A vehicle-mounted low-voltage battery management system supporting intelligent power compensation and fault diagnosis

Abstract

This invention relates to the field of low-voltage power supply technology and discloses an on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis. The system includes a state detection module, a charging control module, a parameter calculation module, a fault determination module, and a power replenishment handling module. When power replenishment is needed and the load is stable, a backflow current step is executed, causing the power replenishment path current to sequentially reach a first charging current target, a second charging current target, and a third charging current target, forming a first reference interval, an intermediate interval, and a second reference interval, and obtaining a data validity flag. If the data validity flag is valid, identification parameters are calculated; otherwise, the identification parameters are set to invalid, fault candidates are generated and verified, and the power replenishment path current is limited according to the allowable charging current and written into the data storage. This invention, by forming three intervals through a backflow current step and obtaining a data validity flag, achieves the calculation of path resistance, instantaneous internal resistance, and polarization voltage only when the step data is valid.

Description

Technical Field

[0001] This invention relates to the field of low-voltage power supply technology, specifically to an on-board low-voltage battery management system that supports intelligent power replenishment and fault diagnosis. Background Technology

[0002] When the vehicle is parked, locked, or in sleep mode, the battery management controller periodically wakes up to detect the low-voltage battery terminal voltage, state of charge (SCC), temperature, and quiescent current. When the terminal voltage or SCC falls below a threshold, the high-voltage system and the high-low voltage DC-DC converter are powered on to replenish the low-voltage battery. During sleep mode, the quiescent current can be monitored using a battery current sensor; if it exceeds a set range, it is determined that the vehicle is not in sleep mode, has been abnormally woken up by the network, or is continuously consuming power from electrical appliances. Existing methods can also apply current pulses to the battery, calculate the internal resistance based on the ratio of voltage change to current change, or evaluate the health status based on cumulative charge, voltage change, and usage records; and collect data on the voltage of the replenishment power supply side, output current, low-voltage battery voltage, and the status of electrical appliances, and determine whether there is a wiring harness open circuit, loose connection, or insufficient power supply based on node voltage difference and communication status.

[0003] Parking-based charging is initiated when the terminal voltage or state of charge exceeds its limit. The charging path current does not change back according to the known charging current target; the low-voltage load changes accordingly after the charging power supply is powered on. The existing process does not determine the validity of the data based on whether the load is stable, whether the current keeps up with the target, or the direction of the voltage response; instead, it uses the terminal voltage change for internal resistance or health status calculations.

[0004] When only the low-voltage battery terminal voltage and battery current are collected, the path voltage drop, transient overvoltage, and polarization response are superimposed in the same set of voltage changes. Existing internal resistance detection and fixed differential voltage diagnosis do not separate path parameters from the power supply side node and the low-voltage battery side node, nor do they calculate battery parameters in segments according to steady state, initial response, and continuous response.

[0005] Static current monitoring, path differential voltage diagnosis, and threshold-based charging are independent of each other. Diagnostic results are not used to limit the charging current for this or subsequent charging operations. If the fault cause is not eliminated, the system will continue to charge at the original threshold. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an on-board low-voltage battery management system that supports intelligent charging and fault diagnosis. It solves the problems in existing technologies where parking charging does not organize return excitation based on a known charging current target, nor does it first determine whether step data is valid, and voltage changes are difficult to use for calculating and identifying parameters.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] This invention provides an on-board low-voltage battery management system that supports intelligent charging and fault diagnosis, including a status detection module, a charging control module, a parameter calculation module, a fault determination module, and a charging handling module.

[0009] The status detection module reads the next detection time from the data storage and performs a power replenishment permission determination. If the power replenishment permission determination is passed, a power replenishment requirement determination is performed.

[0010] When the charging control module determines that charging is required, it obtains the current limited charging result and the load stability determination. If the load stability determination is passed and the current limited charging result is greater than 0, it reads two charging current targets from the data storage; if there are no two charging current targets, it determines two charging current targets based on the current limited charging result.

[0011] The two charging current targets are a first charging current target and a second charging current target, both of which are no greater than the current limited charging result. The charging control module executes a reversal current step, causing the charging path current to sequentially reach the first charging current target, the second charging current target, and the first charging current target, forming a first reference interval, an intermediate interval, and a second reference interval. Initial response data, continuous response data, and steady-state data for each interval are obtained. Then, based on the initial response data, continuous response data, steady-state data, and low-voltage load current, the load residual, current tracking residual, node current imbalance, and voltage response direction are obtained.

[0012] The data is valid if the load residual, current tracking residual, and node current imbalance are all no greater than the upper limit of the allowable range, and the voltage response direction is equal to the current change direction; otherwise, the data is invalid. The first and second reference intervals both execute the first charging current target, while the intermediate interval executes the second charging current target. The steady-state difference between the two reference intervals provides the unrecovered amount.

[0013] The parameter calculation module calculates identification parameters when the data validity flag is valid. Identification parameters include path resistance, instantaneous internal resistance, and polarization voltage. When the data validity flag is invalid, the identification parameters are set to invalid.

[0014] The fault determination module generates and verifies fault candidates based on the data validity flags, and obtains the fault verification results.

[0015] The power replenishment module obtains the allowable charging current based on the identification parameters. When the identification parameters are invalid, the current limited power replenishment result is used as the allowable charging current. The module confirms the fault based on the fault candidate and fault verification results, limits the power replenishment path current according to the allowable charging current, and writes the next detection time and allowable charging current into the data storage.

[0016] Furthermore, the permission determination unit reads the next detection time, measurement uncertainty, power supply restart permission, current number of consecutive power replenishments, and effective zero-point calibration results from the data storage to obtain the sampling channel self-test. If there is no power supply restart permission, the power replenishment permission is not deemed unsuccessful due to the lack of such permission.

[0017] If the power supply restart permission is not granted, or if the current consecutive power supply count reaches the preset limit for consecutive power supply counts, the power supply permission is deemed unsuccessful. When the power supply permission is deemed successful and the power supply remains off, the demand determination unit reads the charge state as the initial charge state and performs a power supply demand determination. If the power supply demand is determined to be unnecessary, the next detection time is obtained and written to the data storage; if the power supply demand is determined to be necessary, unconfirmed fault candidates are read.

[0018] Furthermore, when power replenishment is required, the load monitoring unit obtains the current limited power replenishment result, completes the load stability determination, and accumulates the amount of power replenished. If the difference between the maximum and minimum values ​​of the low-voltage load current within the preset continuous sampling window is not greater than the upper limit of the preset allowable range, the load stability determination is passed. If the time from starting the power replenishment supply until low-voltage load stabilization is not achieved is greater than the preset load stabilization waiting time, the load stability determination is failed, and the data validity flag is invalidated.

[0019] Preferably, when the data validity flag is valid, the path calculation unit calculates the path resistance using the path resistance method based on steady-state data, and obtains the path hysteresis and deviation sign. The battery calculation unit calculates the instantaneous internal resistance using the instantaneous internal resistance method based on the initial response data, and calculates the polarization voltage using the polarization recursive method based on the continuous response data, obtaining the polarization resistance and polarization time constant. The path resistance is taken from the node voltage drop and the charging path current under steady state, the instantaneous internal resistance is taken from the initial overvoltage after current switching, and the polarization voltage is taken from the overvoltage recursion during the continuous response period.

[0020] Furthermore, the candidate generation unit generates fault candidates based on data validity flags, path resistance, instantaneous internal resistance, polarization voltage, load residual, and current tracking residual.

[0021] Fault candidates include path anomaly candidates, battery body anomaly candidates, charging power supply anomaly candidates, low voltage load anomaly candidates, and sampling link anomaly candidates.

[0022] A path anomaly candidate is obtained when the path resistance or path hysteresis is outside the allowable range; a battery body anomaly candidate is obtained when the instantaneous internal resistance, polarization resistance, or polarization time constant is outside the allowable range; a supplementary power supply anomaly candidate is obtained when the current tracking residual is greater than the upper limit of the allowable range; a low-voltage load anomaly candidate is obtained when the load stability judgment fails or the load residual is greater than the upper limit of the allowable range; a sampling link anomaly candidate is obtained when the sampling channel self-test is abnormal or the data validity flag is invalid.

