A motor drive system health monitoring method and system based on discharge time

Abstract

The application provides a motor drive system health monitoring method and system based on discharge time, which utilizes the discharge characteristics of the DC bus capacitor after the motor drive system is turned off, measures the discharge time, i.e. the time required for the DC bus voltage to decrease from a first fixed value to a second fixed value, as a health indicator, and realizes the evaluation of the overall health state of the system. This method does not require additional hardware investment, only relies on the existing DC bus voltage information in the system, has a simple calculation process, does not require complex algorithms, and is monitored after the system is disabled, without disturbing the normal operation. Experimental verification shows that this method has high sensitivity to the degradation of the DC bus capacitor and power semiconductor devices, can effectively reflect the change of the health state of the system, and is suitable for health management and predictive maintenance of motor drive systems in the fields of electric vehicles, industrial drives and aerospace.

Description

Technical Field

[0001] This invention belongs to the field of power electronic system status and health monitoring, and particularly relates to a method and system for health monitoring of motor drive systems based on discharge time. Background Technology

[0002] Motor drive systems play a crucial role in applications such as automotive, rail, and aerospace, where zero-failure operation is essential to ensure the safety of people and property. However, due to long-term operation and environmental stress, the DC bus capacitors and power semiconductor devices in motor drive systems can degrade, leading to unexpected failures. Therefore, monitoring the health status of these two critical reliability components is essential to improve system safety and availability and reduce maintenance costs.

[0003] In existing technologies, a significant amount of research has been devoted to implementing condition and health monitoring at the component level. Based on the consensus on the physical degradation mechanisms of DC bus capacitors and power semiconductor devices, many health indicators have been proposed, such as the equivalent series resistance (ESR) and capacitance value of DC bus capacitors, and the switching time and on-saturation voltage of power semiconductor devices. Correspondingly, various measurement or estimation methods for obtaining these health indicators have also been developed. These component-level monitoring schemes are highly effective in identifying the degradation of individual components, but their applicability in practical applications requires further consideration. In field applications, depending on the specific maintenance strategy, the entire motor drive system can be considered as the smallest unit for maintenance and replacement, without explicitly distinguishing which specific component in the motor drive system has degraded. Therefore, it is necessary to treat the motor drive system as a whole for health assessment to achieve more comprehensive health management.

[0004] Currently, several system-level health monitoring strategies have been proposed. Efficiency-based methods use the efficiency of the motor drive system as a health indicator, simultaneously monitoring the DC bus capacitor and power semiconductor devices within the system. When these components degrade, their power loss increases, leading to a decrease in the efficiency of the motor drive system. However, the system's DC input current and AC output voltage required for efficiency calculations are not existing information within the motor drive system, necessitating the addition of extra sensors and high-performance measurement equipment. Harmonic-based methods consider the sensitivity of harmonics in the controller output variables and system input voltage of the motor drive system to the degradation of power semiconductor modules and DC bus capacitors, using these harmonics as a health indicator. While harmonic-based methods do not involve additional sensors, hardware upgrades are needed to improve sampling rate and resolution for accurate harmonic signal capture. Complex spectrum analysis algorithms are also required, undoubtedly increasing monitoring costs and complexity. Furthermore, both efficiency-based and harmonic-based methods share a common problem: their sensitivity to component degradation is relatively low. This means that changes in health indicators caused by degradation are easily masked by noise and variable operating conditions, making accurate identification difficult. Algorithm-based methods inject excitation signals into system loops and then analyze the system response using specific identification algorithms to identify system degradation. These techniques are widely used in control optimization, and the algorithms can be based on predefined model structures or independent of specific models. Digital twins, as an emerging technology, are also a powerful tool for health monitoring. Some studies have extracted health indicators by building digital virtual models of motor drive systems and combining computational methods such as particle swarm optimization and genetic algorithms. With the rapid development of artificial intelligence, data-driven health monitoring methods have also gained widespread attention. These methods excel in solving classification and regression problems and are particularly suitable for scenarios where system models are difficult to construct accurately. These algorithm-based monitoring schemes can utilize existing hardware sensors in motor drive systems without additional hardware investment. However, these methods typically require high-speed, high-precision data sampling, and the computational burden of system modeling and artificial intelligence algorithms is considerable, increasing the complexity of implementation.

[0005] In summary, existing system-level health monitoring methods rely on hardware upgrades and complex algorithms in practical applications, and generally face challenges of high cost and complex implementation. Whether it is a component-level or system-level health monitoring method, it is performed in real time during normal system operation, which can interfere with normal operation due to additional hardware and software requirements, and bring about reliability issues. Summary of the Invention

[0006] Purpose of the invention: This invention provides a method and system for health monitoring of motor drive systems based on discharge time. It aims to utilize the discharge characteristics of the DC bus capacitor after the system is shut down, and to use the time required for the DC bus voltage to drop from one preset fixed value to another preset fixed value, i.e., discharge time, as a new health indicator. This achieves health status assessment without incremental hardware investment, without complex algorithms, and without interfering with the normal operation of the system.

[0007] Technical solution: This invention provides a health monitoring method for a motor drive system based on discharge time, comprising:

[0008] After the motor drive system stops running, the DC bus discharge operation mode is triggered;

[0009] Monitor the DC bus voltage change process and record the discharge time; the discharge time is the time from the DC bus voltage changing from a preset first fixed value. Reduced to the second fixed value Required time; first fixed value Below the minimum initial DC bus voltage, the second fixed value Higher than the discharge termination safety voltage; first fixed value and the second fixed value All are within the steady-state discharge range;

[0010] Discharge time data at different stages were collected, and probability density analysis, median filtering and per-unit processing were performed to obtain the median per-unit discharge time.

