An offline multi-parameter calibration method and system for embedded BMS

Abstract

The application provides an offline multi-parameter calibration method and system of an embedded BMS, the method comprising: temperature calibration, collecting reported temperature values and standard temperature values of each temperature calibration point and fitting to generate temperature calibration coefficients written into a non-volatile memory; voltage calibration, applying standard voltage values of each voltage calibration point to a voltage sampling channel and fitting to generate voltage calibration coefficients written into a non-volatile memory; current calibration, making the current sampling channel pass through standard current values of each current calibration point and fitting to generate current calibration coefficients written into a non-volatile memory; state of charge model parameter updating, performing charge and discharge operations on a battery and obtaining open-circuit voltage measured values of each state of charge point to update state of charge model parameters written into a non-volatile memory. The system comprises a calibration board, an incubator, a standard thermometer, a programmable stabilized power supply, a standard voltmeter, an electronic load instrument, a standard ammeter and a communication interface. The application relates to the technical field of battery management system calibration.

Description

Technical Field

[0001] This invention relates to the field of battery management system calibration technology, and in particular to an offline multi-parameter calibration method and system for an embedded BMS. Background Technology

[0002] Embedded battery management systems (BMS) are core components in new energy vehicles, energy storage, and industrial batteries. They are responsible for acquiring and managing key parameters such as cell voltage, total voltage, charge / discharge current, temperature, and estimation and equalization control of state of charge (SOC) and state of health (SOH). Their measurement accuracy directly affects battery safety, lifespan, and energy utilization efficiency. Currently, mainstream BMS calibration methods primarily employ single-point or two-point hardware calibration combined with simple software compensation. This involves zero-point and full-scale bias corrections for parameters such as voltage and current under normal temperature conditions, storing the corrections as fixed offset values ​​in memory for runtime recall. For temperature measurement, a voltage divider lookup table using negative temperature coefficient thermistors (NTCs) is typically used, supplemented by bias correction at a single temperature point. SOC estimation generally relies on the ampere-hour integration method combined with a factory-calibrated open-circuit voltage-state of charge (standard open-circuit voltage-state of charge table) lookup table. When the battery is at rest, the table is used to perform limited correction of integral drift. This hierarchical, independent, and item-by-item calibration architecture was widely adopted in early BMS applications due to its simplicity and low computational overhead.

[0003] However, the aforementioned traditional calibration schemes have revealed significant shortcomings in practical applications. At the hardware level, the analog-to-digital converter (ADC) in the analog front-end (AFE) exhibits nonlinear errors and offsets. Factors such as the tolerance and temperature drift of the voltage divider resistors, parasitic parameters of the printed circuit board (PCB), and contact resistance of the wiring harness and connectors collectively lead to voltage acquisition deviations of ±5mV to ±15mV. Current and temperature measurements also become inaccurate simultaneously. Simple single-point or two-point bias corrections cannot compensate for the full-range nonlinearity and temperature drift characteristics of the ADC, rendering the calibrated accuracy effective only under limited operating conditions. At the environmental and operating condition level, temperatures vary widely from -40℃ to 85℃, and charging... The parameter drift caused by discharge rate switching and battery aging makes single-point calibration at room temperature completely inadequate for covering the full temperature range and multiple operating conditions of actual operation scenarios. Furthermore, the inaccuracy of temperature measurement itself further deteriorates the temperature drift compensation effect of voltage and current, creating a cascading amplification of errors. At the system level, there is a strong coupling relationship between voltage, current, temperature, and SOC model parameters. Existing serial independent calibration methods do not consider this coupling; after calibrating a single parameter, the cross-influence of temperature on voltage and current is not eliminated before proceeding to the next parameter calibration, leading to the superposition of errors among parameters and limiting the overall system accuracy. In addition, the offline calibration process heavily relies on external high-precision equipment such as standard sources and multimeters, resulting in cumbersome operation procedures, low automation, and difficulty in meeting the high-efficiency requirements of embedded BMS mass production and after-sales maintenance scenarios. Moreover, the lack of a systematic full-condition calibration model leads to continuous error accumulation over long-term operation, directly threatening the accuracy of SOC estimation and the reliability of equalization control.

[0004] Therefore, the inventors urgently need an offline multi-parameter collaborative calibration method and system for embedded BMS to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of the prior art, this invention provides an offline multi-parameter calibration system and method for embedded BMS, aiming to solve the problems of insufficient accuracy, low efficiency, and accumulation of multi-parameter coupling errors caused by single-point or two-point bias correction, serial independent calibration, and manual reliance on high-precision equipment operation in the existing embedded battery management system calibration schemes.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: an offline multi-parameter calibration method for an embedded BMS, comprising the following steps: S1: Provide a number of preset temperature calibration points, which are distributed within the temperature range to be calibrated. Place the battery management system to be calibrated in a controllable temperature environment. Collect the reported temperature value and standard temperature value of the battery management system at each of the temperature calibration points. Perform linear fitting based on the standard temperature value and the reported temperature value at each of the temperature calibration points to generate a temperature calibration coefficient. Write the temperature calibration coefficient into the non-volatile memory of the battery management system. S2: Provide a number of preset voltage calibration points, which are distributed within the voltage range to be calibrated. After writing the temperature calibration coefficient, apply the standard voltage value corresponding to each voltage calibration point to the voltage sampling channel of the battery management system, collect the reported voltage value of the battery management system at each voltage calibration point, perform linear fitting between the standard voltage value and the reported voltage value of each voltage calibration point to generate a voltage calibration coefficient, and write the voltage calibration coefficient into the non-volatile memory. S3: Provide a number of preset current calibration points, which are distributed within the current range to be calibrated. After writing the voltage calibration coefficient, the current sampling channel of the battery management system collects the reported current value of the battery management system at each current calibration point by passing the standard current value corresponding to each current calibration point. Based on the standard current value of each current calibration point and the reported current value, a current calibration coefficient is generated by linear fitting. The current calibration coefficient is then written into the non-volatile memory. S4: Provide several preset state of charge (POC) points, which are distributed within the POC range. After writing the current calibration coefficient, use the voltage sampling value collected by the calibrated voltage sampling channel to perform charging and discharging operations on the battery electrically connected to the battery management system. Obtain the corresponding open-circuit voltage measurement value at each POC point. Update the POC model parameters of the battery management system according to the open-circuit voltage measurement value of each POC point. Write the updated POC model parameters into the non-volatile memory.

