Battery string voltage acquisition system with calibration system and voltage acquisition method

Abstract

The invention relates to the technical field of voltage acquisition, in particular to a battery string voltage acquisition system with a calibration system and a voltage acquisition method. Which comprises a multi-path voltage input end, a voltage signal adjusting end, a voltage follower, an analog-to-digital converter, a main control center and a communication circuit, and is characterized in that the voltage signal adjusting end comprises a dynamic voltage division network, consists of a digital potentiometer and a high-voltage operational amplifier, and dynamically adjusts the voltage division ratio to suppress common-mode voltage; the main control center is integrated with a multi-channel synchronous sampling calibration module and generates a global synchronous signal through the main MCU to control the ADC to perform synchronous acquisition; a temperature compensation module is arranged between the voltage follower and the ADC, and a digital temperature sensor is adopted to correct the gain drift of the operational amplifier in real time; the voltage division ratio of the dynamic voltage division network is dynamically adjusted according to the real-time common-mode voltage and a preset safety threshold value. According to the scheme, dynamic partial pressure adjustment, multi-channel self-calibration, nonlinear temperature compensation and a rapid protection mechanism can be integrated.

Description

Technical Field

[0001] The present invention relates to the technical field of voltage acquisition, and in particular to a battery string voltage acquisition system with a calibration system and a voltage acquisition method. Background Art

[0002] As electric vehicles and energy storage systems continue to increase their requirements for automated battery testing and management accuracy, traditional voltage acquisition technology faces multiple bottlenecks. In high-voltage battery packs, the common-mode voltage increases with the number of connected battery strings and fluctuates dramatically during the charge and discharge process, causing breakdown and damage to the operational amplifier input stage. Multi-channel ADCs experience inter-channel gain errors due to reference source drift and PCB layout differences, forcing the system to shut down for manual calibration. Temperature-induced nonlinear gain drift in the op amp severely limits the system's full-temperature measurement accuracy. Existing technologies use fixed voltage divider ratios that cannot adapt to sudden voltage changes. Conventional linear temperature compensation can multiply residual errors several times at extreme temperatures. Fuse-based overcurrent protection suffers from response lag and non-recoverable defects, reducing system availability by over 40%.

[0003] Even more serious is the fact that electromagnetic interference caused by transient surges in the battery pack superimposes ±50mV of noise on the sampled signal. Traditional single-stage filtering solutions are insufficient in filtering frequencies above 1MHz, resulting in SOC estimation errors exceeding 3%. Timing deviations caused by clock jitter in multi-channel synchronous acquisition systems further amplify voltage acquisition errors. The industry urgently needs a technical solution that integrates dynamic voltage divider adjustment, multi-channel self-calibration, nonlinear temperature compensation, and a fast protection mechanism. Summary of the Invention

[0004] Electromagnetic interference caused by common-mode voltage and transient surges in current battery strings can add ±50mV of noise to the sampled signal. Traditional single-stage filtering solutions are insufficient in the frequency band above 1MHz, resulting in SOC estimation errors exceeding 3%. Timing deviations caused by clock jitter in multi-channel synchronous acquisition systems further amplify the battery voltage acquisition errors. This application provides a battery string voltage acquisition system and method with a calibration system to address these issues.

[0005] To achieve the above object, the present invention is implemented through the following technical solutions: The present application discloses a battery string voltage acquisition system with a calibration system, including a multi-channel voltage input terminal, a voltage signal adjustment terminal, a voltage follower, an analog-to-digital converter, a main control center, and a communication circuit. The voltage signal regulating terminal includes a dynamic voltage divider network composed of a digital potentiometer and a high-voltage operational amplifier, which dynamically adjusts the voltage divider ratio to suppress the common-mode voltage; The main control center integrates a multi-channel synchronous sampling and calibration module, and generates a global synchronization signal through the main MCU to control ADC synchronous acquisition; A temperature compensation module is provided between the voltage follower and the ADC, and a digital temperature sensor is used to correct the gain drift of the operational amplifier in real time; The voltage division ratio of the dynamic voltage division network is dynamically adjusted according to the real-time common mode voltage and a preset safety threshold.

