A three-phase full-bridge inverter fault diagnosis method, system and device

Abstract

The present application relates to the technical field of fault type diagnosis, and particularly relates to a three-phase full-bridge inverter fault diagnosis method, system and device. The method obtains PWM instruction signals, electrical signals and temperature signals of IGBT devices in a three-phase full-bridge inverter; based on the bridge arm on-off logic performance of the IGBT devices, the electrical signals and the temperature signals within a PWM instruction signal cycle, obtains on-off logic analysis coefficients and conduction efficiency coefficients of the IGBT devices, obtains IGBT fault type labels by combining fault determination thresholds and constraint conditions to construct a fault sample set, trains a fault type identification model, and obtains a trained fault type identification model; by accurately obtaining the on-off logic analysis coefficients and the conduction efficiency coefficients, the accuracy of fault type label identification is effectively improved, which is conducive to real-time and accurate identification of the fault type of the IGBT devices, timely corresponding processing, and effective reduction of the fault loss of the three-phase full-bridge inverter.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault type diagnosis, in particular to a three-phase full-bridge inverter fault diagnosis method, system and device. BACKGROUND

[0002] As a core device of power conversion, the three-phase full-bridge inverter is widely used in key fields such as new energy power generation, motor drive and uninterruptible power supply. It contains multiple core power devices such as IGBT (Insulated Gate Bipolar Transistor) inside, and is long-term operated under the working conditions of high-frequency switching, large current and high voltage. It is easy to be affected by factors such as load mutation, device aging and poor heat dissipation, thereby causing IGBT device failure. Common fault types include short circuit, open circuit, overload and overheating. If these faults cannot be identified and processed in time, it may cause equipment downtime, production interruption, permanent damage to IGBT devices, and even serious safety accidents such as fire. Therefore, real-time and accurate fault diagnosis of the three-phase full-bridge inverter is of great significance to ensure the reliability, safety and economy of the power electronic system.

[0003] At present, the current, voltage and temperature signals of the IGBT device are analyzed to determine whether the IGBT device has a fault, but it cannot further determine whether the specific fault type is short circuit, open circuit, overload or overheating. Because the causes, risk levels and processing methods of the four types of faults are completely different, this ambiguous diagnosis result forces the maintenance personnel to rely on personal experience to blindly and one-by-one disassemble and troubleshoot. This not only greatly prolongs the fault downtime and increases the unplanned maintenance cost, but also hides the risk of fault expansion because the correct measures cannot be taken in time to address the specific fault root cause, ultimately affecting the reliability and availability of the entire inverter system operation. SUMMARY

[0004] In order to solve the technical problem of being unable to accurately evaluate the fault type of the IGBT device in real time, the purpose of the present application is to provide a three-phase full-bridge inverter fault diagnosis method, system and device, and the technical solution adopted is as follows:

[0005] In the first aspect, an embodiment of the present application provides a three-phase full-bridge inverter fault diagnosis method, which comprises the following steps:

[0006] Obtaining PWM instruction signals in the three-phase full-bridge inverter, electrical signals and temperature signals of each IGBT device at each time; the electrical signals include collector current and collector-emitter voltage; the temperature signals include IGBT junction temperature;

[0007] The on-off logic analysis coefficient of each IGBT device is obtained based on the on-off logic performance of each IGBT device and the change of the collector current in a preset number of PWM instruction signal periods.

[0008] The conduction efficiency coefficient of each IGBT device is obtained based on the collector-emitter voltage fluctuation of each IGBT device and the size of the collector current, the collector-emitter voltage and the IGBT junction temperature in a preset number of PWM instruction signal periods.

[0009] The fault type label of each IGBT is obtained based on the on-off logic analysis coefficient and the conduction efficiency coefficient of each IGBT, combined with the fault determination threshold and the constraint condition; the fault type label includes short circuit, open circuit, overload, overheating and no fault.

[0010] The fault sample set is constructed based on the fault type label, the fault type recognition model is trained, and the trained fault type recognition model is obtained; the fault type recognition model is used to output the fault type of each IGBT device in the three-phase full-bridge inverter.

[0011] Further, the on-off logic analysis coefficient is obtained by:

[0012] The on-off logic compliance degree of each IGBT device is obtained according to the number of periods of on-off logic compliance and the number of periods of continuous on-off logic violation of each IGBT device in a preset number of PWM instruction signal periods.

[0013] The current-related correction coefficient of each IGBT device is obtained according to the change of the collector current of each IGBT device in a preset number of PWM instruction signal periods.

[0014] The product of the on-off logic compliance degree and the current-related correction coefficient of each IGBT device is taken as the on-off logic analysis coefficient of each IGBT device.

[0015] Further, the on-off logic compliance degree is obtained by:

[0016] For any IGBT device, the ratio of the number of periods of on-off logic compliance of the IGBT device to the preset number is taken as the first compliance analysis value of the IGBT device.

[0017] The number of periods of continuous on-off logic violation of the IGBT device is taken as the first number.

[0018] The result of negative correlation between the ratio of the maximum first number and the preset number is taken as the compliance correction weight of the IGBT device.

[0019] The product of the first compliance analysis value of the IGBT device and the compliance correction weight is taken as the compliance degree of the turn-on and turn-off logic of the IGBT device.

[0020] Further, the method for obtaining the current correlation correction coefficient is:

[0021] For any IGBT device, the ratio of the collector current rising rate of the IGBT device to the preset rated current rising rate in each PWM instruction signal period is taken as the current impact degree;

[0022] The ratio of the collector current of the IGBT device to the preset rated current in a preset number of PWM instruction signal periods is taken as the current excess analysis value;

[0023] The product of the maximum current impact degree and the mean of the current excess analysis value is negatively correlated and normalized, and the result is taken as the current correlation correction coefficient of the IGBT device.

[0024] Further, the method for obtaining the conduction efficiency coefficient is:

[0025] For any IGBT device, the ratio of the collector-emitter voltage of the IGBT device to the preset rated voltage in a preset number of PWM instruction signal periods is taken as the voltage excess analysis value;

[0026] The standard deviation of the voltage excess analysis value is negatively correlated and normalized, and the result is taken as the voltage stability degree of the IGBT device;

[0027] For any PWM instruction signal period, the product of the average collector-emitter voltage and the average collector current of the IGBT device in the PWM instruction signal period is taken as the actual loss of the IGBT device in the PWM instruction signal period;

[0028] The mean of the actual loss of the IGBT device in a preset number of PWM instruction signal periods is taken as the overall actual loss of the IGBT device;

[0029] The difference between the overall actual loss and the preset rated loss is normalized, and the result is taken as the loss degree analysis value of the IGBT device;

[0030] The mean of the IGBT junction temperature of the IGBT device in a preset number of PWM instruction signal periods is normalized, and the result is taken as the junction temperature correction coefficient of the IGBT device;

[0031] The product of the loss degree analysis value of the IGBT device and the junction temperature correction coefficient is taken as the energy transfer analysis value of the IGBT device;

[0032] The product of the negative correlation result of the energy transfer analysis value of the IGBT device and the voltage stability degree is taken as the turn-on efficiency coefficient of the IGBT device.

[0033] Further, the fault determination threshold value comprises a turn-on-off logic abnormality demarcation threshold value, a short-circuit-overload differentiation threshold value, a turn-on efficiency abnormality demarcation threshold value, an open-circuit-overheat differentiation threshold value and a junction temperature critical threshold value.

[0034] For any IGBT device, the turn-on-off logic abnormality demarcation threshold value of the IGBT device is the difference between the mean value of the turn-on-off logic analysis coefficient of the IGBT device under normal working condition and three times of the standard deviation thereof.

[0035] The short-circuit-overload differentiation threshold value of the IGBT device is the negative correlation result of the ratio of the saturation collector-emitter voltage of the IGBT device to the bus voltage under normal working condition.

[0036] The turn-on efficiency abnormality demarcation threshold value of the IGBT device is the difference between the mean value of the turn-on efficiency coefficient of the IGBT device under normal working condition and three times of the standard deviation thereof.

[0037] The open-circuit-overheat differentiation threshold value of the IGBT device is preset according to experimental data.

[0038] The junction temperature critical threshold value of the IGBT device is the maximum IGBT junction temperature of the IGBT device minus the specified number of times of the three times of the IGBT junction temperature standard deviation of the IGBT device under normal working condition.

[0039] Further, the fault type label acquisition method is:

[0040] For any IGBT device, when the turn-on-off logic analysis coefficient of the IGBT device is less than the short-circuit-overload differentiation threshold value and the turn-on efficiency coefficient is greater than or equal to the turn-on efficiency abnormality demarcation threshold value, the fault type label of the IGBT device is short-circuit.

[0041] When the turn-on-off logic analysis coefficient of the IGBT device is greater than or equal to the short-circuit-overload differentiation threshold value and less than the turn-on-off logic abnormality demarcation threshold value and the turn-on efficiency coefficient is greater than or equal to the turn-on efficiency abnormality demarcation threshold value, the fault type label of the IGBT device is overload.

[0042] When the turn-on-off logic analysis coefficient of the IGBT device is greater than or equal to the turn-on-off logic abnormality demarcation threshold value and the turn-on efficiency coefficient is less than the open-circuit-overheat differentiation threshold value, the fault type label of the IGBT device is open-circuit.