[0011] The health status of the system is assessed by comparing the median per-unit discharge time with the set end-of-life index. If the median per-unit discharge time is lower than the end-of-life index, the motor drive system needs maintenance.

[0012] Furthermore, the triggered DC bus discharge operation mode includes:

[0013] After the motor drive system finishes normal operation, the drive signal to the power semiconductor module is stopped, causing the motor drive current to drop rapidly.

[0014] Once the electronic control unit confirms that the motor speed and drive current have both dropped to zero, it isolates the motor drive system from the front-end power supply by disconnecting the circuit breaker.

[0015] The electronic control unit generates a set of discharge commands in the rotating reference coordinate system, including the q-axis current. and d-axis current This triggers the DC bus discharge operation mode.

[0016] Furthermore, the calculation of the discharge time includes:

[0017] One switching cycle during the discharge process Internally, it calculates the energy consumed by the entire power semiconductor module in the motor drive system. The formula is:

[0018] ;

[0019] in, Indicates power semiconductor devices and its corresponding freewheeling diode A general term; The conduction current of a power semiconductor device is represented by the electrical angle of the motor rotor. and d-axis current Decide; This represents the equivalent on-resistance of a power semiconductor device in its turn-on state. It increases with the degree of device degradation and is also related to the on-current of the power semiconductor device. and junction temperature Related; The duty cycle of a power semiconductor device is represented by the rotor electrical angle. DC bus voltage and d-axis current Joint decision; This represents the switching energy loss of a power semiconductor device, relative to the DC bus voltage during the switching cycle. Motor rotor electrical angle and d-axis current and junction temperature Related;

[0020] One switching cycle during the discharge process Internally, calculate the energy consumed by the load motor of the motor drive system. The formula is:

[0021] ;

[0022] in, , , These represent the equivalent resistances of the stator windings of a three-phase motor, and these resistance values ​​vary with the coolant temperature. change;

[0023] Assuming the DC bus voltage is from the voltage voltage drop If the interval contains n switching cycles, then the total energy consumed by the power semiconductor module and the load motor within this interval is approximately:

[0024] ;

[0025] According to the principle of energy conservation, the total energy consumed by the power semiconductor module and the load motor is equal to the energy released by the DC bus capacitor, as shown in the formula:

[0026] ;

[0027] in, The capacitance value of the DC bus capacitor decreases as the degree of degradation of the DC bus capacitor increases.

[0028] The formula for calculating the discharge time is:

[0029] .

[0030] Furthermore, the processing of the discharge time data includes:

[0031] The collected discharge time data is divided into a healthy data cluster and a current data cluster. The healthy data cluster represents the discharge time data when the motor drive system is in a healthy state, and the current data cluster represents the discharge time data when the motor drive system is not in a healthy state. Probability density analysis and median filtering are performed on the healthy data cluster and the current data cluster to obtain the median healthy discharge time and the median current discharge time.

[0032] The median discharge time is normalized by the median healthy discharge time to obtain the median per-unit discharge time.

[0033] The present invention also provides a health monitoring system for a motor drive system based on discharge time, comprising:

[0034] The switch discharge module is used to trigger the DC bus discharge operation mode after the motor drive system stops running;

[0035] The voltage detection module is used to monitor the change process of DC bus voltage and record the discharge time; the discharge time is the time from the DC bus voltage to a preset first fixed value. Reduced to the second fixed value Required time; first fixed value Below the minimum initial DC bus voltage, the second fixed value Higher than the discharge termination safety voltage; first fixed value and the second fixed value All are within the steady-state discharge range;

[0036] The data collection module is used to collect discharge time data at different stages, perform probability density analysis, median filtering and per-unit processing to obtain the median per-unit discharge time.

[0037] The status assessment module is used to compare the median per-unit discharge time with the set life end index to assess the health status of the system. If the median per-unit discharge time is lower than the life end index, the motor drive system needs maintenance.

[0038] Furthermore, in the switching discharge module, the triggering DC bus discharge operation mode includes:

[0039] After the motor drive system finishes normal operation, the drive signal to the power semiconductor module is stopped, causing the motor drive current to drop rapidly.

[0040] Once the electronic control unit confirms that the motor speed and drive current have both dropped to zero, it isolates the motor drive system from the front-end power supply by disconnecting the circuit breaker.

[0041] The electronic control unit generates a set of discharge commands in the rotating reference coordinate system, including the q-axis current. and d-axis current This triggers the DC bus discharge operation mode.

[0042] Furthermore, in the voltage detection module, the calculation of the discharge time includes:

[0043] One switching cycle during the discharge process Internally, it calculates the energy consumed by the entire power semiconductor module in the motor drive system. The formula is:

[0044] ;

[0045] in, Indicates power semiconductor devices and its corresponding freewheeling diode A general term; The conduction current of a power semiconductor device is represented by the electrical angle of the motor rotor. and d-axis current Decide; This represents the equivalent on-resistance of a power semiconductor device in its turn-on state. It increases with the degree of device degradation and is also related to the on-current of the power semiconductor device. and junction temperature Related; The duty cycle of a power semiconductor device is represented by the rotor electrical angle. DC bus voltage and d-axis current Joint decision; This represents the switching energy loss of a power semiconductor device, relative to the DC bus voltage during the switching cycle. Motor rotor electrical angle and d-axis current and junction temperature Related;

[0046] One switching cycle during the discharge process Internally, calculate the energy consumed by the load motor of the motor drive system. The formula is:

[0047] ;

[0048] in, , , These represent the equivalent resistances of the stator windings of a three-phase motor, and these resistance values ​​vary with the coolant temperature. change;

[0049] Assuming the DC bus voltage is from the voltage voltage drop If the interval contains n switching cycles, then the total energy consumed by the power semiconductor module and the load motor within this interval is approximately:

[0050] ;

[0051] According to the principle of energy conservation, the total energy consumed by the power semiconductor module and the load motor is equal to the energy released by the DC bus capacitor, as shown in the formula:

[0052] ;

[0053] in, The capacitance value of the DC bus capacitor decreases as the degree of degradation of the DC bus capacitor increases.