[0006] Adopting the above technical solution: The present invention can solve the problems in traditional battery voltage acquisition systems, such as common-mode voltage damaging devices as the number of battery strings increases, violent fluctuations in charging and discharging, which easily lead to overvoltage damage to the op amp input, and temperature changes easily causing op amp gain drift. This solution uses a dynamic voltage divider network to adjust the voltage divider ratio in real time, so that the op amp input voltage is always within a safe range; generates a global synchronization signal through the MCU to achieve microsecond-level synchronous triggering of multiple ADC chips; embeds temperature acquisition in the signal chain, and establishes a temperature-gain compensation model.

[0007] The voltage division ratio of the dynamic voltage division network satisfies: ; in, is the real common mode voltage, is the maximum withstand voltage of the op amp, is the dynamic adjustment coefficient, and its value range is .

[0008] Adopt the above technical solution: This solution can solve the problem of the traditional fixed voltage ratio solution when the battery pack load changes instantaneously. The problem of excessively high input voltage of the op amp is solved by introducing When the common-mode voltage change rate exceeds 100V / ms, the voltage divider ratio is automatically increased. By dynamically adjusting the coefficient γ to an optimal range of 0.01-0.1, Monte Carlo simulations have shown that system stability can be improved.

[0009] The synchronous sampling calibration module adopts the cross-redundancy calibration method to construct the error transfer matrix, which is: ; in, For the and Channel coupling error, For the Channel noise standard deviation, .

[0010] Adopting the above technical solution: This solution can solve the problem of traditional multi-channel ADCs where the inter-channel gain error increases due to reference source drift, PCB trace differences, etc., and manual and timely calibration is required.

[0011] Further preferably, the gain correction formula of the temperature compensation module is: ; in, The gain is calibrated at 25°C. is the first-order temperature coefficient, is the second-order temperature coefficient.

[0012] Adopting the above technical solution: This solution can solve the problem that the op amp gain of the traditional technical solution changes with the temperature in a quadratic curve, and the traditional linear compensation has a large residual error at extremely low temperatures.

[0013] Further preferably, in the TVS diode and resettable fuse parallel protection circuit of the dynamic voltage divider network, the breakdown voltage of the TVS diode is set to 120%-130% of the system's maximum common-mode voltage, and the response time is ≤1ns; the resettable fuse uses a high molecular polymer positive temperature coefficient material, which triggers a circuit break when the current exceeds 150% of the rated value, and automatically recovers within 60s after the fault is eliminated, and the rated current value is set to 10mA±5% based on the maximum operating current of the op amp.

[0014] Adopting the above technical solution: This solution uses TVS diodes and polymer self-recovery fuses to form dual protection, which can solve the impact of battery pack surge current on the acquisition system. The self-recovery function of the fuse solves the problem of traditional fuses being unable to recover after melting, causing system downtime.

[0015] Further preferably, the analog-to-digital converter adopts a 16-bit SAR ADC with an input impedance ≥1MΩ, and achieves a nonlinear error of ±0.0015% in the input range of -15V to +15V; the built-in overvoltage protection circuit can automatically cut off the sampling channel when the input exceeds ±18V, and switch to the backup channel within 10μs; in conjunction with the anti-aliasing filter, a signal-to-noise ratio of more than 80dB is achieved at a sampling rate of 1MSPS, where the cutoff frequency of the anti-aliasing filter is set to 1 / 5 of the ADC sampling frequency and the attenuation slope is ≥40dB / decade.

[0016] Adopting the above technical solution: This solution reduces signal attenuation to <0.1% by using a high-impedance 16-bit SAR ADC with a 1MΩ input impedance. This can solve the signal attenuation problem caused by low input impedance of traditional ADCs in scenarios where multiple batteries are connected in series.