[0043] When the on-off logic analysis coefficient of the IGBT device is greater than or equal to the on-off logic abnormal boundary threshold value, the conduction efficiency coefficient is greater than or equal to the open circuit-heat over threshold value and less than the conduction efficiency abnormal boundary threshold value, and the IGBT junction temperature is greater than the junction temperature critical threshold value, the fault type label of the IGBT device is heat over.

[0044] When the on-off logic analysis coefficient of the IGBT device is greater than or equal to the on-off logic abnormal boundary threshold value, the conduction efficiency coefficient is greater than or equal to the conduction efficiency abnormal boundary threshold value, the fault type label of the IGBT device is no fault.

[0045] Further, the fault type identification model is a graph neural network model; the graph structure sample is constructed by taking each IGBT device as an independent node and taking the physical topological correlation between devices as an edge, and the feature vector of each node includes the on-off logic analysis coefficient and the conduction efficiency coefficient, and the node label is the fault type label.

[0046] In a second aspect, another embodiment of the present application provides a three-phase full-bridge inverter fault diagnosis system, which comprises:

[0047] A data acquisition module is configured to acquire PWM instruction signals in the three-phase full-bridge inverter, electrical signals and temperature signals of each IGBT device at each time point; the electrical signals include collector current and collector-emitter voltage; and the temperature signals include IGBT junction temperature.

[0048] An on-off logic analysis coefficient acquisition module is configured to acquire the on-off logic analysis coefficient of each IGBT device based on the on-off logic performance and the change of the collector current of each IGBT device in a preset number of PWM instruction signal periods.

[0049] A conduction efficiency coefficient acquisition module is configured to acquire the conduction efficiency coefficient of each IGBT device based on the collector-emitter voltage fluctuation of each IGBT device in a preset number of PWM instruction signal periods, and the magnitudes of the collector current, the collector-emitter voltage and the IGBT junction temperature.

[0050] A fault type label acquisition module is configured to acquire the fault type label of each IGBT based on the on-off logic analysis coefficient and the conduction efficiency coefficient of each IGBT, in combination with fault determination threshold values and constraint conditions; the fault type label includes short circuit, open circuit, overload, heat over and no fault.

[0051] A data processing module is configured to construct a fault sample set based on the fault type label, train a fault type identification model, and obtain a trained fault type identification model; the fault type identification model is used to output the fault type of each IGBT device in the three-phase full-bridge inverter.

[0052] In a third aspect, the present application provides a device for diagnosing faults of a three-phase full-bridge inverter, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0053] The present application has the following advantages:

[0054] The present application first obtains a turn-on-off logic analysis coefficient of each IGBT device based on the turn-on-off logic performance and the change of the collector current of each IGBT device within a preset number of PWM instruction signal periods, accurately reflects the turn-on-off logic compliance, current shock intensity, and load matching degree of the IGBT device, clearly depicts the fault characteristics caused by turn-on-off logic abnormalities or current over-limiting, and is beneficial to subsequent accurate definition of the risk interval of turn-on-off related faults and provides a core basis for fault type differentiation; in order to comprehensively characterize the turn-on state and the intrinsic operating characteristics of the IGBT device, the present application further obtains a turn-on efficiency coefficient of each IGBT device based on the collector-emitter voltage fluctuation of each IGBT device within a preset number of PWM instruction signal periods, as well as the sizes of the collector current, the collector-emitter voltage, and the IGBT junction temperature, accurately reflects the voltage stability, energy transfer efficiency, and temperature-loss coupling risk of the IGBT device, accurately captures the fault characteristics caused by intrinsic damage or loss accumulation of the device, and is beneficial to subsequent effective differentiation of the types of turn-on related faults and makes up for the defect that a single parameter cannot comprehensively evaluate the turn-on state; then, based on the turn-on-off logic analysis coefficient and the turn-on efficiency coefficient of each IGBT device, in combination with a fault determination threshold and a constraint condition, the present application obtains a fault type label of each IGBT, which is beneficial to subsequent provision of an accurate and unique supervision signal for a fault sample set and ensures the effectiveness and reliability of model training; then, based on the fault type label, the present application constructs a fault sample set, trains a fault type recognition model, and obtains a trained fault type recognition model, so that the model can fully integrate the intrinsic state characteristics of the IGBT device and the circuit topology related information, effectively improves the recognition accuracy, real-time performance, and robustness of the fault type of the IGBT device in the three-phase full-bridge inverter, and realizes accurate determination of five types of states, i.e., short circuit, open circuit, overload, overheating, and no fault, and adapts to the application requirements of complex industrial working conditions. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, a brief introduction will be given to the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without any creative effort.

[0056] Figure 1 A schematic flow chart of a three-phase full-bridge inverter fault diagnosis method provided by an embodiment of the present application;

[0057] Figure 2 A flow chart of a method for obtaining a turn-on logic analysis coefficient provided by an embodiment of the present application;

[0058] Figure 3 A flow chart of a method for obtaining a turn-on efficiency coefficient provided by an embodiment of the present application;

[0059] Figure 4 A structural diagram of a three-phase full-bridge inverter fault diagnosis system provided by an embodiment of the present application;

[0060] Figure 5 A schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined purposes, the following describes in detail the specific embodiments, structures, features and effects of the three-phase full-bridge inverter fault diagnosis method, system and device according to the present application, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0063] The specific scheme of the three-phase full-bridge inverter fault diagnosis method, system and device provided by the present application is described in detail below with reference to the accompanying drawings.

[0064] Embodiment 1:

[0065] The present application proposes a three-phase full-bridge inverter fault diagnosis method, please refer to Figure 1 , which shows a schematic flow chart of a three-phase full-bridge inverter fault diagnosis method provided by an embodiment of the present application. The method includes the following steps:

[0066] Step S1: Obtain the PWM instruction signal in the three-phase full-bridge inverter, the electrical signal and the temperature signal of each IGBT device at each time; the electrical signal includes the collector current and the collector voltage; the temperature signal includes the IGBT junction temperature.

[0067] Specifically, it is known that the fault diagnosis of the three-phase full-bridge inverter depends on the accurate perception of the operating state of the IGBT device, and the four types of faults of the IGBT device, such as short circuit, open circuit, overload and overheating, can be reflected through abnormal changes in electrical signals and temperature signals, for example, the instantaneous current impact when short-circuiting, the on-voltage drop when open-circuiting, the continuous overcurrent when overloading, and the IGBT junction temperature exceeding the standard when overheating. Therefore, in order to accurately detect the fault of the three-phase full-bridge inverter, the embodiment adopts a collection architecture of multi-sensor cooperation and time sequence synchronization, that is, taking the PWM instruction signal as the time sequence reference, the electrical signals and temperature signals of each IGBT device in the three-phase full-bridge inverter are collected by dedicated current sensors, voltage sensors and temperature sensors. Among them, the sampling frequency of all sensors is synchronized with the PWM instruction signal, ensuring that the electrical signals, temperature signals and control signals (PWM) at the same time correspond one by one. Among them, the electrical signals include collector current and collector-emitter voltage, and the temperature signals include IGBT junction temperature.

[0068] It should be noted that the sampling trigger signal of all sensors is synchronized with the starting edge of the PWM instruction, and the sampling frequency is not less than 10 times the PWM switching frequency, for example, if the PWM switching frequency is 10 kHz, the sampling frequency is greater than or equal to 100 kHz, to avoid characteristic calculation deviation caused by time sequence misplacement. In addition, the measurement range of the sensor needs to cover the signal extreme value of the normal operation and fault state of the IGBT device, for example, the current sensor needs to cover the instantaneous impact current when short-circuiting, and the voltage sensor needs to cover the on-voltage drop close to the bus voltage when open-circuiting, and the measurement accuracy is not less than 0.5 level, to ensure that the small fault characteristics can be captured. Each IGBT device is configured with an independent signal collection channel, to avoid fault node positioning ambiguity caused by multi-device signal mixed sampling, and to ensure that the subsequent signal abnormality-IGBT device-fault type can be accurately associated.

[0069] Among them, the PWM (Pulse Width Modulation) instruction signal is the drive control signal output by the three-phase full-bridge inverter control system to the IGBT gate, used to control the on and off timing of the IGBT device, and its high level corresponds to the on instruction of the IGBT device, and its low level corresponds to the off instruction of the IGBT device; it is a digital control signal, which is the basis for judging the compliance of the bridge arm on-off logic, by comparing the on or off state required by the PWM instruction signal with the on or off state reflected by the actual electrical signal of the IGBT device, to judge whether the bridge arm on-off logic is compliant; the embodiment directly collects the PWM instruction signal output by the inverter control unit through the digital signal collection module, and records the starting time, high level duration and low level duration of each instruction period;

[0070] The collector current is the current flowing into the collector (C) and flowing out of the emitter (E) of the IGBT device, which is the core parameter reflecting the load state of the IGBT device. When the IGBT device is in normal operation, the collector current is stable within the rated current range; when short-circuit occurs, the collector current increases explosively; when overload occurs, the collector current continuously exceeds the rated current; and when open-circuit occurs, the collector current is close to 0A. In order to more accurately monitor the collector current, the closed-loop Hall current sensor with model number LT508-S6 is selected to obtain the collector current of each IGBT device at each moment. The closed-loop Hall current sensor with model number LT508-S6 is suitable for the working current range of the IGBT device, has strong anti-electromagnetic interference ability, and is suitable for the strong electromagnetic environment of the inverter. The closed-loop Hall current sensor is connected in series in the collector loop of the IGBT device, is installed close to the IGBT device, and the length of the lead wire is shortened to reduce the influence of stray inductance. At the same time, the electrical isolation between the closed-loop Hall current sensor and the IGBT device is ensured to avoid high-voltage breakdown. In addition, the positive and negative directions of the current (reflecting the current direction) and the change rate are recorded synchronously when the collector current is collected.