[0054] The formula for calculating the discharge time is:

[0055] .

[0056] Furthermore, in the data collection module, the processing of discharge time data includes:

[0057] The collected discharge time data is divided into a healthy data cluster and a current data cluster. The healthy data cluster represents the discharge time data when the motor drive system is in a healthy state, and the current data cluster represents the discharge time data when the motor drive system is not in a healthy state. Probability density analysis and median filtering are performed on the healthy data cluster and the current data cluster to obtain the median healthy discharge time and the median current discharge time.

[0058] The median discharge time is normalized by the median healthy discharge time to obtain the median per-unit discharge time.

[0059] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above method.

[0060] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0061] Beneficial effects: This invention provides a method and system for health monitoring of a motor drive system based on discharge time, which has the following significant effects compared with the prior art:

[0062] (1) Zero hardware incremental investment: This method only utilizes the DC bus voltage information already measured in the motor drive system, without the need to add any additional sensors or measuring devices. In addition, this method has low requirements for DC bus voltage sampling performance, and can be achieved simply by conventional sampling synchronized with the system switching frequency, avoiding hardware upgrade costs. This efficient reuse strategy of the system's inherent resources significantly reduces the implementation cost of the health monitoring system.

[0063] (2) Easy to implement: This method uses discharge time as a health indicator. Its calculation principle is extremely simple. It only requires recording the two time points when the DC bus voltage drops from the first fixed value to the second fixed value and subtracting them. This simple time difference calculation does not require complex signal processing algorithms or spectrum analysis, nor does it rely on high-performance computing resources. It can be easily completed by any level of controller. The simplicity of the method significantly reduces the implementation threshold and is particularly suitable for practical engineering applications.

[0064] (3) System-level monitoring: This method assesses the health status of the motor drive system as a whole without distinguishing which specific component is degrading, providing more comprehensive health status information that is highly consistent with actual maintenance strategies.

[0065] (4) High sensitivity: The discharge time exhibits high sensitivity to the degradation of the DC bus capacitor and power semiconductor module within the system. It can produce a discharge time change of more than 2% for only 2.7% capacitor degradation, which can effectively reflect changes in the health status of the system. In existing system-level monitoring methods, efficiency-based methods only reduce efficiency by about 0.11% even when system components reach the end of their lifespan (see paper MV Kjær, H. Wang, and F. Blaabjerg, “End-of-life detection of power electronic converters by exploiting an application-level health precursor,” IEEE Open J. Power Electron., vol. 3, pp. 549–559, 2022.). This tiny change is easily masked by measurement noise. Compared with this technology, the sensitivity of this invention is nearly 20 times higher, which greatly improves the reliability of monitoring.

[0066] (5) No operational interference: Unlike existing health monitoring methods, the method of this invention is executed after the system is completely shut down, which will not interfere with the normal operation of the system or introduce new reliability issues, thus greatly reducing safety risks. Especially for motor drive applications, this non-intrusive monitoring avoids the impact on the system's operating status and control performance.

[0067] In summary, compared with existing system-level monitoring methods in the background art, existing efficiency-based methods require additional sensors, and harmonic-based and algorithm-based methods both require additional hardware upgrades, while the method of the present invention does not. The implementation complexity of harmonic-based and algorithm-based methods is high, while the implementation complexity of the method of the present invention is low. The efficiency-based, harmonic-based, and algorithm-based methods all monitor during normal operation, resulting in high operational interference, while the method of the present invention monitors after the system is shut down, resulting in low operational interference. The degradation sensitivity of efficiency-based and harmonic-based methods is low, while the degradation sensitivity of the method of the present invention is high. Attached Figure Description

[0068] Figure 1 This is a typical configuration diagram of a motor drive system. In the diagram, C represents the DC bus capacitor, EM represents the load motor, ECU is the electronic control unit, and Q1-Q6 and D1-D6 represent the power devices and freewheeling diodes in the power semiconductor module, respectively.

[0069] Figure 2The figure shows the simulation waveform of DC bus discharge for the motor drive system. The curves of DC bus voltage, three-phase drive current and duty cycle of three-phase switches during the discharge process are displayed. The initial DC bus voltage, the discharge steady-state range, the safe voltage and the defined range of discharge time are marked.

[0070] Figure 3 This is a schematic diagram of the discharge time data processing flow. The diagram shows the processing flow of the raw discharge time data collected over time, including probability density analysis, median filtering, and health status assessment.

[0071] Figure 4 Figure (a) shows the topology of the experimental platform for the motor drive system of electric vehicles, and Figure (b) shows the hardware implementation of the experimental platform.

[0072] Figure 5 The waveforms are from the DC bus discharge experiment of the motor drive system.

[0073] Figure 6 This is a data distribution of the discharge time of the DC bus capacitor under different degradation levels.

[0074] Figure 7 The data distribution of discharge time for IGBT modules under different degradation levels.