[0017] Further preferably, the isolated CAN FD interface circuit supports the ISO 11898-2 standard protocol and achieves a continuous isolation withstand voltage of 3.75kV rms / 1min at a transmission rate of 5Mbps; each frame of data in the communication protocol contains 16-channel voltage values, temperature status words and CRC-32 check codes, and a dynamic priority allocation mechanism is used to ensure that the transmission delay of key data is ≤2ms; when it is detected that the communication bit error rate exceeds the threshold, it automatically switches to the frequency reduction mode and starts the real-time data retransmission strategy.

[0018] Adopting the above technical solution: This solution uses a 16-bit SAR ADC with a 1MΩ input impedance to reduce signal attenuation; it solves the signal attenuation problem caused by low input impedance of traditional ADCs in the scenario of multiple battery cells connected in series.

[0019] Further preferably, the ARIMA prediction model uses 100ms as the basic prediction step, establishes a prediction model of (p, d, q) = (2, 1, 1) through historical common-mode voltage data, and the prediction error band is set to ±3%; when it is predicted that the common-mode voltage change within the next 200ms will exceed 15% of the operational amplifier's withstand voltage value, the voltage divider ratio adjustment instruction is triggered 10ms in advance; the model has a built-in online learning mechanism, which automatically updates the model parameters every 24 hours, and the sliding average method is used to maintain the prediction function during the model update.

[0020] Adopting the above technical solution: This solution establishes an ARIMA (2,1,1) prediction model and predicts the voltage trend in the next 200ms with a step size of 100ms. It can solve the problems of common-mode voltage change rate fluctuations caused by battery aging and prediction deviation caused by traditional fixed threshold voltage adjustment strategy.

[0021] Further preferably, the 16-channel synchronous acquisition system has a sampling clock deviation of ≤10ns between channels within an operating temperature range of -40°C to +85°C, and an integral nonlinear error of ≤±2LSB within the full range; the system self-calibrates the reference source after power-on, the calibration time is <500ms, and the zero drift after calibration is <±5μV / °C; it supports independent refresh rate setting for each channel, which can be configured to a continuous sampling mode of up to 1kHz, and the data output refresh rate is adaptively matched to the communication interface bandwidth.

[0022] Adopting the above technical solution: This solution can solve the problems faced by the traditional technical solution battery collection system, such as 50Hz power frequency interference and switching noise above 10MHz, and insufficient attenuation of traditional single-stage filtering in the 1kHz-100MHz frequency band.

[0023] A voltage acquisition method, applied to a battery string voltage acquisition system with a calibration system as described in any one of the above, comprising: S1: The voltage divider ratio is adjusted in real time through a dynamic voltage divider network. The voltage protection threshold at the input of the operational amplifier is adaptively adjusted based on the instantaneous value of the common-mode voltage and its rate of change. When a transient overvoltage is detected, the resettable fuse and voltage clamp protection are synchronously triggered. S2: Two-stage filtering is applied to the collected signal. The front-stage filter suppresses low-frequency power frequency interference, and the rear-stage filter attenuates high-frequency switching noise. A star grounding layout is used to reduce crosstalk between channels. S3: A multi-channel synchronous acquisition circuit is used to acquire the voltage signal of each battery cell. The inter-channel coupling error is modeled using the error transfer matrix. The maximum error source is located through singular value decomposition and dynamic calibration compensation is performed. S4: Calculate the second-order temperature compensation coefficient in real time according to the ambient temperature, and perform closed-loop correction on the reference voltage of the analog-to-digital converter in combination with the pre-calibrated gain nonlinear curve; S5: Synchronize the sampling timing of each acquisition channel based on the global clock signal. After completing the preset number of data accumulations, output the battery voltage data packet that meets the preset bit error rate requirements through the isolated communication interface. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0025] Figure 1 This is the block diagram of the battery string voltage acquisition system with calibration system; Figure 2 This is a flow chart of the voltage acquisition method of this application. DETAILED DESCRIPTION

[0026] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0027] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, operations, elements, components and / or groups thereof.