[0071] The collector-emitter voltage is the voltage difference between the collector and the emitter of the IGBT device, which is the core parameter reflecting the conduction state of the IGBT device. When normally conducting, the collector-emitter voltage is the saturation voltage drop (usually 1V-2V, the specific value is referred to the IGBT device manual), when short-circuit occurs, the collector-emitter voltage is close to 0V, when open-circuit occurs, the collector-emitter voltage rises to close to the bus voltage, and when overheating occurs, the collector-emitter voltage is higher than the saturation voltage drop. In order to more accurately monitor the collector-emitter voltage, the voltage sensor with device model LV25-P is selected. The voltage sensor with model number LV25-P is suitable for the working voltage range of the IGBT device, has high safety and good linearity. In the embodiment, the voltage sensor is connected in parallel across the collector and the emitter of the IGBT device, the voltage polarity is consistent with the polarity of the IGBT device, a shielded cable is selected to transmit the signal to reduce the voltage fluctuation error caused by electromagnetic interference. At the same time, only the collector-emitter voltage of the IGBT device in the PWM command conduction stage is collected, because the collector-emitter voltage in the off stage is meaningless.

[0072] The IGBT junction temperature is the temperature of the PN junction inside the IGBT device chip, is a direct parameter reflecting the thermal state of the IGBT device, is lower than the maximum allowable IGBT junction temperature in normal operation, exceeds the critical threshold of the IGBT junction temperature in overheating, and the continuous rise of the IGBT junction temperature will be caused by overload and heat dissipation failure; in order to more accurately monitor the IGBT junction temperature, the temperature sensor of the equipment model WZP-230 is selected in the embodiment, because it can completely cover the working and fault temperature of the IGBT device, and has high precision, good stability and small size, the temperature sensor is installed on the heat dissipation substrate of the IGBT device chip in the embodiment, so as to ensure good heat conduction with the chip, in addition, the temperature sensor is electrically isolated from the high-voltage area of the IGBT device, so as to avoid the risk of short circuit.

[0073] In order to ensure the accuracy of the signal, the current sensor, the voltage sensor and the temperature sensor need to be pre-calibrated before signal acquisition, the measurement error is corrected through a standard signal source (such as a standard current source, a standard voltage source and a standard thermostat), and the accuracy is ensured to meet the requirements. When the inverter enters the normal operation state or the preventive maintenance mode, the data acquisition is started, the effective data is screened through the PWM instruction signal time sequence, only the collector current, the collector-emitter voltage of the IGBT device in the conduction stage and the IGBT junction temperature at the corresponding moment in each PWM instruction signal period are reserved. Align the PWM instruction state, collector current, collector-emitter voltage and IGBT junction temperature of the same IGBT device and the same moment based on the time stamp.

[0074] Through the above comprehensive, accurate and synchronous data acquisition scheme, the complete running state data of each IGBT device in the PWM instruction signal period can be obtained, which lays a solid foundation for accurate judgment of subsequent fault types and ensures the accuracy and reliability of fault diagnosis.

[0075] Step S2: Based on the bridge arm on-off logic performance and the change of the collector current of each IGBT device in the preset number of PWM instruction signal periods, the on-off logic analysis coefficient of each IGBT device is obtained.

[0076] Specifically, the IGBT device, as the core switching device of the three-phase full-bridge inverter, bears the key functions of electric energy conversion and high-frequency on-off control, and its working state directly determines the stability and reliability of the inverter system. It is known that the abnormal on-off logic of the bridge arm is the direct cause of the short-circuit fault, and is also the core inducement of the overcurrent surge causing the overload fault. Among them, the abnormal on-off of the bridge arm for a plurality of consecutive instruction signal periods often points to the abnormality of the IGBT device itself, while the sporadic on-off abnormality of the bridge arm is mostly transient electromagnetic interference, and the fault risk of the same on-off violation behavior under heavy load working condition is much higher than that under light load working condition; therefore, based on the on-off logic performance and the change of the collector current of each IGBT device within the preset number of PWM instruction signal periods, the on-off logic analysis coefficient of each IGBT device is obtained, which accurately reflects the potential risk level of each IGBT device causing short-circuit or overload fault due to on-off logic abnormality, and is beneficial to subsequent accurate distinction of short-circuit and overload fault types. In the embodiment, the preset number is set to 10 and the 10 PWM instruction signal periods are uninterrupted, which can not only ensure the stability of the statistical result (effectively filter transient interference and highlight the real running state of the device), but also consider the real-time diagnosis (avoid diagnosis delay due to too many periods), and the implementer can set the size of the preset number according to the actual situation, which is not limited herein.

[0077] Preferably, in one implementation manner of the embodiment, the method for obtaining the on-off logic analysis coefficient can refer to Figure 2 which shows a flow chart of a method for obtaining an on-off logic analysis coefficient provided by the embodiment, and the method comprises the following steps:

[0078] Step S201: According to the number of periods of on-off logic compliance and the number of periods of continuous on-off logic violation of each IGBT device within the preset number of PWM instruction signal periods, the on-off logic compliance degree of each IGBT device is obtained.

[0079] When the number of periods of on-off logic compliance of a certain IGBT device within the preset number of PWM instruction signal periods is more, it means that the on-off logic of the IGBT device is more in line with the normal situation; when the number of periods of continuous on-off logic violation of the IGBT device is more, it indirectly means that the on-off logic of the IGBT device is less in line with the normal situation; and then, according to the number of periods of on-off logic compliance and the number of periods of continuous on-off logic violation of each IGBT device within the preset number of PWM instruction signal periods, the on-off logic compliance degree of each IGBT device is obtained. The greater the on-off logic compliance degree, the more in line with the normal situation of the on-off of the corresponding IGBT device, the higher the overall compliance, and the more stable the on-off state.

[0080] In an implementable manner of the embodiment, the obtaining method of the compliance degree of the on-off logic is as follows: for any IGBT device, a ratio of the number of cycles in which the bridge arm on-off logic of the IGBT device is compliant to a preset number (10 PWM instruction signal cycles in the embodiment) is taken as a first compliance analysis value of the IGBT device; the larger the first compliance analysis value, the higher the overall compliance level of the bridge arm on-off logic of the IGBT device in the statistical cycle, and the higher the proportion of cycles that meet the operation rules; the number of cycles in which the continuous bridge arm on-off logic of the IGBT device is non-compliant is further obtained, and all the numbers are taken as first numbers; the larger the first numbers, the more likely that the IGBT device has persistent on-off logic abnormalities, which are more likely to be caused by performance degradation or driving failure of the device itself rather than transient electromagnetic interference, and the fault risk is significantly higher than that of sporadic non-compliance; to highlight the impact of continuous non-compliant cycles on the fault risk, the embodiment takes a ratio of the largest first number to the preset number as a compliance correction weight of the IGBT device, which is negatively correlated; the larger the compliance correction weight, the shorter the longest continuous non-compliant cycle of the IGBT device, which indirectly indicates that the fault risk caused by continuous non-compliance is lower; the embodiment negatively correlates the ratio of the largest first number to the preset number by taking 1-the ratio of the largest first number to the preset number; to represent the compliance state of the bridge arm on-off logic of the IGBT device, the product of the first compliance analysis value and the compliance correction weight of the IGBT device is taken as the compliance degree of the on-off logic of the IGBT device.

[0081] At this point, the compliance degree of the on-off logic of each IGBT device is obtained.

[0082] Step S202: Obtain a current-related correction coefficient of each IGBT device according to the collector current change of each IGBT device in the preset number of PWM instruction signal cycles.

[0083] When the collector current of a certain IGBT device is larger and the rising rate is faster in the preset number of PWM instruction signal cycles, it indicates that the current impact intensity borne by the IGBT device is stronger, which indirectly reflects that the bridge arm on-off state of the IGBT device deviates from the normal operation requirement, and the risk of causing short circuit or overload failure is higher. Therefore, the embodiment obtains a current-related correction coefficient of each IGBT device according to the collector current change of each IGBT device in the preset number of PWM instruction signal cycles, and the smaller the current-related correction coefficient, the stronger the current impact of the corresponding IGBT device, and the higher the risk of causing short circuit or overload failure caused by the abnormality of the bridge arm on-off logic.