[0023] When a path anomaly candidate exists, the fault verification unit calls back the current excitation unit and the path calculation unit. If the path resistance or path hysteresis obtained again is not within the allowable range and the deviation sign is the same, the path anomaly verification result is obtained as the fault verification result.

[0024] When a candidate for a faulty charging power supply exists, the fault verification result is obtained based on whether the charging power supply can track the corresponding charging current target. When a candidate for a faulty sampling link exists, independent verification is performed, and the fault verification result is obtained.

[0025] The power-off verification unit reduces the output of the power supply to zero, obtains the discharge current and the residual voltage of the low-voltage battery terminal, and obtains the battery body abnormality verification result and the low-voltage load abnormality verification result as the fault verification result.

[0026] Furthermore, when the identification parameters are valid, the current limiting unit obtains the upper limit of the battery-side current and the upper limit of the power supply-side current based on the path resistance, instantaneous internal resistance, and polarization voltage, and takes the minimum of the two as the allowable charging current.

[0027] The fault handling unit completes fault confirmation when the intersection of the set of fault candidates and fault verification results is not empty, and writes the first charging current target, the second charging current target, the power supply restart permission, the unconfirmed fault candidates, the identification parameters, the measurement uncertainty, the current number of consecutive power replenishments, the amount of power replenished this time, and the effective zero-point calibration results into the data storage.

[0028] This invention provides an on-board low-voltage battery management system that supports intelligent charging and fault diagnosis. It offers the following advantages:

[0029] 1. This invention forms a first reference interval, an intermediate interval, and a second reference interval by sequentially making the current in the charging path reach the first charging current target, the second charging current target, and the first charging current target. The data validity flag is obtained from the load residual, current tracking residual, node current imbalance, and voltage response direction, and the identification parameters are calculated only when the return current step data is valid.

[0030] 2. This invention achieves separate calculation of the charging path parameters and the battery body parameters by calculating the path resistance using the path resistance method based on steady-state data, calculating the instantaneous internal resistance using the instantaneous internal resistance method based on the initial response data, and calculating the polarization voltage using the polarization recursive method based on the continuous response data.

[0031] 3. This invention performs path callback verification, power supply tracking verification, sampling link independent verification, and power failure verification on fault candidates. When the intersection of the fault candidate and the fault verification result is not empty, the power supply path current is limited by the minimum value of the upper limit of the battery side current and the upper limit of the power supply side current, thereby realizing the constraint of the subsequent power supply current on the confirmed fault result. Attached Figure Description

[0032] Figure 1 This is a system architecture diagram of the present invention;

[0033] Figure 2 This is a flowchart of the method of the present invention;

[0034] Figure 3 This is a working logic diagram of the state detection module of the present invention;

[0035] Figure 4 This is a working logic diagram of the charging control module of the present invention;

[0036] Figure 5 This is a working logic diagram of the parameter calculation module of the present invention;

[0037] Figure 6 This is a working logic diagram of the fault determination module of the present invention;

[0038] Figure 7 This is a working logic diagram of the power replenishment module of the present invention;

[0039] Figure 8 Box plots of path resistance and instantaneous internal resistance under different states in the application embodiments of the present invention. Detailed Implementation

[0040] The technical solutions in 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.

[0041] Please see the appendix Figure 1 This invention provides an on-board low-voltage battery management system that supports intelligent charging and fault diagnosis, comprising:

[0042] The status detection module is used to complete the self-test of the sampling channel, the determination of power supply permission, and the determination of power supply requirement.

[0043] The charging control module is used to determine the current limited charging result, perform load stability judgment, control the charging path current to complete the return current step, and generate a data validity flag based on the collected results.

[0044] The parameter calculation module is used to calculate identification parameters, including path resistance, instantaneous internal resistance, and polarization voltage, according to path resistance, instantaneous internal resistance, and polarization recursion methods when the data validity flag is valid, and to determine whether the calculated changes are consistent with the measured changes.

[0045] The fault determination module is used to generate fault candidates and obtain the corresponding fault verification results by re-executing the return current step, sampling link independent verification, and power-off verification.

[0046] The power replenishment module is used to determine the allowable charging current, confirm the fault based on the fault candidate and fault verification results, implement power replenishment limits or stop power replenishment, and write the data to be used in the next cycle into the data storage.

[0047] See appendix Figure 2 This invention provides a method for managing on-board low-voltage batteries that supports intelligent charging and fault diagnosis, comprising the following steps:

[0048] S1, the status detection module completes the sampling channel self-test, power replenishment permission determination, and power replenishment requirement determination.

[0049] S2, the charging control module determines the current limited charging result, performs load stability judgment, controls the charging path current to complete the return current step, and generates a data validity flag based on the collected results.

[0050] S3, when the data validity flag is valid, the parameter calculation module uses path resistance, instantaneous internal resistance and polarization recursion to obtain identification parameters including path resistance, instantaneous internal resistance and polarization voltage, and judges whether the calculated changes are consistent with the measured changes.

[0051] S4, the fault determination module generates fault candidates, and obtains the corresponding fault verification results by re-executing the return current step, sampling link independent verification, or power-off verification.

[0052] S5, the power replenishment module determines the allowable charging current, completes fault confirmation based on fault candidates and fault verification results, executes power replenishment limit or stops power replenishment, and writes the data to be used in the next cycle into the data storage.

[0053] The technical solutions in the embodiments of the present invention will be further explained below in conjunction with the specific processing procedures.

[0054] In this embodiment, the on-board low-voltage power supply device includes a low-voltage battery, a charging power supply, a charging power supply side node, a low-voltage battery side node, a charging path, a charging switch, and a battery management controller. This system is housed within the battery management controller. The charging switch uses a relay or a MOS switch; the charging power supply uses a high-low voltage DC-DC converter powered by the high-voltage battery, or a DC power supply within the vehicle with controllable low-voltage output.

[0055] Low-voltage electrical loads are connected to the supplementary power supply side node, and the supplementary power path connects the supplementary power supply side node and the low-voltage battery side node. When there are power branches in the supplementary power path, the supplementary power path current is directly collected. The low-voltage load current is read by the low-voltage load branch current sampling device or the vehicle power distribution controller; when there are multiple low-voltage load branches, the currents of each low-voltage load branch are summed. When the parallel low-voltage power supply is in the off state or does not exist, its output current is taken as 0; when the parallel low-voltage power supply is in the working state, the output current of the parallel low-voltage power supply is read. The battery management controller obtains system operation data through voltage sampling, current sampling, temperature sampling, and vehicle network communication.

[0056] See appendix Figure 3 In this embodiment, step S1 is executed by a state detection module, which includes a permission determination unit and a requirement determination unit.

[0057] S11, the permission determination unit reads the next detection time, measurement uncertainty, power supply restart permission, current consecutive power supply count, and valid zero-point calibration result from the data storage. If there is no next detection time, it enters the current state detection; if there is a next detection time but the current time has not yet arrived, it maintains a low-power state. If there is no current consecutive power supply count, it initializes the current consecutive power supply count to 0; if there is no power supply restart permission, this item does not prevent the power supply permission determination.

[0058] After the vehicle is in parked, locked, or a predetermined low-power state, the sampled values, update times, and hardware diagnostic results of the voltage, current, and temperature samples are read. If the sampled value is within the range of the corresponding sampling channel, the update time or sample count changes continuously, and the sampling channel hardware diagnostic is normal, the sampling channel self-test passes; if any condition is not met, the sampling channel self-test fails. Open-circuit, short-circuit, and range diagnostics of the sampling channels can be performed using the sampling chip's self-diagnostics and controller input diagnostics, which are well-known technologies in the field and will not be elaborated upon here.

