Transistor aging parameter detection device, aging degree prediction system and method

Abstract

The present invention provides a transistor aging parameter detection device, aging degree prediction system, and method, belonging to the field of transistor technology, comprising: a switching power supply that converts an external AC power supply into a DC power supply, a power module connected to the switching power supply, a power switch disposed between the switching power supply and the adjustable power module; a capacitor, a potentiometer, a voltage digital display, a voltage divider, and a current digital display connected to the power module; an output button disposed between the switching power supply and the adjustable power module, the output button being connected to a first relay. The present invention can flexibly perform parameter detection of the cutoff voltage and cutoff current of an IGBT in accordance with actual engineering environments, and predict the degree of IGBT aging based on the measured cutoff voltage and cutoff current; the detected cutoff voltage and cutoff current are static parameters of the IGBT, which can reflect the performance, failure status, and health of the IGBT; and can provide guidance to on-site personnel in the early stage selection process of the IGBT, failure identification during operation and maintenance, and preventive maintenance.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit equipment, and in particular to an insulated gate bipolar transistor aging parameter detection device, an aging degree prediction system and a method. Background Art

[0002] At present, there are still many problems in the device selection process and vehicle operation and maintenance process of urban rail traction-grade IGBTs. The main problems are as follows: First, since the spurious parameters, junction temperature, humidity and other factors in the data manual measurement environment are different from the actual environment, the data in the data manual and the actual working conditions are quite different, and the parameter values ​​referenced in the selection process are difficult to guide actual applications; second, the test equipment is heavy and complicated to operate, and it is difficult to evaluate the device performance, aging and failure level in the actual engineering environment, which is not conducive to the testing work of on-site operation and maintenance personnel; third, there is a lack of ability to evaluate and predict the aging degree of on-site IGBT modules, making it difficult to carry out predictive maintenance and operation, which easily leads to a waste of time and economic costs, and is not conducive to improving system reliability. Summary of the Invention

[0003] The object of the present invention is to provide an insulated gate bipolar transistor aging parameter detection device, an aging degree prediction system and a method to solve at least one technical problem existing in the above-mentioned background technology.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] In one aspect, the present invention provides a transistor aging parameter detection device, comprising:

[0006] A switching power supply for converting an external AC power supply into a DC power supply, a power module being connected to the switching power supply, and a power switch being provided between the switching power supply and the adjustable power module;

[0007] The power supply module is connected to a capacitor, a potentiometer, a voltage digital display, a voltage divider and a current digital display; an output button is provided between the switching power supply and the adjustable power supply module, and the output button is connected to a first relay.

[0008] Optionally, an input-side fuse is provided between the power switch and the power module, and the power module is connected to an output-side ceramic fuse.

[0009] Optionally, the power module is connected to a discharge resistor, a discharge button is connected between the power switch and the switching power supply, and the discharge button is connected to a second relay.

[0010] Optionally, it also includes a casing, which is provided with a power input socket connected to the switching power supply for connecting to an external AC power supply, and a voltage adjustment knob connected to the voltage divider. The output button and the discharge button are both provided on the casing. The casing is also provided with an output socket connected to the power module.

[0011] Optionally, the housing is provided with a USB interface, an Ethernet interface and a data save and upload button.

[0012] In a second aspect, the present invention provides a transistor aging degree prediction system, comprising:

[0013] The transistor aging parameter detection device as described above is used to obtain the cut-off voltage and cut-off current of the transistor to be predicted;

[0014] A prediction module is used to process the obtained cutoff voltage and cutoff current using a pre-trained prediction model to obtain the aging status of the transistor to be predicted in the next time interval; wherein the prediction model is trained by a training set, and the training set includes multiple groups of data, wherein each group of data includes a set of cutoff voltage and cutoff current, and a label indicating the aging status of the transistor corresponding to the set of cutoff voltage and cutoff current.

[0015] In a third aspect, the present invention provides a method for predicting transistor aging, comprising:

[0016] Obtaining a cutoff voltage and a cutoff current of a transistor to be predicted;

[0017] The obtained cutoff voltage and cutoff current are processed using a pre-trained prediction model to obtain the aging status of the transistor to be predicted in the next time interval; wherein the prediction model is trained by a training set, and the training set includes multiple groups of data, wherein each group of data includes a set of cutoff voltage and cutoff current, and a label that marks the aging status of the transistor corresponding to the set of cutoff voltage and cutoff current.

[0018] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium for storing computer instructions, which, when executed by a processor, implements the transistor aging degree prediction method as described above.

[0019] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed on one or more processors, is used to implement the transistor aging degree prediction method as described above.