[0028] See also Figure 1 and Figure 2 Traditional battery voltage acquisition systems have the following technical problems: the common-mode voltage fluctuates dramatically with battery pack charging and discharging, causing overvoltage damage to the op amp input; multi-channel sampling timing deviations cause SOC calculation errors; temperature changes cause op amp gain drift, and the error increases by more than three times in low-temperature environments. Based on this, the present application discloses a battery string voltage acquisition system with a calibration system, including multiple voltage input terminals, a voltage signal conditioning terminal, a voltage follower, an analog-to-digital converter, a main control center, and a communication circuit. The voltage signal regulating terminal includes a dynamic voltage divider network composed of a digital potentiometer and a high-voltage operational amplifier, which dynamically adjusts the voltage divider ratio to suppress the common-mode voltage; The main control center integrates a multi-channel synchronous sampling and calibration module, and generates a global synchronization signal through the main MCU to control ADC synchronous acquisition; A temperature compensation module is provided between the voltage follower and the ADC, and a digital temperature sensor is used to correct the gain drift of the operational amplifier in real time; The voltage division ratio of the dynamic voltage division network is dynamically adjusted according to the real-time common mode voltage and a preset safety threshold.

[0029] It's worth noting that the aforementioned solutions utilize a dynamic voltage divider network to adjust the voltage divider ratio in real time, ensuring that the op amp input voltage remains within a safe range. A global synchronization signal is generated by the main MCU, enabling micro-scale-level synchronous triggering of multiple ADC chips. Furthermore, a digital temperature sensor is embedded in the signal chain to establish a temperature-gain compensation model. This approach improves common-mode voltage rejection by fivefold, reaching ±275V, compared to ±30V in traditional solutions, reducing the op amp damage rate by 98%. The 16-channel synchronous acquisition timing deviation is ≤10ns, doubling the synchronization accuracy compared to the 1μs deviation of traditional RS485 cascade solutions. Gain drift is ≤±0.005% / °C within the -40°C to +85°C temperature range, significantly improving stability compared to uncompensated systems.

[0030] For example, the traditional fixed voltage ratio solution will cause the battery pack to load instantaneously change. Too high will cause the op amp input voltage to exceed the limit, with a typical failure rate of 10 -4 Based on this, the voltage division ratio of the dynamic voltage division network satisfies: ; in, is the real-time common mode voltage, representing the potential fluctuation at the midpoint of the battery pack. The maximum voltage resistance of the op amp is set to ±18V according to the selected op amp device. is the dynamic adjustment coefficient, and its value range is .

[0031] Static voltage divider item: :Can ensure that the input voltage does not exceed ,when =600V, the voltage division ratio ; Dynamic correction items: :For electric vehicle rapid acceleration / braking scenarios, when it is detected , reserve a safety margin by increasing the voltage divider ratio.

[0032] Constraint coefficient: The range of 0.01-0.1 is determined by Monte Carlo simulation. When the system is prone to oscillation; Insufficient response.

[0033] It is worth mentioning that this scheme introduces the differential term When the common-mode voltage change rate exceeds 100V / ms, the voltage divider ratio is automatically increased by 5%-10%. Under a sudden voltage change of 1000V / ms, the overshoot voltage at the op amp input is ≤0.2V, significantly lower than the 3V-5V of traditional solutions. The dynamic adjustment coefficient γ is optimized within the range of 0.01-0.1, and Monte Carlo simulations have shown that this improves system stability by 82%. The voltage divider ratio adjustment response time is ≤2μs, a 25-fold improvement compared to the 50μs of traditional PID algorithms.

[0034] For example, traditional multi-channel ADCs have inter-channel gain errors of up to ±0.1% due to reference source drift and PCB trace differences, requiring manual calibration at regular intervals. Based on this, the synchronous sampling calibration module uses a cross-redundant calibration method to construct an error transfer matrix, which is formulated as follows: ; in, For the and Channel coupling error reflects systematic errors such as PCB routing and reference source differences; is the kth sampling value of the i-th channel, and the sampling frequency is required to be ≥ 1kHz to capture the noise spectrum; For the The standard deviation of channel noise was measured by Allan analysis of variance; .