[0007] Based on the above, the beneficial effects of an offline multi-parameter calibration system and method for embedded BMS are that it solves the problems of insufficient accuracy, low efficiency, and accumulation of multi-parameter coupling errors caused by single-point or two-point bias correction, serial independent calibration, and manual reliance on high-precision equipment operation in existing embedded battery management system calibration schemes; mainly reflected in: 1. This invention eliminates the cross-influence of temperature measurement error on voltage and current calibration by sequentially executing temperature calibration, voltage calibration, current calibration, and state of charge model parameter update steps. After the temperature calibration coefficients are written, voltage calibration is performed; after the voltage calibration coefficients are written, current calibration is performed; and after the current calibration coefficients are written, state of charge model parameters are updated. This solves the problem of multi-parameter coupling error accumulation. 2. This invention collects the temperature values ​​reported by the battery management system at multiple temperature calibration points distributed within the temperature range to be calibrated, and collects the temperature values ​​measured by a standard thermometer in a controllable temperature environment as the standard temperature values. Based on the standard temperature values ​​and the reported temperature values ​​at each temperature calibration point, a temperature calibration coefficient including a temperature gain calibration coefficient and a temperature offset calibration coefficient is generated through linear fitting. Compared with single-point temperature offset correction, this invention can effectively adapt to the nonlinear drift of the temperature sensor in the entire temperature range, thereby solving the problem of insufficient temperature measurement accuracy. 3. This invention collects the voltage values ​​reported by the battery management system at multiple voltage calibration points distributed within the voltage range to be calibrated, and collects the standard voltage value verified by a standard voltmeter applied to the voltage sampling channel, as the standard voltage value. Based on the standard voltage value at each voltage calibration point and the reported voltage value, a voltage calibration coefficient including a voltage gain calibration coefficient and a voltage offset calibration coefficient is generated through linear fitting. Compared with single-point or two-point voltage bias correction, this invention can effectively compensate for the nonlinear error of the analog-to-digital converter across the entire range and the voltage divider network deviation, thereby solving the problem of insufficient voltage measurement accuracy. 4. This invention collects the current value reported by the battery management system at multiple current calibration points distributed within the current range to be calibrated, and collects the standard current value measured by a standard ammeter connected in series in the current loop, as the standard current value. Based on the standard current value at each current calibration point and the reported current value, a current calibration coefficient including a current gain calibration coefficient and a current offset calibration coefficient is generated by linear fitting. Compared with single-point current bias correction, this invention can effectively adapt to the nonlinearity of the entire current sampling range, thereby solving the problem of insufficient current measurement accuracy. 5. This invention automatically performs calibration coefficient generation and writing operations throughout the entire process of temperature calibration, voltage calibration, current calibration, and state of charge model parameter update via a calibration board. The temperature calibration coefficient, voltage calibration coefficient, and current calibration coefficient are replaced with old coefficients using an overwrite method and written to a designated address in the non-volatile memory of the battery management system via a communication interface on the calibration board. The updated state of charge model parameters are also written to the non-volatile memory in batches via the communication interface. After each calibration coefficient writing step, the calibration board automatically re-collects the reported values ​​at each calibration point and compares them with standard values ​​for accuracy verification. Once the device is connected and ready, the calibration process is automatically executed by the calibration board, eliminating the need for manual operation of external high-precision equipment. This solves the inefficiency problem caused by manual operation in traditional calibration methods.

[0008] Furthermore, in step S1, the controllable temperature environment is provided by a constant temperature chamber, the standard temperature value is measured by a standard thermometer set inside the constant temperature chamber, the battery management system is set inside the constant temperature chamber, and at each of the temperature calibration points, the reported temperature value and the standard temperature value are collected after the temperature sensor of the battery management system and the air inside the constant temperature chamber reach thermal equilibrium. Each of the temperature calibration points at least covers the low end, middle end and high end of the operating temperature range of the battery management system.

[0009] Based on the above, the beneficial effect of collecting and reporting temperature values ​​and standard temperature values ​​after reaching thermal equilibrium is that it enables the comparison of reported temperature values ​​and standard temperature values ​​under the same steady-state thermodynamic conditions, eliminating transient measurement deviations caused by NTC thermal response hysteresis and differences in heat capacity. The beneficial effect of covering the low, middle and high ends of the operating temperature range is that it constructs a linear calibration equation through multi-point fitting, comprehensively characterizing the nonlinear drift characteristics of the temperature sensor under all operating conditions, and avoiding inaccuracies in high and low temperature ranges caused by only covering room temperature or single-point calibration.

[0010] Furthermore, in step S2, the standard voltage value is provided by a programmable regulated power supply electrically connected to the voltage sampling channel of the battery management system. The standard voltage value corresponding to each voltage calibration point is the actual output voltage value of the programmable regulated power supply. The actual output voltage value is verified by a standard voltmeter electrically connected to the output terminal of the programmable regulated power supply. Each voltage calibration point at least covers the low end, middle end, and high end of the operating voltage range of the battery management system.

[0011] Based on the above, the beneficial effect of standard voltmeter verification is that the actual output voltage of the programmable regulated power supply can be independently measured and confirmed using a standard voltmeter with a higher accuracy level, thus achieving the accuracy and traceability of the standard voltage value. The beneficial effect of covering the low, middle and high ends of the working voltage range is that multi-point fitting can comprehensively compensate for the nonlinear error of the AFE analog-to-digital converter and the deviation of the voltage divider network across the entire range, avoiding the inaccuracy of other sections after correction due to covering only a single voltage point.

[0012] Furthermore, in step S3, the standard current value is set by an electronic load instrument connected in series with the current sampling channel of the battery management system in the current loop. The standard current value corresponding to each current calibration point is the measured current value of the standard ammeter connected in series in the current loop. Each current calibration point at least covers the low end, middle end and high end of the operating current range of the battery management system. The current calibration coefficients share the same set of coefficients for both the charging direction and the discharging direction.

[0013] Based on the above, the beneficial effect of covering the low, middle and high ends of the operating current range is that it can comprehensively compensate for the comprehensive error of the current sampling circuit across the entire range through multi-point fitting. The comprehensive error includes the sampling resistor tolerance, operational amplifier nonlinearity and analog-to-digital converter offset, avoiding the inaccuracy caused by only covering a single current point after correction in the small signal section or the large current section. The beneficial effect of sharing the same set of coefficients for charging and discharging directions is that it can utilize the bidirectional symmetrical hardware characteristics of current sampling to apply the same set of calibration coefficients to both charging and discharging directions at the same time, reducing the data acquisition workload required for separate calibration of charging and discharging directions, and simplifying the correction logic of bidirectional charging and discharging currents during battery management system operation.

[0014] Further, in steps S1, S2, and S3, the linear fitting involves: collecting the corresponding standard temperature value and the reported temperature value at at least three temperature calibration points, collecting the corresponding standard voltage value and the reported voltage value at at least three voltage calibration points, and collecting the corresponding standard current value and the reported current value at at least three current calibration points; using the least squares method to perform linear fitting on the standard temperature value and the reported temperature value at each temperature calibration point to obtain a temperature calibration equation; performing linear fitting on the standard voltage value and the reported voltage value at each voltage calibration point to obtain a voltage calibration equation; and performing linear fitting on the standard current value and the reported current value at each current calibration point to obtain a current calibration equation. The temperature calibration equation is: Y_t = k_t·X_t + b_t, Where Y_t represents the standard temperature value, X_t represents the reported temperature value, k_t is the temperature gain calibration coefficient, and b_t is the temperature offset calibration coefficient; The voltage calibration equation is as follows: Y_v = k_v·X_v + b_v Where Y_v represents the standard voltage value, X_v represents the reported voltage value, k_v is the voltage gain calibration coefficient, and b_v is the voltage offset calibration coefficient; The current calibration equation is as follows: Y_i = k_i·X_i + b_i, Where Yi represents the standard current value, Xi represents the reported current value, k_i is the current gain calibration coefficient, and b_i is the current offset calibration coefficient.