[0020] In a sixth aspect, the present invention provides an electronic device comprising: a processor, a memory, and a computer program; wherein, the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the transistor aging degree prediction method as described above.

[0021] The beneficial effects of the present invention are as follows: it can flexibly perform parameter detection on the cut-off voltage and cut-off current of the IGBT in combination with the actual engineering environment, and predict the aging degree of the IGBT based on the measured cut-off voltage and cut-off current; the detected cut-off voltage and cut-off current are static parameters of the IGBT, which can reflect the performance, failure and health of the IGBT; it can provide guidance to on-site personnel in the early selection process of the IGBT, operation and maintenance failure judgment, and preventive maintenance process.

[0022] Additional aspects and advantages of the present invention will be set forth in part in the following description, will become apparent from the following description, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 This is a circuit structure topology diagram of the transistor aging parameter detection device according to an embodiment of the present invention.

[0025] Figure 2 This is a three-dimensional structural diagram of a transistor aging parameter detection device according to an embodiment of the present invention.

[0026] Figure 3 This is a diagram of the internal structure of the transistor aging parameter detection device according to an embodiment of the present invention.

[0027] Figure 4 This is a schematic diagram of the connection status of the transistor aging parameter detection device according to an embodiment of the present invention.

[0028] Figure 5 Schematic diagram of the process of predicting the degree of transistor aging according to an embodiment of the present invention.

[0029] Figure 6 This is a flow chart of using two independent LSTM time series prediction networks to predict the cut-off voltage and cut-off current of the IGBT module in the next time interval according to an embodiment of the present invention.

[0030] Among them: 1-switching power supply; 2-voltage adjustment knob; 3-discharge button; 4-output button; 5-first relay; 6-second relay; 7-power module; 8-output socket; 9-ammeter; 10-voltage divider; 11-voltmeter; 12-network port; 13-control interface board; 14-input socket; 15-power switch; 16-data upload and save button; 17-USB port. DETAILED DESCRIPTION

[0031] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and are not to be construed as limiting the present invention.

[0032] Those skilled in the art will understand that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs.

[0033] It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have a meaning consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless as defined herein.

[0034] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a," "an," "said," and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, and / or groups thereof.

[0035] In the description of this specification, reference to the terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and integrate different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless otherwise inconsistent.

[0036] In the description of this specification, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0037] In the description of this specification, the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings. They are only for the convenience of describing the present technology and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, they should not be understood as limiting the present technology.

[0038] Unless otherwise specified or limited, the terms "installed," "connected," "connected," and "disposed" should be understood broadly. For example, they may refer to fixed connection or disposition, detachable connection or disposition, or integral connection or disposition. Those skilled in the art will understand the specific meanings of these terms in this technology based on specific circumstances.

[0039] To facilitate understanding of the present invention, the present invention is further explained below with reference to specific embodiments in conjunction with the accompanying drawings. However, the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0040] Those skilled in the art should understand that the drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily necessary for implementing the present invention.

[0041] Example 1

[0042] like Figure 1 、 Figure 2 As shown, in this embodiment 1, a portable IGBT module aging parameter detection device is provided, and its circuit topology design and function description are as follows:

[0043] The device is divided into three parts: the first part is the test main circuit and measurement data display part, the components used include AC power socket (i.e. power input socket 14), AC220V-DC12V (AC / DC) switching power supply 1 (i.e. AC / DC power supply), rocker power switch S1 (i.e. power switch 15), output button S4 (self-locking) (i.e. output button 4) and high-voltage relay K1 (i.e. first relay 5), potentiometer, 0-5000V high-voltage power supply module (i.e. power module 7), capacitor C1, voltage digital display (i.e. voltmeter 11), voltage divider 10, microampere current digital display (i.e. current meter). Flow meter 9), output socket 8; the second part is the protection part, the devices used include input side fuse FU1, output side ceramic fuse FU2, discharge button S5 (non-self-locking) (i.e. discharge button 3) and high-voltage relay K2 (i.e. second relay 6), discharge resistor R1; the third part is the control and data transmission interface part, the devices used include control interface board 13, button S6 (i.e. data upload and save button 16), wherein the control interface board includes power conditioning part, voltage conditioning part, ARM chip, Ethernet chip and interface (i.e. network port 12), Modbus interface, USB interface 17 and storage chip.

