagnosis. First, BMS needs to monitor the voltage, current, temperature, and capacity of the battery in real time to assess the health status and remaining capacity of the battery. Second, BMS controls the charging and discharging process to prevent overcharging and overdischarging of the battery, thereby prolonging the battery life. In addition, BMS also needs to manage the temperature of the battery to prevent overheating and safety problems. Equalization control is another important function of BMS, which adjusts the voltage difference between individual cells to ensure the consistency and balance of the battery pack. This helps to improve the overall performance and life of the battery pack. The fault diagnosis function can detect abnormal conditions of the battery in real time, such as short circuits, overheating, etc., and take appropriate protective measures. With the development of technology, the design of BMS is becoming more and more complex, and the integration is becoming higher and higher. Modern BMS usually uses advanced algorithms and hardware, such as microcontrollers, sensors, and communication interfaces, to achieve more accurate control and higher intelligent level.

[0004] Lithium batteries generally have high energy density, and improper use can pose a risk of explosion, fire, etc. Therefore, a good BMS should have a protection function against the above hazards. However, the existing BMS system rarely has ESD protection, spark monitoring and other protection functions. The SiC APD photodiode has good photosensitive characteristics and can be used to detect the working state of the lithium battery, thereby protecting the battery.

[0005] At present, the common SOC estimation algorithm can be roughly divided into ampere-hour integration method, open circuit voltage method, filtering method, neural network method, etc. The ampere-hour integration method is the simplest SOC estimation method, which integrates the current of the battery over time to obtain the SOC of the battery. The method is simple, but has the disadvantages of indefinite initial value and error amplification over time. The open circuit voltage method is based on the one-to-one correspondence between the open circuit voltage of the battery and the SOC, and calculates the SOC of the battery from the terminal voltage of the battery. However, the estimation accuracy of this method is related to the calculation accuracy of the open circuit voltage, and the estimation accuracy is usually low. The neural network method inputs the charge and discharge data of the battery into the neural network to estimate the current SOC of the battery, which has high accuracy. However, the data storage and calculation amount required by the algorithm is large, which is not suitable for the embedded hardware platform of the BMS system, and the interpretability of the neural network model is poor, making it difficult to locate errors for abnormal results. The Kalman filter method starts from the equivalent circuit model of the battery, and uses the extended Kalman filter algorithm to estimate the SOC of the battery. This method has the advantages of moderate calculation complexity and high accuracy, and has been widely used. SUMMARY

[0006] The present application aims to overcome the shortcomings of the prior art and provide a lithium battery management and protection system based on silicon carbide diode.

[0007] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows:

[0008] A lithium battery management and protection system based on silicon carbide diode, comprising:

[0009] Lithium battery pack: used to power the system and electrical equipment;

[0010] ARM / FPGA module: ARM and FPGA, used to receive or send signals or instructions to other modules, and judge the system state

[0011] Temperature sensing module: including temperature sensor and first ADC, the temperature sensor is used to measure the surface temperature of each lithium battery in the lithium battery pack, and the analog signal is converted into digital signal by the first ADC and sent to the ARM / FPGA module;

[0012] Lithium battery management chip: for receiving control commands of ARM / FPGA module, to control lithium battery pack in the form of configuring registers;

[0013] Host computer module: for transmitting instructions to ARM / FPGA module, so that users can intervene in the system in real time;

[0014] Serial port and bus communication module: for communication and data transmission between ARM / FPGA module and lithium battery management chip, and between ARM / FPGA module and host computer;

[0015] APD module: including a first APD, which is connected between the lithium battery pack and the electrical equipment, and the anode of the first APD is grounded GND;

[0016] The reverse breakdown voltage of the first APD needs to meet: when the voltage across the lithium battery pack is greater than or equal to the threshold voltage, the first APD is reversely broken down, and the current between the lithium battery and the electrical equipment is bypassed to GND.

[0017] In the above technical solution, further, the system further includes a diode filtering module, the diode filtering module includes a second APD, a filter circuit and a second ADC, the two ends of the second APD are connected to the filter circuit, and the output of the filter circuit is connected to the ARM / FPGA module through the second ADC.

[0018] Further, the system further includes a spark monitoring module, the spark monitoring module includes a third APD, a transimpedance amplifier, a third ADC, the two ends of the third APD are fixed or close to the surface of the lithium battery by insulation wire, and the positive electrode is grounded, the negative electrode is connected to the negative input end of the transimpedance amplifier, and the transimpedance amplifier is connected to the third ADC, and the third ADC is connected to the ARM+FPGA controller.