[0015] Based on the above, the beneficial effects of collecting data at at least three temperature calibration points, at least three voltage calibration points, and at least three current calibration points are: providing redundant data samples to support the statistical validity of the least squares method; enabling multi-point data to participate in the fitting to reduce the disturbance effect of single-point measurement random errors on the calibration coefficient calculation results, making the fitting results more representative globally; the beneficial effects of the least squares method are: performing linear fitting on the standard temperature value and the reported temperature value, the standard voltage value and the reported voltage value, and the standard current value and the reported current value at each calibration point to solve the corresponding calibration equation coefficients, realizing accuracy compensation with a unified linear correction model across the entire range; the beneficial effects of linear fitting are: using the same mathematical model to establish a linear mapping relationship between the measured values ​​and the true values ​​for the three parameters, realizing the unified inclusion of the nonlinear drift and offset errors of the three physical quantities of voltage, current, and temperature into the first-order linear correction framework, and using the same linear fitting formula for the runtime correction of the three parameters of temperature, voltage, and current, the battery management system only needs to perform two multiplication operations and one addition operation to complete a single correction, with a unified algorithm structure and no need for additional storage or computing resources.

[0016] Furthermore, in step S4, the specific steps for updating the parameters of the state of charge model are as follows: S41: Based on the voltage sampling value collected by the calibrated voltage sampling channel, charge the battery at a constant current of 0.5C to the charging cut-off voltage of the battery, then switch to constant voltage charging until the charging current of the battery is less than 0.02C. After resting, mark the state of charge as 100% and record the current open circuit voltage value. S42: Discharge the battery in segments at a discharge rate of 0.05C. Stop discharging after each 5% of the rated capacity is discharged and allow the battery to stand until the terminal voltage stabilizes. Record the measured value of the open circuit voltage corresponding to the current state of charge. Repeat this process until the terminal voltage drops to the discharge cutoff voltage of the battery and mark the state of charge as 0%. S43: Compare the measured open-circuit voltage value corresponding to each state of charge point with the corresponding value in the pre-stored standard open-circuit voltage-state of charge table point by point. If there is a state of charge point with a deviation greater than or equal to the preset OCV deviation threshold, then replace the standard voltage value of the corresponding state of charge point in the standard open-circuit voltage-state of charge table with the measured open-circuit voltage value. If the deviation of all state of charge points is less than the preset OCV deviation threshold, then keep the standard open-circuit voltage-state of charge table unchanged. S44: Write the updated standard open-circuit voltage-state-of-charge table into the non-volatile memory.

[0017] Based on the above, the beneficial effect of S41 is that by using a three-stage charging process of "constant current charging to cutoff voltage + constant voltage current reduction to 0.02C + resting," the battery is pushed to a true full charge state one by one before the open-circuit voltage value is obtained. This achieves accurate calibration of the open-circuit voltage at the 100% SOC anchor point, avoiding the polarization false fullness error introduced by simply assuming full charge based on the voltage reaching the cutoff value. The beneficial effect of S42 is that by using a very low discharge rate combined with a segmented step-by-step method of "stopping immediately after discharge, resting immediately after stopping, and reading after stabilization," the measured open-circuit voltage values ​​of each node in the entire range of 100% to 0% state of charge are obtained point by point. This achieves complete calibration of the actual standard open-circuit voltage-state-of-charge table characteristic curve of the battery while maximally suppressing the influence of discharge polarization, providing a high-quality measured data source for subsequent comparison with the standard table. The beneficial effect of S43 is that by setting a deviation tolerance... By comparing the measured curve with the factory standard curve point by point and selectively replacing only the values ​​of the nodes that exceed the tolerance, the system can correct only the segments that deviate due to individual cell differences or aging while retaining the accurate nodes in the factory standard table. This avoids losing the original accurate calibration data by covering the entire table. The beneficial effect of S44 is that the standard open circuit voltage-state of charge table, which has been judged point by point and selectively updated, is persistently stored in non-volatile memory. This enables the battery management system to load the latest calibration results for daily SOC estimation after power failure and restart, completing the final closed loop from online measured calibration to offline hardened storage.

[0018] Further, in steps S1, S2, and S3, the temperature calibration coefficient, the voltage calibration coefficient, and the current calibration coefficient are all written to a designated address of the non-volatile memory via a communication interface on a calibration board connected to the battery management system, using an overwrite write method. After writing the temperature calibration coefficient, the reported temperature value of the battery management system at each temperature calibration point is re-acquired and compared with the standard temperature value. If the deviation of each temperature calibration point is less than a preset temperature deviation threshold, the temperature calibration is deemed successful and proceeds to the next step. If the deviation of any temperature calibration point is greater than or equal to the preset temperature deviation threshold, the temperature calibration is deemed unsuccessful. After writing the voltage calibration coefficient, the reported temperature value is re-acquired... The reported voltage values ​​of the battery management system at each voltage calibration point are collected and compared with the standard voltage value. If the deviation of each voltage calibration point is less than the preset voltage deviation threshold, the voltage calibration is deemed successful and proceeds to the next step. If the deviation of any voltage calibration point is greater than or equal to the preset voltage deviation threshold, the voltage calibration is deemed unsuccessful. After writing the current calibration coefficient, the reported current values ​​of the battery management system at each current calibration point are collected again and compared with the standard current value. If the deviation of each current calibration point is less than the preset current deviation threshold, the current calibration is deemed successful and proceeds to the next step. If the deviation of any current calibration point is greater than or equal to the preset current deviation threshold, the current calibration is deemed unsuccessful.

[0019] Based on the above, the beneficial effects of writing to a specified address of non-volatile memory using the overwrite method are that the new temperature calibration coefficient, voltage calibration coefficient, and current calibration coefficient are directly written to the same storage address where the old coefficients are located. This achieves complete replacement of the old coefficients by the new coefficients rather than cumulative storage, avoiding the possibility of the battery management system reading the wrong version or confusing the superposition of old and new coefficients when they exist at different addresses at the same time. The beneficial effects of the temperature deviation threshold, voltage deviation threshold, and current deviation threshold are that independent acceptance tolerances are set for the measurement characteristics and accuracy requirements of the three physical quantities of temperature, voltage, and current, respectively. This achieves differentiated criterion management for the calibration quality of the three parameters, avoiding the situation where the same threshold is used, which may lead to overly strict or lenient acceptance of some parameters.

[0020] Furthermore, the communication interface is a UART interface, the non-volatile memory is the EEPROM of the battery management system, the gain calibration coefficient and offset calibration coefficient of the temperature calibration coefficient, the voltage calibration coefficient and the current calibration coefficient are stored in the EEPROM in floating-point format and little-endian, and are written through command code 0x02; the state of charge model parameters are written in batches through command code 0x03.

