A method and system for detecting short circuit of electrically controlled MOS

Abstract

The present invention belongs to the field of brushless motor technology, and specifically relates to a method and system for detecting short circuits in an electrically controlled MOSFET. The present invention initializes the system and sets a dynamic threshold, sending extremely short pulses to the lower and upper MOSFETs of each phase in sequence while simultaneously shutting down all MOSFETs on the other side, thereby forming a specific current path. Low-side current sampling resistors or Hall sensors are used to capture the bus current peak in real time and compare it with the dynamic threshold to accurately determine whether a short circuit fault exists. This solves the problem of high misjudgment rates caused by asymmetric grounding of the motor neutral point or parasitic capacitance in existing phase voltage floating detection methods. This method directly relies on current signals to reflect physical short circuit conditions, avoiding the uncertainty of floating potentials in traditional voltage methods and improving detection accuracy and reliability. Phase-by-phase scanning prevents multi-phase interference and enhances fault location accuracy.

Description

Technical Field

[0001] The present invention belongs to the technical field of brushless motors, and in particular relates to a method and system for detecting short circuits in electrically controlled MOS devices. Background Art

[0002] In brushless motor control systems, MOSFET lower-side short circuits (such as body diode breakdown or gate shoot-through) are common hardware faults. These faults can cause busbar short circuits, device burnout, or even system fires at power-up. Current detection methods for lower-side short circuits rely primarily on static diagnostics based on phase voltage floating detection. Specifically, at power-up, all MOSFETs are turned off, and the voltage at each phase terminal is measured using an ADC. If a phase lower-side short circuit occurs, its terminal voltage should be close to ground (0V), while normally the terminal voltage is floating. However, this method has significant drawbacks: it relies heavily on a balanced motor neutral point. If the motor neutral point is not symmetrically grounded or has parasitic capacitance, the floating voltage may approach 0V, leading to false positives.

[0003] The main problem with existing technologies is that methods based on phase voltage floating detection are susceptible to asymmetric grounding of the motor neutral point or parasitic capacitance, resulting in a high rate of false positives. Furthermore, this method cannot accurately distinguish between true short-circuit faults and false positives caused by variations in electrical characteristics, reducing system reliability and safety. Summary of the Invention

[0004] The present invention aims to provide a method and system for detecting electrically controlled MOSFET short-circuit failures, resolving the high false positive rate inherent in existing technologies due to reliance on floating phase voltage detection. By employing active current injection combined with real-time current monitoring, this method not only efficiently and accurately detects MOSFET short-circuit failures but also avoids the risk of false positives associated with voltage-based methods, significantly improving system reliability and safety.

[0005] To achieve the above object, the present invention adopts the following technical solution: a method for detecting an electrically adjustable MOS short circuit, comprising the following steps:

[0006] Initialize the system and set the dynamic threshold, which is calculated based on bus voltage, motor inductance and pulse width;

[0007] Pulses are sent to the lower-bridge MOSFET of each phase in sequence, while all upper-bridge MOSFETs are turned off simultaneously. If the upper-bridge MOSFET of a phase is short-circuited, a current path is formed short-circuiting the upper bridge, lower bridge, and ground. The bus current peak is captured in real time through a low-side current sampling resistor or Hall sensor. If the current peak exceeds the dynamic threshold, the upper-bridge MOSFET of the corresponding phase is determined to be short-circuited.

[0008] Pulses are sent to the upper bridge MOSFET of each phase in turn, and all lower bridge MOSFETs are turned off at the same time. If the lower bridge MOSFET of a phase is short-circuited, a current path is formed from the upper bridge, the short-circuited lower bridge, and the ground. The bus current peak is captured in real time through a low-side current sampling resistor or Hall sensor. If the current peak exceeds the dynamic threshold, it is determined that the lower bridge MOSFET of the corresponding phase is short-circuited.

[0009] Preferably, the calculation formula of the dynamic threshold is:

[0010] in, is the bus voltage, is the pulse width, is the motor inductance; pulse width is 5 microseconds, and the bus voltage The range is 40-50 volts, the motor inductance The range is 50-200 microhenries.

[0011] Preferably, the width of the pulse is 5 microseconds, and only one phase lower bridge MOSFET or upper bridge MOSFET is turned on each time, and the lower bridge MOSFET or upper bridge MOSFET of other phases remains in the off state; the triggering time of the pulse is synchronized with the current sampling time.

[0012] Preferably, the real-time capture of the bus current peak value by a low-side current sampling resistor or a Hall sensor includes:

[0013] The current peak is processed by a sliding average filter or a digital filter; the window length of the sliding average filter is 3 sampling periods, and the cutoff frequency of the digital filter is 10 kHz.