[0035] Used to calculate the mean square error of the two-channel measurement values, including systematic error and random noise; Item 2 It is used to deduct the influence of the inherent noise of each channel and extract the pure system coupling error.

[0036] For a 16-channel system, you need to calculate Group The largest error source is located by singular value decomposition (SVD).

[0037] It is worth mentioning that this solution achieves channel consistency of ±0.005% by constructing a sampling data matrix with N ≥ 1000 and solving the inter-channel coupling error Eij through the least squares method. Traditional manual calibration is ±0.05%, which can extend the calibration cycle to 2 years and reduce maintenance costs by 90%. The dimensionality compression algorithm of the error transfer matrix reduces the computational complexity by 75% and can run in real time on the MCU.

[0038] For example, the gain of a traditional op amp changes with temperature in a quadratic curve, and the residual error of traditional linear compensation reaches 0.3% at -40°C. The gain correction formula of the temperature compensation module is: ; in, The gain is calibrated at 25°C using a 6.5-digit multimeter. is the first-order temperature coefficient, is the second-order temperature coefficient.

[0039] In the above formula is the actual gain at temperature T, and the compensation target is ; Linear term Able to correct the main temperature drift in the range of -40℃~+85℃, contributing about 85% of the compensation; nonlinear terms It can achieve curvature compensation for high temperature ranges above 80°C and eliminate the 0.3% residual error of the traditional linear model.

[0040] This program adopts The α and β coefficients are precisely calibrated through high-temperature aging tests to compensate for nonlinear drift. The system can achieve a compensation residual of ≤±0.002% over the entire temperature range, a 15-fold improvement over linear compensation. The α and β coefficients are optimized using cubic spline interpolation, reducing the gain deviation from 0.25% to 0.0018% at 125°C. The compensation parameters are programmed into FRAM memory, achieving a radiation interference resistance of 100 krad (Si). For example, in traditional high-voltage battery module monitoring scenarios, common-mode voltage fluctuations can easily cause operational amplifier input stage breakdown. Traditional fuses suffer from delayed operation and lack of automatic recovery. When a battery pack experiences a transient high-voltage shock, existing protection solutions cannot achieve both fast response (<1ms) and fault recovery capabilities, resulting in frequent system downtime for maintenance. The TVS diode and resettable fuse in the dynamic voltage divider network are connected in parallel for protection. The TVS diode's breakdown voltage is set to 120%-130% of the system's maximum common-mode voltage, with a response time of ≤1ns. The resettable fuse, made of a high-molecular-weight polymer positive temperature coefficient material, trips when the current exceeds 150% of its rated value and automatically recovers within 60s after the fault is resolved. The rated current is set to 10mA±5% based on the op amp's maximum operating current. It is worth mentioning that the above solution uses a TVS diode with a breakdown voltage set to 120%-130% of the system's maximum common-mode voltage, achieving voltage clamping with a 1ns response speed; the resettable fuse uses a high-molecular polymer material, triggering a circuit break at an overcurrent of 150%, and automatically recovering within 60s, eliminating manual intervention; the rated current value is set at 10mA±5% based on the maximum operating current of the op amp, achieving precise overcurrent protection. For example, in the scenario of multiple battery cells connected in series, traditional ADCs have signal attenuation problems caused by low input impedance, and the nonlinear error is greater than 0.01% within the ±10V input range, making it difficult to meet the requirements of high-precision acquisition and control of the average battery voltage. Based on this, the analog-to-digital converter adopts a 16-bit SAR ADC with an input impedance ≥1MΩ, achieving a nonlinear error of ±0.005% within the input range of -15V to +15V. The built-in overvoltage protection circuit can automatically cut off the sampling channel when the input exceeds ±18V and switch to the backup channel within 10μs. In conjunction with the anti-aliasing filter, a signal-to-noise ratio of more than 80dB is achieved at a sampling rate of 1MSPS, where the cutoff frequency of the anti-aliasing filter is set to 1 / 5 of the ADC sampling frequency and the attenuation slope is ≥40dB / decade. It is worth mentioning that this solution uses a 16-bit SAR ADC with a 1MΩ input impedance to reduce signal attenuation to <0.1%. The built-in overvoltage protection switches to the backup channel within 10μs when the input is ±15V. The anti-aliasing filter cutoff frequency is set to 1 / 5 of the sampling frequency, and the 40dB / decade attenuation slope suppresses aliasing noise. It can achieve a nonlinear error of ±0.005% and a voltage detection accuracy of ±0.5mV, which is 4 times higher than the traditional solution. The 80dB signal-to-noise ratio ensures effective resolution of 1μV voltage fluctuations at a 1MSPS sampling rate. The channel switching mechanism shortens the overvoltage fault recovery time to 50μs, compared to >1ms in the original system. For example, the traditional battery management system has high voltage common ground interference, and the traditional CAN bus has a bit error rate of >10 at a rate of 5Mbps. -5, and the communication delay is >5ms, which cannot meet the real-time battery voltage acquisition and balancing control requirements. Based on this, the isolated CAN FD interface circuit supports the ISO 11898-2 standard protocol and achieves a continuous isolation withstand voltage of 3.75kV rms / 1min at a transmission rate of 5Mbps; each frame of data in the communication protocol contains 16 channel voltage values, temperature status words and CRC-32 check codes, and a priority dynamic allocation mechanism is used to ensure that the transmission delay of key data is ≤2ms; when it is detected that the communication bit error rate exceeds the threshold, it automatically switches to the frequency reduction mode and starts the real-time data retransmission strategy. It is worth mentioning that the present invention achieves 3.75kVrms isolation withstand voltage through the transceiver circuit adopted; each frame of data is embedded with a CRC-32 check code, and the bit error rate is >10 -6 Real-time retransmission is initiated immediately; a dynamic priority allocation mechanism ensures that critical data latency is ≤2ms‌; communication availability can be increased to 99.999%, compared to 97.5% with traditional solutions; 3.75kV isolation withstand voltage meets the ISO 6469-3 electric vehicle safety standard; and a data retransmission mechanism increases effective bandwidth utilization by 40%.

