Power supply current detection method based on BMS (Battery Management System)

Abstract

The invention relates to a power supply current detection method based on a BMS system, and relates to the technical field of battery management systems, current sensors are configured at the positive electrode input end and the negative electrode output end of a battery respectively to construct hardware redundancy, and the method comprises the following steps that the reading Ipos of the positive electrode current sensor and the reading Ineg of the negative electrode current sensor are collected in real time and preprocessed; respectively carrying out validity judgment on the preprocessed Ipos and Ineg and a preset normal working current range [Imin, Imax]; according to the validity judgment result, multi-mode arbitration is carried out based on a fault-safety principle, and one of a single-sensor failure mode, a double-sensor consistent mode, a double-sensor conflict mode and a double-sensor failure mode is automatically selected; and outputting the final system current value Ifinal for SOC estimation, thermal management and / or power control of the BMS. According to the method, the problem that the system is out of control possibly caused by failure of a single-point sensor or data conflict of double sensors in the existing BMS current detection technology is solved.

Description

Technical Field

[0001] This invention relates to the field of battery management system technology, and more specifically to a current detection method for battery management systems used in electric vehicles, energy storage systems and other fields, and in particular a current signal arbitration method based on redundant current sensors and the fail-safe principle. Background Technology

[0002] The battery management system (BMS) is the core control unit of the battery pack. It is responsible for monitoring the battery status (such as voltage, current, and temperature) and implementing protection strategies accordingly to prevent overcharging, over-discharging, and overheating, thereby ensuring battery safety and extending battery life. Current detection is fundamental to the BMS's state estimation (such as SOC and SOH), thermal management, and power control; its accuracy directly affects the safety and performance of the entire system.

[0003] Traditional battery management systems (BMS) typically rely on a single high-precision current sensor (such as a Hall sensor or shunt) for detection. This single-point detection scheme inherently carries a safety hazard: if the sensor malfunctions (e.g., drift, damage, signal harness interruption, or electromagnetic interference), the BMS will acquire incorrect current information. If the incorrect current value is lower than the true value, the BMS may fail to take protective measures when the battery is actually overcharged or over-discharged, potentially leading to catastrophic consequences such as thermal runaway, fire, or explosion. This is a typical "single-point failure" pattern leading to system-wide failure.

[0004] To address this issue, existing technologies employ redundant current sensor designs, such as installing one sensor in each of the positive and negative circuits of the battery pack. However, simply adding hardware does not completely solve the problem; the key lies in how the system arbitrates when the two sensor readings are inconsistent. Simply averaging or randomly selecting one may mask the true fault under the most dangerous operating conditions (such as high-current charging and discharging), failing to fundamentally improve safety. Therefore, there is an urgent need for a current sensing arbitration method that can be embedded in BMS software and is based on rigorous safety logic to ensure that the system's decisions lead to the safest state under any sensor failure mode. Summary of the Invention

[0005] The purpose of this invention is to overcome the system runaway problem that may be caused by single-point sensor failure or data conflict between dual sensors in existing BMS current detection technology, and to provide a current detection method based on the fail-safe principle. This method not only identifies sensor faults, but more importantly, when the true current value cannot be determined, it can force the system into a conservative, restrictive protection state through preset safety arbitration logic, thereby absolutely avoiding the two most dangerous situations of battery overcharging and over-discharging.

[0006] To achieve the above objectives, the present invention provides a power current detection method based on a BMS system, which configures current sensors at the positive input terminal and negative output terminal of the battery to build hardware redundancy, including the following steps:

[0007] The readings I_pos from the positive current sensor and I_neg from the negative current sensor are acquired in real time and preprocessed. The preprocessed I_pos and I_neg are compared with the preset normal operating current range [I_min, I_max] to determine their validity. Based on the validity judgment result, multi-mode arbitration is performed based on the fail-safe principle, and one of the following is automatically selected: single sensor failure mode, dual sensor consistency mode, dual sensor conflict mode, or dual sensor failure mode. The final system current value I_final is output for use by the BMS for SOC estimation, thermal management and / or power control.

[0008] Preferably, the two current sensors are installed at the positive input terminal and the negative output terminal of the battery pack, respectively, forming a full-loop redundant detection architecture.

[0009] Preferably, the preprocessing includes at least one of moving average filtering, first-order low-pass digital filtering, and rationality verification based on the battery dynamic model, wherein the rationality verification combines battery voltage, temperature, and load state information for joint judgment.