[0014] Preferably, the triggering time of the pulse is synchronized with the current sampling time, including:

[0015] Synchronization is achieved through the trigger signal of the hardware timer or PWM controller to ensure that the timing error between pulse injection and current sampling is less than 1 microsecond.

[0016] Preferably, the method further includes: if the current peak values ​​measured three times in succession all exceed the dynamic threshold, determining that the corresponding phase MOSFET is short-circuited; wherein the time interval between the three measurements is 10 milliseconds, and the pulse parameters of each measurement are consistent.

[0017] Preferably, the method further includes: dynamically adjusting the pulse width or the estimated value of the motor inductance when the bus voltage fluctuates in real time to optimize the dynamic threshold, specifically including: collecting the bus voltage in real time through the ADC module Update the motor inductance according to the preset motor inductance model An estimate of and Recalculate the dynamic threshold.

[0018] Preferably, the method also includes: cross-validation in combination with the sampling data of the low-side current sampling resistor and the Hall sensor, specifically including: fusing the two-way sampling data through a weighted average algorithm, and dynamically adjusting the weight coefficient according to the stability of the historical data; if the difference between the two-way data exceeds the preset tolerance range, the sampling circuit self-test program is triggered.

[0019] In another aspect, the present invention provides a system for detecting short circuits in an electrically adjustable MOS, comprising:

[0020] PWM controller, used to send pulses to the lower bridge MOSFET and upper bridge MOSFET of each phase in sequence;

[0021] Current sampling module, including low-side current sampling resistor and Hall sensor, used to capture bus current peak value in real time;

[0022] a comparator for comparing the current peak value with a dynamic threshold;

[0023] a processing unit, configured to determine whether the corresponding phase MOSFET is short-circuited based on the comparison result;

[0024] Memory, used to record historical detection data;

[0025] An alarm module, used to trigger a fault alarm when a short circuit is detected;

[0026] Safety shutdown module, used to turn off all MOSFETs and enter safety mode when a short circuit is detected.

[0027] Technical effects and advantages of the present invention: Compared with the prior art, the method and system for detecting short circuits in electrically controlled MOS devices proposed in the present invention have the following advantages:

[0028] The present invention initializes the system and sets dynamic thresholds, sending extremely short pulses to each phase's lower and upper MOSFETs in sequence while simultaneously shutting down all MOSFETs on the other side, forming a specific current path. Using a low-side current sampling resistor or Hall effect sensor to capture the busbar current peak in real time and compare it with the dynamic threshold, it accurately determines whether a short-circuit fault exists. This solves the high misjudgment rate caused by asymmetric grounding of the motor neutral point or parasitic capacitance in existing phase voltage floating detection methods. This method directly relies on current signals to reflect physical short-circuit conditions, avoiding the uncertainty of floating potentials in traditional voltage methods and improving detection accuracy and reliability. The phase-by-phase scanning method prevents multiphase interference and enhances fault location accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a flow chart of the method for detecting short circuits in an electrically adjustable MOS according to the present invention;

[0030] Figure 2 This is a block diagram of the system for electrically adjusting MOS short circuit detection of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0032] The present invention provides Figure 1 A method for detecting short circuits in an electrically controlled MOS is shown, comprising the following steps:

[0033] Initialize the system and set the dynamic threshold. The dynamic threshold is calculated based on the bus voltage, motor inductance, and pulse width. The calculation formula for the dynamic threshold is: in, is the bus voltage, is the pulse width, is the motor inductance; pulse width is 5 microseconds, and the bus voltage The range is 40-50 volts, the motor inductance The range is 50-200 microhenries.

[0034] Furthermore, when the bus voltage fluctuates in real time, the pulse width or the estimated value of the motor inductance is dynamically adjusted to optimize the dynamic threshold, specifically including: collecting the bus voltage in real time through the ADC module Update the motor inductance according to the preset motor inductance model An estimate of and Recalculate the dynamic threshold.

[0035] Send pulses to the lower bridge MOSFET of each phase in turn, and turn off all the upper bridge MOSFETs at the same time. If the upper bridge MOSFET of a phase is short-circuited, a current path is formed that shorts the upper bridge, lower bridge, and ground;

[0036] The bus current peak is captured in real time through a low-side current sampling resistor or Hall sensor. If the current peak exceeds the dynamic threshold, the corresponding phase upper bridge MOSFET is determined to be short-circuited.

[0037] Send pulses to each phase's upper bridge MOSFET in turn, and turn off all lower bridge MOSFETs at the same time. If a phase's lower bridge MOSFET is short-circuited, a current path is formed from the upper bridge, the short-circuited lower bridge, and the ground.

[0038] The bus current peak is captured in real time through a low-side current sampling resistor or a Hall sensor. If the current peak exceeds the dynamic threshold, it is determined that the corresponding phase lower bridge MOSFET is short-circuited.