[0084] In an implementable manner of the embodiment, the current-related correction coefficient is obtained by: for any IGBT device, obtaining the ratio of the collector current rise rate of the IGBT device in each PWM instruction signal period to the preset rated current rise rate, as the current impact degree; the greater the current impact degree, the more intense the current mutation of the IGBT device in the corresponding instruction signal period, the stronger the instantaneous electric stress, and the more prone to short-circuit failure caused by abnormal on-off logic; wherein the preset rated current rise rate is the rated current rise rate marked in the IGBT device manual, which is an objective criterion for judging whether the current impact is abnormal;

[0085] It should be noted that the collector current rise rate of the IGBT device in each PWM instruction signal period is obtained by: for any PWM instruction signal period, fitting the collector current of the IGBT device in the PWM instruction signal period into a curve in time sequence, obtaining the maximum point and the minimum point on the curve; for any maximum point, the difference between the collector current corresponding to the maximum point and its previous adjacent minimum point is taken as a first value; the time period formed by the time points corresponding to the maximum point and its previous adjacent minimum point is taken as a first time period; the ratio of the first value to the first time period is taken as the current rise rate corresponding to the maximum point; if the maximum point does not have a previous adjacent minimum point, the collector current at the start time of the PWM instruction signal period is taken as the collector current corresponding to the previous adjacent minimum point of the maximum point; the current rise rates corresponding to all maximum points are obtained, and the maximum current rise rate is taken as the collector current rise rate of the IGBT device in the PWM instruction signal period; wherein the method of fitting the curve is a known technology and will not be described in detail;

[0086] In order to more accurately analyze the risk of continuous overloading of the current load, the ratio of each collector current of the IGBT device in the preset number of PWM instruction signal periods to the preset rated current is taken as a current overage analysis value; the greater the current overage analysis value, the more the actual load of the IGBT device exceeds the rated range, and the longer the IGBT device is in this state, which is prone to cause thermal accumulation and further trigger an overload failure; wherein the preset rated current is the rated working current marked in the IGBT device manual, which is a core parameter for determining whether the load is overrated;

[0087] In order to comprehensively represent the intensity of current impact and the continuous level of load overage, realize the adaptive fusion with the compliance degree of on-off logic, and further take the product of the maximum current impact degree and the mean value of the current overage analysis value as the current-related correction coefficient of the IGBT device. The embodiment realizes the adaptive fusion of the current-related correction coefficient with the compliance degree of the on-off logic by The product of the maximum current impact degree and the average of the current excess analysis value is negatively correlated and normalized, where x denotes the product of the maximum current impact degree and the average of the current excess analysis value, and norm is a linear normalization function.

[0088] At this point, the current-related correction coefficient of each IGBT device is obtained.

[0089] Step S203: The product of the on-off logic compliance degree of each IGBT device and the current-related correction coefficient is taken as the on-off logic analysis coefficient of each IGBT device.

[0090] The greater the on-off logic compliance degree, the higher the overall compliance level of the bridge arm on-off logic of the corresponding IGBT device, the shorter the continuous violation period, the more stable the on-off state, and the lower the potential risk of short circuit and overload failure caused by on-off logic abnormalities. The greater the current-related correction coefficient, the weaker the current impact intensity of the IGBT device, the lower the degree of continuous overcurrent, and the smaller the amplification effect of the current working condition on the failure risk. In order to comprehensively represent the compliance level of the bridge arm on-off logic and the failure risk of the current working condition, realize the organic integration of the two types of core features, and then accurately quantify the comprehensive risk level of each IGBT device causing short circuit and overload failure due to on-off logic abnormalities, the product of the on-off logic compliance degree of each IGBT device and the current-related correction coefficient is taken as the on-off logic analysis coefficient of each IGBT device in the embodiment, which directly reflects the failure risk of each IGBT device. The greater the on-off logic analysis coefficient, the more compliant the on-off logic of the corresponding IGBT device, the lower the current impact risk, and the safer the running state. The smaller the on-off logic analysis coefficient, the higher the probability of short circuit or overload failure of the corresponding IGBT device.

[0091] At this point, the on-off logic analysis coefficient of each IGBT device is obtained.

[0092] Step S3: Based on the collector-emitter voltage fluctuation of each IGBT device within a preset number of PWM instruction signal periods, as well as the collector current, collector-emitter voltage, and IGBT junction temperature, the conduction efficiency coefficient of each IGBT device is obtained.

[0093] Specifically, it is known that the turn-on efficiency of the IGBT device directly determines the energy conversion efficiency and thermal stability, and is the core basis for judging the two types of faults of open circuit and overheating. From the fault mechanism, the collector-emitter voltage drop of the IGBT device should be stable in the rated range (usually 1V-2V) marked in the IGBT device manual in the preset number (10 in this embodiment) of PWM instruction signal periods under the normal turn-on state. If the collector-emitter voltage drop fluctuates greatly, it is probably directed to two types of abnormalities: one is intermittent open circuit fault (such as intermittent conduction caused by poor contact of IGBT device pin), and the other is performance degradation of IGBT device caused by overheating (long-term high temperature leads to deterioration of turn-on characteristics); from the nature of energy transfer, the energy transfer efficiency is the core index for quantifying the turn-on efficiency of the IGBT device, and the turn-on loss is the main source of heat of the IGBT device. The higher the loss, the faster the temperature rise rate of the IGBT device, which is the direct cause of the overheating fault; at the same time, even the intermittent open circuit fault will also cause current waveform distortion, thereby indirectly increasing the turn-on loss.

[0094] The turn-on loss of the IGBT device is determined by the collector-emitter voltage drop and the collector current, and the nature of the energy transfer efficiency is the ratio of the effective energy transfer to the total energy consumption. When the actual turn-on loss is closer to the rated loss of the IGBT device, the less invalid energy consumption caused by abnormal voltage drop or current overrun in the conduction process, the better the turn-on efficiency of the IGBT device, and the lower the risk of overheating. Therefore, the difference between the rated loss and the actual loss can be used to quantify the energy transfer efficiency. It should be noted that the turn-on voltage drop of the IGBT device will change with the change of the IGBT junction temperature. As the IGBT junction temperature rises, the turn-on voltage drop shows a linear downward trend. This characteristic will directly affect the calculation accuracy of the turn-on loss. Therefore, a correction term corresponding to the IGBT junction temperature needs to be introduced to compensate for the influence of temperature on the turn-on characteristics, so as to ensure the accuracy of the quantification result of the energy transfer efficiency.

[0095] Therefore, in this embodiment, the turn-on efficiency coefficient of each IGBT device is obtained based on the collector-emitter voltage fluctuation of each IGBT device in the preset number of PWM instruction signal periods, as well as the collector current, collector-emitter voltage and IGBT junction temperature, so as to realize accurate evaluation of the turn-on state of the IGBT device.

[0096] Preferably, in an implementable manner of this embodiment, the method for obtaining the turn-on efficiency coefficient is as follows: Figure 3 which shows a flowchart of a method for obtaining a turn-on efficiency coefficient provided by this embodiment. The method comprises the following steps:

[0097] Step S301: Obtain the voltage stability degree.

[0098] The more stable the collector-emitter voltage of an IGBT device is, the more stable the conduction state of the IGBT device is, and there is no intermittent open circuit or poor contact abnormality, and the voltage loss fluctuation in the energy conversion process is small, and the conduction efficiency is better. Then, for any IGBT device, the ratio of each collector-emitter voltage of the IGBT device in a preset number of PWM command signal periods to a preset rated voltage is taken as a voltage excess analysis value. The preset rated voltage is the rated conduction voltage drop (usually 1V-2V) marked in the IGBT device manual. The more the voltage excess analysis value tends to 1, the closer the collector-emitter voltage of the IGBT device is to the rated value, the smaller the deviation of the conduction voltage drop is, and the less the voltage abnormality caused by obvious overheating or open circuit is.

[0099] To quantify the overall fluctuation level of the collector-emitter voltage, and then the result of the negative correlation and normalization of the standard deviation of the voltage excess analysis value is taken as the voltage stability degree of the IGBT device. The greater the voltage stability degree is, the smaller the fluctuation amplitude of the collector-emitter voltage of the IGBT device in the preset PWM command signal period is, the stronger the consistency of the conduction state is, and the lower the risk of open circuit and overheating failure caused by voltage abnormality is. The embodiment takes the opposite of the above standard deviation as the power of an exponential function with a natural constant as the base number. The output result of the exponential function is the result of the negative correlation and normalization of the above standard deviation.

[0100] For better description, the IGBT device is taken as an example for analysis in the following.

[0101] Step S302: Obtain the loss degree analysis value.

[0102] Since the conduction loss of the IGBT device is essentially the power loss in its conduction state, it can be characterized by the product of voltage and current. Then, for any PWM command signal period, the product of the average collector-emitter voltage and the average collector current of the IGBT device in the PWM command signal period is taken as the actual loss of the IGBT device in the PWM command signal period. It should be noted that the average collector-emitter voltage and the average collector current of the IGBT device in the PWM command signal period are both data in the conduction phase of the PWM command signal period. In order to comprehensively evaluate the conduction loss level of the IGBT device in a preset number of PWM command signal periods (10 periods in this embodiment), eliminate the interference of instantaneous fluctuation in a single PWM command signal period on loss evaluation, and then take the average value of the actual loss of the IGBT device in the preset number of PWM command signal periods as the overall actual loss of the IGBT device.

[0103] To quantify the deviation degree of the overall actual loss from the rated loss, and then linearly normalize the difference between the overall actual loss and the preset rated loss as the loss quota analysis value of the IGBT device. The preset rated loss is the rated on-state power loss marked in the IGBT device manual, which is the core reference to determine whether the IGBT device loss is over-standard. The larger the loss quota analysis value is, the higher the overall actual loss of the IGBT device is than the preset rated loss, the more invalid consumption in the energy transfer process, the worse the on-state efficiency, and the higher the risk of heating due to loss accumulation under long-term operation, which is more likely to cause overheating failure; on the contrary, the smaller the loss quota analysis value is, the closer the actual loss is to the rated loss, the better the on-state efficiency, and the lower the risk of overheating.

[0104] Step S303: Obtain the junction temperature correction coefficient.