[0021] Based on the above, the benefits of the UART interface include: reliable asynchronous serial transmission of calibration coefficients and instruction data, requiring only two signal lines for bidirectional communication and reducing hardware connection complexity; the benefits of EEPROM include: long-term retention of calibration coefficients and state-of-charge model parameters even after the battery management system is completely powered off, and support for byte-level addressing and erasing / writing, facilitating the overwriting and updating of coefficients at specific addresses during subsequent recalibration; the benefits of the floating-point format include: lossless storage of calibration coefficients with decimal places generated from fitting calculations, avoiding decimal truncation errors caused by using integer or fixed-point formats; and the benefits of little-endian byte order include: arranging multi-byte floating-point numbers in memory with the least significant byte stored at the lowest address, achieving compatibility with the battery management system. The microcontroller maintains a consistent default data access format to avoid errors in coefficient value parsing during reading due to byte order mismatch. The beneficial effect of command code 0x02 is that it serves as a unified command identifier for writing calibration coefficients of voltage, current, and temperature parameters, enabling the same command code to execute coefficient writing processes with completely consistent formats for different parameters, simplifying the parsing logic of the write command on the battery management system side. The beneficial effect of command code 0x03 for batch writing is that it completes the writing of the entire table of state of charge model parameters (multiple open-circuit voltage values) in one go with a dedicated command code that is different from single-parameter coefficients. It enables the transmission of the open-circuit voltage values ​​of twenty state of charge nodes as an atomic operation in one go, avoiding the increased communication overhead caused by multiple single-parameter writes and the risk of data inconsistency during the whole table update.

[0022] Furthermore, this invention also provides an offline multi-parameter calibration system for an embedded BMS, including a calibration board, a constant temperature chamber, a standard thermometer, a programmable power supply, a standard voltmeter, an electronic load meter, a standard ammeter, a battery, and a communication interface. The calibration board is equipped with a microcontroller and a memory electrically connected to the microcontroller. The constant temperature chamber houses the battery management system to be calibrated and provides a controllable temperature environment. The standard thermometer is located inside the constant temperature chamber and electrically connected to the calibration board to transmit the collected standard temperature values ​​to the calibration board. The programmable power supply is electrically connected to both the calibration board and the battery. The calibration board controls the programmable power supply to provide charging voltage to the battery. The standard voltmeter is connected to both the programmable power supply and the battery. The calibration board is electrically connected to transmit the collected standard voltage value to the calibration board. The voltage sampling channel of the battery management system is electrically connected to the battery. The temperature sensor in the battery management system transmits the collected reported temperature value to the calibration board via the communication interface. The electronic load is electrically connected to the battery. The electronic load, the standard ammeter, and the current sampling channel of the battery management system are connected in series in the discharge circuit of the battery. The calibration board controls the electronic load to provide a discharge load to the battery. The standard ammeter is electrically connected to the calibration board to transmit the collected standard current value to the calibration board. The communication interface is located on the calibration board. The battery management system is electrically connected to the calibration board through the communication interface.

[0023] Based on the above, the beneficial effects of the calibration board are as follows: as the core control unit of the entire calibration system, it coordinates the timing and data interaction between peripheral devices such as the constant temperature chamber, programmable regulated power supply, and electronic load instrument and the battery management system, realizing the automatic serial execution of each calibration step in a preset order without the need for manual switching of each device; the beneficial effects of the microcontroller are that it runs the calibration process control program and processes the measurement data returned by each standard instrument in real time, realizing the integrated processing of multi-device collaborative operation, data acquisition, and coefficient calculation; the beneficial effects of the memory are that it stores the preset calibration process program, standard open-circuit voltage-state-of-charge table curves, and calibration parameters of each parameter. The results show that the calibration board can independently complete the entire calibration process without the host computer. The constant temperature chamber provides a uniform temperature field with programmable and closed-loop temperature control, accommodating the battery management system to be calibrated and maintaining the temperature environment at the preset calibration point during calibration, allowing data acquisition only after thermal equilibrium has stabilized. The standard thermometer transmits the measured standard temperature value back to the calibration board as an electrical signal, enabling the calibration board to automatically acquire the standard temperature value of the current calibration point without manual reading. The programmable regulated power supply regulates the charging voltage output to the battery according to the calibration board's programmable commands, enabling the charging voltage to be adjusted. The standard voltage value of the calibration point is automatically switched via programmable control, eliminating the need for manual adjustment. The beneficial effect of the standard voltmeter is that it transmits the measured standard voltage value back to the calibration board in real time, enabling the calibration board to automatically acquire the standard voltage value verified by the standard voltmeter and use it as a comparison benchmark. The beneficial effect of the electronic load instrument is that it adjusts the absorbed power according to the calibration board's programmable commands to establish the set discharge current, realizing the automatic establishment and switching of the standard current value of the current calibration point via programmable control, and safely dissipating the discharge energy as heat. The beneficial effect of the standard voltmeter is that it transmits the measured standard voltage value back to the calibration board in real time, enabling the calibration board to automatically acquire the standard voltage value verified by the standard voltmeter. The voltage value is used as a comparison benchmark; the beneficial effects of the battery are that it serves as the actual carrier of charging and discharging current and the voltage sampling object, providing a real current path for the current sampling channel and a real voltage source for the voltage sampling channel, and serving as the actual test object for the standard open-circuit voltage-state-of-charge table calibration, realizing the authenticity verification of each parameter under load conditions during the calibration process; the beneficial effects of the communication interface are that it establishes a bidirectional data channel between the calibration board and the battery management system, realizing the uplink reading of reported temperature values, reported voltage values, and reported current values, as well as the downlink writing of temperature calibration coefficients, voltage calibration coefficients, current calibration coefficients, and state-of-charge model parameters.

[0024] Furthermore, the calibration board's memory pre-stores a standard open-circuit voltage-state-of-charge table and a calibration process control program. The calibration board reads the reported temperature value, reported voltage value, and reported current value reported by the battery management system through the communication interface, and writes the temperature calibration coefficient, the voltage calibration coefficient, the current calibration coefficient, and the updated state-of-charge model parameters into the non-volatile memory of the battery management system through the communication interface.

[0025] Based on the above, the beneficial effects of the standard open-circuit voltage-state-of-charge table are that it enables the calibration board to call up the standard table and compare it with the measured values ​​point by point during the SOC calibration step without connecting to an external host computer or database, thus avoiding the dependence of on-site calibration on network or external storage devices. The beneficial effects of the calibration process control program are that the timing logic, device control instructions and judgment conditions of the four-step calibration of temperature, voltage, current and SOC are solidified in the local memory of the calibration board in firmware form, so that the calibration board can automatically perform a complete multi-parameter collaborative calibration according to the preset process after power-on without the intervention of an external computer, and supports offline repeated operation in mass production scenarios.