[0039] The width of the above-mentioned pulse is 5 microseconds, and only one phase lower bridge MOSFET or upper bridge MOSFET is turned on at a time, and the lower bridge MOSFET or upper bridge MOSFET of other phases remains in the off state; the trigger time of the pulse is synchronized with the current sampling time; further, synchronization is achieved through the trigger signal of the hardware timer or PWM controller to ensure that the timing error between pulse injection and current sampling is less than 1 microsecond.

[0040] The bus current peak is captured in real time through a low-side current sampling resistor or a Hall sensor, including: processing the current peak through a sliding average filter or a digital filter; the window length of the sliding average filter is 3 sampling cycles, and the cutoff frequency of the digital filter is 10 kHz.

[0041] The method further includes: if the current peak values ​​measured three times in succession all exceed the dynamic threshold, determining that the corresponding phase MOSFET is short-circuited; wherein the time interval between the three measurements is 10 milliseconds, and the pulse parameters of each measurement are consistent.

[0042] The method further includes: performing cross-validation on the sampling data of the low-side current sampling resistor and the Hall sensor, specifically including: fusing the two sampling data through a weighted average algorithm, and dynamically adjusting the weight coefficient according to the stability of the historical data; if the difference between the two data exceeds a preset tolerance range, triggering a sampling circuit self-test program.

[0043] In this embodiment, a system for detecting short circuits of an electrically adjustable MOS is also provided. Figure 2 ,include:

[0044] PWM controller, used to send pulses to the lower bridge MOSFET and upper bridge MOSFET of each phase in sequence;

[0045] Current sampling module, including low-side current sampling resistor and Hall sensor, used to capture bus current peak value in real time;

[0046] a comparator for comparing the current peak value with a dynamic threshold;

[0047] a processing unit, configured to determine whether the corresponding phase MOSFET is short-circuited based on the comparison result;

[0048] Memory, used to record historical detection data;

[0049] An alarm module, used to trigger a fault alarm when a short circuit is detected;

[0050] Safety shutdown module, used to turn off all MOSFETs and enter safety mode when a short circuit is detected.

[0051] In addition, when executed, each of the above modules is also used to implement other steps of the above-mentioned method for detecting short circuits in an electrically adjustable MOS, as follows:

[0052] Step 1: Initialize the system and set parameters

[0053] Before starting the brushless motor control system, the system initialization needs to be completed, including configuring the PWM control logic, setting the current sampling circuit parameters, and calculating the dynamic threshold. The determination is based on the bus voltage , motor inductance and pulse width , the formula is:

[0054] in, is the bus voltage (unit: volts), is the pulse width (unit: seconds), is the motor inductance (in henries). For example, if the bus voltage is 48 volts, the pulse width is set to 5 microseconds, and the motor inductance is 100 microhenries, the dynamic threshold is: This threshold is used to determine whether a short circuit exists. Furthermore, the PWM controller's duty cycle must be initialized to 0% to ensure that all MOSFETs are off, preventing short circuits caused by misoperation during power-up.

[0055] Step 2: Actively stimulate the fault path and send a pulse to the lower bridge

[0056] After the system is initialized, pulses (e.g., 5 microseconds) are sent to the lower bridge MOSFET of each phase in turn, while keeping all upper bridge MOSFETs off. During this process, if a phase upper bridge MOSFET has a short circuit problem, a current path of "short circuit bridge-lower bridge-ground" will be formed, causing the bus current to Rapidly rise. The current peak is captured in real time through a low-side current sampling resistor or Hall sensor, and compared with the dynamic threshold calculated in the first step. For example, if the monitored bus current is 3A, which exceeds the threshold of 2.4A, it indicates that there may be a short circuit.

[0057] If the pulse is too long, the injected energy may be too large, thus damaging the MOSFET; if the pulse is too short, it may not trigger enough current for detection. and bus voltage Dynamically adjust pulse width To ensure the accuracy and safety of detection. In addition, the phase-by-phase scanning method avoids coupling interference between multiple phases and improves the accuracy of fault location.

[0058] Step 3: Actively stimulate the fault path and send pulses to the upstream bridge

[0059] After confirming that the upper bridge MOSFET is not short-circuited, the next step is to send a pulse (e.g., 5 microseconds) to each phase’s upper bridge MOSFET while keeping all lower bridge MOSFETs off. If a phase’s lower bridge MOSFET is short-circuited, a current path of “upper bridge - short-circuited lower bridge - ground” will be formed, which will also cause a sharp increase in bus current. By monitoring the bus current in real time and comparing it with the dynamic threshold By comparing, it can be determined whether there is a short circuit fault.