[0010] Preferably, the method for determining the normal operating current range [I_min, I_max] includes: Based on the maximum allowable current capability of the battery cell, and combined with the maximum charge and discharge current that the battery system may encounter in actual application scenarios, [I_min, I_max] are set; In the actual application scenario, the maximum charging and discharging current is determined by the output / input characteristics of the external power supply or load, and [I_min, I_max] does not exceed the upper limit of the current sensor's range.

[0011] Preferably, the consistency threshold δ used in multi-mode arbitration is calculated based on the full scale and accuracy class of the two sensors, and can be fine-tuned according to system noise and dynamic characteristics. Its basic value satisfies δ=√[(FS×Accuracy_A)²+(FS×Accuracy_B)²], where FS is the full scale of the sensor.

[0012] Preferably, in the single sensor failure mode, when only one of the readings of I_pos and I_neg falls within the range of [I_min, I_max], the reading of the normal sensor is directly used as I_final, and the corresponding sensor fault code is recorded.

[0013] Preferably, in the dual-sensor consistency mode, when both sensor readings are valid and |I_pos-I_neg| is less than the consistency threshold δ, I_final is the arithmetic mean of I_pos and I_neg.

[0014] Preferably, in dual-sensor conflict mode, when both sensor readings are valid but |I_pos-I_neg| is greater than or equal to δ: Calculate the temporary current value I_temp = (I_pos + I_neg) / 2; If I_temp > 0, it is determined to be a discharge state, then I_final = max(I_pos, I_neg); If I_temp < 0, it is determined to be in a charging state, then I_final = min(I_pos, I_neg).

[0015] Preferably, in the dual-sensor failure mode, when both I_pos and I_neg are abnormal values: The scene recognition mechanism determines whether the current scene is a driving scene or a charging scene. In the driving scenario, I_final uses a discharge current value that is sufficient to trigger the BMS discharge power limit, such as the maximum discharge current value allowed by the cell or system. In the charging scenario, I_final is set to 0A and the charging stop mechanism is triggered.

[0016] Preferably, it also includes a fault diagnosis and recording step: when entering a single sensor failure, dual sensor conflict, or dual sensor failure mode, the corresponding fault code, conflict event, and scene information are recorded in the non-volatile memory of the BMS.

[0017] Compared with the prior art, the present invention has the following beneficial technical effects: 1. Fundamentally Enhanced System Safety: This invention deeply integrates the "fail-safe" design concept into the core of the current detection algorithm. In uncertain scenarios such as sensor reading conflicts or double failures, the algorithm does not pursue data accuracy, but prioritizes the safest behavioral decision (limiting power / stopping charging and discharging), fundamentally eliminating thermal runaway causes such as battery overcharging and over-discharging due to current detection errors.

[0018] 2. High Robustness and Reliability: Through a dual-redundant sensor design with positive and negative poles and a multi-mode arbitration mechanism, the system is fault-tolerant to single-point sensor failures, ensuring the continuity of system function under most failure conditions. Simultaneously, in complex failures involving dual-sensor conflicts, there are clear and safe handling strategies, greatly enhancing the reliability of the BMS system in complex electromagnetic environments and long-term use.

[0019] 3. Proactive Fault Warning: This method can not only handle hard faults (signal anomalies) but also effectively identify soft faults (signal conflicts), which are often precursors to sensor performance degradation or abnormal internal connection impedance. By recording and analyzing conflict events, it can provide data support for preventative maintenance, eliminating potential safety hazards in their infancy.

[0020] 4. Good engineering feasibility: This invention is mainly a software algorithm, which does not require changes to the BMS hardware platform. It can be implemented by simply upgrading the software on the basis of existing redundant sensor hardware. It is low cost and easy to promote and apply in existing products. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the overall process of a power supply current detection method based on a BMS system according to the present invention. Figure 2 This is a detailed flowchart of the safety arbitration logic in Mode 3 (dual sensor conflict mode) of the power supply current detection method based on BMS system of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] This invention discloses a power current detection method based on a BMS system. This method constructs a hardware redundancy foundation by configuring two current sensors installed at the positive input terminal and negative output terminal of the battery in a traditional battery management system, and embeds a complete software arbitration logic on this basis, fundamentally solving the system-level safety risks that may be caused by the failure of a single current sensor or data conflict between two sensors.