[0041] For example, traditional fixed-threshold voltage divider adjustment strategies can easily experience prediction errors exceeding 15% due to fluctuations in the common-mode voltage change rate caused by battery aging. Based on this, the proposed ARIMA prediction model uses historical common-mode voltage data with a basic prediction step size of 100ms to establish a (p, d, q) = (2, 1, 1) prediction model with a prediction error band of ±3%. When the predicted common-mode voltage change within the next 200ms exceeds 15% of the op amp's withstand voltage, a voltage divider ratio adjustment instruction is triggered 10ms in advance. The model also incorporates a built-in online learning mechanism that automatically updates model parameters every 24 hours, using a sliding average method to maintain prediction performance during model updates. This solution, by establishing an ARIMA (2, 1, 1) prediction model with a 100ms step size to predict voltage trends 200ms into the future, addresses both the common-mode voltage change rate fluctuations caused by battery aging and the prediction errors associated with traditional fixed-threshold voltage divider adjustment strategies. It's worth noting that this solution uses an ARIMA (2,1,1) prediction model to predict voltage trends 200ms into the future with a 100ms step size. An online learning mechanism updates model parameters every 24 hours. A sliding average method maintains a ±3% error band during model updates, increasing prediction accuracy to 97%. Adjusting instructions 10ms in advance increases the voltage divider network's response margin by 50%. The parameter update mechanism enables the model to adapt to battery aging cycles three times faster. For example, traditional multi-channel battery voltage acquisition suffers from clock deviation (>50ns) and temperature drift (>20μV / °C), resulting in SOC estimation errors >3%. The 16-channel synchronous acquisition system has a sampling clock deviation of ≤10ns between channels within the operating temperature range of -40°C to +85°C, and an integral nonlinearity error of ≤±2LSB over the full range. After power-on, the system self-calibrates the reference source, with a calibration time of <500ms and a zero-point drift of <±5μV / °C after calibration. It supports independent refresh rate settings for each channel, up to a configurable 1kHz continuous sampling mode, and the data output refresh rate adaptively matches the communication interface bandwidth. It is worth noting that this solution generates a global synchronization signal through the MCU, with clock deviation of ≤10ns between channels. The internal reference source self-calibrates to achieve zero-point drift of <±5μV / °C. The refresh rate adaptively matches the communication bandwidth, supporting up to 1kHz continuous sampling. It can achieve micro-level synchronization accuracy, improving multi-channel sampling time consistency by 5 times. The 500ms fast calibration improves efficiency by 75% compared to traditional solutions. The adaptive mechanism increases effective data throughput by 30%. A voltage acquisition method, applied to a battery string voltage acquisition system with a calibration system as described in any one of the above, comprising: S1: The voltage divider ratio is adjusted in real time through a dynamic voltage divider network. The voltage protection threshold at the input of the operational amplifier is adaptively adjusted based on the instantaneous value of the common-mode voltage and its rate of change. When a transient overvoltage is detected, the resettable fuse and voltage clamp protection are synchronously triggered. S2: Two-stage filtering is applied to the collected signal. The front-stage filter suppresses low-frequency power frequency interference, and the rear-stage filter attenuates high-frequency switching noise. A star grounding layout is used to reduce crosstalk between channels. S3: A multi-channel synchronous acquisition circuit is used to acquire the voltage signal of each battery cell. The inter-channel coupling error is modeled using the error transfer matrix. The maximum error source is located through singular value decomposition and dynamic calibration compensation is performed. S4: Calculate the second-order temperature compensation coefficient in real time according to the ambient temperature, and perform closed-loop correction on the reference voltage of the analog-to-digital converter in combination with the pre-calibrated gain nonlinear curve; S5: Synchronize the sampling timing of each acquisition channel based on the global clock signal. After completing the preset number of data accumulations, output the battery voltage data packet that meets the preset bit error rate requirements through the isolated communication interface.