[0060] The key to this step lies in pulse synchronization and isolation. Since only one phase's high-side MOSFET is turned on at a time, the high-side MOSFETs of the other phases must remain off to avoid the risk of short circuits caused by multiple phases being turned on simultaneously. Furthermore, the pulse triggering time must be strictly synchronized with the current sampling time to reduce measurement errors. For example, if the measured current peak is 2.7A during a test cycle, exceeding the 2.4A threshold, this indicates a short circuit fault in the corresponding phase.

[0061] Step 4: Real-time current monitoring and fault determination

[0062] Based on the operations in the second and third steps, a low-side current sampling resistor or Hall sensor is used to monitor the current change in real time. When the monitored current peak exceeds the dynamic threshold When the current peak exceeds 2.4A for three consecutive measurements, a short circuit fault is detected.

[0063] Use a low-side current sampling resistor or Hall sensor to monitor current changes in real time. The specific method is as follows:

[0064] Use low-side current sampling resistor: When current flows through the low-side current sampling resistor According to Ohm's law, the voltage drop Expressed as: By measuring , combined with the known The actual bus current can be calculated by .

[0065] Use Hall sensor: Hall sensor directly senses the change of magnetic field strength to measure current. Its output signal With the measured current Proportional, proportional coefficient Depending on the specific model, the relationship is: ; By reading , and use The actual bus current can be obtained by conversion .

[0066] When the detected current peak exceeds the dynamic threshold To further verify the reliability of the results, the above steps can be repeated multiple times to ensure data consistency.

[0067] The key to this step lies in the accuracy and response speed of current sampling. The low-side current sampling resistor must have high precision and low thermal drift to reduce measurement errors. Furthermore, the Hall effect sensor's bandwidth must be wide enough to capture microsecond-level current changes. Furthermore, a filtering algorithm must be introduced to eliminate noise interference, such as using a sliding average filter or digital filter to process the sampled data to improve detection stability.

[0068] The low-side current sampling resistor must have high precision and low thermal drift to reduce measurement errors. Furthermore, the Hall effect sensor's bandwidth must be wide enough to capture microsecond-level current changes. Furthermore, a filtering algorithm must be introduced to eliminate noise interference. For example, a sliding average filter or digital filter must be used to process the sampled data to improve detection stability. The following describes the specific steps for these two methods.

[0069] Moving average filter:

[0070] The sliding average filter is a simple and effective digital filtering technique that smooths the signal and reduces noise interference by calculating the average value of a series of consecutive sampling points. The following are the specific steps to use the sliding average filter:

[0071] 1. Determine the window length: Select an appropriate window length N (usually 3 sampling periods), that is, take N consecutive current sampling values ​​for averaging each time.

[0072] 2. Initialize the accumulator: Before starting filtering, initialize an accumulator variable It is 0 and is used to store the sum of all sample values ​​in the current window.

[0073] 3. Cumulative sampling value: For each new sampling value , adding it to the accumulator: ;

[0074] 4. Remove old sample values: When the window moves, subtract the oldest sample value from the accumulator ;

[0075] 5. Calculate the average: the updated average The current accumulator value is divided by the window length: ; This method can effectively remove high-frequency noise while retaining the main trend of the signal.

[0076] Digital filter:

[0077] Digital filters provide more flexible options and can be used to design filters with specific frequency responses as needed. Here, a low-pass filter is used as an example, whose purpose is to allow frequencies below a certain cutoff frequency to be The following are the design steps for an IIR (Infinite Impulse Response) low-pass filter:

[0078] 1. Select cutoff frequency: Set the cutoff frequency according to actual needs For example, in this case, a cutoff frequency of 10kHz can be selected.

[0079] 2. Determine the filter coefficients: Based on the selected cutoff frequency and sampling rate , calculate the filter coefficients. For a first-order IIR low-pass filter, the transfer function is expressed as: ;in, is the filter coefficient, which is related to the cutoff frequency and the sampling rate and is calculated by the following formula: ;

[0080] 3. Implement the filtering process: For each new input sample , output Calculate according to the following recursive formula: ; is the filter output at the previous moment.

[0081] The two filtering methods described above significantly improve the stability and accuracy of current sampling, thereby enhancing the reliability of the entire system for MOS short-circuit detection. While sliding average filters are suitable for removing random noise, digital filters are more suitable for precisely controlling signal components within a frequency band. The choice of method depends on the specific project requirements and system design goals.

[0082] Step 5: Troubleshooting and system recovery

[0083] Once the specific MOSFET short circuit fault location is identified, the next step is to take appropriate measures to address the fault and attempt to restore normal operation. For minor short circuits, this may simply require disconnecting the power supply for a period of time to allow the device to cool. For more severe short circuits, however, the damaged component may need to be replaced. After the fault is resolved, restart the system and repeat the above steps to confirm that the system has fully recovered.