[0075] Figure 8 This data shows the discharge time distribution at different degradation levels when both the DC bus capacitor and the IGBT module degrade simultaneously. Detailed Implementation

[0076] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading the present invention, any modifications of the present invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0077] Example 1

[0078] Please see Figures 1 to 3 As shown, the present invention provides a health monitoring method for a motor drive system based on discharge time, comprising:

[0079] Step 1, as follows Figure 1 As shown, after the motor drive system finishes normal operation, the drive signal to the power semiconductor module is stopped, causing the motor drive current to drop rapidly.

[0080] Step 2: Once the electronic control unit (ECU) confirms that the motor speed and drive current have both dropped to zero, it isolates the motor drive system from the front-end power supply by disconnecting the circuit breaker.

[0081] Step 3: The Electronic Control Unit (ECU) generates a set of discharge commands in the rotating reference coordinate system, including the q-axis current. and d-axis current This is used to trigger the DC bus discharge operation mode. It is typically used to prevent torque generation. Set to 0, It can be set to any value below the system rating, but it must be ensured that the discharge time complies with relevant regulations. The EU standard requires ≤5 seconds, and the Chinese standard requires ≤3 seconds.

[0082] Step 4: Using the existing current control loop of the motor drive system, discharge the energy stored in the DC bus capacitor according to the discharge command. , and motor rotor electrical angle Released through power semiconductor devices and load motor windings.

[0083] Step 5: Monitor the DC bus voltage change process and record the voltage change from the preset first fixed value. Reduced to the second fixed value The required time is defined as the discharge time, such as... Figure 2 As shown. The first fixed value With the second fixed value The determination of the first fixed value needs to be reasonably set according to the voltage operating range of the specific application. The voltage should be lower than the minimum initial DC bus voltage Vinitial before system discharge, with a certain margin, to avoid the transient process at the beginning of discharge. The second fixed value... The discharge voltage should be higher than the safe discharge termination voltage and have sufficient margin to avoid the transient process at the end of the discharge. This ensures that the discharge time is only related to the system degradation state and stable operating conditions, and is not affected by the transient process during the discharge start and end stages. The discharge stage should be determined based on the three-phase drive current to determine whether it is in the discharge steady-state range.

[0084] The calculation of the discharge time specifically includes:

[0085] During DC bus discharge, the energy stored in the DC bus capacitor is mainly consumed by power semiconductor devices and the load motor windings. The energy consumption of power semiconductor devices includes conduction losses and switching losses. The downstream motor primarily consumes the energy stored in the DC bus capacitor as heat during discharge.

[0086] One switching cycle during the discharge process Internally, it calculates the energy consumed by the entire power semiconductor module in the motor drive system. The formula is:

[0087] ;

[0088] in, Indicates power semiconductor devices and its corresponding freewheeling diode A general term; The conduction current of a power semiconductor device is represented by the electrical angle of the motor rotor. and d-axis current Decide; This represents the equivalent on-resistance of a power semiconductor device in its turn-on state. It increases with the degree of device degradation and is also related to the on-current of the power semiconductor device. and junction temperature Related; The duty cycle of a power semiconductor device is represented by the rotor electrical angle. DC bus voltage and d-axis current Joint decision; This represents the switching energy loss of a power semiconductor device, relative to the DC bus voltage during the switching cycle. Motor rotor electrical angle and d-axis current and junction temperature Related;

[0089] During the discharge process of the motor drive system, the back electromotive force is zero because the rotor electrical angle of the motor remains constant. At this time, the stator winding of the load motor can be regarded as a three-phase RL load, and the energy consumed by the load motor during the discharge process can be calculated through the equivalent resistance.

[0090] One switching cycle during the discharge process Internally, calculate the energy consumed by the load motor of the motor drive system. The formula is:

[0091] ;

[0092] in, , , These represent the equivalent resistances of the stator windings of a three-phase motor, and these resistance values ​​vary with the coolant temperature. change;

[0093] Assuming the DC bus voltage is from the voltage voltage drop If the interval contains n switching cycles, then the total energy consumed by the power semiconductor module and the load motor within this interval is approximately:

[0094] ;

[0095] According to the principle of energy conservation, the total energy consumed by the power semiconductor module and the load motor is equal to the energy released by the DC bus capacitor, as shown in the formula:

[0096] ;

[0097] in, The capacitance value of the DC bus capacitor decreases as the degree of degradation of the DC bus capacitor increases.

[0098] The formula for calculating the discharge time is:

[0099] ;

[0100] For a given voltage range to Because the degradation of the DC bus capacitor will affect its capacitance value. The decrease in capacitance reduces the numerator on the right side of the equation, thus the discharge time decreases with the degradation of the DC bus capacitance; this is because the degradation of power semiconductor devices leads to a decrease in their equivalent on-resistance. The increase in the value of the power semiconductor device increases the denominator on the right side of the formula, so the discharge time will decrease as the power semiconductor device degrades; the discharge time can simultaneously reflect the degradation of both the DC bus capacitance and the power semiconductor device.

[0101] Step 6: Collect discharge time data clusters from different stages, and divide them into healthy data clusters and current data clusters. The healthy data cluster represents the discharge time data when the motor drive system is in a healthy state, and the current data cluster represents the discharge time data when the motor drive system is not in a healthy state. Establish a historical discharge time database.