[0059] When the power supply switch is off and there are no other conductive branches in the power supply path, putting the current sampling in a hardware zero-input state, current sampling zero-point calibration is performed. Current sampling values ​​are continuously acquired within a preset calibration sampling time. The average value of the current sampling values ​​and the difference between the maximum and minimum values ​​are calculated, with the average value used as the zero-point offset. A valid zero-point calibration result is generated when both the zero-point offset and the difference between the maximum and minimum values ​​are within preset allowable ranges; otherwise, the sampling channel self-test is set to abnormal. When a hardware zero-input state is not available, the valid zero-point calibration result stored in the data is used; if no valid zero-point calibration result exists, the sampling channel self-test is set to abnormal. The calibration sampling time is set according to the sampling frequency and filter period, for example, 1s to 5s; the allowable ranges for the zero-point offset and sampling dispersion are set according to the current sampling device's range, sampling resolution, and allowable error, for example, 0.1% to 0.5% and 0.05% to 0.2% of the full scale, respectively.

[0060] The measurement uncertainty is updated based on the zero-point calibration of the current sampling. The measurement uncertainty corresponding to the current in the supplementary power path is synthesized from the quantization error corresponding to the sampling resolution, the dispersion of the zero-point acquisition value, and the allowable error of the current sampling device; the measurement uncertainty corresponding to the node voltage on the supplementary power supply side and the terminal voltage of the low-voltage battery is synthesized from the quantization error corresponding to the voltage sampling resolution and the allowable error of the voltage sampling device.

[0061] When the error components are independent, the square root of the sum of the squares of the error components is taken; when their independence cannot be determined, the sum of the absolute values ​​of the error components is taken. When obtaining a related combined quantity by summing, subtracting, averaging, or weighting multiple sample quantities, the measurement uncertainties corresponding to each sample quantity are synthesized using the same operational coefficients; when the error components are independent, the square root of the sum of the squares of the weighted error components is taken; when their independence cannot be determined, the sum of the absolute values ​​of the weighted error components is taken.

[0062] The collision status and vehicle status are read from the vehicle controller; the high-voltage interlock status and high-voltage battery status are read from the high-voltage battery management controller; the low-voltage battery status is read from the battery management controller; and the charging power supply status is read from the charging power supply controller. The vehicle status condition is considered passed when the collision status is "no collision," the high-voltage interlock status is "closed," the vehicle status is in a charging-allowed state, the high-voltage battery status is "allowed to output," the low-voltage battery status is "allowed to charge," and the charging power supply status is "allowed to start." If the vehicle status condition fails, the charging power supply restart permission is not given, or the current consecutive charging count reaches the preset consecutive charging count limit, the charging permission is deemed failed, the corresponding reason is recorded, and the system remains in a low-power state; otherwise, the charging permission is deemed passed. The consecutive charging count limit is set based on the allowed charging frequency of the low-voltage battery, the vehicle's static power consumption, and the high-voltage battery energy management requirements, for example, 2 to 5 times.

[0063] S12, when the charging permission is approved and the charging power supply remains off, the demand determination unit reads the existing state of charge (SOC) of the battery management controller and uses it as the initial SOC. The SOC can be obtained by ampere-hour integration and open-circuit voltage correction, which is a well-known technology in the field and will not be described in detail here.

[0064] The current replenishment power is read from the data storage. The current replenishment power read in the current cycle is the data written at the end of the previous replenishment cycle. If no corresponding data exists, the current replenishment power is not used as the input for the pre-replenishment baseline. Within the preset baseline sampling time, the low-voltage battery terminal voltage, replenishment path current, and battery temperature are collected according to a unified timestamp. Sampling data with invalid timestamps or sampling channel self-test anomalies are discarded. When the effective sampling quantity of each input quantity reaches the preset minimum effective sampling quantity, the average value of the remaining sampling data is calculated, and the difference between the maximum and minimum values ​​is recorded. If the effective sampling quantity of any input quantity does not reach the minimum effective sampling quantity, the pre-replenishment baseline is not established, the reason for the sampling channel self-test anomaly is recorded, the next detection time is obtained according to the shortest detection interval in the detection cycle table, and after being written to the data storage, the low-power state is restored.

[0065] Each average value serves as a stable representative value for the corresponding input quantity, and together with the initial state of charge, the number of effective samples, the amount of power replenished, and the sampling time, constitutes the baseline before replenishment. The baseline sampling time is set based on the low-voltage battery terminal voltage stabilization time and the controller's low-power wake-up cycle, for example, 5s to 30s; the minimum number of effective samples is set based on the sampling frequency and the baseline sampling time, for example, 10 to 100.

[0066] Based on the baseline before recharging, the existing conclusion of whether recharging is needed is obtained using the threshold recharging method. If recharging is determined not to be needed, the corresponding detection interval is retrieved from a preset detection cycle table based on the state of charge and the low-voltage battery terminal voltage. The shorter detection interval is added to the current time to obtain the next detection time, which is then written to the data storage. Subsequently, the low-power state is restored. The detection cycle table is set based on the low-voltage battery self-discharge characteristics, vehicle static power consumption, and allowable state of charge range, for example, the detection interval is 0.5h to 24h. If recharging is determined to be needed, unconfirmed fault candidates are read from the data storage. The initial state of charge is passed to the parameter calculation module, the unconfirmed fault candidates are passed to the fault determination module, and the result indicating that recharging is needed is output to the charging control module.

[0067] See appendix Figure 4 In this embodiment, step S2 is executed by the charging control module, which includes a load monitoring unit and a current excitation unit.

[0068] S21, after determining that power replenishment is needed, the load monitoring unit activates the controller and power supply required for power replenishment, and reads the low-voltage load current, power replenishment path current, and battery temperature according to a unified timestamp. The low-voltage load current is read by the low-voltage load branch current sampling device or the vehicle power distribution controller; when there are multiple low-voltage load branches, the currents of each branch at the same sampling time are added together.

[0069] The charging characteristic table is consulted based on battery temperature to obtain the temperature-limited current and rated current, and the maximum output current is read from the charging power supply. The low-voltage load current is subtracted from the maximum output current; if the resulting current is not greater than 0, it is set to 0. The minimum value between this current and the temperature-limited current and rated current is then used to obtain the current limited charging result. If the current limited charging result is not greater than 0, it is set to 0, the charging power supply output is stopped, and a power-off verification is initiated. The establishment and lookup of the charging characteristic table are well-known techniques in this field and will not be elaborated upon here.

[0070] If there is a valid instantaneous internal resistance in the data storage and the instantaneous internal resistance is greater than 0, the current low-voltage battery terminal voltage is subtracted from the preset terminal voltage upper limit to obtain the terminal voltage margin for the startup phase. If the terminal voltage margin is greater than 0, the terminal voltage margin is divided by the instantaneous internal resistance to obtain the allowable increase in the charging path current during the charging startup phase. The smaller value between this charging path current and the current restricted charging result is taken as the output constraint for the startup phase. If the terminal voltage margin is not greater than 0, the current restricted charging result is set to 0 and the power-off verification is initiated.

[0071] If the effective instantaneous internal resistance is not greater than 0, the data validity flag is set to invalid, the current limited power supply result is set to 0, the power supply output is stopped, and the power-off verification is initiated; if there is no effective instantaneous internal resistance, the current limited power supply result is used.

[0072] Within a preset continuous sampling window, the difference between the maximum and minimum values ​​of the low-voltage load current is calculated. If this difference is not greater than the upper limit of the allowable range, the load stability is deemed successful. If the above condition is not met after a preset load stability waiting time, the load stability is deemed unsuccessful, and the data validity flag is set to invalid. The continuous sampling window is set based on the load switching cycle after the vehicle controller is woken up, for example, 1s to 5s; the allowable range is set based on the low-voltage load current sampling error and the normal fluctuation range of the parking load, for example, its upper limit is 0.2A to 2A; the load stability waiting time is set based on the power supply startup time and the vehicle controller wake-up time, for example, 5s to 30s.

[0073] After the charging power supply is started, the charging path current in each sampling interval is multiplied by the corresponding sampling interval and accumulated to obtain the current charging amount. The charging path current in the non-charging direction is not included in the current charging amount. When the load stability is determined to be unsuccessful, the current charging ends when the current charging amount reaches the preset single charging amount. Low-voltage load anomaly candidates and low-voltage load faults generated in this cycle are subject to subsequent restrictions by the fault handling unit according to the single charging amount. The single charging amount is set based on the low-voltage battery rated capacity and parking charging energy management requirements, for example, 2% to 10% of the low-voltage battery rated capacity.