[0084] For example, if a short circuit is detected in the lower-bridge MOSFET of Phase A, the system first removes power, replaces the damaged MOSFET, and then restarts the system, repeating the previous four steps to ensure the newly installed MOSFET is functioning properly. Furthermore, redundant mechanisms, such as automatic switching to a backup MOSFET after a short circuit is detected, are required to maintain continuous system operation.

[0085] The specific composition of the redundancy mechanism in the electronically controlled MOS short-circuit detection system mainly focuses on the following aspects, aiming to improve the reliability and safety of the system and ensure that the system can still operate normally or enter a safe mode to avoid greater damage even if some components fail.

[0086] Redundant current sampling path:

[0087] 1. Dual Current Monitoring: Bus current is monitored simultaneously using both a low-side current sampling resistor and a Hall effect sensor. This not only improves current measurement accuracy but also reduces errors that may be introduced by a single measurement method through cross-validation.

[0088] Low-side current sampling resistor: used to directly measure the current flowing through the circuit and calculate the actual current value using Ohm's law.

[0089] Hall sensor: uses the principle of magnetic field induction to indirectly measure current, suitable for high-precision and fast-response applications.

[0090] 2. Data fusion and processing: The two sets of collected current data are fused using digital signal processing algorithms (such as weighted averaging). The weight coefficients are dynamically adjusted based on the stability of historical data to optimize the reliability of the final output.

[0091] 3. Self-check routine: An automated check function is designed to periodically verify the integrity of the current sampling path. For example, this can be done by injecting a known current into the system and comparing the expected current with the actual measured current to confirm that the sampling circuit is functioning properly.

[0092] Fault recording and self-learning mechanism:

[0093] 1. Historical Data Analysis: The system records the results of each test (including but not limited to current peak values, pulse parameters, bus voltage, etc.) and stores them in non-volatile memory. Analysis of this data can identify potential patterns in short-circuit faults, such as fault frequency under specific load conditions or temperature-related failure modes.

[0094] 2. Parameter optimization: Based on the results of the above historical data analysis, the system is able to dynamically adjust some key parameters, such as pulse width or threshold, to adapt to motor aging or environmental changes, thereby extending the service life of the system and improving detection accuracy.

[0095] Safety protection and emergency shutdown function:

[0096] 1. Rapid response measures:

[0097] Immediately turn off all MOSFETs: Once a short-circuit condition is detected, the system quickly switches off all high-side and low-side MOSFETs to interrupt the fault current path.

[0098] Trigger alarm notification: Alarms are issued to users through LED indicators, buzzers, or communication interfaces (such as CAN bus) to promptly convey fault information.

[0099] Switch to backup options: If the system design supports redundancy, switch to the backup MOSFET or limit the motor output power to maintain basic operating functions.

[0100] Log record: Detailed record of fault time, current peak value, pulse parameters and other information to facilitate subsequent fault diagnosis and maintenance work.

[0101] Through the construction of the above aspects, this redundancy mechanism can greatly improve the robustness and safety of the system while ensuring the efficient operation of the electronically controlled MOS short-circuit detection system, ensuring that various possible problems can be effectively prevented and responded to even under adverse conditions.

[0102] Step 6: Introducing a dynamic adaptive threshold adjustment mechanism

[0103] In traditional methods, dynamic threshold Usually based on a fixed bus voltage , motor inductance and pulse width However, in actual applications, the bus voltage may change due to load fluctuations or power instability, resulting in misjudgment of the fixed threshold. To this end, this embodiment uses a dynamic adaptive threshold adjustment mechanism to monitor the bus voltage in real time. changes and dynamically adjusts the pulse width or motor inductance To ensure that the threshold Always reflects the actual status of the current system.

[0104] Specifically, the system can collect bus voltage before each detection The instantaneous value of the motor inductance is combined with the preset Model, dynamically calculate new thresholds For example, if the bus voltage drops from 48V to 45V, and the motor inductance If it remains unchanged, the new threshold is: ;

[0105] This mechanism can effectively deal with the impact of bus voltage fluctuations and improve the robustness of detection. In addition, if the motor inductance Due to aging or temperature changes, the system can be updated in real time through online identification algorithms (such as least squares method or Kalman filter) The estimated value of , thereby further optimizing the accuracy of the threshold.

[0106] Step 7: Multi-stage pulse injection and cross-validation

[0107] To further improve detection accuracy, this embodiment introduces a multi-stage pulse injection and cross-validation strategy. Traditional methods typically use a single pulse injection for detection, but improper selection of pulse width or voltage can lead to misjudgment. To address this issue, the system performs multiple injections with pulses of varying parameters at different stages, and uses cross-validation to determine whether a short circuit fault is present.