[0102] Step 7, as follows Figure 3 As shown, probability density analysis and median filtering are performed on the collected healthy discharge time data clusters and the current discharge time data clusters to reduce the impact of operating conditions and measurement noise, and to obtain the median healthy discharge time and the median current discharge time that better represent the true health state of the motor drive system. Using the median healthy discharge time of the motor drive system in a healthy state as a benchmark, the median current discharge time is normalized using the median healthy discharge time to obtain the median per unit discharge time.

[0103] Step 8: Compare the median per-unit discharge time with the set end-of-life index (EOL) to assess the health status of the motor drive system and determine whether maintenance is required. If the median per-unit discharge time is lower than the end-of-life index, the motor drive system needs maintenance.

[0104] Regarding the above technical solution in this embodiment, the following is a detailed explanation. Figures 4 to 8This will detail the experimental process of applying the technical solution to a specific motor drive test experiment and the technical effects of the technical solution.

[0105] To verify the effectiveness of the method of the present invention, this embodiment conducted experimental tests on a full-power motor drive system for an electric vehicle, such as... Figure 4 As shown. The system configuration includes a custom automotive-grade film capacitor and an Infineon HybridPACK. TM The IGBT power module (model FS380R12A6T4B) is mounted on a liquid-cooled base plate. The load motor is a vehicle-grade asynchronous motor with a rated continuous power of 60kW and a rated torque of 120Nm. The coolant circulates simultaneously through the motor windings and the IGBT module to ensure good thermal management.

[0106] Figure 5 The experimental waveforms of DC bus discharge are shown. In this embodiment, the operating parameters are set as follows: q-axis current reference. 0, d-axis current reference The current is 100A, the initial DC bus voltage (Vinitial) is 400V, and the motor rotor electrical angle is... The value is 3π / 5, and the coolant temperature is... The temperature was 24°C. As shown in the figure, the discharge mode activated at 0.01s, followed by a gradual decrease in the DC bus voltage. The three-phase drive current reached a steady state at 0.02s, and the device duty cycle remained around 50% during steady-state operation. In this embodiment, the initial DC bus voltage Vinital was set to vary from 350V to 500V; therefore, the discharge time was defined as the time from 330V (…) to 24°C. ) dropped to 160V ( The time required, such as Figure 5 As indicated by the markings, the selection of these two voltage values ​​ensures that they always fall within the steady-state discharge range and are unaffected by transient processes.

[0107] To verify the effectiveness of discharge time as a health monitoring indicator, experiments need to be conducted on DC bus capacitors and power module samples with different degradation levels. Therefore, this embodiment first conducts a degradation simulation experiment. For the DC bus capacitor, the degradation level is simulated by connecting additional capacitors of different values ​​in parallel with the DC bus. For the IGBT module, since cutting the bond wire will increase the equivalent resistance of the device, the degradation level is simulated by manually cutting the bond wire between IGBT Q3 and diode D1. The simulation results are shown in Table 2.

[0108] Table 2. Degradation Level Settings for DC Bus Capacitors and IGBT Modules

[0109]

[0110] Table 2 shows that for the DC bus capacitor, a 38μF film capacitor can be connected in parallel to simulate degradation level 0, i.e., a healthy state; a 20μF film capacitor in parallel is equivalent to a 2.7% reduction in capacitance, simulating degradation level 1; without additional capacitors in parallel, the capacitance is equivalent to a 5.7% reduction, simulating degradation level 2. For the IGBT module, all bond wires remain intact, simulating degradation level 0; IGBT Q3 and diode D1 each have one bond wire cut, simulating degradation level 1; at degradation level 2, IGBT Q3 and diode D1 each have two bond wires cut, simulating degradation level 2.

[0111] Next, the feasibility of the discharge time under various degradation combinations and actual operating conditions was comprehensively evaluated. A comprehensive test scheme as shown in Table 3 was designed, recording the DC bus voltage from 330V to... ) dropped to 160V ( (Time required)

[0112] Table 3 Discharge test conditions at different degradation levels

[0113]

[0114] The test scheme comprises seven test conditions: A1, A2, and A3 test the DC bus capacitor at different degradation levels; A1, A4, and A5 test the IGBT module at different degradation levels; and A1, A6, and A7 test the complex scenario of simultaneous degradation of both the DC bus capacitor and the IGBT module. Each test condition is tested under multiple actual operating conditions, including seven motor rotor electrical angles (ranging from 17π / 30 to 23π / 30), three coolant temperatures (20°C, 25°C, and 30°C), and two initial DC bus voltages (400V and 500V), totaling 42 different combinations of operating conditions. This means that the same level of DC bus capacitor degradation and IGBT module degradation is tested under 42 different operating conditions, and the DC bus voltage is recorded from 330V (…). ) dropped to 160V ( The comprehensive test matrix design ensures the reliability and applicability of the experimental results, simulating various working scenarios that motor drive systems may encounter in practical applications.

[0115] To ensure the statistical validity and reliability of the experimental results, this embodiment performed 100 random sampling analyses on 42 sets of data under each test condition (A1 to A7). In each random sampling, 50% of the data, i.e., 21 sets, were randomly selected from the 42 sets. Median filtering was then applied to these 21 sets of data to obtain the median discharge time, which represents the true health state of the motor drive system, including the median healthy discharge time and the current median discharge time. After 100 random samplings and median filtering, 100 median discharge times were finally obtained for each test condition. The distribution of the median discharge times is as follows: Figures 6 to 8 As shown in the box plot, the dispersion of discharge time at the same degradation level is mainly caused by varying operating conditions.

[0116] When the capacitance value decreases by 5%, the film capacitor reaches the end-of-life (EoL) standard, at which point the motor drive system needs to be replaced.