[0019] Further, the amplification factor of the transimpedance amplifier is greater than or equal to 10 times, for amplifying the high voltage generated by the optical signal to the voltage size that can be read by the third ADC, and sending to the ARM / FPGA module after reading by the third ADC, if the voltage size exists instantaneous rise, it indicates that the lithium battery pack generates spark due to some abnormality, then the ARM / FPGA module sends instructions to control the lithium battery pack switch to close, cutting off the power supply of the lithium battery pack.

[0020] Further, the SOC estimation algorithm is built in the ARM / FPGA module, and the SOC estimation algorithm is an extended Kalman filter SOC estimation method based on temperature correction.

[0021] Further, the method specifically includes:

[0022] The identified and temperature-corrected parameters are brought into the state equation of extended Kalman filter determined by the second-order RC equivalent circuit model, and the system state quantity at the current time is determined through the classical iteration equation of extended Kalman filter, and then the battery SOC at the current time is separated from the state quantity.

[0023] Further, the parameter identification and correction method before filtering comprises the following:

[0024] The lithium battery pack is subjected to pulse experiments at different temperatures, and the experimental data are brought into the recursive least square method to determine the model parameters of the lithium battery pack at different temperatures.

[0025] The maximum capacity, open-circuit voltage and capacity relationship of the lithium battery pack at different temperatures are tested to determine the maximum capacity and open-circuit voltage-capacity curve of the lithium battery pack at different temperatures.

[0026] The obtained measured parameters are respectively corrected by using linear interpolation method.

[0027] Further, the state equation of extended Kalman filter determined by the second-order RC equivalent circuit model is as follows:

[0028] x k+1 =Ax k +Bu k +w k

[0029] y k =Cx k +Du k +v k

[0030] Wherein x k is the state variable of the system; u k is the system excitation; y k is the observation variable of the system; w k and v k represent the process excitation noise and observation noise of the system respectively; according to the state space equation of the second-order RC model, the state variable x k , the system excitation u k and the matrix A, B, C, D are obtained:

[0031]

[0032]

[0033]

[0034]

[0035]

[0036] Δt is the time interval of each iteration of Kalman filtering, C N is the maximum capacity of the battery, η is the coulomb efficiency of the battery, R0 is the internal resistance of the battery, R1, C1 are the polarization resistance and polarization capacitance in the model, R2, C2 are the diffusion resistance and diffusion capacitance in the model, U1 is the voltage across the first RC element in the model, U2 is the voltage across the second RC element, I is the discharge current of the lithium battery, and subscript k represents the corresponding value of the kth iteration.

[0037] The beneficial effects of the present application are:

[0038] The present application utilizes the single conduction, photosensitivity and other characteristics of silicon carbide diodes, and connects several silicon carbide diodes to the traditional lithium battery management system, so that the functions of ESD protection, filtering, spark detection, etc. are realized. ESD protection and spark detection can effectively monitor abnormal phenomena such as high voltage and battery spark, and timely protect against harm to the system, while the filtering function can effectively convert the AC component of the system into a DC signal. The temperature-corrected extended Kalman filter algorithm for estimating the SOC of the lithium battery proposed in the present application can converge after 600 iterations, and the SOC estimation accuracy after iteration convergence using the parameter correction algorithm is more than 0.5%, and the accuracy is improved by 0.2% compared with the method without parameter correction. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 is a schematic diagram of the overall architecture of the system of the present application.

[0040] Figure 2 is a schematic diagram of the communication link between ARM and FPGA in the present application deployed through AXI bus.

[0041] Figure 3 is a schematic diagram of the SOC estimation results of EKF under the contrast correction model parameters and uncorrected model parameters. DETAILED DESCRIPTION

[0042] The technical solutions of the present application will be further described in detail below in combination with the drawings and specific implementation examples.

[0043] The overall architecture of the system is shown in Figure 1 According to a specific example of the present application, the system includes the following modules:

[0044] Lithium battery pack: the object to be managed and protected, used to power the system and electrical equipment.

[0045] ARM / FPGA control module: used to receive system instructions, judge system state, and send instructions to control the system. At the same time, the SOC estimation algorithm is built-in to estimate the SOC of the lithium battery pack.

[0046] Temperature sensor and ADC: The temperature sensor should be placed on the surface of each lithium battery to measure its surface temperature, and the analog signal is converted into a digital signal by the ADC and sent to the ARM / FPGA module.

[0047] Lithium battery management chip: used to receive control commands from the ARM / FPGA module to control the lithium battery by configuring the register.