[0108] For example, the first phase uses a 5μs, 48V pulse for testing. If no abnormality is found, the system proceeds to the second phase, using a 10μs, 45V pulse. If the two test results are consistent, the system is confirmed to be normal. If the results are inconsistent, a comprehensive assessment is necessary, combining historical data or introducing other detection methods (such as voltage float detection). Furthermore, by comparing current response curves under different pulse parameters, characteristic patterns of short-circuit faults, such as differences in current rise slope or peak value, can be identified, thereby improving detection reliability.

[0109] Step 8: Introduce redundant current sampling paths to improve anti-interference capabilities

[0110] In complex electromagnetic environments, the current sampling circuit may be affected by external noise or parasitic capacitance, resulting in measurement errors. To address this, this embodiment designs a redundant current sampling path, using both a low-side current sampling resistor and a Hall effect sensor for cross-validation to reduce the uncertainty of a single sampling path.

[0111] For example, during the detection process, the system can simultaneously collect the voltage signal from the low-side current sampling resistor and the magnetic induction signal from the Hall effect sensor, and fuse the two data streams through digital signal processing algorithms (such as weighted averaging or differential filtering) to improve anti-interference capabilities. In addition, a self-check mechanism can be designed to regularly verify the integrity of the sampling path, for example by injecting a known current and comparing the sampling results to determine whether there is a sampling circuit fault.

[0112] The specific composition of the self-check mechanism:

[0113] 1. Regularly verify sampling path integrity

[0114] Injecting a known current: The system periodically injects a known current into the circuit. This current value is pre-set and should be large enough to be accurately measured by the low-side current sampling resistor and Hall sensor.

[0115] Comparing expected and actual measurements: After injecting a known current, the system compares the actual measurement obtained through the low-side current sensing resistor and Hall effect sensor with the expected value. If the difference between the two is outside the preset tolerance range, it indicates a possible sampling circuit failure.

[0116] 2. Weighted average algorithm for data fusion

[0117] Dynamically adjust weighting factors: Based on the stability of historical data, the system can dynamically adjust the weighting factors of the two sampling data sources (i.e., the data from the low-side current sampling resistor and the data from the Hall effect sensor). For example, if a sensor has demonstrated higher stability and accuracy over a period of time, the data from that sensor will receive a greater weight in the weighted averaging process.

[0118] Handling discrepancies between two streams of data: When the difference between two streams of data exceeds a preset tolerance, the system not only triggers an alarm to notify the user but also initiates further diagnostic procedures to determine the cause of the problem.

[0119] 3. Trigger the sampling circuit self-test program

[0120] Self-test start conditions: Once a significant discrepancy is detected between the two data paths, the system will automatically initiate a self-test of the sampling circuit. This includes, but is not limited to, a comprehensive check of circuit connection status and component functionality.

[0121] Detailed inspection steps:

[0122] Check that all connections are secure to avoid measurement errors caused by poor contact.

[0123] Test the functions of various electronic components, such as confirming whether the resistance of the low-side current sampling resistor meets the specification requirements and whether the Hall sensor is working normally.

[0124] If a problem is detected, the system logs detailed error information and attempts to take corrective action, such as recalibrating the affected component or prompting the user to perform maintenance.

[0125] This self-checking mechanism not only effectively detects and locates potential problems in the sampling circuit, but also enables timely resolution, ensuring the long-term stable operation of the entire electronically controlled MOS short-circuit detection system. Furthermore, this mechanism enhances the system's robustness, enabling it to maintain high-precision current monitoring capabilities in complex and changing operating environments.

[0126] Step 9: Fault Recording and Self-Learning Mechanism

[0127] To improve the long-term reliability of the system, this embodiment incorporates a fault logging and self-learning mechanism. The system records the results of each test (such as current peak, pulse parameters, bus voltage, etc.) and stores them in non-volatile memory. By analyzing historical data, the system can identify potential patterns in short-circuit faults, such as fault frequency under specific load conditions or temperature-related failure modes.

[0128] In order to identify the potential patterns of short circuit faults by analyzing historical data, follow these specific steps:

[0129] 1. Data collection and initialization

[0130] Data recording: The system records current peak value, pulse parameters, bus voltage and other data during each detection process, and stores this information in non-volatile memory.

[0131] Timestamp: Each record is timestamped, allowing subsequent analysis to accurately correlate environmental conditions (such as temperature) and other factors that may affect system performance.

[0132] 2. Data Preprocessing

[0133] Data cleaning: Clean the collected data to remove outliers or erroneous data. For example, if a measurement shows a current peak that is significantly out of the normal range and cannot be explained, consider deleting the record or further investigating the cause.

[0134] Format conversion: Ensure that all data is stored in a unified format to facilitate subsequent processing and analysis. This may include unit conversion (such as converting time to seconds) and normalization of values.