[0074] S22, when the load stability is determined to be satisfactory and the current restricted power supply result is greater than 0, the current excitation unit reads two charging current targets, including a first charging current target and a second charging current target, from the data storage. If no corresponding data exists, the first charging current target is determined within the current restricted power supply result according to a preset step target rule. The first charging current target is added to the preset target difference, and the second charging current target is obtained with the current restricted power supply result as the upper limit. If corresponding data exists, the first and second charging current targets are modified in combination with the current restricted power supply result so that neither the first nor the second charging current target is greater than the current restricted power supply result and the target difference is maintained.

[0075] If the target difference cannot be maintained, or if the target difference is not greater than the measurement uncertainty corresponding to the supplementary charging path current, the data validity flag is set to invalid, and supplementary charging continues according to the current limited supplementary charging result. The return current step is not executed, and the path resistance, instantaneous internal resistance, and polarization voltage are invalid. The step target rule is set based on the low-voltage battery's allowable charging current and the supplementary charging power supply's adjustment range. For example, the first charging current target is 20% to 60% of the current limited supplementary charging result; the target difference is set based on the current sampling resolution and the minimum adjustable current of the supplementary charging power supply, for example, 1A to 10A. Both the first and second charging current targets are in the charging direction.

[0076] The current in the supplementary charging path is controlled to sequentially reach the first charging current target, the second charging current target, and the first charging current target again, forming a first reference interval, an intermediate interval, and a second reference interval. When using total output current control, the first or second charging current target corresponding to the current interval is taken as the charging current target for the current interval. The initial total output current target is the sum of the charging current target for the current interval and the low-voltage load current during the load stability determination stage. Subsequently, the real-time low-voltage load current is added to the charging current target for the current interval, and the product of the difference between the charging current target for the current interval and the supplementary charging path current and the preset adjustment coefficient is added to update the total output current target.

[0077] The adjustment coefficient is set based on the closed-loop response characteristics of the power supply, for example, from 0.1 to 1. When output voltage control is used, the output voltage target is adjusted through the existing closed-loop control of the power supply to bring the current in the power supply path into the corresponding allowable range.

[0078] The three intervals use the same preset relative sampling time. After each current switch, the sampled value close to the switching time but not yet in steady state is used as the initial response data; the sampled value from the end of the initial response data to the start of the steady-state data is used as the continuous response data; and the sampled value at the end of the interval is used as the steady-state data. The three data segments do not overlap. The duration of each interval is greater than the sum of the end time of the continuous response data and the duration of the steady-state data.

[0079] The relative sampling time is set according to the current sampling period, the response time of the power supply, and the polarization response time of the low-voltage battery. For example, the initial response data corresponds to 5ms to 100ms, the continuous response data corresponds to more than 100ms and no more than 10s, and the steady-state data lasts for 1s to 5s after 10s.

[0080] Timestamp alignment is considered complete when the timestamp difference of the sampled data is not greater than the preset timestamp alignment upper limit. The timestamp alignment upper limit is set based on the sampling period and communication delay, for example, from 1ms to 10ms. For each alignment sampling moment, the total output current of the supplementary power supply is added to the output current of the parallel low-voltage power supply, and then the supplementary path current and the low-voltage load current are subtracted. The absolute value is taken, and the maximum absolute value obtained within the three intervals is taken as the node current imbalance. The node current relationship is considered valid when the node current imbalance is not greater than the upper limit of the allowable range.

[0081] Calculate the average low-voltage load current for each of the three intervals, and use the larger of the absolute values ​​of the differences between the middle interval and the second reference interval relative to the first reference interval as the load residual. Calculate the average value of the supplementary charging path current for each of the three intervals, and use the larger of the absolute values ​​of the differences between the average supplementary charging path current and the corresponding charging current target as the current tracking residual. Use the smaller of the absolute values ​​of the differences between the average supplementary charging path currents of the middle interval and the two reference intervals as the path current step amplitude, and use the smaller of the absolute values ​​of the differences between the average low-voltage battery terminal voltages as the low-voltage battery terminal voltage response amplitude. When the supplementary charging path current increases or decreases, record the direction of current change as increasing or decreasing, and determine the voltage response direction according to the corresponding change in low-voltage battery terminal voltage.

[0082] The data validity flag is set to valid only when the load residual and current tracking residual are not greater than the upper limit of the allowable range, the node current relationship is valid, the path current step amplitude and the low-voltage battery terminal voltage response amplitude are greater than the corresponding measurement uncertainty, the timestamp alignment is completed, the sampling channel self-test passes, the battery temperature is within the allowable range, and the voltage response direction of both switching is equal to the current change direction. If any condition is not met, the data validity flag is set to invalid, and the path resistance, instantaneous internal resistance, and polarization voltage are invalid. Each allowable range is set based on the sampling error, the power supply control error, and the allowable operating temperature of the low-voltage battery.

[0083] See appendix Figure 5 In this embodiment, step S3 is executed by a parameter calculation module, which includes a path calculation unit and a battery calculation unit.

[0084] S31, when the data validity flag is valid, the path calculation unit reads the power supply side node voltage, low-voltage battery terminal voltage, and power supply path current from the steady-state data of the first reference interval, the intermediate interval, and the second reference interval. It subtracts the low-voltage battery terminal voltage from the power supply side node voltage at the same sampling time to obtain the path voltage drop, and calculates the average path voltage drop and the average power supply path current for the three intervals respectively. The voltage difference between two nodes corresponds to the voltage drop formed at the wiring harness, connector, and power supply switch along the power supply path. The path resistance is calculated using a path resistance method.

[0085] ;

[0086] In the formula, For path resistance; , and The average path pressure drop values ​​are, in order, the first reference interval, the middle interval, and the second reference interval; , and The values ​​are the average currents of the supplementary power path for the three intervals, respectively.

[0087] If the absolute value of the value within the brackets in the denominator of the path resistance formula is not greater than the measurement uncertainty corresponding to the current combination quantity in that denominator, the data validity flag is set to invalid, and the path resistance, instantaneous internal resistance, and polarization voltage are invalid. The path hysteresis is obtained by subtracting the average path voltage drop of the first reference interval from the average path voltage drop of the second reference interval.

[0088] Calculate the difference between the average values ​​of the supplementary current path in the intermediate section and the two reference sections to obtain the current changes in the two supplementary current paths. Multiply the path resistance by the corresponding current change in the supplementary current path to obtain the voltage drop changes in the two paths. Subtract the average voltage drop of the path in the two reference sections from the average voltage drop of the intermediate section to obtain the measured voltage drop changes in the two paths. The two segments are considered consistent if the absolute value of the difference is not greater than the upper limit of the allowable range; otherwise, they are considered inconsistent. The allowable range is set based on the normal fluctuation range of the node voltage measurement uncertainty, the supplementary current measurement uncertainty, and the conduction voltage drop of the supplementary switch; for example, its upper limit is 10mV to 200mV.

[0089] The allowable range of path resistance is set based on the wiring harness resistance, connector contact resistance, and the conduction resistance of the supplementary switch; the allowable range of path hysteresis is centered at 0 and is set based on the uncertainty of node voltage measurements in two reference intervals and the repeated conduction error of the supplementary switch. When the path resistance or path hysteresis is lower than the lower limit of the allowable range, the deviation sign is "below the lower limit"; when it is higher than the upper limit of the allowable range, the deviation sign is "above the upper limit".

[0090] S32, the battery calculation unit updates the state of charge (SOC) through ampere-hour integration based on the initial SOC and the current in the charging path, and obtains the open-circuit voltage by looking up the open-circuit voltage table based on the SOC and battery temperature. The ampere-hour integration and open-circuit voltage lookup can be performed by those skilled in the art based on the rated capacity and open-circuit voltage characteristics of the low-voltage battery; the implementation method is well-known in the field and will not be elaborated here.

[0091] When the data validity flag is valid, the battery overvoltage is obtained by subtracting the corresponding open-circuit voltage from the low-voltage battery terminal voltage. The representative value of the initial response for each interval is the average of the valid sampled data within the same relative sampling time, and the instantaneous internal resistance is calculated using the instantaneous internal resistance formula:

[0092] ;

[0093] In the formula, Instantaneous internal resistance; , and The following are representative values ​​of battery overvoltage from the initial response data of the three intervals, in order. , and The values ​​are, in order, the representative values ​​of the current in the compensation path from the corresponding initial response data.