[0024] The execution body of this method is the BMS main control chip (MCU). Key parameters need to be preset when the system is powered on and initialized, including the normal operating current range [I_min, I_max] determined according to the cell specifications and sensor range constraints, the consistency threshold δ calculated based on sensor accuracy and system noise level, and the default substitution value I_default set for different application scenarios. These parameters provide a benchmark framework for subsequent signal processing and fault arbitration.

[0025] The method of the present invention includes the following steps: Step S1, Signal Acquisition and Preprocessing The system acquires real-time readings from the positive current sensor (I_pos) and the negative current sensor (I_neg), and performs multi-dimensional preprocessing on these raw signals to eliminate high-frequency noise, spikes, and other interference. Preprocessing methods include, but are not limited to: Moving average filtering effectively smooths random noise and has a low computational burden by maintaining a queue of length N to store the most recent N sampled values ​​and outputting their arithmetic mean (e.g., N=10). First-order low-pass digital filtering, using the formula y(k)=α x(k)+(1 α) y(k 1) Filtering is performed to suppress high-frequency noise with low memory usage, where x(k) is the current sampled value, and y(k) and y(k) are... 1) represents the current and previous filtered output values, and α is an adjustable filter coefficient (0 < α < 1). When combined with a moving average filter, it can maximize signal purity while preserving the dynamic characteristics of the current.

[0026] Based on the rationality verification of the battery dynamic model, combined with real-time acquired battery voltage, temperature and load status information (such as motor torque and charger handshake signal) for joint judgment, transient current values ​​that obviously exceed physical possibilities are limited or marked, thereby preventing abnormal jump signals from interfering with subsequent judgments.

[0027] Step S2, preliminary judgment of signal validity After preprocessing, I_pos and I_neg are compared with the preset normal operating current range [I_min, I_max] for preliminary validity assessment. The accurate determination of this range directly affects the accuracy of fault identification. The method for determining the normal operating current range is as follows: The determination of the lower limit I_min mainly considers the maximum continuous charging current of the battery at the highest permissible charging temperature. Its value is the maximum permissible charging current of the battery at the highest permissible charging temperature, conservatively set by multiplying the value defined in the cell specifications by 1.1 times the safety margin. It is usually a negative value, such as -1C, i.e., Imin = 1 (Maximum continuous charging current of the battery cell) (Safety margin). For example, if the maximum continuous charging rate is 0.5C for a 100Ah battery cell, then I_min should be set to -55A.

[0028] The upper limit I_max is determined by calculating the maximum continuous discharge current of the battery at the minimum allowable discharge temperature, plus a safety margin. According to the cell specification sheet, it is usually a positive value, i.e., I_max = maximum continuous discharge current of the cell. Safety margin. For example, if the maximum continuous discharge rate of a 100Ah cell is 3C, then I_max can be set to 330A.

[0029] Sensor range constraints: the final determined [I_min, I_max] must strictly fall within the range of the two current sensors. For example, when the sensor range is ±500A, the above calculation result is the valid feasible interval. This dual constraint mechanism ensures the physical rationality of the validity judgment.

[0030] Step S3, Multi-mode arbitration based on the fault-safe principle Based on the above validity judgment results, the core multi-mode arbitration stage based on the fault-safe principle is entered. This stage automatically selects one of four working modes according to the sensor state combination, and finally outputs a stable and reliable system current value I_final for subsequent functions such as SOC estimation, thermal management and power control of BMS.

[0031] Mode 1 (Single Sensor Failure Mode): When only one of I_pos and I_neg readings falls within the normal range while the other is abnormal, the system enters single sensor failure mode. In this case, the reading of the normal sensor is directly adopted as I_final. This design ensures that the system can still maintain basic functions and continuously provide protection when a single sensor drifts, is damaged, or experiences signal interruption. This avoids the risk of complete loss of control under traditional single sensor solutions and achieves preliminary fault tolerance capability with hardware redundancy.

[0032] Mode 2 (Dual Sensor Consistency Mode): When both sensor readings are valid and the absolute value of their difference, |I_pos - I_neg|, is less than a preset consistency threshold δ (the threshold selection needs to be adjusted according to battery characteristics and the supporting environment), the system enters dual sensor consistency mode. This mode represents that the system is in a healthy operating state. At this time, the two sensor readings are considered reliable and consistent. I_final is taken as the arithmetic mean of I_pos and I_neg. The most accurate current measurement value is obtained through sensor data fusion technology, significantly improving SOC estimation accuracy and power control finesse.