[0135] 3. Historical Data Analysis

[0136] Trend analysis: Use statistical methods to analyze data trends over time. For example, observe whether the current peak increases as the motor runs longer, which may be due to component aging.

[0137] Pattern Recognition:

[0138] Load correlation analysis: Analyze the frequency of short-circuit faults under specific load conditions. For example, whether faults occur more frequently under high load than under low load;

[0139] Temperature sensitivity assessment: Evaluate the impact of ambient temperature changes on the short-circuit failure rate. If a significant increase in failure rate is observed in high-temperature environments, consider improving the heat dissipation design or adjusting the threshold settings.

[0140] Periodic behavior detection: Using Fourier transform or other frequency domain analysis tools, we can look for periodic patterns in fault occurrence. For example, certain faults may recur at regular intervals, which can help predict future faults.

[0141] 4. Parameter Optimization

[0142] Dynamically adjust detection parameters: Based on the above analysis results, automatically adjust some key parameters of the system to optimize the detection process. For example:

[0143] If it is found that the current peak under certain load conditions is close to but does not exceed the dynamic threshold, it can be appropriately reduced to increase the detection sensitivity.

[0144] The calculated dynamic threshold is updated in real time based on the actual bus voltage fluctuations; this ensures accurate detection of short-circuit faults even in the presence of voltage fluctuations.

[0145] 5. System maintenance suggestion generation

[0146] Alarm mechanism: When the system detects an abnormal pattern or predicts an impending failure, the alarm mechanism is triggered, sending a notification to the user to remind them to perform necessary inspections and maintenance.

[0147] Maintenance reports: Generate regular maintenance reports summarizing the operating status over time, any problems detected, and their resolution. The reports should also include recommendations for future operations, such as when to replace specific components or adjust operating parameters.

[0148] 6. Implementation and Feedback Loop

[0149] Implement recommendations: Take actions based on the recommendations generated by the system, such as adjusting inspection parameters or replacing aging components.

[0150] Effectiveness evaluation: After implementation, collect data again and evaluate the effectiveness of the new strategy to confirm whether it effectively reduces the failure rate or improves detection accuracy.

[0151] Continuous optimization: Establish a feedback loop to continuously adjust and optimize the system's behavior based on the latest data to ensure it is always in optimal condition.

[0152] Through the above steps, not only can the potential patterns of short-circuit faults be identified, but measures can also be taken in time to prevent the occurrence of faults, thereby greatly improving the reliability of the electronically controlled MOS short-circuit detection system.

[0153] In addition, the system can optimize detection parameters based on historical data, such as dynamically adjusting pulse width or threshold To adapt to motor aging or environmental changes. For example, if the system detects a significant increase in the incidence of short-circuit faults in a high-temperature environment, it can automatically increase the pulse width. To improve detection sensitivity, or adjust the threshold This mechanism not only improves the adaptability of detection but also extends the service life of the system.

[0154] Step 10: Integrate safety protection and emergency shutdown functions

[0155] After detecting a short circuit fault, the system needs to take safety measures immediately to prevent further damage. To this end, this embodiment integrates safety protection and emergency shutdown functions, including the following steps:

[0156] Quickly turn off all MOSFETs: Once a short circuit is detected, the system immediately turns off all high-side and low-side MOSFETs to cut off the fault current path.

[0157] Trigger fault alarm: Alert the user via LED indicator, buzzer or communication interface (such as CAN bus) to indicate the fault location and type.

[0158] Entering safe mode: If the system supports a redundant design, it can switch to the backup MOSFET or limit the motor output power to maintain basic functions.

[0159] Record fault logs: Store fault occurrence time, current peak value, pulse parameters and other information in the system log for subsequent analysis and maintenance.

[0160] For example, if the lower-bridge MOSFET of phase A shorts, the system first shuts down all MOSFETs, then illuminates a red LED and sounds an alarm. Finally, it logs the fault and enters safe mode, allowing only phases B and C to operate at a reduced rating. This feature not only improves system safety but also provides a valuable basis for subsequent fault diagnosis and repair.

[0161] This embodiment significantly improves the accuracy and reliability of MOSFET short-circuit detection through active fault path stimulation, dynamic adaptive threshold adjustment, multi-stage pulse injection, redundant current sampling, fault recording and self-learning mechanism, and safety protection functions. Specifically:

[0162] By triggering the short-circuit path with a short pulse, the current signal is directly used to reflect the physical short-circuit condition, avoiding the uncertainty of floating potentials in traditional voltage methods. Real-time threshold adjustment adapts to bus voltage fluctuations and motor parameter changes, improving detection robustness. Combining a low-side current sampling resistor with a Hall effect sensor reduces noise interference and improves interference immunity.