[0048] Host computer module: used to send and receive instructions to the ARM / FPGA module so that users can intervene in the system in real time.

[0049] Serial port and bus communication module: communication and data transmission between the ARM / FPGA module and the lithium battery management chip, and between the ARM / FPGA module and the host computer.

[0050] 3 APD diodes: one APD is used in the spark monitoring module to determine the state of the lithium battery by reading the light-sensitive characteristics of the APD; the spark monitoring module includes the APD, a transimpedance amplifier, an ADC, and the APD is fixed or close to the surface of the lithium battery with an insulating wire across its terminals, with the positive terminal connected to ground and the negative terminal connected to the negative input terminal of the transimpedance amplifier, and the transimpedance amplifier is connected to the ADC, which is connected to the ARM+FPGA controller.

[0051] Another APD is used in the diode filter module to achieve filtering function through the one-way conduction characteristics of the APD; the diode filter module includes the APD, a filter circuit, and an ADC, the two ends of the APD are connected to the filter circuit, and the output of the filter circuit is connected to the ARM / FPGA module through the ADC.

[0052] The last APD is connected between the lithium battery pack and the electrical equipment, with its anode connected to ground GND, and a diode switch is set by the reverse breakdown characteristics of the APD to protect the system, and the reverse breakdown voltage of the APD needs to meet: when the voltage across the lithium battery pack is greater than or equal to the threshold voltage, the APD is reverse breakdown, and the current between the lithium battery and the electrical equipment is bypassed to GND.

[0053] As Figure 1As shown, the role of the APD in the system in the application mainly embodies in the following three aspects: 1. ESD protection: APD diode has a diode reverse conduction characteristic. To achieve ESD protection for lithium batteries, an APD device with a reverse breakdown voltage close to the threshold voltage of the lithium battery pack needs to be selected. Reverse APD1 is connected to GND and placed between the lithium battery and the system power supply end. When the voltage across the lithium battery is greater than or equal to the threshold voltage, the APD is reversely broken down, and the current between the lithium battery and the power supply system is bypassed to GND. This setting can prevent the application of high voltage at the system power supply end when the voltage at the lithium battery end is too high, thereby avoiding the risk of power consumption.

[0054] 2. Filtering: Some circuit functions may require converting AC signals containing AC components into DC signals. Diode filter circuit relies on the unidirectional conductivity of diode to convert AC into DC through rectification filtering process. In the rectification process, diode allows current to flow in one direction, thereby cutting off the negative half cycle of AC and allowing only the positive half cycle of current to pass, thereby generating pulsating DC. Connect the two APD2 across the corresponding filter circuit, and connect the output of the filter circuit to the ADC, and the ADC is connected to the ARM+FPGA controller, and the detection filtering function effect is read through the ARM / FPGA module.

[0055] 3. Spark monitoring: APD has photosensitive characteristics and can respond to weak light signals to generate a high voltage of tens of mV. Fix the APD across the lithium battery box with an insulating wire at a distance of 2 cm from the surface of the lithium battery, and make its positive level connected to ground, and the negative pole connected to the negative input end of the operational amplifier, and the transimpedance amplifier connected to the subsequent ADC, and the ADC connected to the ARM+FPGA controller. Set the amplification factor of the transimpedance amplifier to be greater than or equal to 10 times to amplify the high voltage generated by the light signal to a voltage size that can be read by the ADC, and send it to the ARM / FPGA module through the ADC reading. If the voltage size is temporarily increased, it indicates that the lithium battery in the lithium battery box has generated sparks due to some abnormalities, and then the ARM / FPGA can send instructions to control the lithium battery switch MOS tube to be closed to cut off the power supply of the lithium battery.

[0056] After the system is powered on, the lithium battery starts to supply power to various power-consuming devices. The ARM / FPGA module initializes and configures the lithium battery management chip, temperature sensor and ADC through pre-burned instructions, and then the lithium battery management chip, temperature sensor and ADC will read the corresponding state of the lithium battery according to the configuration state of its register.

[0057] The battery management chip and the ARM / FPGA module communicate through IIC protocol used by the battery management system, the protocol stipulates that 16 bits of data are read each time, 8 bits of data are written each time, and 8-bit CRC8 check code is attached after each read and write to check. The ADC and the upper computer communicate with the ARM / FPGA module through UART serial communication protocol, the serial port baud rate is set to 115200, the data bit is 8 bits, the stop bit is one bit, and no check bit is set. The data between the ARM / FPGA is transmitted through AXI4 bus, such as Figure 2 .