[0026] To make the above features of the present invention and the objectives to be achieved clearer, the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. Attached Figure Description

[0027] Figure 1 : This is a flowchart of the method of the present invention; Figure 2 : This is a signal block diagram of the present invention. Detailed Implementation

[0028] See Figures 1-2 As shown, This invention discloses an offline multi-parameter calibration method for an embedded BMS, comprising the following steps: S1: Provide a number of preset temperature calibration points, which are distributed within the temperature range to be calibrated. Place the battery management system to be calibrated in a controllable temperature environment. Collect the reported temperature value and standard temperature value of the battery management system at each of the temperature calibration points. Perform linear fitting based on the standard temperature value and the reported temperature value at each of the temperature calibration points to generate a temperature calibration coefficient. Write the temperature calibration coefficient into the non-volatile memory of the battery management system. S2: Provide a number of preset voltage calibration points, which are distributed within the voltage range to be calibrated. After writing the temperature calibration coefficient, apply the standard voltage value corresponding to each voltage calibration point to the voltage sampling channel of the battery management system, collect the reported voltage value of the battery management system at each voltage calibration point, perform linear fitting between the standard voltage value and the reported voltage value of each voltage calibration point to generate a voltage calibration coefficient, and write the voltage calibration coefficient into the non-volatile memory. S3: Provide a number of preset current calibration points, which are distributed within the current range to be calibrated. After writing the voltage calibration coefficient, the current sampling channel of the battery management system collects the reported current value of the battery management system at each current calibration point by passing the standard current value corresponding to each current calibration point. Based on the standard current value of each current calibration point and the reported current value, a current calibration coefficient is generated by linear fitting. The current calibration coefficient is then written into the non-volatile memory. S4: Provide several preset state of charge (POC) points, which are distributed within the POC range. After writing the current calibration coefficient, use the voltage sampling value collected by the calibrated voltage sampling channel to perform charging and discharging operations on the battery electrically connected to the battery management system. Obtain the corresponding open-circuit voltage measurement value at each POC point. Update the POC model parameters of the battery management system according to the open-circuit voltage measurement value of each POC point. Write the updated POC model parameters into the non-volatile memory.

[0029] In this embodiment, there are five temperature calibration points in step S1, set as -20℃, 0℃, 25℃, 45℃ and 60℃ respectively. The temperature control accuracy of the constant temperature chamber is ±0.5℃. The standard thermometer is a platinum resistance thermometer with an accuracy of ±0.1℃. The static heat preservation time at each temperature calibration point is 30 minutes, and the static heat preservation time at the two extreme points of -20℃ and 60℃ is 45 minutes. The arithmetic mean of the reported temperature values ​​is collected 20 times continuously at each calibration point. The least squares method is used to fit the five sets of data to obtain the temperature calibration equation Y_t=0.975·X_t-0.29.

[0030] In this embodiment, there are five voltage calibration points in step S2, which are set to 2.5V, 3.0V, 3.2V, 3.4V and 3.65V respectively. The accuracy of the standard voltmeter is ±0.02%. The voltage value is collected 10 times for each calibration point and the arithmetic mean is taken. The voltage calibration equation is obtained by fitting the five sets of data as Y_v=1.004·X_v-0.012. The voltage gain calibration coefficient k_v and the voltage offset calibration coefficient b_v are written to the EEPROM address 0x0010 to 0x0017 in floating point format.

[0031] In this embodiment, there are five current calibration points in step S3, set to 0.5A, 1.0A, 2.0A, 5.0A and 10.0A respectively. The accuracy of the electronic load instrument is ±0.05%, and the accuracy of the standard ammeter is ±0.02%. The arithmetic mean of the reported current value is taken 10 times for each calibration point. The current calibration equation is obtained by fitting the five sets of data as Y_i=0.978·X_i-0.003. The current gain calibration coefficient k_i and the current offset calibration coefficient b_i are written to the address 0x0018 to 0x001F of the EEPROM in floating-point format. Three current calibration points are selected in the charging direction for verification. After verification, the charging direction and the discharging direction share the same set of coefficients.

[0032] In this embodiment, the battery in step S4 is a lithium iron phosphate battery. The charging cutoff voltage is 3.65V, and the discharging cutoff voltage is 2.5V. When fully charged, it is charged at a constant current of 0.5C to 3.65V, then switched to constant voltage until the current is less than 0.02C. After standing for 2 hours, the SOC is marked as 100%, and the open circuit voltage value of 3380mV is recorded. It is discharged in segments at 0.05C. After discharging 5% of the charge, it is left to stand for 30 minutes until the terminal voltage stabilizes, and the measured value of the open circuit voltage is recorded. This process continues until the voltage drops to 2.5V and is marked as SOC. With OC=0%, a total of 21 measured open-circuit voltage values ​​were obtained for each state of charge (OCC) point. The OCV deviation threshold is 5mV. For OCC points with a deviation greater than or equal to 5mV, the measured values ​​were used to replace the corresponding values ​​in the standard table. The updated OCC model parameters for the 20 OCC points (5% to 100%) were written to the EEPROM at addresses 0x0028 to 0x004F in uint16 format, in mV units, with 2 bytes per point, for a total of 40 bytes. After writing, the data was read back to verify and confirm that the data was consistent.

[0033] In step S1 of this embodiment, the controllable temperature environment is provided by a constant temperature chamber, the standard temperature value is measured by a standard thermometer set inside the constant temperature chamber, the battery management system is set inside the constant temperature chamber, and at each of the temperature calibration points, the reported temperature value and the standard temperature value are collected after the temperature sensor of the battery management system and the air inside the constant temperature chamber reach thermal equilibrium. Each of the temperature calibration points at least covers the low end, middle end and high end of the operating temperature range of the battery management system.

[0034] In this embodiment, the static heat preservation time at the two extreme temperature points of -20℃ and 60℃ is 45 minutes, and at other temperature points it is 30 minutes. Twenty consecutive temperature calibration values ​​are collected at each temperature calibration point, and the arithmetic mean is taken as the reported temperature value for that point to eliminate random noise interference from single sampling. Specifically, at the -20℃ point, the battery management system reported a temperature of -18.5℃, while the standard thermometer reading was -19.8℃, a deviation of +1.3℃. At the 60℃ point, the battery management system reported a temperature of 61.2℃, while the standard thermometer reading was 60.0℃, a deviation of +1.2℃. The deviation directions in the high-temperature and low-temperature regions are consistent and the values ​​are close, indicating that the gain deviation of the temperature sensor is the main source of error.

[0035] In step S2 of this embodiment, the standard voltage value is provided by a programmable regulated power supply electrically connected to the voltage sampling channel of the battery management system. The standard voltage value corresponding to each voltage calibration point is the actual output voltage value of the programmable regulated power supply. The actual output voltage value is verified by a standard voltmeter electrically connected to the output terminal of the programmable regulated power supply. Each voltage calibration point at least covers the low end, middle end and high end of the operating voltage range of the battery management system.

[0036] In this embodiment, the accuracy of the standard voltmeter is ±0.02%. At each voltage calibration point, the reported voltage values ​​are collected 10 times consecutively, and the arithmetic mean is taken as the reported voltage value of that point, so as to reduce the impact of random quantization noise during analog-to-digital converter reading on the fitting results.

[0037] In step S3 of this embodiment, the standard current value is set by an electronic load instrument connected in series with the current sampling channel of the battery management system in the current loop. The standard current value corresponding to each current calibration point is the measured current value of the standard ammeter connected in series in the current loop. Each current calibration point covers at least the low end, middle end and high end of the operating current range of the battery management system. The current calibration coefficients share the same set of coefficients for both the charging direction and the discharging direction.