[0163] This method requires no specialized hardware, only standard current sampling circuits and PWM control logic, and is applicable to brushless motor controllers of any topology. Through phase-by-phase scanning and cross-validation, it achieves independent detection of all phases, accurately localizing fault locations while reducing false positives. Furthermore, dynamic adjustment of pulse parameters and thresholds enables the system to adapt to different operating conditions, improving its versatility and cost-effectiveness.

[0164] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting short circuits in electrically controlled MOS, characterized in that: The following steps are involved: Initialize the system and set the dynamic threshold, which is calculated based on bus voltage, motor inductance and pulse width; Send pulses to the lower bridge MOSFET of each phase in turn, and turn off all the upper bridge MOSFETs at the same time. If the upper bridge MOSFET of a phase is short-circuited, a current path is formed that shorts the upper bridge, lower bridge, and ground; The bus current peak is captured in real time through a low-side current sampling resistor or Hall sensor. If the current peak exceeds the dynamic threshold, the corresponding phase upper bridge MOSFET is determined to be short-circuited. Send pulses to each phase's upper bridge MOSFET in turn, and turn off all lower bridge MOSFETs at the same time. If a phase's lower bridge MOSFET is short-circuited, a current path is formed from the upper bridge, the short-circuited lower bridge, and the ground. The bus current peak is captured in real time through a low-side current sampling resistor or a Hall sensor. If the current peak exceeds the dynamic threshold, it is determined that the corresponding phase lower bridge MOSFET is short-circuited.

2. The method for detecting short circuit of an electrically controlled MOS according to claim 1, wherein: The calculation formula of the dynamic threshold is: ; in, is the bus voltage, is the pulse width, is the motor inductance; pulse width is 5 microseconds, and the bus voltage The range is 40-50 volts, the motor inductance The range is 50-200 microhenries.

3. The method for detecting short circuit of an electrically controlled MOS according to claim 1, wherein: The pulse width is 5 microseconds, and only one phase lower bridge MOSFET or upper bridge MOSFET is turned on each time, while the lower bridge MOSFETs or upper bridge MOSFETs of other phases remain in the off state; the triggering time of the pulse is synchronized with the current sampling time.

4. The method for detecting short circuit of an electrically controlled MOS according to claim 1, wherein: The method of capturing the bus current peak value in real time through the low-side current sampling resistor or the Hall sensor includes: The current peak is processed by a sliding average filter or a digital filter; the window length of the sliding average filter is 3 sampling periods, and the cutoff frequency of the digital filter is 10 kHz.

5. The method for detecting short circuit of an electrically controlled MOS according to claim 3, wherein: The triggering time of the pulse is synchronized with the current sampling time, including: Synchronization is achieved through the trigger signal of the hardware timer or PWM controller to ensure that the timing error between pulse injection and current sampling is less than 1 microsecond.

6. The method for detecting short circuit of an electrically controlled MOS according to claim 1, wherein: The method further comprises: If the current peak values ​​measured three times in a row all exceed the dynamic threshold, the corresponding phase MOSFET is determined to be short-circuited; wherein the time interval between the three measurements is 10 milliseconds, and the pulse parameters of each measurement are consistent.

7. The method for detecting short circuit of an electrically controlled MOS according to claim 2, wherein: The method further comprises: When the bus voltage fluctuates in real time, the pulse width or estimated value of the motor inductance is dynamically adjusted to optimize the dynamic threshold, including: Real-time acquisition of bus voltage through ADC module The instantaneous value of Update the motor inductance according to the preset motor inductance model estimated value of; Based on the updated and Recalculate the dynamic threshold.

8. The method for detecting short circuit of an electrically controlled MOS according to claim 1, wherein: The method further comprises: Cross-validation is performed using the sampling data of the low-side current sampling resistor and the Hall sensor, specifically including: The two sampling data are fused through the weighted average algorithm, and the weight coefficient is dynamically adjusted according to the stability of historical data; If the difference between the two data channels exceeds the preset tolerance range, the sampling circuit self-test procedure is triggered.

9. A system for implementing the method according to any one of claims 1 to 8 for detecting short circuits in electrically adjustable MOS, characterized in that: include: PWM controller, used to send pulses to the lower bridge MOSFET and upper bridge MOSFET of each phase in sequence; Current sampling module, including low-side current sampling resistor and Hall sensor, used to capture bus current peak value in real time; a comparator for comparing the current peak value with a dynamic threshold; The processing unit is used to determine whether the corresponding phase MOSFET is short-circuited according to the comparison result.

10. The system for detecting short circuits of electrically adjustable MOS according to claim 9, characterized in that: The system further comprises: Memory, used to record historical detection data; An alarm module, used to trigger a fault alarm when a short circuit is detected; Safety shutdown module, used to turn off all MOSFETs and enter safety mode when a short circuit is detected.

Images