[0094] If the absolute value of the value within the brackets in the denominator of the instantaneous internal resistance formula is not greater than the measurement uncertainty corresponding to the initial response current combination in the denominator, the data validity flag is set to invalid, and the path resistance, instantaneous internal resistance, and polarization voltage are invalid. Calculate the average battery overvoltage for each of the three intervals. Subtract the average battery overvoltage for the first reference interval from the average battery overvoltage for the second reference interval to obtain the battery hysteresis. The allowable range of instantaneous internal resistance is set based on the low-voltage battery type, rated capacity, battery temperature, and state of charge; the allowable range of battery hysteresis is centered at 0 and set based on the measurement uncertainty of the low-voltage battery terminal voltage and the calculation error of the open-circuit voltage.

[0095] The polarization overvoltage is obtained by subtracting the product of the instantaneous internal resistance and the current in the charging path from the battery overvoltage. The polarization overvoltage is a measured derived quantity, while the polarization voltage is a recursive calculated quantity. The polarization recursive formula is:

[0096] ;

[0097] In the formula, For the first Polarization voltage at each sampling time; The polarization voltage at the previous sampling time; It is a natural constant; The sampling interval between adjacent sampling times; The polarization time constant; Polarization resistor; For the first The current in the supplementary current path at each sampling time.

[0098] The initial polarization voltage is obtained by reading a valid polarization voltage from the data storage; if no corresponding data is available, a preset initial polarization voltage is used. The initial polarization voltage is set based on the low-voltage battery type, the charging path current in the baseline before charging, and the low-voltage battery resting time; when the low-voltage battery resting time reaches the preset polarization decay time, the initial polarization voltage is set to 0.

[0099] The polarization voltage is continuously recursively calculated based on the sampling time, with the polarization voltage at the last sampling time of the previous interval serving as the starting point for the recursion of the next interval. Initial response data is used to maintain the continuity of the recursive state and is not included in the fitting error. The polarization decay time is set according to the polarization time constant range of the low-voltage battery.

[0100] Within the preset range of polarization resistance and polarization time constant values, parameter combinations are formed according to the preset search steps for polarization resistance and polarization time constant, respectively. Each parameter combination is iterated, and the sum of squares of the difference between polarization voltage and polarization overvoltage at the continuous response sampling time under each parameter combination is calculated. The parameter combination with the smallest sum of squares is taken as the polarization resistance and polarization time constant. When there are the same minimum values, the values ​​are selected in the order of polarization resistance from small to large and polarization time constant from small to large.

[0101] The two value ranges are set according to the resistance and capacitance characteristics of the low-voltage battery, for example, 0.1mΩ to 100mΩ and 0.1s to 300s respectively; the polarization resistance search step size and polarization time constant search step size are set according to the parameter value range, the controller's computing power and parameter identification resolution, for example, 0.1mΩ to 1mΩ and 0.1s to 1s respectively.

[0102] If the absolute value of the difference between the polarization voltage and the polarization overvoltage at each consecutive response sampling time is not greater than the upper limit of the allowable range, the polarization resistance, polarization time constant, and polarization voltage are retained; if any difference is greater than the upper limit, the data validity flag is set to invalid, and the path resistance, instantaneous internal resistance, and polarization voltage are invalid. The allowable range is set based on the measurement uncertainty of the low-voltage battery terminal voltage, the calculation error of the open-circuit voltage, and the first-order RC discrepancy error.

[0103] The difference between the average polarization voltage values ​​corresponding to the steady-state data of the intermediate interval and the two reference intervals is calculated to obtain the polarization voltage changes in two segments. The product of the instantaneous internal resistance and the corresponding current change in the replenishment path is added to the polarization voltage change to obtain the battery overvoltage changes in two segments. The average battery overvoltage value of the intermediate interval is subtracted from the average battery overvoltage value of the two reference intervals to obtain the measured battery overvoltage changes in two segments. If the absolute value of the difference between the two segments is not greater than the upper limit of the allowable range, they are considered consistent; if either segment is greater than this upper limit, they are considered inconsistent. The path resistance, path hysteresis, deviation sign, instantaneous internal resistance, battery hysteresis, polarization resistance, polarization time constant, and the consistency / inconsistency determination results are transmitted to the fault determination module; the path resistance, instantaneous internal resistance, and polarization voltage are transmitted to the replenishment processing module.

[0104] See appendix Figure 6 In this embodiment, step S4 is executed by the fault determination module, which includes a candidate generation unit, a fault verification unit, and a power-off verification unit.

[0105] S41, when a recursive current step has been executed, the candidate generation unit reads the load residual, current tracking residual, node current imbalance, node current relationship, network status, timestamp alignment, sampling channel self-test, data validity flag, and load stability determination. It also reads the state of charge, low-voltage battery terminal voltage, path resistance, path hysteresis, path voltage drop change, battery hysteresis, polarization resistance, polarization time constant, and battery overvoltage change, generating fault candidates accordingly. When a recursive current step has not been executed, it only generates low-voltage load anomaly candidates or sampling link anomaly candidates based on the network status, load stability determination, and sampling channel self-test. Fault candidates indicate that the current detection cycle meets the corresponding anomaly conditions and are not considered fault confirmation results.

[0106] When the data validity flag is valid, the load stability is passed, the current tracking residual is no greater than the upper limit of the allowable range, the path resistance or path hysteresis is not within the allowable range, and the path voltage drop change is determined to be consistent, a path anomaly candidate is generated, and the anomaly parameters and their deviation signs are recorded. When the data validity flag is valid and any of the following conditions are met, a battery-related anomaly candidate is generated:

[0107] The determination of battery overvoltage changes is inconsistent;

[0108] The battery hysteresis, polarization resistance, or polarization time constant is not within the allowable range.

[0109] Both path anomaly candidates and battery body anomaly candidates are allowed to coexist.

[0110] A low-voltage load anomaly candidate is generated when the load residual exceeds the upper limit of the allowable range, the network status is an abnormal wake-up, or the load stability is deemed unsatisfactory. The network status is determined based on the wake-up source, wake-up time, and message duration recorded in the vehicle network communication log; if the wake-up source during parking is not a permitted wake-up source specified by the vehicle's sleep policy, the network status is considered an abnormal wake-up. The preset permitted wake-up sources and permitted wake-up periods are set according to the vehicle's sleep policy.

[0111] A candidate for a charging power supply anomaly is generated when the load stability is deemed satisfactory, the state of charge (SOC) is below the preset SOC charging limit, the low-voltage battery terminal voltage is below the preset terminal voltage charging limit, and the current tracking residual is greater than the upper limit of the allowable range. The SOC charging limit and terminal voltage charging limit are set based on the allowable charging range of the low-voltage battery to exclude cases where the low-voltage battery has reached its charging limit. A candidate for a sampling link anomaly is generated when the node current imbalance is greater than the upper limit of the allowable range, timestamp alignment is incomplete, or the sampling channel self-test is abnormal.

[0112] Path anomaly candidates, power supply anomaly candidates, and sampling link anomaly candidates, along with their generation time, triggering conditions, and related parameters, are transmitted to the fault verification unit; battery body anomaly candidates and low-voltage load anomaly candidates are transmitted to the power failure verification unit. Path anomaly candidates and battery body anomaly candidates are also associated with a record of the data validity flag at the time of candidate generation, for the fault handling unit to perform fault confirmation.

[0113] S42, the fault verification unit reads unconfirmed fault candidates from the data storage, processes the unconfirmed fault candidates first, and then processes the fault candidates generated this time. If there are path anomaly candidates or power supply anomaly candidates, the callback current excitation unit and the path calculation unit are called back to execute the return current step again.

[0114] Recalculate the current limited power supply result. If the data storage contains the next allowable charging current, take the smaller value between the next allowable charging current and the current limited power supply result to continue power supply; if no corresponding data exists, use the current limited power supply result. If the current limited power supply result is 0, or cannot form a target difference greater than the measurement uncertainty, do not perform a return current step, and save the corresponding fault candidate as an unconfirmed fault candidate.