[0117] Figure 6 The discharge time distribution is shown for DC bus capacitor degradation only, corresponding to test numbers A1, A2, and A3. A1 shows the median discharge time distribution for the healthy state, i.e., degradation level 0. A2 shows the median discharge time distribution for the DC bus capacitor at degradation level 1, with a capacitance reduction of 2.7%. A3 shows the median discharge time distribution for the DC bus capacitor at degradation level 2, with a capacitance reduction of 5.7%. When the capacitance decreases by 5%, the film capacitor reaches the end-of-life (EoL) standard, requiring replacement of the motor drive system. Figure 6 As can be seen, the discharge time shortens significantly with the increase of the DC bus capacitance degradation level. When the capacitance value decreases by 2.7%, i.e., degradation level 1, the median discharge time is at least 2.1% lower than the median discharge time in the healthy state; when the capacitance value decreases by 5.7%, i.e., degradation level 2, the median discharge time is at least 5.2% lower than the median discharge time in the healthy state. This indicates that the health monitoring method provided by the present invention has high sensitivity and can meet the needs of actual monitoring.

[0118] Figure 7 The diagram shows the discharge time distribution under IGBT module degradation conditions only, corresponding to test numbers A1, A4, and A5. A1 shows the median discharge time distribution for the healthy state, i.e., degradation level 0; A4 shows the median discharge time distribution for the IGBT module at degradation level 1; and A5 shows the median discharge time distribution for the IGBT module at degradation level 2. From... Figure 7As can be seen, when the IGBT module is at degradation level 1, the median discharge time is reduced by at least 0.9% compared to the median discharge time in the healthy state. When the IGBT module is at degradation level 2, the median discharge time is reduced by at least 2.5% compared to the median discharge time in the healthy state. This indicates that the discharge time also has good sensitivity to the degradation of power semiconductor devices.

[0119] Figure 8 The discharge time distributions of both the DC bus capacitor and the IGBT module under simultaneous degradation are shown, corresponding to test numbers A1, A6, and A7. A1 shows the median discharge time distribution for the healthy state, i.e., degradation level 0. A6 shows the median discharge time distribution for both the DC bus capacitor and the IGBT module at degradation level 1. A7 shows the median discharge time distribution for both the DC bus capacitor and the IGBT module at degradation level 2. Figure 8 As can be seen, under this combined degradation condition, the median discharge time is significantly lower than the median discharge time in the healthy state. When the DC bus capacitor and IGBT module are both at degradation level 1, the median discharge time is at least 3.9% lower than the median discharge time in the healthy state. When the DC bus capacitor and IGBT module are both at degradation level 2, the median discharge time is at least 8.9% lower than the median discharge time in the healthy state. This indicates that the discharge time is more sensitive to the combined degradation of multiple components in the motor drive system. This characteristic is particularly important in practical applications because multiple components often degrade simultaneously in real-world systems.

[0120] from Figures 6 to 8 It can be seen that even under the worst-case scenario considering data randomness, i.e., using the lower adjacent value of A1 as the benchmark, the upper adjacent values ​​of A2 to A7 still show a significant decrease compared to this benchmark, effectively eliminating the influence of random factors and fully demonstrating the stability and reliability of the method of the present invention in practical application scenarios. Experiments show that this method can not only effectively detect the individual degradation of capacitors and IGBT modules, but also accurately identify their combined degradation state, providing highly sensitive technical support for the health status assessment of motor drive systems.

[0121] Example 2

[0122] Please see Figures 1 to 3 As shown in Embodiment 1, the present invention provides a health monitoring system for a motor drive system based on discharge time, comprising:

[0123] The speed reduction module is used to stop supplying drive signals to the power semiconductor module after the motor drive system has finished normal operation, causing the motor drive current to drop rapidly, such as... Figure 1 As shown.

[0124] The isolation module isolates the motor drive system from the front-end power supply by disconnecting the circuit breaker after the electronic control unit (ECU) confirms that the motor speed and drive current have both dropped to zero.

[0125] The discharge command module generates a set of discharge commands, including the q-axis current, in the rotating reference coordinate system via the electronic control unit (ECU). and d-axis current This is used to trigger the DC bus discharge operation mode. It is typically used to prevent torque generation. Set to 0, It can be set to any value below the system rating, but it must be ensured that the discharge time complies with relevant regulations. The EU standard requires ≤5 seconds, and the Chinese standard requires ≤3 seconds.

[0126] The release module, through the existing current control loop of the motor drive system, discharges the energy stored in the DC bus capacitor according to the discharge command. , and motor rotor electrical angle Released through power semiconductor devices and load motor windings.

[0127] The monitoring and calculation module is used to monitor the DC bus voltage change process and record the voltage change from a preset first fixed value. Reduced to the second fixed value The required time is defined as the discharge time, such as... Figure 2 As shown. The first fixed value With the second fixed value The determination of the first fixed value needs to be reasonably set according to the voltage operating range of the specific application. The voltage should be lower than the minimum initial DC bus voltage Vinitial before system discharge, with a certain margin, to avoid the transient process at the beginning of discharge. The second fixed value... The discharge voltage should be higher than the safe discharge termination voltage and have sufficient margin to avoid the transient process at the end of the discharge. This ensures that the discharge time is only related to the system degradation state and stable operating conditions, and is not affected by the transient process during the discharge start and end stages. The discharge stage should be determined based on the three-phase drive current to determine whether it is in the discharge steady-state range.