[0105] Considering that the IGBT junction temperature will affect the calculation accuracy of the on-state voltage drop and the on-state loss, the higher the IGBT junction temperature is, the more serious the internal heat accumulation of the IGBT device is, and the greater the linear decline of the on-state voltage drop is. If the IGBT junction temperature is ignored, the loss calculation result will be distorted, and then the embodiment linearly normalizes the average value of the IGBT junction temperature of the IGBT device in the preset number of PWM instruction signal periods as the junction temperature correction coefficient of the IGBT device. The larger the junction temperature correction coefficient is, the closer the average IGBT junction temperature of the IGBT device is to its rated maximum junction temperature (usually 150-175℃), the higher the thermal stress level is, and the more significant the influence of the IGBT junction temperature on the on-state voltage drop and the loss is, which needs to be given a higher correction weight in the subsequent on-state efficiency coefficient calculation. The rated maximum junction temperature is the junction temperature parameter marked in the IGBT device manual.

[0106] Step S304: Obtain the energy transfer analysis value.

[0107] The greater the loss rating analysis value is, the higher the overall actual loss of the IGBT device is than the preset rated loss, the more invalid consumption in the energy transfer process is, the poorer the conduction efficiency is, and the higher the risk of heating and overheating failure under long-term operation is. The greater the junction temperature correction coefficient is, the closer the average junction temperature of the IGBT device is to its rated maximum junction temperature, the higher the thermal stress level is, and the more significant the influence of the junction temperature on the conduction voltage drop and loss calculation is. In order to comprehensively characterize the actual loss exceeding degree and the correction weight of the junction temperature on the loss, eliminate the deviation caused by the temperature influence on the simple loss evaluation, and then multiply the loss rating analysis value of the IGBT device by the junction temperature correction coefficient, the product is taken as the energy transfer analysis value of the IGBT device, which accurately reflects the real energy transfer efficiency and failure risk level of the IGBT device under the dual action of temperature-coupled loss. Because the loss rating analysis value and the junction temperature correction coefficient are both normalized data, the value range of the loss rating analysis value and the junction temperature correction coefficient is 0 to 1, and then the value range of the energy transfer analysis value is also 0 to 1. The greater the energy transfer analysis value is, the more the loss exceeding problem of the IGBT device is further amplified by high temperature, the lower the energy transfer efficiency is, the more serious the conduction efficiency degradation is, and the higher the probability of causing overheating failure is.

[0108] Step S305: Obtain the conduction efficiency coefficient.

[0109] The greater the voltage stability degree is, the smaller the collector-emitter voltage fluctuation amplitude of the IGBT device is, the stronger the consistency of the conduction state is, and the less the abnormal situation such as intermittent open circuit or poor contact is. The greater the energy transfer analysis value is, the more the loss exceeding problem of the IGBT device is further amplified by high temperature, the lower the energy transfer efficiency is, the more serious the conduction efficiency degradation is, and the higher the probability of causing overheating failure is. In order to comprehensively characterize the voltage stability characteristics and the comprehensive influence of temperature-loss coupling risk on the conduction state of the IGBT device, realize the organic integration of the two types of core indicators, and accurately quantify the conduction efficiency level of the IGBT device, the product of the negative correlation result of the energy transfer analysis value of the IGBT device and the voltage stability degree is taken as the conduction efficiency coefficient of the IGBT device. The greater the conduction efficiency coefficient is, the smaller the voltage fluctuation and the lower the temperature-loss coupling risk of the IGBT device are, the more stable the conduction state is, the higher the energy transfer efficiency is, and the lower the probability of open circuit and overheating failure is. Based on the normalized value range of the energy transfer analysis value of 0 to 1, the result of subtracting the energy transfer analysis value from 1 is taken as the negative correlation processing result of the energy transfer analysis value.

[0110] At this point, the conduction efficiency coefficient of each IGBT device is obtained.

[0111] Step S4: based on the on-off logic analysis coefficient and the conduction efficiency coefficient of each IGBT device, combined with the fault judgment threshold and the constraint condition, the fault type label of each IGBT is obtained; the fault type label includes short circuit, open circuit, overload, overheating and no fault.

[0112] Specifically, it is known that the value range of the on-off logic analysis coefficient is 0 to 1, the smaller the on-off logic analysis coefficient, the worse the on-off logic compliance of the corresponding IGBT device, the greater the current impact strength, and the higher the probability of short circuit or overload failure; on the contrary, the larger the on-off logic analysis coefficient, the more stable the on-off state of the corresponding IGBT device, the lower the risk of short circuit and overload; at the same time, the value range of the conduction efficiency coefficient is also 0 to 1, the smaller the conduction efficiency coefficient, the greater the voltage fluctuation of the corresponding IGBT device, the higher the risk of temperature-loss coupling, and the higher the probability of open circuit or overheating failure; on the contrary, the larger the conduction efficiency coefficient, the better the conduction state of the corresponding IGBT device, the lower the risk of open circuit and overheating. Therefore, based on the on-off logic analysis coefficient and the conduction efficiency coefficient of each IGBT device, combined with the fault judgment threshold and the constraint condition, the fault type label of each IGBT is obtained; wherein the fault type label includes short circuit, open circuit, overload, overheating and no fault.

[0113] Wherein, the fault judgment threshold includes the on-off logic abnormal boundary threshold, the short circuit-overload distinction threshold, the conduction efficiency abnormal boundary threshold, the open circuit-overheating distinction threshold and the junction temperature critical threshold. It should be noted that for any IGBT device, the on-off logic abnormal boundary threshold of the IGBT device is the difference between the mean value of the on-off logic analysis coefficient of the IGBT device under normal working condition and three times the standard deviation thereof; this embodiment uses the statistical 3σ principle to determine the boundary between normal working condition and abnormal working condition, and the value of the mean value of the above on-off logic analysis coefficient minus three times the standard deviation can cover 99.73% of the normal working condition coefficient fluctuation, only when the on-off logic analysis coefficient is lower than the on-off logic abnormal boundary threshold, it is determined that the on-off logic is abnormal, which avoids false judgment caused by instantaneous fluctuation and accurately captures the on-off logic analysis coefficient mutation caused by short circuit and overload failure; it should be noted that if the difference between the mean value of the on-off logic analysis coefficient of the IGBT device under normal working condition and three times the standard deviation thereof is less than 0, 0 is defaulted as the on-off logic abnormal boundary threshold of the IGBT device; wherein, the 3σ principle is a known content and will not be described in detail; this embodiment sets the normal working condition as 10 consecutive PWM instruction signal periods without any fault, and the implementer can set the normal working condition according to the actual situation, which is not limited here;

[0114] The short-circuit-overload distinguishing threshold of the IGBT device is a result of negative correlation of the ratio of the saturation collector-emitter voltage to the bus voltage under normal working conditions; wherein, 1 minus the difference of the ratio of the saturation collector-emitter voltage to the bus voltage is a result of negative correlation processing of the ratio of the saturation collector-emitter voltage to the bus voltage; wherein, the saturation collector-emitter voltage and the bus voltage are explicitly marked in the IGBT device manual and can be directly used; the core feature of the short-circuit fault is that the collector-emitter voltage of the IGBT device drops to the saturation voltage drop, which is much lower than the bus voltage; and the overload fault only shows current overrun, and the voltage is still in the normal range. By 1 minus the difference of the ratio of the saturation collector-emitter voltage to the bus voltage, the short-circuit working condition can be mapped to a high value, and the overload working condition can be mapped to a low value, and then the short-circuit-overload distinguishing threshold is used to accurately distinguish the two types of faults;

[0115] The conduction efficiency abnormality dividing threshold of the IGBT device is a difference between the mean value of the conduction efficiency coefficient of the IGBT device under normal working conditions and three times the standard deviation thereof; consistent with the design logic of the on-off logic abnormality dividing threshold, based on the 3σ principle, the small amplitude change of the conduction efficiency coefficient caused by voltage fluctuation and loss fluctuation under normal working conditions is filtered out, and only when the conduction efficiency coefficient is lower than the conduction efficiency abnormality dividing threshold, it is determined that the conduction efficiency is abnormal, corresponding to the risk interval of open circuit or overheating fault, which guarantees the statistical significance of fault judgment;

[0116] The open circuit-overheating distinguishing threshold of the IGBT device is preset after fitting and calibration according to experimental data under different fault simulation conditions; the core feature of the open circuit fault is that the collector-emitter voltage fluctuates greatly, which is relatively obvious; the core feature of the overheating fault is that the loss is out of standard and the IGBT junction temperature rises, and the influence degree of the two types of faults on the conduction efficiency coefficient is different. By collecting the coefficient data of open circuit and overheating fault simulation experiments, the optimal distinguishing threshold is determined through fitting and calibration, which can avoid the confusion of the judgment of the two types of faults and improve the classification accuracy; wherein, the larger the open circuit-overheating distinguishing threshold, the more likely the IGBT device has an overheating fault; the smaller the open circuit-overheating distinguishing threshold, the more likely the IGBT device has an open circuit fault; because the deterioration degree of the open circuit fault on the conduction efficiency of the IGBT device is much higher than that of the overheating fault;

[0117] The junction temperature critical threshold of the IGBT device is the maximum IGBT junction temperature of the IGBT device minus a specified number of three times the IGBT junction temperature standard deviation of the IGBT device under normal working conditions. The maximum IGBT junction temperature marked in the IGBT device manual is the limit value for safe operation, and minus three times the standard deviation of the IGBT junction temperature under normal working conditions, a reasonable temperature safety margin can be reserved to avoid false alarms due to normal temperature fluctuations. At the same time, the junction temperature critical threshold is directly related to the core mechanism of the overheat fault, when the actual IGBT junction temperature is close to the junction temperature critical threshold, it is determined that there is an overheat fault risk, and early warning of the fault is realized. The specified number in this embodiment is 100, which can meet the significance requirement of statistical analysis, and the implementer can set the size of the specified number according to the actual situation, which is not limited here.