[0115] After re-executing the current step, the data validity flag, current tracking residual, and node current relationship are obtained again, and the path resistance, path hysteresis, and voltage drop changes of the two path segments are calculated. If the data validity flag is valid, the abnormal parameters recorded during candidate generation are still outside the allowable range, the deviation sign remains the same, the voltage drop changes of the two path segments are still considered consistent, and the current tracking residual is not greater than the upper limit of the allowable range, a path anomaly verification result is generated. If the path voltage drop changes are considered inconsistent, a sampled link anomaly candidate is generated, and unconfirmed path anomaly candidates are saved.

[0116] If the current tracking residuals of the two consecutive current step jumps are both greater than the upper limit of the allowable range, and the sampling channel self-test passes, the timestamp alignment is completed, the node current relationship is valid, and the load stability judgment passes, and the state of charge is less than the state of charge charging limit, the low-voltage battery terminal voltage is less than the terminal voltage charging limit, and the path resistance is invalid or within the allowable range, then a power supply anomaly verification result is generated. Repeated current tracking failure is used to exclude single load fluctuations; the remaining conditions are used to exclude current limitations caused by sampling link anomalies, power supply path anomalies, and charging limits.

[0117] When a sampling link anomaly candidate exists, the corresponding current or voltage data is repeatedly collected, and sampling channel hardware diagnostics are performed. If the node current imbalance still exceeds the upper limit of the allowable range after repeated sampling, the sampling channel hardware diagnostics return an anomaly. The sampling result is not within the allowable range when the power supply switch is disconnected to form a known zero input; the sampling result is not within the allowable range when the reference source inside the sampling chip forms a known stable input; or the timestamp alignment is still incomplete after re-performing time synchronization. A sampling link anomaly verification result is then generated. The allowable ranges corresponding to the known zero input and known stable input are set based on the sampling resolution, reference source error, and sampling channel allowable error.

[0118] S43, after the power-off verification unit completes the return current step and parameter calculation, or if the return current step is not executed because the current limited power supply result is 0 and the target difference cannot be formed, before the final end of this power supply, the total output current of the power supply is gradually reduced according to the preset rate of decrease of the total output current of the power supply, so that the power supply path current reaches 0 or turns into normal discharge, and the power supply path current and the low-voltage battery terminal voltage are recorded.

[0119] When the current in the charging path is less than 0, its absolute value is used to obtain the discharge current; when the current in the charging path is not less than 0, the discharge current is taken as 0. The rate of decrease of the total output current of the charging power supply is set according to the allowable output change rate and sampling period of the charging power supply, for example, from 1A / s to 20A / s. After the power failure verification is completed, the fault verification result is transmitted to the charging handling module, which determines whether to stop charging or resume charging according to the allowable charging current.

[0120] When both the instantaneous internal resistance and polarization voltage are valid, the polarization voltage is updated using a polarization recursive method based on the current in the replenishment path and the sampling interval during the power outage verification period. The product of the open-circuit voltage, the current in the replenishment path, and the instantaneous internal resistance, along with the polarization voltage, is added to obtain the calculated value of the low-voltage battery terminal voltage. The low-voltage battery terminal voltage residual is obtained by subtracting the calculated value from the low-voltage battery terminal voltage.

[0121] The data when the current in the charging path reaches 0 is used as the zero-current reference. After the current in the charging path transitions to normal discharge, within a preset discharge stabilization sampling time, if the difference between the maximum and minimum values ​​of the discharge current is not greater than the upper limit of the allowable range, the end of this sampling time is determined as the discharge stabilization moment. The absolute value of the voltage difference between the low-voltage battery terminals between the zero-current reference and the discharge stabilization moment is taken to obtain the low-voltage battery terminal voltage change. The product of the discharge current and the instantaneous internal resistance is added to the polarization voltage change between the two moments to obtain the calculated voltage change.

[0122] When the difference between the maximum and minimum discharge current exceeds the upper limit of the allowable range, no battery-related abnormality verification results or low-voltage load abnormality verification results are generated, and the corresponding fault candidate is saved as an unconfirmed fault candidate. The discharge stabilization sampling time is set based on the low-voltage load response time and sampling period, for example, from 1s to 5s.

[0123] If the instantaneous internal resistance or polarization voltage is invalid, both are treated as invalid. If the discharge current exceeds the upper limit of the allowable range, a low-voltage load anomaly verification result is generated. If both the instantaneous internal resistance and polarization voltage are valid, and the discharge current does not exceed the upper limit of the allowable range while the absolute value of the low-voltage battery terminal voltage residual is greater than the upper limit of the allowable range, a battery body anomaly verification result is generated. If the discharge current exceeds the upper limit of the allowable range while the absolute value of the difference between the low-voltage battery terminal voltage change and the calculated voltage change does not exceed the upper limit of the allowable range, a low-voltage load anomaly verification result is generated.

[0124] The allowable range for discharge current is set based on the normal parking load current, for example, its upper limit is 0.5A to 5A; the allowable range for low-voltage battery terminal voltage residual and voltage variation difference is set based on the measurement uncertainty of low-voltage battery terminal voltage, instantaneous internal resistance calculation error, and polarization voltage calculation error, for example, its upper limit is 50mV to 300mV. Each fault verification result is transmitted to the charging module, where the path anomaly verification result and the battery body anomaly verification result are linked and the data validity flag used for verification is recorded.

[0125] See appendix Figure 7 In this embodiment, step S5 is executed by a power replenishment module, which includes a current limiting unit and a fault handling unit.

[0126] S51, the current limiting unit reads the open circuit voltage, polarization voltage, instantaneous internal resistance, path resistance, low-voltage load current, temperature current limit value, calibrated current and maximum output current to determine the allowable charging current.

[0127] When the path resistance, instantaneous internal resistance, and polarization voltage are all valid, and the path resistance is not less than 0 and the instantaneous internal resistance is greater than 0, the upper limit of the terminal voltage is subtracted from the open-circuit voltage and polarization voltage to obtain the battery-side voltage margin. If the battery-side voltage margin is greater than 0, it is divided by the instantaneous internal resistance to obtain the upper limit of the battery-side current; otherwise, the upper limit of the battery-side current is set to 0. When the path resistance is less than 0 or the instantaneous internal resistance is not greater than 0, the allowable charging current is set to 0 and charging is stopped.

[0128] The system reads the path resistance calculated in this operation and the valid path resistance stored in the data. If both exist, the larger value is used; otherwise, the path resistance calculated in this operation is used. The open-circuit voltage and polarization voltage are subtracted from the preset upper limit of the power supply-side node voltage to obtain the power supply-side voltage margin. If the power supply-side voltage margin is greater than 0, and the sum of the instantaneous internal resistance and the path resistance is greater than 0, the power supply-side voltage margin is divided by the sum of the two resistances to obtain the upper limit of the power supply-side current; otherwise, the upper limit of the power supply-side current is set to 0. A larger path resistance is used to ensure that the node voltage limit covers the path voltage drop state recorded in the current detection result and the previous power supply cycle.

[0129] The maximum output current minus the low-voltage load current yields the current that the power supply can provide to the low-voltage battery; this current is set to 0 if it is not greater than 0. The minimum value among the temperature current limit, rated current, the current that the power supply can provide to the low-voltage battery, the upper limit of the battery-side current, and the upper limit of the power supply-side current is used to obtain the allowable charging current. If any of the path resistance, instantaneous internal resistance, or polarization voltage is invalid, the current limited power supply result is used as the allowable charging current. The upper limit of the terminal voltage is set based on the allowable charging voltage of the low-voltage battery, and the upper limit of the node voltage on the power supply side is set based on the rated output voltage of the power supply and the allowable voltage of the low-voltage load.

[0130] S52, the fault handling unit constructs a fault candidate set based on path anomaly candidates, battery body anomaly candidates, low voltage load anomaly candidates, charging power supply anomaly candidates and sampling link anomaly candidates, and constructs a verification result set based on the corresponding five types of fault verification results.