[0058] The ARM / FPGA module function area is divided into: the FPGA end reads and writes the power management chip and the ADC, and deploys the control algorithm of the battery management system. The ARM end deploys the Kalman filter algorithm for SOC estimation, and obtains the data read by the FPGA end through the AXI4 bus, so as to calculate the SOC of the battery.

[0059] According to a specific example of the application, the SOC estimation in the application is realized by using an extended Kalman filter algorithm for estimating the SOC of the lithium battery based on temperature correction, mainly based on the temperature correction of the second-order RC model using the linear interpolation method. This method can quickly and effectively estimate the SOC with high precision.

[0060] The SOC estimation algorithm models the electrical characteristics of the lithium battery through the second-order RC model, and estimates the SOC of the lithium battery using the extended Kalman filter with corrected parameters. The state equation and the observation equation of the extended Kalman filter system can be written as follows:

[0061] x k+1 =Ax k +Bu k +w k

[0062] y k =Cx k +Du k +v k

[0063] Where x k is the state variable of the system; u k is the system excitation; y k is the observation variable of the system; w k and v k represent the process excitation noise and observation noise of the system respectively; according to the state space equation of the second-order RC model, the state variable x k , the system excitation u k and the matrix A, B, C, D are obtained as follows:

[0064]

[0065]

[0066]

[0067]

[0068]

[0069] Δt is the time interval of each iteration of Kalman filter, C N is the maximum capacity of the battery, η is the coulomb efficiency of the battery, R0 is the internal resistance of the battery, R1, C1 are the polarization resistance, polarization capacitance in the model, R2, C2 are the diffusion resistance, diffusion capacitance in the model, U1 is the voltage across the first RC element in the model, U2 is the voltage across the second RC element, I is the discharge current of the lithium battery, subscript k represents the corresponding value of the kth iteration.

[0070] The parameter identification before filtering and the parameter correction are carried out in the following steps:

[0071] 1) HPPC pulse experiment is carried out on the lithium battery under different temperatures / different capacities, and the model parameters of the lithium battery under different temperatures / different capacities are determined by bringing the experimental data into the recursive least squares method.

[0072] 2) The maximum capacity and the open circuit voltage-capacity relationship of the lithium battery under different temperatures are tested, and the maximum capacity and the open circuit voltage-capacity curve of the lithium battery under different temperatures are determined.

[0073] 3) For the relationship between temperature, capacity and battery model parameters, the capacitance and resistance parameters in the lithium battery model parameters are two-dimensionally linearly interpolated using the data measured in 1) under different temperatures / different capacities, so as to obtain a complete data model of the capacitance and resistance parameters changing with temperature and capacity. In this way, the capacitance and resistance parameters of the measured battery under any temperature can be determined.

[0074] 4) For the relationship between temperature, maximum capacity and open circuit voltage-capacity, the maximum capacity and open circuit voltage-capacity in the lithium battery model parameters are two-dimensionally linearly interpolated using the data measured in 2) under different temperatures / different capacities, so as to obtain a complete data model of the maximum capacity and open circuit voltage-capacity relationship changing with temperature. In this way, the maximum capacity and open circuit voltage-capacity relationship data of the measured battery under any temperature can be determined.

[0075] 5) The revised parameters are brought into the state equation of the extended Kalman filter determined by the second-order RC equivalent circuit model, and the system state quantity at the current time is determined through the classical iteration equation of the extended Kalman filter, and then the battery SOC at the current time is separated from the state quantity.

[0076] The SOC of the lithium battery is estimated using the extended Kalman filter with parameter revision. The algorithm converges at about 600 iterations (i.e., after 600 seconds), and the convergence speed is relatively fast. The estimation accuracy of the extended Kalman filter with parameter revision after iteration convergence reaches more than 0.5%, and the accuracy is improved by 0.2% compared with the method without parameter revision, as shown in Figure 3 The SOC estimation can be well performed.

[0077] The control algorithm of the battery management system mainly includes charge balance and temperature control algorithms, and realizes the following functions: 1. Passive balancing function: passive balancing is started when the voltage difference between any two batteries is 50mV or the voltage of a certain battery is higher than 4.25v. 2. Overcharge and overdischarge protection function: when the battery is discharging and the total voltage of the four batteries is lower than 10.8V or other set value, the FPGA controls the discharge MOS switch to be closed to realize overdischarge protection. When the battery is charging and the total voltage of the four batteries is higher than 16.8V or other set value, the FPGA controls the charging MOS switch to be closed to realize overcharge protection. 3. SOC abnormal protection function: when the SOC of a certain battery reaches the threshold value, the battery switch is turned off / on to disable / enable the certain battery.