[0118] It should be noted that the short-circuit-overload distinguishing threshold is less than the on-off logic abnormality dividing threshold; and the open-circuit-overheat distinguishing threshold is less than the conduction efficiency abnormality dividing threshold. The on-off logic abnormality dividing threshold is the lower limit critical value of the on-off logic analysis coefficient under normal working conditions; and the conduction efficiency abnormality dividing threshold is the lower limit critical value of the conduction efficiency coefficient under normal working conditions.

[0119] Preferably, in an implementable manner of this embodiment, the acquisition method of the fault type label is as follows: for any IGBT device, when the on-off logic analysis coefficient of the IGBT device is less than the short-circuit-overload distinguishing threshold, and the conduction efficiency coefficient is greater than or equal to the conduction efficiency abnormality dividing threshold, the fault type label of the IGBT device is short circuit; because when the bridge arm on-off logic is abnormal, the actual voltage and current are mismatched, but the IGBT device itself is not damaged and can normally conduct;

[0120] When the on-off logic analysis coefficient of the IGBT device is greater than or equal to the short-circuit-overload distinguishing threshold and less than the on-off logic abnormality dividing threshold, and the conduction efficiency coefficient is greater than or equal to the conduction efficiency abnormality dividing threshold, the fault type label of the IGBT device is overload; because overload is a cumulative fault of long-term current exceeding the rated range of the IGBT device, and the IGBT device can normally conduct, only the current waveform is distorted due to the over-standard load;

[0121] When the on-off logic analysis coefficient of the IGBT device is greater than or equal to the on-off logic abnormality dividing threshold, and the conduction efficiency coefficient is less than the open-circuit-overheat distinguishing threshold, the fault type label of the IGBT device is open circuit; because the open circuit fault is caused by the damage of the conduction ability of the IGBT device itself, which leads to the failure to form an effective current path, and belongs to the device fault; although the PWM instruction (a signal for controlling the turn-on of the IGBT) is normally output, the actual conduction time of the IGBT device is much lower than the instruction requirement, and the conduction matching degree is reduced.

[0122] When the on-off logic analysis coefficient of the IGBT device is greater than or equal to the on-off logic abnormality demarcation threshold, the on performance coefficient is greater than or equal to the open circuit-heat over threshold and less than the on performance abnormality demarcation threshold, and the IGBT junction temperature is greater than the junction temperature critical threshold, the fault type label of the IGBT device is heat over. Because heat over is a temperature-induced cumulative failure, the on loss and switching loss generated when the IGBT device is working, in the scenario of insufficient heat dissipation capacity, the heat cannot be dissipated in time, causing the IGBT junction temperature to continue to rise, eventually exceeding the safety threshold. Due to the increase of the IGBT junction temperature, the on performance of the IGBT device decreases slightly, but does not reach the serious failure degree of open circuit, and the PWM instruction is normal, excluding the failure caused by logic abnormality or load exceeding, and it is clear that the failure is caused by internal loss accumulation; the heat continues to accumulate and gradually deviates from the normal fluctuation range;

[0123] When the on-off logic analysis coefficient of the IGBT device is greater than or equal to the on-off logic abnormality demarcation threshold, the on performance coefficient is greater than or equal to the on performance abnormality demarcation threshold, the fault type label of the IGBT device is no fault.

[0124] At this point, the fault type label of each IGBT device is obtained.

[0125] Step S5: constructing a fault sample set based on the fault type label, training a fault type recognition model, and obtaining a trained fault type recognition model; the fault type recognition model is used to output the fault type of each IGBT device in the three-phase full-bridge inverter.

[0126] Specifically, in order to realize efficient and accurate identification of the fault type of the IGBT device in the three-phase full-bridge inverter, the embodiment constructs a fault sample set based on the fault type label of the IGBT device analyzed historically and the corresponding on-off logic analysis coefficient and on performance coefficient, to train a fault type recognition model; wherein the fault type recognition model can directly output the fault type of each IGBT device in the three-phase full-bridge inverter, including short circuit, open circuit, overload, heat over and no fault.

[0127] The construction of the fault sample set takes data integrity, label accuracy, and feature effectiveness as core principles to ensure that the sample data can comprehensively cover typical operating conditions and fault modes of the IGBT device, and to provide reliable supervised data for model training. The sample data in the embodiment is derived from two parts of laboratory fault simulation experiments and field operation historical data of the three-phase full-bridge inverter. The laboratory fault simulation experiment is to build a three-phase full-bridge inverter test platform consistent with actual application. By simulating different fault types (such as short-circuit fault by bridge arm through simulation, open-circuit fault by device pin circuit simulation, overload fault by loading excess load simulation, and overheat fault by blocking heat dissipation channel simulation), the full-cycle operation data of each IGBT device before, during and after the fault occurs are collected. The field operation historical data is to collect the operation data of the inverter in the actual industrial scene, and to filter the fault data and normal operation data with clear labels to supplement the scene limitations of the laboratory simulation data.

[0128] The feature vector of each sample is composed of the state features and topology correlation features of the IGBT device. The state features are the on-off logic analysis coefficient, conduction efficiency coefficient and real-time IGBT junction temperature of each IGBT device. The topology correlation features are the state features of the adjacent devices (such as upper and lower tubes on the same bridge arm, parallel devices on the same layer) that are physically connected to the target IGBT device, which reflect the influence of the topology correlation between devices on fault propagation. Finally, the sample feature vector dimension of each IGBT device is 5, which ensures that the feature information contains not only the state of the device itself, but also the correlation characteristics of the circuit topology.

[0129] Each sample corresponds to only one fault type label to avoid ambiguity of labeling multiple labels for the same sample. The labels of the laboratory simulation data are directly labeled based on the fault simulation method, and the labels of the field historical data are cross-validated combined with operation records and fault determination rules to ensure the accuracy of the labels. To balance the number of each type of fault sample (such as fewer short-circuit fault samples and more no-fault samples), a combination of oversampling (data augmentation for scarce fault samples) and undersampling (random selection of no-fault samples) is used to balance the number ratio of each type of fault sample and avoid model training bias. To eliminate the influence of the dimension difference of different feature parameters on model training, all feature parameters in the sample set are normalized to map the feature values to the [0, 1] interval. The linear normalization method is used to normalize the feature parameters in the embodiment, and the normalization method is not limited here.

[0130] The fault type recognition model of the embodiment adopts a graph neural network (GNN) architecture, preferably a graph convolution network (GCN) as the core model, which has the advantage of effectively utilizing the topological correlation information of the IGBT devices in the three-phase full-bridge inverter and breaking through the limitation of traditional machine learning models that ignore the physical connection between devices. The overall architecture of the model includes an input layer, a graph convolution layer, and an output layer. The core function of the input layer is to adapt the graph structure data to the model, and the input data includes a node feature matrix and an adjacency matrix. The dimension of the node feature matrix is ; N is the total number of IGBT devices in the three-phase full-bridge inverter, and F is the feature dimension of a single node (5 dimensions, including turn-on and turn-off logic analysis coefficients, conduction efficiency coefficients, real-time IGBT junction temperature, and the average turn-on and turn-off logic analysis coefficients and conduction efficiency coefficients of the corresponding IGBT device and adjacent IGBT devices). The dimension of the adjacency matrix is , which is used to represent the physical topological correlation between IGBT devices. If two IGBT devices have a physical connection in series or parallel, the corresponding position element of the adjacency matrix is 1, otherwise it is 0. The adjacency matrix is also normalized to avoid numerical bias in the feature aggregation process. The graph convolution layer is the core part of the model, usually with 2-3 layers, which is not limited here. Its role is to update the feature representation of the target node by aggregating the neighbor node features of the target node through multiple layers, achieving deep fusion of local topological information and node state. Through multiple graph convolution operations, the feature representation of each IGBT node gradually integrates the state information of adjacent devices, effectively capturing the fault propagation characteristics caused by topological correlation. The output layer consists of a fully connected layer and a Softmax classifier, which maps the node features to the fault type probability.

[0131] In addition, the training target of the fault type recognition model is to minimize the difference between the predicted fault type and the true label. Therefore, the cross-entropy loss function is used as the optimization objective function. The graph convolution network and the cross-entropy loss function are well-known and will not be described in detail.

[0132] Thus, a trained fault type recognition model is obtained.

[0133] The trained fault type recognition model is deployed in the fault diagnosis system of the three-phase full-bridge inverter, and the specific application process is as follows: for the real-time running inverter circuit, the feature parameters of each IGBT device are extracted according to the sample set construction rule to construct the node feature matrix; the adjacency matrix is generated according to the inverter topological structure, and the node feature matrix and the adjacency matrix together form the graph structure input data; the input data is input into the trained fault type recognition model, and the model outputs the fault type probability distribution of each IGBT node; the class with the maximum probability value is taken as the final fault type of the IGBT device, and the fault type recognition of all IGBT devices in the inverter is realized.