[0131] Path anomaly candidates and path anomaly verification results are mapped to replenishment path faults; battery body anomaly candidates and battery body anomaly verification results are mapped to battery body faults; low-voltage load anomaly candidates and low-voltage load anomaly verification results are mapped to low-voltage load faults; replenishment power supply anomaly candidates and replenishment power supply anomaly verification results are mapped to replenishment power supply faults; and sampling link anomaly candidates and sampling link anomaly verification results are mapped to sampling link faults. When the fault candidate set and the verification result set contain the same fault type, that fault type is included in the intersection of the two sets.

[0132] If the intersection includes either a charging path fault or a battery body fault, and the data validity flag used for the corresponding fault candidate and fault verification result is valid, the fault confirmation is completed. If the intersection includes a charging power supply fault, a low-voltage load fault, or a sampling link fault, the fault confirmation is completed directly. Fault candidates that do not enter the intersection, as well as charging path faults and battery body faults that are not confirmed due to invalid data validity flags, are saved as unconfirmed fault candidates.

[0133] When a fault is confirmed in the charging path, if the allowable charging current is greater than 0, the charging path current is limited according to the allowable charging current; otherwise, charging is stopped, and inspection information for the wiring harness, connectors, relays, or MOS switches is output. When a fault is confirmed in the battery itself, the charging path current is limited or charging is stopped in the same manner, and inspection information for the low-voltage battery is output. The inspection information is written to the data storage and sent to the vehicle fault diagnosis device via the vehicle network communication.

[0134] When a fault is confirmed in the power supply, power supply replenishment is stopped, and the power supply restart permission is set to "Not Given". If the power supply completes fault reset or maintenance confirmation, and there are no active faults in the power supply's self-diagnostic status, the fault status is "Cleared", and the power supply restart permission is set to "Given"; otherwise, it remains "Not Given". If a power supply fault is not confirmed, the power supply restart permission remains unchanged; if there is no corresponding data and the power supply's self-diagnostic status is normal, the power supply restart permission is set to "Given".

[0135] Record the network status when a low-voltage load fault is confirmed. If a low-voltage load anomaly candidate exists or a low-voltage load fault has been confirmed, the current power replenishment will stop once the single power replenishment amount is reached; if the single power replenishment amount has not been reached and other stopping conditions are not met, power replenishment will continue according to the allowable charging current.

[0136] If the load stability assessment fails, a low-voltage load anomaly is detected, or a low-voltage load fault is confirmed, the current consecutive charging count is incremented by one. If none of the above situations occur during the current charging process, the current consecutive charging count is reset to 0. When the current consecutive charging count reaches the upper limit, the next charging permission is deemed unsuccessful. After the vehicle exits parking mode or locks down and the low-voltage battery is fully charged by the vehicle charging process, the current consecutive charging count is reset to 0.

[0137] When a sampling link failure is confirmed, it is determined whether the abnormal sampling quantity is included in the calculation of the current limited charging result. If the abnormal sampling quantity is not included in the calculation and the input of the current limited charging result is valid, charging continues according to the current limited charging result; if the abnormal sampling quantity is low-voltage load current, battery temperature, maximum output current, or charging path current, the allowable charging current is set to 0 and charging is stopped. The sampling link inspection information is written to the data storage and output through vehicle network communication.

[0138] If the power outage verification does not confirm the need to stop power replenishment, power replenishment will resume according to the allowable charging current until the threshold power replenishment or whole vehicle power replenishment gives a conclusion of power replenishment completion, or the amount of power replenished in this operation reaches the amount of power replenished in a single operation. When any termination condition is met, this power replenishment operation will finally end, and the next detection time will be calculated and data write-back will be performed.

[0139] Based on the state of charge at the end of the charging cycle, the confirmed fault type, load stability assessment, unconfirmed fault candidates, and the amount of charge generated during the current charging cycle, a detection interval is selected from the detection cycle table and added to the current time to obtain the next detection time; when multiple states correspond to different detection intervals, the shortest value is taken. The detection cycle table is set based on the low-voltage battery self-discharge characteristics, vehicle static power consumption, and fault review cycle, for example, the detection interval is from 0.5h to 24h.

[0140] The allowable charging current after fault handling is used as the allowable charging current for the next cycle. New values ​​are written when a new first and second charging current target are formed during this cycle; otherwise, the original values ​​are retained. The following data are written to the data storage: next detection time, next allowable charging current, first charging current target, second charging current target, power supply restart permission, unconfirmed fault candidates, valid path resistance, valid instantaneous internal resistance, valid polarization voltage, measurement uncertainty, current number of consecutive power replenishments, current power replenishment amount, and valid zero-point calibration results.

[0141] In the next charging cycle, the permission determination unit reads the next detection time, measurement uncertainty, charging power supply restart permission, current number of consecutive charging cycles, and valid zero-point calibration results; the demand determination unit reads the current charging capacity and unconfirmed fault candidates; the load monitoring unit reads the valid instantaneous internal resistance; the current excitation unit reads the first charging current target and the second charging current target; the battery calculation unit reads the valid polarization voltage; the fault verification unit reads the next allowed charging current; and the current limiting unit reads the valid path resistance.

[0142] To further aid in understanding the technical solutions of the embodiments of the present invention, a specific application example is given below.

[0143] A verification device was built using a 12V low-voltage battery with a rated capacity of 60Ah, an adjustable charging power supply, a charging switch, an adjustable electronic load, and a battery management controller. The low-voltage load current was maintained at around 6A, the battery temperature was 25℃, and the open-circuit voltage before charging was 12.42V. The current excitation unit sequentially controlled the charging path current to 8A, 16A, and 8A, forming the first reference interval, the intermediate interval, and the second reference interval. The average charging path currents in the three intervals were 8.0A, 16.0A, and 8.1A, respectively; the average low-voltage battery terminal voltages were 12.55V, 12.67V, and 12.56V, respectively; and the average node voltages on the charging power supply side were 12.67V, 12.911V, and 12.682V, respectively.

[0144] Subtracting the low-voltage battery terminal voltage from the voltage at the power supply node yields the average path voltage drop for the three intervals, which are 0.120V, 0.241V, and 0.122V, respectively. Calculations are then performed using the path resistance formula.

[0145] ;

[0146] The calculated path resistance is 15.09 mΩ. Subtracting the average path voltage drop of the first reference interval from the average path voltage drop of the second reference interval yields a path hysteresis of 2 mV. Both reference intervals use the same first charging current target; this path hysteresis is used to determine the recovery state of the charging path after a return current step.

[0147] The battery overvoltage values ​​in the initial response data are 0.112V, 0.191V, and 0.113V, respectively, with corresponding charge path current values ​​of 8.0A, 15.8A, and 8.1A. Calculations are performed based on the instantaneous internal resistance formula.

[0148] ;

[0149] The instantaneous internal resistance was calculated to be 10.13 mΩ. Subtracting the product of the instantaneous internal resistance and the current in the charging path from the battery overvoltage yielded the polarization overvoltage. Using continuous response data and a polarization recursive formula for least-squares fitting, the polarization resistance was found to be 6.8 mΩ and the polarization time constant to be 18.5 s. When the sampling interval was 0.1 s, the polarization voltage at the previous sampling moment was 0.080 V, and the current charging path current was 16 A, substituting these values ​​into the polarization recursive formula yielded a current polarization voltage of approximately 0.0802 V.

[0150] The upper limit of the terminal voltage is 14.4V, the upper limit of the charging power supply node voltage is 14.8V, and the current polarization voltage is 0.08V. Based on the open-circuit voltage, instantaneous internal resistance, path resistance, and polarization voltage, the upper limit of the battery-side current is approximately 188.1A, and the upper limit of the power supply-side current is approximately 91.3A. The temperature current limit is 30A, the rated current is 25A, and the maximum output current minus the low-voltage load current is 34A. Taking the minimum value among all limits, the allowable charging current is 25A.

[0151] To verify the differentiation effect between path parameters and battery parameters, three states were set up: normal state, abnormal state of the charging path, and abnormal state of the battery itself. Each verification state was subjected to 20 back-current step jumps. The abnormal state of the charging path was formed by adding an additional resistor in the charging path; the abnormal state of the battery itself was formed by a low-voltage battery with a momentary increase in internal resistance.