[0078] The above-described embodiments are only some of the preferred schemes of the present application, and are not intended to limit the present application. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application. Therefore, any technical scheme obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present application.

Claims

1. A silicon carbide diode based lithium battery management and protection system, characterized in that, The application relates to a lithium battery system, which comprises the following modules: a lithium battery pack for supplying power to the system and the powered device; an ARM / FPGA module for receiving or sending signals or instructions to other modules and judging the system state; a temperature sensing module comprising a temperature sensor and a first ADC, wherein the temperature sensor is used for measuring the surface temperature of each lithium battery in the lithium battery pack and sending the analog signal converted into a digital signal by the first ADC to the ARM / FPGA module; a lithium battery management chip for receiving the control command of the ARM / FPGA module to control the lithium battery pack by configuring the register; a host computer module for sending and receiving instructions of the ARM / FPGA module so that the user can intervene in the system in real time; a serial port and bus communication module for communicating and transmitting data between the ARM / FPGA module and the lithium battery management chip and between the ARM / FPGA module and the host computer; an APD module comprising a first APD, wherein the first APD is connected between the lithium battery pack and the powered device, the anode of the first APD is connected to the ground GND, and the reverse breakdown voltage of the first APD meets the following condition: when the voltage between the lithium battery pack is greater than or equal to the threshold voltage, the first APD is reversely broken down, and the current between the lithium battery and the powered device is bypassed to the GND; a diode filtering module comprising a second APD, a filtering circuit and a second ADC, wherein the two ends of the second APD are connected to the filtering circuit, and the output of the filtering circuit is connected to the ARM / FPGA module through the second ADC; a spark monitoring module comprising a third APD, a trans-impedance amplifier and a third ADC, wherein the two ends of the third APD are fixed or close to the surface of the lithium battery by using an insulating wire, the positive electrode is connected to the ground, the negative electrode is connected to the negative input end of the trans-impedance amplifier, the trans-impedance amplifier is connected to the third ADC, and the third ADC is connected to the ARM+FPGA controller; the SOC estimation algorithm built in the ARM / FPGA module is a temperature correction-based extended Kalman filtering SOC estimation method, which comprises the following steps: the identified and temperature-corrected parameters are brought into the state equation of the extended Kalman filtering determined by the second-order RC equivalent circuit model, the system state quantity at the current time is determined through the classical iteration equation of the extended Kalman filtering, and then the battery SOC at the current time is separated from the state quantity; the parameter identification and correction method before filtering comprises the following steps: pulse experiments are conducted on the lithium battery pack at different temperatures, the experimental data are brought into the recursive least square method, the model parameters of the lithium battery pack at different temperatures are determined, the maximum capacity, open circuit voltage and capacity relationship tests are conducted on the lithium battery pack at different temperatures, the maximum capacity and open circuit voltage-capacity curves of the lithium battery pack at different temperatures are determined, and the obtained measurement parameters are respectively corrected by using the linear interpolation method.

2. The silicon carbide diode based lithium battery management and protection system of claim 1, wherein, The amplification multiple of the transimpedance amplifier is greater than or equal to 10 times, which is used for amplifying the high voltage generated by the optical signal to the voltage size that can be read by the third ADC, and then sending to the ARM / FPGA module after being read by the third ADC, if the voltage size exists instantaneous rise, it indicates that the lithium battery pack generates sparks due to some abnormalities, then the ARM / FPGA module sends a command to control the lithium battery pack switch to be closed, and cut off the power supply of the lithium battery pack.

3. The silicon carbide diode based lithium battery management and protection system of claim 1, wherein, The state equation of the extended Kalman filter determined by the second-order RC equivalent circuit model is as follows: ； ； where is the state variable of the system; is the excitation of the system; is the observation variable of the system; and represent the process excitation noise and the observation noise of the system, respectively; according to the state space equation of the second-order RC model, the state variable , the excitation of the system and the matrix can be obtained ; ； ； ； ； T is the time interval for each iteration of the Kalman filter, Cmax is the maximum capacity of the battery, Ceff is the coulombic efficiency of the battery, R is the internal resistance of the battery, , Rpolar is the polarization resistance and Cpolar is the polarization capacitance in the model, , Rdiff is the diffusion resistance and Cdiff is the diffusion capacitance in the model, V1 is the voltage across the first RC element in the model, V2 is the voltage across the second RC element in the model, I is the discharge current of the lithium battery, subscript k denotes the corresponding value for the kth iteration.

Images