[0033] The consistency threshold δ is set as follows: The baseline value is calculated based on the sensor accuracy. The value of δ should be greater than the maximum combined error of the two sensors at full scale. The formula δ=√[(FS×Accuracy_A)²+(FS×Accuracy_B)²] is used, where FS is the sensor's full scale, and Accuracy_A and Accuracy_B are the accuracy classes of the two sensors, respectively. For example, for a sensor with a range of 500A and an accuracy of ±1%, δ≈7.07A is calculated.

[0034] Based on adjustments to system noise and dynamic characteristics, the theoretical values ​​calculated with high precision can be fine-tuned according to the actual system noise and dynamic characteristics. In applications with complex electromagnetic environments or high vibration, the threshold can be appropriately relaxed (e.g., set to 10A) to prevent false judgments. In scenarios with extremely high requirements for dynamic response, the threshold can be appropriately tightened while ensuring safety. This flexible setting mechanism balances the contradiction between false judgment rate and response speed.

[0035] Mode 3 (Dual Sensor Conflict Mode): When both sensor readings I_pos and I_neg are valid, but the absolute value of the difference |I_pos-I_neg| is greater than or equal to δ, it indicates that the two reliable sensors have conflicting readings. The system determines that it has entered dual sensor conflict mode, indicating that a soft fault or data drift has occurred. At this time, a safety arbitration logic based on the fail-safe principle is initiated, rather than simply relying on confidence level or taking the average value.

[0036] First, calculate the temporary current value I_temp = (I_pos + I_neg) / 2. The battery operating state is determined by judging the sign of I_temp. If I_temp > 0, it is determined to be in a discharging state. In this case, the worst-case scenario is that the system underestimates the actual discharge current, leading to over-discharge or power exceeding limits. Therefore, the arbitration strategy is I_final = max(I_pos, I_neg), which uses a larger current value to make the BMS tend to overestimate the discharge power, forcing the BMS to trigger the discharge power limit or low-voltage protection earlier, ensuring the battery is not damaged by over-discharge. If I_temp < 0, it is determined to be in a charging state. In this case, the worst-case scenario is that the system underestimates the absolute value of the charging current, leading to overcharging, or overestimates the absolute value of the charging current. Therefore, the arbitration strategy is I_final = min(I_pos, I_neg). Since the charging current is negative, the min function actually selects a negative current with a larger absolute value, making the BMS tend to overestimate the charging intensity and trigger overcurrent protection or SOC upper limit protection earlier, absolutely avoiding the risk of overcharging. This design, which dynamically adjusts the arbitration rules based on the direction of the current, ensures that decisions are always directed towards the safest state in conflict scenarios, fundamentally eliminating the risk of thermal runaway caused by data uncertainty.

[0037] Mode 4 (Dual Sensor Failure Mode): In more extreme cases, when both I_pos and I_neg are abnormal values, the system enters dual sensor failure mode. In this mode, no valid current information can be obtained, and the system must rely on the preset default replacement value I_default as I_final, combined with a scene recognition mechanism to implement the highest level of safety protection. The setting of I_default follows safety principles, and its scene recognition and value retrieval logic is as follows: Scene recognition determines the vehicle driving scenario by detecting the Ready signal from the vehicle controller (VCU), the torque request from the motor controller, or the gear position signal from the transmission (such as D / R gear). When these signals are detected as valid, the vehicle is determined to be in a driving or ready-to-drive state. The charging scenario is determined by detecting the charging connection confirmation signal CC2, CP, or CAN communication with the charger. Specifically, it is determined by detecting the charging connection confirmation signal (such as the CC2 signal for standard DC charging and the CP signal for AC charging) or establishing communication with the charger (such as the charger message on the CAN bus). When a valid charging connection signal is detected, the vehicle is determined to be in a charging state.

[0038] In vehicle driving scenarios, I_default is set to a discharge current value sufficient to immediately trigger the BMS discharge power limit. Specifically, it is a dynamic value (e.g., 120%) that is greater than or equal to a certain percentage of the maximum allowable continuous discharge current under the current battery conditions (considering SOC and temperature). For example, when the system calculates the instantaneous maximum allowable discharge current to be 250A, I_default is set to 300A to ensure that the BMS protection logic is forcibly activated and enters a safe power-limiting state. In charging scenarios, the safest strategy is to stop charging, prioritizing setting I_default to 0A. Simultaneously, the BMS sends an immediate stop-charging request to the charger via communication messages (e.g., CAN signals) and actively disconnects the charging relay in the high-voltage circuit, fundamentally eliminating the possibility of overcharging. When this mode is triggered, a "current sensing system failure" fault alarm is immediately reported via the CAN bus or other communication interfaces, enabling rapid fault notification.