[0152] Under normal conditions, the path resistance ranges from 13.8mΩ to 16.5mΩ, with a median of 15.1mΩ; the instantaneous internal resistance ranges from 9.2mΩ to 11.4mΩ, with a median of 10.3mΩ. Under abnormal conditions in the power supply path, the path resistance ranges from 42.5mΩ to 47.8mΩ, with a median of 45.2mΩ; the instantaneous internal resistance ranges from 9.4mΩ to 11.8mΩ, with a median of 10.6mΩ.

[0153] Under abnormal battery conditions, the path resistance ranges from 14.1 mΩ to 16.9 mΩ, with a median of 15.4 mΩ; the instantaneous internal resistance ranges from 25.6 mΩ to 31.2 mΩ, with a median of 28.4 mΩ. The verification results indicate that abnormal conditions in the charging path mainly correspond to changes in path resistance, while abnormal conditions in the battery itself mainly correspond to changes in instantaneous internal resistance, providing parameter basis for distinguishing between these two types of fault candidates.

[0154] See appendix Figure 8 In each verification state, the unfilled box line on the left represents the path resistance, and the gray-filled box line on the right represents the instantaneous internal resistance. The horizontal line in the middle of the box represents the median, the upper and lower boundaries of the box represent the upper and lower quartiles, respectively, and the end lines on the outside of the box represent the range of values ​​for non-outlier data.

[0155] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these specific embodiments are merely illustrative. Those skilled in the art can omit, substitute, and modify the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, combining the above method steps to perform substantially the same function and achieve substantially the same result using substantially the same method falls within the scope of the present invention.

Claims

1. A vehicle-mounted low-voltage battery management system supporting intelligent power replenishment and fault diagnosis, characterized in that, include: The status detection module is used to read the next detection time from the data storage and perform a power replenishment permission determination. The power replenishment permission determination is performed when the power replenishment requirement is passed. The charging control module is used to obtain the current limited charging result and load stability determination when the charging demand is determined to be required. When the load stability determination is passed and the current limited charging result is greater than 0, two charging current targets are obtained and a return current step is executed to make the charging path current change according to the two charging current targets, thereby obtaining a data validity flag. The parameter calculation module is used to calculate the identification parameters when the data validity flag is valid, and otherwise set the identification parameters to invalid. The fault determination module is used to generate and verify fault candidates based on the data validity flags, and obtain fault verification results; The power replenishment module is used to obtain the allowable charging current according to the identification parameters. When the identification parameters are invalid, the current limited power replenishment result is used as the allowable charging current. The module completes the fault confirmation according to the fault candidate and the fault verification result, limits the power replenishment path current according to the allowable charging current, and writes the next detection time and the allowable charging current into the data storage.

2. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 1, characterized in that, The status detection module includes: The permission determination unit is used to read the next detection time, measurement uncertainty, power supply restart permission, current number of consecutive power replenishment cycles, and effective zero-point calibration results from the data storage to obtain the sampling channel self-test. If the power supply restart permission is not available, the power supply permission will not be deemed unsuccessful due to the lack of the power supply restart permission. The power supply restart permission will be deemed unsuccessful if it is not provided, or if the current number of consecutive power supply restarts reaches the preset limit for consecutive power supply restarts. The demand determination unit is used to read the charge state as the initial charge state and perform the power replenishment demand determination when the power replenishment permission is determined to be passed and the power replenishment power supply is kept off. When the power replenishment requirement is determined to be not required, the next detection time is obtained and written into the data storage; when the power replenishment requirement is determined to be required, unconfirmed fault candidates are read.

3. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 2, characterized in that, The charging control module includes: The load monitoring unit is used to obtain the current limited power supply result when the power supply demand is determined to be required, complete the load stability determination, and accumulate the power supply amount for this power supply. The load stability is determined to be satisfactory when the difference between the maximum and minimum values ​​of the low-voltage load current within the preset continuous sampling window is not greater than the upper limit of the preset allowable range. If the time from the start of the power supply until the low-voltage load is not stabilized is greater than the preset load stabilization waiting time, the load stabilization is deemed unsuccessful, and the data validity flag is deemed invalid.

4. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 3, characterized in that, The charging control module includes: A current excitation unit is used to read the two charging current targets from the data storage when the load stability is determined to be passable and the current limited power supply result is greater than 0; and to determine the two charging current targets based on the current limited power supply result when the two charging current targets do not exist. The two charging current targets include a first charging current target and a second charging current target, both of which are not greater than the current limited power supply result. The power supply path current sequentially reaches the first charging current target, the second charging current target, and the first charging current target, forming a first reference interval, an intermediate interval, and a second reference interval. Initial response data, continuous response data, and steady-state data for each interval are obtained. Based on the initial response data, the continuous response data, the steady-state data, and the low-voltage load current, the load residual, current tracking residual, node current imbalance, and voltage response direction are obtained. The data validity flag is valid when the load residual, the current tracking residual, and the node current imbalance are all no greater than the upper limit of the allowable range and the voltage response direction is equal to the current change direction; otherwise, the data validity flag is invalid.

5. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 4, characterized in that, The parameter calculation module includes: The path calculation unit is used to calculate the path resistance using the path resistance method based on the steady-state data when the data validity flag is valid, and to obtain the path hysteresis and deviation sign.

6. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 5, characterized in that, The parameter calculation module includes a battery calculation unit; The battery calculation unit is used to calculate the instantaneous internal resistance based on the initial response data using the instantaneous internal resistance method when the data validity flag is valid, and to calculate the polarization voltage based on the continuous response data using the polarization recursive method, thereby obtaining the polarization resistance and polarization time constant. The identification parameters include path resistance, instantaneous internal resistance, and polarization voltage.

7. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 6, characterized in that, The fault determination module includes: The candidate generation unit is used to generate the fault candidate based on the data validity flag, the path resistance, the instantaneous internal resistance, the polarization voltage, the load residual, and the current tracking residual. The fault candidates include path anomaly candidates, battery body anomaly candidates, charging power supply anomaly candidates, low-voltage load anomaly candidates, and sampling link anomaly candidates. When the path resistance or the path hysteresis is outside the allowable range, a path anomaly candidate is obtained; When the instantaneous internal resistance, the polarization resistance, or the polarization time constant are not within the allowable range, a candidate for battery body abnormality is obtained; When the current tracking residual is greater than the upper limit of the allowable range, a candidate for power supply anomaly is obtained; When the load stability determination fails or the load residual exceeds the upper limit of the allowable range, a low-voltage load anomaly candidate is obtained. When the sampling channel self-test is abnormal or the data validity flag is invalid, a sampling link abnormality candidate is obtained.

8. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 7, characterized in that, The fault determination module includes: A fault verification unit is used to verify the fault candidate and obtain the fault verification result; When the path anomaly candidate exists, the current excitation unit and the path calculation unit are called back. If the path resistance or the path hysteresis obtained again is not within the allowable range and the deviation sign is the same, the path anomaly verification result is obtained as the fault verification result. When there is a candidate for the abnormality of the power supply, the abnormality verification result of the power supply is obtained based on whether the power supply can track the corresponding charging current target, and this is used as the fault verification result. When the sampling link anomaly candidate exists, independent verification is performed to obtain the sampling link anomaly verification result, which is then used as the fault verification result.

9. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 8, characterized in that, The fault determination module also includes: The power-off verification unit is used to reduce the output of the power supply to zero, obtain the discharge current and the residual voltage of the low-voltage battery terminal, and obtain the battery body abnormality verification result and the low-voltage load abnormality verification result as the fault verification result.

10. The on-board low-voltage battery management system supporting intelligent power replenishment and fault diagnosis according to claim 9, characterized in that, The power replenishment module includes a current limiting unit and a fault handling unit; A current limiting unit is used to obtain the upper limit of the battery-side current and the upper limit of the power supply-side current based on the path resistance, the instantaneous internal resistance and the polarization voltage when the identification parameter is valid, and the minimum of the two is taken as the allowable charging current. The fault handling unit is used to confirm the fault when the intersection of the set of fault candidates and the set of fault verification results is not empty, and writes the first charging current target, the second charging current target, the power supply restart permission, the unconfirmed fault candidates, the identification parameters, the measurement uncertainty, the current number of consecutive power replenishments, the amount of power replenished this time, and the effective zero-point calibration result into the data storage.

Images