[0038] In this embodiment, the accuracy of the electronic load meter is ±0.05%, and the accuracy of the standard ammeter is ±0.02%. The arithmetic mean of the reported current values ​​is taken 10 times for each current calibration point as the reported current value of that point. Three current calibration points are selected in the charging direction for verification. After verification, the charging direction and the discharging direction share the same set of coefficients.

[0039] In steps S1, S2, and S3 of this embodiment, the linear fitting is as follows: At least three temperature calibration points are used to collect corresponding standard temperature values ​​and reported temperature values; at at least three voltage calibration points are used to collect corresponding standard voltage values ​​and reported voltage values; and at at least three current calibration points are used to collect corresponding standard current values ​​and reported current values. The least squares method is then used to perform linear fitting on the standard temperature values ​​and reported temperature values ​​at each temperature calibration point to obtain a temperature calibration equation; the standard voltage values ​​and reported voltage values ​​at each voltage calibration point are then used to perform linear fitting on the standard current values ​​and reported current values ​​at each current calibration point to obtain a current calibration equation. The temperature calibration equation is: Y_t = k_t·X_t + b_t, Where Y_t represents the standard temperature value, X_t represents the reported temperature value, k_t is the temperature gain calibration coefficient, and b_t is the temperature offset calibration coefficient; The voltage calibration equation is as follows: Y_v = k_v·X_v + b_v Where Y_v represents the standard voltage value, X_v represents the reported voltage value, k_v is the voltage gain calibration coefficient, and b_v is the voltage offset calibration coefficient; The current calibration equation is as follows: Y_i = k_i·X_i + b_i, Where Yi represents the standard current value, Xi represents the reported current value, k_i is the current gain calibration coefficient, and b_i is the current offset calibration coefficient.

[0040] In this embodiment, the calibration of the three parameters—temperature, voltage, and current—all employs a least-squares fitting process with five calibration points. The optimal gain calibration coefficient k and offset calibration coefficient b are determined by minimizing the sum of squared residuals at each calibration point. The calibration equations for the three parameters have a unified structure, all in the form Y=k·X+b. During battery management system operation, the correction logic for the three parameters—temperature, voltage, and current—is the same. Only two multiplication operations and one addition operation are required to complete the correction of a single parameter. The calculation time for a single correction of a single parameter is less than 1 microsecond (based on an ARM Cortex-M0 core with a main frequency of 48MHz). No table lookup or iterative calculation is required, and it does not have any additional impact on the timing of the original real-time sampling and control tasks of the battery management system.

[0041] In step S4 of this embodiment, the specific steps for updating the state of charge model parameters are as follows: S41: Based on the voltage sampling value collected by the calibrated voltage sampling channel, charge the battery at a constant current of 0.5C to the charging cut-off voltage of the battery, then switch to constant voltage charging until the charging current of the battery is less than 0.02C. After resting, mark the state of charge as 100% and record the current open circuit voltage value. S42: Discharge the battery in segments at a discharge rate of 0.05C. Stop discharging after each 5% of the rated capacity is discharged and allow the battery to stand until the terminal voltage stabilizes. Record the measured value of the open circuit voltage corresponding to the current state of charge. Repeat this process until the terminal voltage drops to the discharge cutoff voltage of the battery and mark the state of charge as 0%. S43: Compare the measured open-circuit voltage value corresponding to each state of charge point with the corresponding value in the pre-stored standard open-circuit voltage-state of charge table point by point. If there is a state of charge point with a deviation greater than or equal to the preset OCV deviation threshold, then replace the standard voltage value of the corresponding state of charge point in the standard open-circuit voltage-state of charge table with the measured open-circuit voltage value. If the deviation of all state of charge points is less than the preset OCV deviation threshold, then keep the standard open-circuit voltage-state of charge table unchanged. S44: Write the updated standard open-circuit voltage-state-of-charge table into the non-volatile memory.

[0042] In this embodiment, the resting time in step S41 is 2 hours. During the resting period, the open-circuit voltage value is continuously monitored. Once the voltage change rate is less than a preset threshold, the current open-circuit voltage value is recorded as the open-circuit voltage value corresponding to 100% state of charge, to avoid reading deviations introduced by the voltage rebound not yet fully converging in the early stage of resting. In step S42, the deviations between the measured open-circuit voltage values ​​at each state of charge point and the corresponding values ​​in the standard open-circuit voltage-state of charge table are as follows: -2mV at 90% state of charge, -2mV at 70% state of charge, -2mV at 50% state of charge, -6mV at 30% state of charge, -10mV at 20% state of charge, and -... At 15mV and 5% state of charge, the deviation is -20mV. After the updated standard open-circuit voltage-state of charge table is written into the non-volatile memory in step S44, the battery management system estimates the state of charge during operation as follows: when the battery is idle for more than 30 minutes, the current open-circuit voltage value is read and the state of charge value obtained from the open-circuit voltage lookup table is obtained by reverse calculation. The final state of charge value is weighted and fused by combining the ampere-hour integral value with the final state of charge value being equal to 0.7 times the ampere-hour integral value plus 0.3 times the state of charge value obtained from the open-circuit voltage lookup table. When the battery is detected to be fully charged or discharged, it is forced to return to 100% state of charge or 0% state of charge respectively to eliminate long-term integral accumulation error.

[0043] In steps S1, S2, and S3 of this embodiment, the temperature calibration coefficient, the voltage calibration coefficient, and the current calibration coefficient are all written to a designated address of the non-volatile memory via a communication interface on a calibration board connected to the battery management system, using an overwrite method. After writing the temperature calibration coefficient, the reported temperature value of the battery management system at each temperature calibration point is re-acquired and compared with the standard temperature value. If the deviation of each temperature calibration point is less than a preset temperature deviation threshold, the temperature calibration is deemed successful and proceeds to the next step. If the deviation of any temperature calibration point is greater than or equal to the preset temperature deviation threshold, the temperature calibration is deemed unsuccessful. After writing the voltage calibration coefficient, the temperature calibration coefficient is re-acquired and the current calibration coefficient is compared with the standard temperature value. The reported voltage values ​​of the battery management system at each voltage calibration point are collected and compared with the standard voltage value. If the deviation of each voltage calibration point is less than the preset voltage deviation threshold, the voltage calibration is deemed successful and proceeds to the next step. If the deviation of any voltage calibration point is greater than or equal to the preset voltage deviation threshold, the voltage calibration is deemed unsuccessful. After writing the current calibration coefficient, the reported current values ​​of the battery management system at each current calibration point are collected again and compared with the standard current value. If the deviation of each current calibration point is less than the preset current deviation threshold, the current calibration is deemed successful and proceeds to the next step. If the deviation of any current calibration point is greater than or equal to the preset current deviation threshold, the current calibration is deemed unsuccessful.

[0044] In this embodiment, the temperature deviation threshold is ±0.5℃, the voltage deviation threshold is ±1mV, and the current deviation threshold is ±10mA. If any preceding step fails to calibrate, the subsequent steps are terminated and an error code is reported.

[0045] In this embodiment, the communication interface is a UART interface, the non-volatile memory is the EEPROM of the battery management system, the gain calibration coefficient and offset calibration coefficient of the temperature calibration coefficient, the voltage calibration coefficient and the current calibration coefficient are stored in the EEPROM in floating-point format and little-endian, and are written through command code 0x02; the state of charge model parameters are written in batches through command code 0x03.