[0134] For the IGBT device fault label output by the fault type identification model, combined with the danger and cause of different faults, a hierarchical disposal strategy is adopted to realize the rapid response and accurate investigation of the fault, and to avoid the damage of equipment or safety risk caused by the expansion of the fault; if the model outputs a short-circuit fault label, the instantaneous large current impact caused by the short-circuit fault will easily lead to serious consequences such as bus voltage collapse and device burnout, so the shutdown protection mechanism needs to be triggered immediately to cut off the power supply of the inverter main circuit; the fault investigation focuses on three aspects: first, detect whether the IGBT chip is broken down by using a multimeter or an insulation tester to check the conduction state between the collector and the emitter; second, check whether there is a short-circuit point in the bridge arm circuit, and focus on checking whether there are problems such as insulation damage, metal lapping, etc. in the bus capacitor, power cable and terminal; third, detect whether the IGBT drive signal is abnormal, observe the PWM waveform output by the drive board through the oscilloscope, confirm whether the drive voltage amplitude, pulse width and timing meet the rated standard, and check whether the drive chip and isolation optocoupler are damaged;

[0135] If the model outputs an open-circuit fault label, the machine needs to be shut down and the power supply needs to be cut off to avoid the imbalance of three-phase output caused by single-phase open circuit, which will cause abnormal vibration or damage of the load such as motor; the fault investigation steps are as follows: first, check the pin connection of the corresponding IGBT device to confirm whether there are problems such as virtual welding, falling off or oxidation corrosion; second, detect the drive circuit to check whether the drive power supply is normal, whether the drive resistor is burned out, and whether the drive cable is broken; finally, test the IGBT chip individually, if it is confirmed that the chip is internally broken, replace the IGBT device of the same type, and perform a conduction test after replacement to ensure that the circuit connection is normal;

[0136] If the model outputs an overload fault label, it does not need to be shut down immediately, and a temporary load reduction operation can be performed first to gradually reduce the output power of the inverter and reduce the current load of the IGBT device to prevent device thermal aging caused by long-term overload; the fault investigation direction includes: first, check whether there is an abnormality at the load end, such as whether the mechanical equipment is stuck or locked, whether the load cable is short-circuited or poorly contacted; second, check the transmission system to confirm whether the reducer, coupling and other components are running normally to avoid excessive load current caused by excessive mechanical resistance; third, verify the current protection parameter setting of the inverter to confirm whether the overload protection threshold matches the actual load demand, and recalibrate the parameters if necessary. After the load abnormality problem is solved, gradually restore the rated power operation, and continuously monitor the current waveform and temperature change of the IGBT to ensure stable operation;

[0137] If the model output fault label is overheating, the heat dissipation system needs to be checked first to avoid the continuous rise of junction temperature, which leads to the increase of on-state voltage drop of IGBT device, the attenuation of service life, and even the initiation of thermal breakdown fault. The specific troubleshooting measures are as follows: first, clean the dust accumulated in the heat dissipation system, check whether the surface of the heat sink is covered with dust and oil stains, and whether the heat dissipation efficiency is affected; second, detect whether the cooling fan is running normally, check whether the fan motor is damaged and whether the rotating speed meets the standard, and replace the faulty fan if necessary; for the liquid cooling heat dissipation system, check whether the cooling liquid level, flow rate and cooling pipeline are blocked to ensure the smoothness of the heat dissipation circulation. At the same time, the surface temperature of the IGBT module can be monitored by an infrared thermometer to confirm whether the temperature falls back to the rated range after the heat dissipation system returns to normal.

[0138] If the model output fault label is no fault, it indicates that the on-off logic and on-state efficiency of the IGBT device are in normal working condition, and no additional intervention operation is needed. Only the normal monitoring needs to be maintained, and the running parameters (current, voltage and junction temperature) of the IGBT are collected regularly to record the equipment running state and provide data support for subsequent preventive maintenance.

[0139] In summary, the PWM instruction signal in the three-phase full-bridge inverter, the electrical signal and the temperature signal of the IGBT device are obtained. Based on the bridge arm on-off logic performance, the electrical signal and the temperature signal of the IGBT device in the PWM instruction signal period, the on-off logic analysis coefficient and the on-state efficiency coefficient of the IGBT device are obtained. Combined with the fault determination threshold and the constraint condition, the fault type label of the IGBT is obtained to construct the fault sample set, the fault type recognition model is trained, and the trained fault type recognition model is obtained. By accurately obtaining the on-off logic analysis coefficient and the on-state efficiency coefficient, the accuracy of the fault type label recognition is effectively improved, which is beneficial to the real-time and accurate identification of the fault type of the IGBT device, timely processing, and effectively reducing the fault loss of the three-phase full-bridge inverter.

[0140] Embodiment 2:

[0141] The application also provides a three-phase full-bridge inverter fault diagnosis system, please refer to Figure 4 which shows a three-phase full-bridge inverter fault diagnosis system structure diagram provided by an embodiment of the application. The system comprises a data acquisition module 10, an on-off logic analysis coefficient acquisition module 20, an on-state efficiency coefficient acquisition module 30, a fault type label acquisition module 40 and a data processing module 50.

[0142] The data acquisition module 10 is used for acquiring the PWM instruction signal in the three-phase full-bridge inverter, the electrical signal and the temperature signal of each IGBT device at each time; the electrical signal comprises the collector current and the collector voltage; and the temperature signal comprises the IGBT junction temperature.

[0143] The on-off logic analysis coefficient acquisition module 20 is configured to acquire the on-off logic analysis coefficient of each IGBT device based on the on-off logic performance of the bridge arm of each IGBT device and the change of the collector current in a preset number of PWM instruction signal periods.

[0144] The conduction efficiency coefficient acquisition module 30 is configured to acquire the conduction efficiency coefficient of each IGBT device based on the collector-emitter voltage fluctuation of each IGBT device and the magnitudes of the collector current, the collector-emitter voltage and the IGBT junction temperature in a preset number of PWM instruction signal periods.

[0145] The fault type label acquisition module 40 is configured to acquire the fault type label of each IGBT device based on the on-off logic analysis coefficient and the conduction efficiency coefficient of each IGBT device, in combination with a fault determination threshold and a constraint condition; the fault type label includes short circuit, open circuit, overload, overheat and no fault.

[0146] The data processing module 50 is configured to construct a fault sample set based on the fault type label, train a fault type recognition model, and obtain the trained fault type recognition model; the fault type recognition model is configured to output the fault type of each IGBT device in the three-phase full-bridge inverter.

[0147] It should be noted that the system provided in the above embodiments is only exemplified by the division of the above functional modules, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the three-phase full-bridge inverter fault diagnosis system and the three-phase full-bridge inverter fault diagnosis method provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0148] Embodiment 3:

[0149] The application further provides a three-phase full-bridge inverter fault diagnosis device, which includes a memory and a processor, wherein the memory stores executable program code, and the processor is configured to call and execute the executable program code to execute the three-phase full-bridge inverter fault diagnosis method provided in the embodiments. The device can be a chip, an assembly or a module, and the chip can include a processor and a memory connected thereto; wherein the memory is configured to store instructions, and when the processor calls and executes the instructions, the chip can execute the three-phase full-bridge inverter fault diagnosis method provided in the above embodiments.

[0150] In addition, the embodiments of the present application also protect a computer device, please refer to Figure 5The computer device comprises a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402, wherein the processor 402 executes the computer program 403, so that the computer device can execute any one of the three-phase full-bridge inverter fault diagnosis methods described above.

[0151] Embodiment 4:

[0152] The embodiment also provides a computer readable storage medium, which stores computer program codes, and when the computer program codes run on a computer, the computer executes the related method steps to realize the three-phase full-bridge inverter fault diagnosis method provided in the above embodiment.

[0153] Embodiment 5:

[0154] The embodiment also provides a computer program product, and when the computer program product runs on a computer, the computer executes the related steps to realize the three-phase full-bridge inverter fault diagnosis method provided in the above embodiment.

[0155] The device, the computer readable storage medium, the computer program product or the chip provided in the embodiment are used to execute the corresponding method provided above, so the beneficial effects achieved by the device, the computer readable storage medium, the computer program product or the chip can refer to the beneficial effects of the corresponding method provided above, which will not be described here.

[0156] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0157] Each of the embodiments in the specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the difference from other embodiments.