[0042] Unless otherwise specified, the device components involved in the above embodiments are all conventional device components, and the connection methods and control methods involved are all conventional connection methods and control methods unless otherwise specified.

[0043] The present invention has been described in detail above with reference to the embodiments. However, those skilled in the art will appreciate that, without departing from the spirit of the present invention, the specific parameters in the above embodiments may be modified to form multiple specific embodiments, which are all within the common variation range of the present invention and will not be described in detail here.

Claims

1. A battery string voltage acquisition system with a calibration system, comprising multiple voltage input terminals, a voltage signal conditioning terminal, a voltage follower, an analog-to-digital converter, a main control center, and a communication circuit, characterized in that: The voltage signal regulating terminal includes a dynamic voltage divider network composed of a digital potentiometer and a high-voltage operational amplifier; and is used to dynamically adjust the voltage divider ratio to suppress the common-mode voltage. The main control center integrates a multi-channel synchronous sampling and calibration module, and generates a global synchronization signal through the main MCU to control ADC synchronous acquisition; A temperature compensation module is provided between the voltage follower and the ADC, and a digital temperature sensor is used to correct the gain drift of the operational amplifier in real time; The voltage division ratio of the dynamic voltage division network is dynamically adjusted according to the real-time common mode voltage and a preset safety threshold.

2. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: The voltage division ratio of the dynamic voltage division network satisfies: ； in, is the real common mode voltage, is the maximum withstand voltage of the op amp, is the dynamic adjustment coefficient, and its value range is .

3. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: The synchronous sampling calibration module adopts the cross-redundancy calibration method to construct the error transfer matrix, which is: ； in, For the and Channel coupling error, is the kth sampling value of the i-th channel; For the Channel noise standard deviation.

4. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: The gain correction formula of the temperature compensation module is: ； in, The gain is calibrated at 25°C. is the first-order temperature coefficient, is the second-order temperature coefficient.

5. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: In the TVS diode and resettable fuse parallel protection circuit of the dynamic voltage divider network, the breakdown voltage of the TVS diode is set to 120%-130% of the system's maximum common-mode voltage, and the response time is ≤1ns; the resettable fuse uses a high-molecular polymer positive temperature coefficient material, triggers a circuit break when the current exceeds 150% of the rated value, and automatically recovers within 60s after the fault is eliminated. The rated current value is set to 10mA±5% based on the maximum operating current of the op amp.

6. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: The analog-to-digital converter uses a 16-bit SAR ADC with an input impedance of ≥1MΩ, achieving a nonlinear error of ±0.0015% within the input range of -15V to +15V; the built-in overvoltage protection circuit can automatically cut off the sampling channel when the input exceeds ±18V and switch to the backup channel within 10μs; combined with an anti-aliasing filter, a signal-to-noise ratio of more than 80dB is achieved at a sampling rate of 1MSPS, where the cutoff frequency of the anti-aliasing filter is set to 1 / 5 of the ADC sampling frequency and the attenuation slope is ≥40dB / decade.

7. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: The isolated CAN FD interface circuit supports the ISO 11898-2 standard protocol and achieves a continuous isolation withstand voltage of 3.75kV rms / 1min at a transmission rate of 5Mbps. Each frame of data in the communication protocol contains 16-channel voltage values, a temperature status word, and a CRC-32 checksum. A dynamic priority allocation mechanism is used to ensure that the transmission delay of critical data is ≤2ms. When the communication bit error rate is detected to exceed the threshold, it automatically switches to a frequency reduction mode and initiates a real-time data retransmission strategy.

8. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: The ARIMA prediction model uses a basic prediction step of 100 ms and establishes a prediction model with (p, d, q) = (2, 1, 1) based on historical common-mode voltage data. The prediction error band is set to ±3%. When the common-mode voltage change amplitude is predicted to exceed 15% of the op amp's withstand voltage within the next 200 ms, the voltage divider ratio adjustment instruction is triggered 10 ms in advance. The model has a built-in online learning mechanism that automatically updates the model parameters every 24 hours. During the model update, the sliding average method is used to maintain the prediction function.

9. The battery string voltage acquisition system with a calibration system according to claim 1, characterized in that: The 16-channel synchronous acquisition system has an operating temperature range of -40°C to +85°C, with a sampling clock deviation of ≤10ns between channels and an integral nonlinearity error of ≤±2LSB over the full range. After powering on, the system self-calibrates the reference source, with a calibration time of <500ms and a zero-point drift of <±5μV / °C after calibration. The system supports independent refresh rate setting for each channel, with a maximum configurable continuous sampling mode of 1kHz, and the data output refresh rate adaptively matches the communication interface bandwidth.

10. A voltage acquisition method, applied to a battery string voltage acquisition system with a calibration system according to any one of claims 1 to 9, characterized in that: include: S1: The voltage divider ratio is adjusted in real time through a dynamic voltage divider network. The voltage protection threshold at the input of the operational amplifier is adaptively adjusted based on the instantaneous value of the common-mode voltage and its rate of change. When a transient overvoltage is detected, the resettable fuse and voltage clamp protection are synchronously triggered. S2: Two-stage filtering is applied to the collected signal. The front-stage filter suppresses low-frequency power frequency interference, and the rear-stage filter attenuates high-frequency switching noise. A star grounding layout is used to reduce crosstalk between channels. S3: A multi-channel synchronous acquisition circuit is used to acquire the voltage signal of each battery cell. The inter-channel coupling error is modeled using the error transfer matrix. The maximum error source is located through singular value decomposition and dynamic calibration compensation is performed. S4: Calculate the second-order temperature compensation coefficient in real time according to the ambient temperature, and perform closed-loop correction on the reference voltage of the analog-to-digital converter in combination with the pre-calibrated gain nonlinear curve; S5: Synchronize the sampling timing of each acquisition channel based on the global clock signal. After completing the preset number of data accumulations, output the battery voltage data packet that meets the preset bit error rate requirements through the isolated communication interface.

Images