[0046] In this embodiment, the baud rate of the UART interface is 115200bps. The temperature gain calibration coefficient k_t and temperature offset calibration coefficient b_t are written to addresses 0x0020 to 0x0027, the voltage gain calibration coefficient k_v and voltage offset calibration coefficient b_v are written to addresses 0x0010 to 0x0017, and the current gain calibration coefficient k_i and current offset calibration coefficient b_i are written to addresses 0x0018 to 0x001F. Each coefficient is stored in IEEE 754 single-precision floating-point format, occupying 4 bytes, little-endian. The gain calibration coefficient and offset calibration coefficient are written via command code 0x02, and the state of charge model parameters are written in batches via command code 0x03. The communication frame format is frame header 0xAA plus command code plus address plus data plus checksum plus frame tail 0x55. The battery management system replies with ACK to indicate successful writing.

[0047] In this embodiment, the present invention also discloses an offline multi-parameter calibration system for an embedded BMS, including a calibration board, a constant temperature chamber, a standard thermometer, a programmable power supply, a standard voltmeter, an electronic load meter, a standard ammeter, a battery, and a communication interface. The calibration board is equipped with a microcontroller and a memory electrically connected to the microcontroller. The constant temperature chamber is used to house the battery management system to be calibrated and provide a controllable temperature environment. The standard thermometer is disposed inside the constant temperature chamber and electrically connected to the calibration board to transmit the collected standard temperature values ​​to the calibration board. The programmable power supply is electrically connected to both the calibration board and the battery. The calibration board controls the programmable power supply to provide charging voltage to the battery. The standard voltmeter is connected to both the programmable power supply and the battery. The calibration board is electrically connected to transmit the collected standard voltage value to the calibration board. The voltage sampling channel of the battery management system is electrically connected to the battery. The temperature sensor in the battery management system transmits the collected reported temperature value to the calibration board via the communication interface. The electronic load is electrically connected to the battery. The electronic load, the standard ammeter, and the current sampling channel of the battery management system are connected in series in the discharge circuit of the battery. The calibration board controls the electronic load to provide a discharge load to the battery. The standard ammeter is electrically connected to the calibration board to transmit the collected standard current value to the calibration board. The communication interface is located on the calibration board. The battery management system is electrically connected to the calibration board through the communication interface.

[0048] In this embodiment, the microcontroller runs a calibration process control program, processes the measurement data returned by the standard thermometer, the standard voltmeter, and the standard ammeter in real time, and coordinates the timing between the constant temperature chamber, the programmable regulated power supply, and the electronic load instrument. The memory stores the standard open-circuit voltage-state-of-charge table and the calibration process control program, enabling the calibration board to independently complete the entire calibration task without the host computer. In each calibration step, after reading the data reported by the battery management system through the communication interface, the calibration board first compares it with the corresponding standard value in real time. If the deviation of any calibration point is greater than or equal to the preset corresponding deviation threshold, the calibration step is determined to have failed and subsequent steps are terminated.

[0049] In this embodiment, the calibration board's memory pre-stores a standard open-circuit voltage-state-of-charge table and a calibration process control program. The calibration board reads the reported temperature value, reported voltage value, and reported current value reported by the battery management system through the communication interface, and writes the temperature calibration coefficient, the voltage calibration coefficient, the current calibration coefficient, and the updated state-of-charge model parameters into the non-volatile memory of the battery management system through the communication interface.

[0050] In this embodiment, the standard open-circuit voltage-state-of-charge table is pre-stored in the memory of the calibration board in the form of an array, and the calibration process control program is fixed in the memory of the calibration board in the form of firmware. After the calibration board is powered on, it automatically executes the calibration process control program and outputs control signals to the constant temperature chamber, the programmable regulated power supply and the electronic load instrument in a preset sequence to complete the calculation and writing of each calibration coefficient and the updating and writing of the state of charge model parameters. After calibration, the calibration board packages all calibration results into a test record and stores it in the memory for subsequent production traceability and quality analysis.

[0051] The above description is merely the optimal embodiment of the present invention and is not intended to limit the present invention. Any modifications or substitutions made by those skilled in the art without departing from the essence and scope of protection of the present invention should also be within the scope of protection of the present invention.

Claims

1. An offline multi-parameter calibration method for an embedded BMS, characterized in that, Includes the following steps: S1: Provide a number of preset temperature calibration points, which are distributed within the temperature range to be calibrated. Place the battery management system to be calibrated in a controllable temperature environment. Collect the reported temperature value and standard temperature value of the battery management system at each of the temperature calibration points. Perform linear fitting based on the standard temperature value and the reported temperature value at each of the temperature calibration points to generate a temperature calibration coefficient. Write the temperature calibration coefficient into the non-volatile memory of the battery management system. S2: Provide a number of preset voltage calibration points, which are distributed within the voltage range to be calibrated. After writing the temperature calibration coefficient, apply the standard voltage value corresponding to each voltage calibration point to the voltage sampling channel of the battery management system, collect the reported voltage value of the battery management system at each voltage calibration point, perform linear fitting between the standard voltage value and the reported voltage value of each voltage calibration point to generate a voltage calibration coefficient, and write the voltage calibration coefficient into the non-volatile memory. S3: Provide a number of preset current calibration points, which are distributed within the current range to be calibrated. After writing the voltage calibration coefficient, the current sampling channel of the battery management system collects the reported current value of the battery management system at each current calibration point by passing the standard current value corresponding to each current calibration point. Based on the standard current value of each current calibration point and the reported current value, a current calibration coefficient is generated by linear fitting. The current calibration coefficient is then written into the non-volatile memory. S4: Provide several preset state of charge points, which are distributed within the state of charge interval. After writing the current calibration coefficient, use the voltage sampling value collected by the calibrated voltage sampling channel to perform charging and discharging operations on the battery electrically connected to the battery management system. Obtain the corresponding open-circuit voltage measured value at each state of charge point. Update the state of charge model parameters of the battery management system according to the open-circuit voltage measured value of each state of charge point. Write the updated state of charge model parameters into the non-volatile memory. The specific steps for updating the parameters of the state of charge model are as follows: S41: Based on the voltage sampling value collected by the calibrated voltage sampling channel, charge the battery at a constant current of 0.5C to the charging cut-off voltage of the battery, then switch to constant voltage charging until the charging current of the battery is less than 0.02C. After resting, mark the state of charge as 100% and record the current open circuit voltage value. S42: Discharge the battery in segments at a discharge rate of 0.05C. Stop discharging after each 5% of the rated capacity is discharged and allow the battery to stand until the terminal voltage stabilizes. Record the measured value of the open circuit voltage corresponding to the current state of charge. Repeat this process until the terminal voltage drops to the discharge cutoff voltage of the battery and mark the state of charge as 0%. S43: Compare the measured open-circuit voltage value corresponding to each state of charge point with the corresponding value in the pre-stored standard open-circuit voltage-state of charge table point by point. If there is a state of charge point with a deviation greater than or equal to the preset OCV deviation threshold, then replace the standard voltage value of the corresponding state of charge point in the standard open-circuit voltage-state of charge table with the measured open-circuit voltage value. If the deviation of all state of charge points is less than the preset OCV deviation threshold, then keep the standard open-circuit voltage-state of charge table unchanged. S44: Write the updated standard open-circuit voltage-state-of-charge table into the non-volatile memory.