[0128] The calculation of the discharge time specifically includes:

[0129] During DC bus discharge, the energy stored in the DC bus capacitor is mainly consumed by power semiconductor devices and the load motor windings. The energy consumption of power semiconductor devices includes conduction losses and switching losses. The downstream motor primarily consumes the energy stored in the DC bus capacitor as heat during discharge.

[0130] One switching cycle during the discharge process Internally, it calculates the energy consumed by the entire power semiconductor module in the motor drive system. The formula is:

[0131] ;

[0132] in, Indicates power semiconductor devices and its corresponding freewheeling diode A general term; The conduction current of a power semiconductor device is represented by the electrical angle of the motor rotor. and d-axis current Decide; This represents the equivalent on-resistance of a power semiconductor device in its turn-on state. It increases with the degree of device degradation and is also related to the on-current of the power semiconductor device. and junction temperature Related; The duty cycle of a power semiconductor device is represented by the rotor electrical angle. DC bus voltage and d-axis current Joint decision; This represents the switching energy loss of a power semiconductor device, relative to the DC bus voltage during the switching cycle. Motor rotor electrical angle and d-axis current and junction temperature Related;

[0133] During the discharge process of the motor drive system, the back electromotive force is zero because the rotor electrical angle of the motor remains constant. At this time, the stator winding of the load motor can be regarded as a three-phase RL load, and the energy consumed by the load motor during the discharge process can be calculated through the equivalent resistance.

[0134] One switching cycle during the discharge process Internally, calculate the energy consumed by the load motor of the motor drive system. The formula is:

[0135] ;

[0136] in, , , These represent the equivalent resistances of the stator windings of a three-phase motor, and these resistance values ​​vary with the coolant temperature. change;

[0137] Assuming the DC bus voltage is from the voltage voltage drop If the interval contains n switching cycles, then the total energy consumed by the power semiconductor module and the load motor within this interval is approximately:

[0138] ;

[0139] According to the principle of energy conservation, the total energy consumed by the power semiconductor module and the load motor is equal to the energy released by the DC bus capacitor, as shown in the formula:

[0140] ;

[0141] in, The capacitance value of the DC bus capacitor decreases as the degree of degradation of the DC bus capacitor increases.

[0142] The formula for calculating the discharge time is:

[0143] ;

[0144] For a given voltage range to Because the degradation of the DC bus capacitor will affect its capacitance value. The decrease in capacitance reduces the numerator on the right side of the equation, thus the discharge time decreases with the degradation of the DC bus capacitance; this is because the degradation of power semiconductor devices leads to a decrease in their equivalent on-resistance. The increase in the value of the power semiconductor device increases the denominator on the right side of the formula, so the discharge time will decrease as the power semiconductor device degrades; the discharge time can simultaneously reflect the degradation of both the DC bus capacitance and the power semiconductor device.

[0145] The data collection module is used to collect discharge time data clusters at different stages, which are divided into a healthy data cluster and a current data cluster. The healthy data cluster represents the discharge time data when the motor drive system is in a healthy state, while the current data cluster represents the discharge time data when the motor drive system is not in a healthy state, and a historical discharge time database is established.

[0146] Data processing module, such as Figure 3 As shown, probability density analysis and median filtering are performed on the collected healthy discharge time data clusters and the current discharge time data clusters to reduce the impact of operating conditions and measurement noise, and to obtain the median healthy discharge time and the median current discharge time that better represent the true health state of the motor drive system. Using the median healthy discharge time of the motor drive system in a healthy state as a benchmark, the median current discharge time is normalized using the median healthy discharge time to obtain the median per unit discharge time.

[0147] The status assessment module evaluates the health status of the motor drive system by comparing the median per-unit discharge time with the set end-of-life index (EOL) and determines whether maintenance is required. If the median per-unit discharge time is lower than the end-of-life index, the motor drive system needs maintenance.

[0148] Given that Example 2 is a modular product representation of the monitoring method in Example 1, the test experiments corresponding to Example 2 are the same as those in Example 1 (combined). Figures 4 to 8 The same applies, so I will not repeat it here.

[0149] The computer device of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above method.

[0150] The computer-readable storage medium of the present invention stores a computer program thereon, which, when executed by a processor, implements the steps of the above-described method.

Claims

1. A method for health monitoring of a motor drive system based on discharge time, characterized in that, include: After the motor drive system stops running, the DC bus discharge operation mode is triggered; Monitor the DC bus voltage change process and record the discharge time; The discharge time is when the DC bus voltage drops from a preset first fixed value. Reduced to the second fixed value Required time; first fixed value Below the minimum initial DC bus voltage, the second fixed value Higher than the discharge termination safety voltage; first fixed value and the second fixed value All are within the steady-state discharge range; the calculation of the discharge time includes: one switching cycle during the discharge process. Internally, it calculates the energy consumed by the entire power semiconductor module in the motor drive system. The formula is: ； in, Indicates power semiconductor devices and its corresponding freewheeling diode A general term; The conduction current of a power semiconductor device is represented by the electrical angle of the motor rotor. and d-axis current Decide; This represents the equivalent on-resistance of a power semiconductor device in its turn-on state. It increases with the degree of device degradation and is also related to the on-current of the power semiconductor device. and junction temperature Related; The duty cycle of a power semiconductor device is represented by the rotor electrical angle. DC bus voltage and d-axis current Joint decision; This represents the switching energy loss of a power semiconductor device, relative to the DC bus voltage during the switching cycle. Motor rotor electrical angle and d-axis current and junction temperature Related; One switching cycle during the discharge process Internally, calculate the energy consumed by the load motor of the motor drive system. The formula is: ； in, , , These represent the equivalent resistances of the stator windings of a three-phase motor, and these resistance values ​​vary with the coolant temperature. change; Assuming the DC bus voltage is from the voltage voltage drop If the interval contains n switching cycles, then the total energy consumed by the power semiconductor module and the load motor within this interval is approximately: ； According to the principle of energy conservation, the total energy consumed by the power semiconductor module and the load motor is equal to the energy released by the DC bus capacitor, as shown in the formula: ； in, The capacitance value of the DC bus capacitor decreases as the degree of degradation of the DC bus capacitor increases. The formula for calculating the discharge time is: ； Discharge time data at different stages were collected, and probability density analysis, median filtering and per-unit processing were performed to obtain the median per-unit discharge time. The health status of the system is assessed by comparing the median per-unit discharge time with the set end-of-life index. If the median per-unit discharge time is lower than the end-of-life index, the motor drive system needs maintenance.