[0039] Step S4, Fault Diagnosis and Recording Throughout the arbitration process, once a single-sensor failure, dual-sensor conflict, or dual-sensor failure mode is entered, the corresponding sensor fault code or conflict event is recorded. Particularly for dual-sensor conflict modes, the number of safety arbitration triggers and specific scenario information are also recorded. This diagnostic data is stored in the BMS's non-volatile memory, providing crucial information for subsequent preventative maintenance. This allows maintenance personnel to identify early signs of sensor performance degradation and replace sensors before complete failure, transforming traditional reactive maintenance into proactive maintenance and further improving the system's long-term reliability. This fault recording function is deeply integrated with the arbitration logic, forming a complete closed loop from fault detection to safety decision-making and historical tracing.

[0040] During system initialization, the specific values ​​of key parameters need to be fixed in the BMS main control chip. These parameters need to be customized according to the battery pack characteristics and application scenarios. The normal operating current range needs to be accurately calculated based on the cell capacity, maximum continuous charge / discharge rate, and operating temperature boundary, with an added safety margin. For example, for a 100Ah cell with a maximum continuous charge rate of 0.5C and a maximum continuous discharge rate of 3C, it can be set to [-55A, 330A]. The consistency threshold δ is set according to the sensor accuracy and system noise level, usually set to 0.5%~1% of the sensor range, with a typical value of 10A. The default replacement value I_default is often set to a discharge current value that can trigger power limiting, such as 300A (discharge direction), in electric vehicle applications. The reasonable configuration of these parameters is a prerequisite for the effective operation of the algorithm. It is recommended to verify their rationality through experiments during the system calibration phase.

[0041] To more clearly illustrate the practical application effects of this invention, the following scenario is illustrated using a battery management system (BMS) for a pure electric vehicle with a rated voltage of 400V and a capacity of 100Ah. A Hall current sensor with a range of ±500A is installed at each of the positive and negative terminals of the battery pack, with preset parameters I_min=-55A, I_max=330A, δ=10A, and I_default=300A.

[0042] In scenario 1 (mode 2), during constant speed cruising, I_pos = 85.2A and I_neg = 85.0A. Both values ​​are within the valid range and the difference of 0.2A is less than δ. The system determines that it is in dual-sensor consistency mode and outputs the average value I_final = 85.1A. At this time, the BMS can accurately calculate the SOC and optimize the power distribution based on this.

[0043] In Scenario 2 (Mode 1), during vehicle startup, I_pos=150.5A is normal, while I_neg=450A exceeds the range and is abnormal. The system enters single sensor failure mode, adopts I_pos as I_final, and records the negative electrode sensor over-range fault code, ensuring continuous monitoring capability during startup.

[0044] Scenario 3 (Mode 3 - Discharge State): In a scenario of rapid vehicle acceleration, both I_pos=280A and I_neg=260A are normal, but the difference of 20A is greater than δ. The system calculates I_temp=270A>0 and determines that it is in a discharge state. It enters the conflict mode and outputs I_final=max(280,260)=280A. This conservative estimate may have approached or reached the peak discharge current allowed by the battery, triggering the peak power limit of the BMS. This effectively avoids potential damage from the actual high current of 280A. Even if the actual current is 260A, the system only performs a conservative, non-destructive, and safe power limit.

[0045] Scenario 4 (Mode 3 - Charging State): In a DC fast charging scenario, both I_pos=-96A and I_neg=-108A are normal, but the difference of 12A is greater than δ. The system calculates I_temp=-102A<0 and determines that it is in a charging state, outputting I_final=min(-96,-108)=-108A. This negative current with a large absolute value may have exceeded the maximum allowable charging current of the battery cell at the current SOC and temperature, prompting the BMS to intervene in advance to reduce the charger's output, triggering the charging current limit or communicating with the charger to reduce the output, thus absolutely avoiding the risk of overcharging. Even if the actual current is -96A, it only reduces the charging speed and ensures safety.