2. The offline multi-parameter calibration method for an embedded BMS according to claim 1, characterized in that, In step S1, the controllable temperature environment is provided by a constant temperature chamber, and the standard temperature value is measured by a standard thermometer set inside the constant temperature chamber. The battery management system is set inside the constant temperature chamber. At each of the temperature calibration points, the reported temperature value and the standard temperature value are collected after the temperature sensor of the battery management system and the air inside the constant temperature chamber reach thermal equilibrium. Each of the temperature calibration points covers at least the low end, middle end and high end of the operating temperature range of the battery management system.

3. The offline multi-parameter calibration method for an embedded BMS according to claim 1, characterized in that, In step S2, the standard voltage value is provided by a programmable regulated power supply electrically connected to the voltage sampling channel of the battery management system. The standard voltage value corresponding to each voltage calibration point is the actual output voltage value of the programmable regulated power supply. The actual output voltage value is verified by a standard voltmeter electrically connected to the output terminal of the programmable regulated power supply. Each voltage calibration point at least covers the low end, middle end and high end of the operating voltage range of the battery management system.

4. The offline multi-parameter calibration method for an embedded BMS according to claim 1, characterized in that, In step S3, the standard current value is set by an electronic load instrument connected in series with the current sampling channel of the battery management system in the current loop. The standard current value corresponding to each current calibration point is the measured current value of the standard ammeter connected in series in the current loop. Each current calibration point covers at least the low end, middle end and high end of the operating current range of the battery management system. The current calibration coefficients share the same set of coefficients for both the charging direction and the discharging direction.

5. The offline multi-parameter calibration method for an embedded BMS according to claim 1, characterized in that, In steps S1, S2, and S3, the linear fitting involves: collecting the corresponding standard temperature value and the reported temperature value at at least three temperature calibration points, collecting the corresponding standard voltage value and the reported voltage value at at least three voltage calibration points, and collecting the corresponding standard current value and the reported current value at at least three current calibration points; using the least squares method to perform linear fitting on the standard temperature value and the reported temperature value at each temperature calibration point to obtain a temperature calibration equation; performing linear fitting on the standard voltage value and the reported voltage value at each voltage calibration point to obtain a voltage calibration equation; and performing linear fitting on the standard current value and the reported current value at each current calibration point to obtain a current calibration equation. The temperature calibration equation is: Y_t = k_t·X_t + b_t, Where Y_t represents the standard temperature value, X_t represents the reported temperature value, k_t is the temperature gain calibration coefficient, and b_t is the temperature offset calibration coefficient; The voltage calibration equation is as follows: Y_v = k_v·X_v + b_v Where Y_v represents the standard voltage value, X_v represents the reported voltage value, k_v is the voltage gain calibration coefficient, and b_v is the voltage offset calibration coefficient; The current calibration equation is as follows: Y_i = k_i·X_i + b_i, Where Yi represents the standard current value, Xi represents the reported current value, k_i is the current gain calibration coefficient, and b_i is the current offset calibration coefficient.

6. The offline multi-parameter calibration method for an embedded BMS according to claim 1, characterized in that, In steps S1, S2, and S3, the temperature calibration coefficient, the voltage calibration coefficient, and the current calibration coefficient are all written to a designated address of the non-volatile memory via a communication interface on a calibration board connected to the battery management system, using an overwrite method. After writing the temperature calibration coefficient, the reported temperature values ​​of the battery management system at each temperature calibration point are re-acquired and compared with the standard temperature value. If the deviation of each temperature calibration point is less than a preset temperature deviation threshold, the temperature calibration is deemed successful and proceeds to the next step. If the deviation of any temperature calibration point is greater than or equal to the preset temperature deviation threshold, the temperature calibration is deemed unsuccessful. After writing the voltage calibration coefficient, the reported temperature values ​​at each temperature calibration point are re-acquired. The battery management system reports the voltage value at each voltage calibration point and compares it with the standard voltage value. If the deviation of each voltage calibration point is less than a preset voltage deviation threshold, the voltage calibration is deemed successful and proceeds to the next step. If the deviation of any voltage calibration point is greater than or equal to the preset voltage deviation threshold, the voltage calibration is deemed unsuccessful. After writing the current calibration coefficient, the battery management system reports the current value at each current calibration point and compares it with the standard current value. If the deviation of each current calibration point is less than a preset current deviation threshold, the current calibration is deemed successful and proceeds to the next step. If the deviation of any current calibration point is greater than or equal to the preset current deviation threshold, the current calibration is deemed unsuccessful.

7. The offline multi-parameter calibration method for an embedded BMS according to claim 6, characterized in that, The communication interface is a UART interface, the non-volatile memory is the EEPROM of the battery management system, the gain calibration coefficient and offset calibration coefficient of the temperature calibration coefficient, the voltage calibration coefficient and the current calibration coefficient are stored in the EEPROM in floating-point format and little-endian, and are written through command code 0x02; the state of charge model parameters are written in batches through command code 0x03.

8. An offline multi-parameter calibration system for an embedded BMS, used to implement the offline multi-parameter calibration method for an embedded BMS as described in any one of claims 1 to 7, characterized in that, The system includes a calibration board, a constant temperature chamber, a standard thermometer, a programmable power supply, a standard voltmeter, an electronic load meter, a standard ammeter, a battery, and a communication interface. The calibration board is equipped with a microcontroller and a memory electrically connected to the microcontroller. The constant temperature chamber houses the battery management system to be calibrated and provides a controlled temperature environment. The standard thermometer is located inside the constant temperature chamber and electrically connected to the calibration board to transmit the collected standard temperature values. The programmable power supply is electrically connected to both the calibration board and the battery. The calibration board controls the programmable power supply to provide charging voltage to the battery. The standard voltmeter is electrically connected to both the programmable power supply and the calibration board to transmit the collected standard temperature values. The voltage value is transmitted to the calibration board. The voltage sampling channel of the battery management system is electrically connected to the battery. The temperature sensor in the battery management system transmits the collected reported temperature value to the calibration board via the communication interface. The electronic load is electrically connected to the battery. The electronic load, the standard ammeter, and the current sampling channel of the battery management system are connected in series in the discharge circuit of the battery. The calibration board controls the electronic load to provide a discharge load to the battery. The standard ammeter is electrically connected to the calibration board to transmit the collected standard current value to the calibration board. The communication interface is located on the calibration board. The battery management system is electrically connected to the calibration board through the communication interface.

9. The offline multi-parameter calibration system for an embedded BMS according to claim 8, characterized in that, The calibration board's memory pre-stores a standard open-circuit voltage-state-of-charge table and a calibration process control program. The calibration board reads the reported temperature, voltage, and current values ​​reported by the battery management system through the communication interface, and writes the temperature calibration coefficient, voltage calibration coefficient, current calibration coefficient, and updated state-of-charge model parameters into the non-volatile memory of the battery management system through the communication interface.

Images