Claims

1. A fault diagnosis method for a three-phase full-bridge inverter, characterized in that, The method includes the following steps: The system acquires the PWM command signal, the electrical signal and temperature signal of each IGBT device at each moment in the three-phase full-bridge inverter; the electrical signal includes the collector current and collector-emitter voltage; the temperature signal includes the IGBT junction temperature. Based on the bridge arm switching logic performance and collector current changes of each IGBT device within a preset number of PWM command signal cycles, the switching logic analysis coefficients of each IGBT device are obtained. Based on the collector-emitter voltage fluctuation of each IGBT device within a preset number of PWM command signal cycles, as well as the magnitude of collector current, collector-emitter voltage and IGBT junction temperature, the conduction efficiency coefficient of each IGBT device is obtained. Based on the on / off logic analysis coefficients and conduction efficiency coefficients of each IGBT device, combined with fault determination thresholds and constraints, a fault type label for each IGBT is obtained; the fault type label includes short circuit, open circuit, overload, overheating, and no fault. A fault sample set is constructed based on fault type labels, and a fault type identification model is trained to obtain a trained fault type identification model; the fault type identification model is used to output the fault type of each IGBT device in the three-phase full-bridge inverter. The method for obtaining the on / off logic analysis coefficients is as follows: The compliance level of the on / off logic of each IGBT device is obtained based on the number of cycles in which the bridge arm on / off logic of each IGBT device complies with the preset number of PWM command signal cycles and the number of cycles in which the bridge arm on / off logic of each device violates the preset number of PWM command signal cycles. Based on the collector current change of each IGBT device within a preset number of PWM command signal cycles, obtain the current correlation correction coefficient for each IGBT device. The product of the on / off logic compliance level and the current-related correction coefficient of each IGBT device is used as the on / off logic analysis coefficient of each IGBT device. The method for obtaining the compliance level of the on / off logic is as follows: For any IGBT device, the ratio of the number of cycles in which the bridge arm switching logic of the IGBT device complies with the predetermined number is used as the first compliance analysis value of the IGBT device. The number of cycles in which the continuous bridge arm switching logic of the IGBT device violates the rule is used as the first quantity. The result of negatively correlating the ratio of the largest first quantity to the preset quantity is used as the compliance correction weight for the IGBT device. The product of the first compliance analysis value and the compliance correction weight of the IGBT device is taken as the compliance level of the IGBT device's on / off logic. The method for obtaining the conduction efficiency coefficient is as follows: For any IGBT device, the ratio of each collector-emitter voltage of the IGBT device to the preset rated voltage within a preset number of PWM command signal cycles is used as the voltage over-limit analysis value. The standard deviation of the voltage excess analysis value is negatively correlated and normalized to obtain the voltage stability of the IGBT device. For any PWM command signal period, the product of the average collector-emitter voltage and the average collector current of the IGBT device during that PWM command signal period is taken as the actual loss of the IGBT device during that PWM command signal period. The average actual loss of the IGBT device within a preset number of PWM command signal cycles is taken as the overall actual loss of the IGBT device. The result of normalizing the difference between the overall actual loss and the preset rated loss is used as the loss rating analysis value of the IGBT device. The result of normalizing the average IGBT junction temperature of the IGBT device within a preset number of PWM command signal cycles is used as the junction temperature correction coefficient of the IGBT device. The product of the loss rating analysis value of the IGBT device and the junction temperature correction factor is used as the energy transfer analysis value of the IGBT device. The product of the negative correlation result of the energy transfer analysis value of the IGBT device and the voltage stability is used as the conduction efficiency coefficient of the IGBT device.

2. The fault diagnosis method for a three-phase full-bridge inverter as described in claim 1, characterized in that, The method for obtaining the current-related correction coefficient is as follows: For any IGBT device, the ratio of the collector current rise rate to the preset rated current rise rate within each PWM command signal cycle is used as the current surge level. The ratio of each collector current of the IGBT device to the preset rated current within a preset number of PWM command signal cycles is used as the current over-current analysis value. The product of the maximum current surge and the mean current excess analysis value is negatively correlated and normalized, and the result is used as the current correlation correction coefficient for this IGBT device.

3. The fault diagnosis method for a three-phase full-bridge inverter as described in claim 1, characterized in that, The fault determination thresholds include the on / off logic abnormality boundary threshold, the short circuit-overload distinction threshold, the conduction performance abnormality boundary threshold, the open circuit-overheat distinction threshold, and the junction temperature critical threshold. For any IGBT device, the abnormal threshold of the on / off logic of the IGBT device is the difference between the mean of the on / off logic analysis coefficients of the IGBT device under normal operating conditions and three times its standard deviation. The short-circuit-overload distinction threshold of this IGBT device is the result of negatively correlated with the ratio of the saturated collector-emitter voltage of this IGBT device to the bus voltage under normal operating conditions. The abnormal threshold for the conduction performance of this IGBT device is the difference between the mean of the conduction performance coefficient of the IGBT device under normal operating conditions and three times its standard deviation. The open-circuit / overheating distinction threshold of this IGBT device is preset based on experimental data; The critical junction temperature threshold of this IGBT device is the maximum IGBT junction temperature minus three times the standard deviation of the IGBT junction temperature under a specified number of normal operating conditions.

4. The fault diagnosis method for a three-phase full-bridge inverter as described in claim 3, characterized in that, The method for obtaining the fault type label is as follows: For any IGBT device, when the on / off logic analysis coefficient of the IGBT device is less than the short-circuit-overload distinction threshold and the conduction efficiency coefficient is greater than or equal to the conduction efficiency abnormal boundary threshold, the fault type label of the IGBT device is short circuit. When the on / off logic analysis coefficient of the IGBT device is greater than or equal to the short-circuit-overload distinction threshold and less than the on / off logic abnormal boundary threshold, and the conduction performance coefficient is greater than or equal to the conduction performance abnormal boundary threshold, the fault type label of the IGBT device is overload. When the on / off logic analysis coefficient of the IGBT device is greater than or equal to the on / off logic abnormality threshold, and the conduction efficiency coefficient is less than the open circuit-overheating distinction threshold, the fault type label of the IGBT device is open circuit. When the on / off logic analysis coefficient of the IGBT device is greater than or equal to the on / off logic abnormality threshold, the conduction performance coefficient is greater than or equal to the open circuit-overheating distinction threshold and less than the conduction performance abnormality threshold, and the IGBT junction temperature is greater than the junction temperature critical threshold, the fault type label of the IGBT device is overheating. When the on / off logic analysis coefficient of the IGBT device is greater than or equal to the on / off logic abnormality threshold, and the conduction performance coefficient is greater than or equal to the conduction performance abnormality threshold, the fault type label of the IGBT device is no fault.

5. The fault diagnosis method for a three-phase full-bridge inverter as described in claim 1, characterized in that, The fault type identification model is a graph neural network model; the fault sample set uses each IGBT device as an independent node and the physical topological relationship between devices as edges to construct a graph structure sample. The feature vector of each node includes the on / off logic analysis coefficient and the conduction efficiency coefficient, and the node label is the fault type label.

6. A fault diagnosis system for a three-phase full-bridge inverter, characterized in that, The system includes: The data acquisition module is used to acquire the PWM command signal, the electrical signal and temperature signal of each IGBT device at each moment in the three-phase full-bridge inverter; the electrical signal includes the collector current and collector-emitter voltage; the temperature signal includes the IGBT junction temperature. The on / off logic analysis coefficient acquisition module is used to acquire the on / off logic analysis coefficients of each IGBT device based on the on / off logic performance of the bridge arm and the change of collector current of each IGBT device within a preset number of PWM command signal cycles. The conduction efficiency coefficient acquisition module is used to obtain the conduction efficiency coefficient of each IGBT device based on the collector-emitter voltage fluctuation of each IGBT device within a preset number of PWM command signal cycles, as well as the magnitude of collector current, collector-emitter voltage and IGBT junction temperature. The fault type label acquisition module is used to acquire the fault type label of each IGBT based on the on / off logic analysis coefficient and conduction efficiency coefficient of each IGBT device, combined with the fault judgment threshold and constraints; the fault type label includes short circuit, open circuit, overload, overheat and no fault; The data processing module is used to construct a fault sample set based on fault type labels, train a fault type identification model, and obtain a trained fault type identification model; the fault type identification model is used to output the fault type of each IGBT device in the three-phase full-bridge inverter. The method for obtaining the on / off logic analysis coefficients is as follows: The compliance level of the on / off logic of each IGBT device is obtained based on the number of cycles in which the bridge arm on / off logic of each IGBT device complies with the preset number of PWM command signal cycles and the number of cycles in which the bridge arm on / off logic of each device violates the preset number of PWM command signal cycles. Based on the collector current change of each IGBT device within a preset number of PWM command signal cycles, obtain the current correlation correction coefficient for each IGBT device. The product of the on / off logic compliance level and the current-related correction coefficient of each IGBT device is used as the on / off logic analysis coefficient of each IGBT device. The method for obtaining the compliance level of the on / off logic is as follows: For any IGBT device, the ratio of the number of cycles in which the bridge arm switching logic of the IGBT device complies with the predetermined number is used as the first compliance analysis value of the IGBT device. The number of cycles in which the continuous bridge arm switching logic of the IGBT device violates the rule is used as the first quantity. The result of negatively correlating the ratio of the largest first quantity to the preset quantity is used as the compliance correction weight for the IGBT device. The product of the first compliance analysis value and the compliance correction weight of the IGBT device is taken as the compliance level of the IGBT device's on / off logic. The method for obtaining the conduction efficiency coefficient is as follows: For any IGBT device, the ratio of each collector-emitter voltage of the IGBT device to the preset rated voltage within a preset number of PWM command signal cycles is used as the voltage over-limit analysis value. The standard deviation of the voltage excess analysis value is negatively correlated and normalized to obtain the voltage stability of the IGBT device. For any PWM command signal period, the product of the average collector-emitter voltage and the average collector current of the IGBT device during that PWM command signal period is taken as the actual loss of the IGBT device during that PWM command signal period. The average actual loss of the IGBT device within a preset number of PWM command signal cycles is taken as the overall actual loss of the IGBT device. The result of normalizing the difference between the overall actual loss and the preset rated loss is used as the loss rating analysis value of the IGBT device. The result of normalizing the average IGBT junction temperature of the IGBT device within a preset number of PWM command signal cycles is used as the junction temperature correction coefficient of the IGBT device. The product of the loss rating analysis value of the IGBT device and the junction temperature correction factor is used as the energy transfer analysis value of the IGBT device. The product of the negative correlation result of the energy transfer analysis value of the IGBT device and the voltage stability is used as the conduction efficiency coefficient of the IGBT device.

7. A fault diagnosis device for a three-phase full-bridge inverter, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the fault diagnosis method for a three-phase full-bridge inverter as described in any one of claims 1-5.

Images