[0046] Scenario 5 (Mode 4): In an extreme scenario involving a vehicle collision, the power lines of both sensors short-circuit simultaneously. The readings of the two sensors, 510A and -505A respectively, are abnormally outside the normal range. The system determines this as a dual-sensor failure mode and immediately uses I_default (e.g., set to 300A discharge current) as I_final. Upon receiving this signal, the BMS, in conjunction with the collision signal, commands the contactor to disconnect the high-voltage circuit and issues a serious alarm to the driver, achieving rapid and safe isolation in the event of a catastrophic failure.

[0047] These five typical scenarios fully cover the entire spectrum of operating conditions, from normal operation to multiple failures, and fully verify the safety and reliability of the invention under various boundary conditions.

[0048] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A power supply current detection method based on a BMS system, characterized in that, To create hardware redundancy, current sensors are configured at both the positive input and negative output terminals of the battery, including the following steps: Real-time acquisition of positive current sensor readings I_pos and negative current sensor readings I_neg, followed by preprocessing; The preprocessed I_pos and I_neg are compared with the preset normal operating current range [I_min, I_max] to determine their validity. Based on the validity judgment result, multi-mode arbitration is performed based on the fail-safe principle, and one of the following is automatically selected: single sensor failure mode, dual sensor consistency mode, dual sensor conflict mode, or dual sensor failure mode. The final system current value I_final is output for use by the BMS for SOC estimation, thermal management and / or power control.

2. The power supply current detection method based on a BMS system according to claim 1, characterized in that, Two current sensors are installed at the positive input terminal and the negative output terminal of the battery pack, respectively, forming a full-loop redundant detection architecture.

3. The power supply current detection method based on a BMS system according to claim 2, characterized in that, The preprocessing includes at least one of moving average filtering, first-order low-pass digital filtering, and rationality verification based on the battery dynamic model, wherein the rationality verification combines battery voltage, temperature, and load state information for joint judgment.

4. The power supply current detection method based on a BMS system according to claim 3, characterized in that, The methods for determining the normal operating current range [I_min, I_max] include: The lower limit I_min is determined by multiplying the maximum continuous charging current of the battery cell by 1.1 times the safety margin; The upper limit I_max is determined by multiplying the maximum continuous discharge current of the battery cell by 1.1 times the safety margin; The final range is strictly limited to the measurement range of the two current sensors.

5. The power supply current detection method based on a BMS system according to claim 4, characterized in that, The consistency threshold δ used in multi-mode arbitration is calculated based on the full scale and accuracy class of the two sensors and can be fine-tuned according to system noise and dynamic characteristics. Its basic value satisfies δ=√[(FS×Accuracy_A)²+(FS×Accuracy_B)²], where FS is the full scale of the sensor.

6. The power supply current detection method based on a BMS system according to claim 5, characterized in that, In single-sensor failure mode, when only one of the readings of I_pos and I_neg falls within the range of [I_min, I_max], the reading of the normal sensor is directly used as I_final, and the corresponding sensor fault code is recorded.

7. The power supply current detection method based on a BMS system according to claim 6, characterized in that, In dual-sensor consistency mode, when both sensor readings are valid and |I_pos-I_neg| is less than the consistency threshold δ, I_final takes the arithmetic mean of I_pos and I_neg.

8. The power supply current detection method based on a BMS system according to claim 7, characterized in that, In dual-sensor conflict mode, when both sensor readings are valid but |I_pos-I_neg| is greater than or equal to δ: Calculate the temporary current value I_temp = (I_pos + I_neg) / 2; If I_temp > 0, it is determined to be a discharge state, then I_final = max(I_pos, I_neg); If I_temp < 0, it is determined to be in a charging state, then I_final = min(I_pos, I_neg).

9. The power supply current detection method based on a BMS system according to claim 8, characterized in that, In dual-sensor failure mode, when both I_pos and I_neg are outliers: The scene recognition mechanism determines whether the current scene is a driving scene or a charging scene. In the driving scenario, I_final uses a discharge current value sufficient to trigger the BMS discharge power limit; In the charging scenario, I_final is set to 0A and the charging stop mechanism is triggered.

10. A power supply current detection method based on a BMS system according to any one of claims 1-9, characterized in that, It also includes fault diagnosis and recording steps: when entering a single sensor failure, dual sensor conflict, or dual sensor failure mode, the corresponding fault code, conflict event, and scene information are recorded in the BMS's non-volatile memory.

Images