2. The method for health monitoring of a motor drive system based on discharge time according to claim 1, characterized in that, The triggered DC bus discharge operation mode includes: After the motor drive system finishes normal operation, the drive signal to the power semiconductor module is stopped, causing the motor drive current to drop rapidly. Once the electronic control unit confirms that the motor speed and drive current have both dropped to zero, it isolates the motor drive system from the front-end power supply by disconnecting the circuit breaker. The electronic control unit generates a set of discharge commands in the rotating reference coordinate system, including the q-axis current. and d-axis current This triggers the DC bus discharge operation mode.

3. The method for health monitoring of a motor drive system based on discharge time according to claim 1, characterized in that, Discharge time data processing includes: The collected discharge time data is divided into a healthy data cluster and a current data cluster. The healthy data cluster represents the discharge time data when the motor drive system is in a healthy state, and the current data cluster represents the discharge time data when the motor drive system is not in a healthy state. Probability density analysis and median filtering are performed on the healthy data cluster and the current data cluster to obtain the median healthy discharge time and the median current discharge time. The median discharge time is normalized by the median healthy discharge time to obtain the median per-unit discharge time.

4. A health monitoring system for a motor drive system based on discharge time, characterized in that, include: The switch discharge module is used to trigger the DC bus discharge operation mode after the motor drive system stops running; The voltage detection module is used to monitor the change process of DC bus voltage and record the discharge time; the discharge time is the time from the DC bus voltage to a preset first fixed value. Reduced to the second fixed value Required time; first fixed value Below the minimum initial DC bus voltage, the second fixed value Higher than the discharge termination safety voltage; first fixed value and the second fixed value All are within the steady-state discharge range; The calculation of the discharge time includes: One switching cycle during the discharge process Internally, it calculates the energy consumed by the entire power semiconductor module in the motor drive system. The formula is: ； in, Indicates power semiconductor devices and its corresponding freewheeling diode A general term; The conduction current of a power semiconductor device is represented by the electrical angle of the motor rotor. and d-axis current Decide; This represents the equivalent on-resistance of a power semiconductor device in its turn-on state. It increases with the degree of device degradation and is also related to the on-current of the power semiconductor device. and junction temperature Related; The duty cycle of a power semiconductor device is represented by the rotor electrical angle. DC bus voltage and d-axis current Joint decision; This represents the switching energy loss of a power semiconductor device, relative to the DC bus voltage during the switching cycle. Motor rotor electrical angle and d-axis current and junction temperature Related; One switching cycle during the discharge process Internally, calculate the energy consumed by the load motor of the motor drive system. The formula is: ； in, , , These represent the equivalent resistances of the stator windings of a three-phase motor, and these resistance values ​​vary with the coolant temperature. change; Assuming the DC bus voltage is from the voltage voltage drop If the interval contains n switching cycles, then the total energy consumed by the power semiconductor module and the load motor within this interval is approximately: ； According to the principle of energy conservation, the total energy consumed by the power semiconductor module and the load motor is equal to the energy released by the DC bus capacitor, as shown in the formula: ； in, The capacitance value of the DC bus capacitor decreases as the degree of degradation of the DC bus capacitor increases. The formula for calculating the discharge time is: ； The data collection module is used to collect discharge time data at different stages, perform probability density analysis, median filtering and per-unit processing to obtain the median per-unit discharge time. The status assessment module is used to compare the median per-unit discharge time with the set life end index to assess the health status of the system. If the median per-unit discharge time is lower than the life end index, the motor drive system needs maintenance.

5. The health monitoring system for a motor drive system based on discharge time according to claim 4, characterized in that, In the switching discharge module, the triggering DC bus discharge operation mode includes: After the motor drive system finishes normal operation, the drive signal to the power semiconductor module is stopped, causing the motor drive current to drop rapidly. Once the electronic control unit confirms that the motor speed and drive current have both dropped to zero, it isolates the motor drive system from the front-end power supply by disconnecting the circuit breaker. The electronic control unit generates a set of discharge commands in the rotating reference coordinate system, including the q-axis current. and d-axis current This triggers the DC bus discharge operation mode.

6. The health monitoring system for a motor drive system based on discharge time according to claim 4, characterized in that, The data collection module processes discharge time data, including: The collected discharge time data is divided into a healthy data cluster and a current data cluster. The healthy data cluster represents the discharge time data when the motor drive system is in a healthy state, and the current data cluster represents the discharge time data when the motor drive system is not in a healthy state. Probability density analysis and median filtering are performed on the healthy data cluster and the current data cluster to obtain the median healthy discharge time and the median current discharge time. The median discharge time is normalized by the median healthy discharge time to obtain the median per-unit discharge time.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 3.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

Images