[0044] In this embodiment 1, the circuit topology principle of the portable IGBT module aging parameter detection device is as follows: an external AC220V power supply is connected, which is converted into DC12V through an AC-DC switching power supply to power the main circuit. When the ship-shaped power switch S1 is closed, the 12V power supply is connected to the main circuit. Press the output button S4, the high-voltage relay K1 coil is energized, the corresponding contacts are attracted, and K1 is closed. The high-voltage adjustable power supply module uses a potentiometer voltage regulation method to achieve an adjustable output voltage of the device between 0 and 5000V. The relationship between the module output voltage and input voltage is shown in formula (1), where U o is the output voltage of the high voltage module, U i = is the input voltage of the high-voltage power supply module. Since a 10kΩ potentiometer is used for voltage regulation, the module's potentiometer voltage regulation reference voltage is 5V. Therefore, the relationship between the module's output voltage and the potentiometer's variable resistance is shown in formula (2), where U o is the output voltage of the high voltage module, and r is the resistance value adjusted by the potentiometer.

[0045] U o =1000·U i (0≤U i ≤5) (1)

[0046]

[0047] The input side of the voltage divider is connected in parallel to the positive and negative ends of the output side of the high-voltage adjustable power supply module, the voltage digital display is connected in parallel to the two ends of the output side of the voltage divider, the power supply is provided by the input voltage AC220V, the voltage digital display is used to display the output voltage value of the device, the microampere-level current digital display is connected in series to the negative pole connection of the high-voltage adjustable power supply module, the power supply is provided by DC12V, and the current digital display is used to display the current value flowing through the IGBT.

[0048] The output voltage after being divided by the voltage divider of the voltage digital display is connected to the voltage conditioning part of the control interface board for collecting its voltage value; the Modbus communication interface of the current digital display is connected to the communication interface of the control interface board for reading the current value; when the control interface board detects that the save data upload button S6 (non-self-locking) is pressed, the test data is transmitted to the host computer software via Ethernet. If a USB flash drive is inserted, it will also be stored in the specified file of the USB flash drive.

[0049] When the output button S4 is disconnected and the high-voltage relay K1 is de-energized and disconnected, the discharge button S5 is pressed, and the coil of the high-voltage relay K2 is energized, the corresponding contacts are attracted, K2 is closed, and the discharge process begins. The purpose of discharge is to prevent the output side of the device from being energized before the next use, which would pose a safety hazard to equipment and personnel. It is worth mentioning that in order to avoid the situation where the discharge button is not closed after the discharge is completed, which may cause an abnormality during the next use, the discharge button S5 adopts a non-self-locking button.

[0050] The ceramic fuses on the input and output sides of the power module are selected with an operating current value of 1A to provide overcurrent protection for various components in the main circuit.

[0051] In summary, this embodiment utilizes the aforementioned device to detect parameters such as the cutoff voltage and cutoff current of the IGBT module, displaying kilovolt-level voltage values ​​and microampere-level current values. This testing takes into account actual operating conditions and the behavior of actual IGBTs. The measured data represents static parameters of the IGBT module under actual application conditions, providing guidance for both IGBT selection and identifying IGBT failures and aging levels during on-site maintenance and repair.

[0052] In this first embodiment, the circuit topology of the portable IGBT aging parameter detection and evaluation device enables adjustable voltages over a wide range of 0 to 5000V, displaying high voltage values ​​and microampere-level current values. It provides high-voltage, discharge, protection, and measurement display functions, meeting the requirements for measuring the cutoff voltage and current of IGBT modules.

[0053] Example 2

[0054] like Figure 1 As shown, in this embodiment 2, a portable IGBT module aging parameter detection device is provided, and its circuit topology design and function description are as follows:

[0055] The device is divided into three parts: the first part is the test main circuit and measurement data display part, the components used include AC power socket (i.e. power input socket 14), AC220V-DC12V (AC / DC) switching power supply 1 (i.e. AC / DC power supply), rocker power switch S1 (i.e. power switch 15), output button S4 (self-locking) (i.e. output button 4) and high-voltage relay K1 (i.e. first relay 5), potentiometer, 0-5000V high-voltage power supply module (i.e. power module 7), capacitor C1, voltage digital display (i.e. voltmeter 11), voltage divider 10, microampere current digital display (i.e. current meter). Flow meter 9), output socket 8; the second part is the protection part, the devices used include input side fuse FU1, output side ceramic fuse FU2, discharge button S5 (non-self-locking) (i.e. discharge button 3) and high-voltage relay K2 (i.e. second relay 6), discharge resistor R1; the third part is the control and data transmission interface part, the devices used include control interface board 13, button S6 (i.e. data upload and save button 16), wherein the control interface board includes power conditioning part, voltage conditioning part, ARM chip, Ethernet chip and interface (i.e. network port 12), Modbus interface, USB interface 17 and storage chip.

[0056] In this second embodiment, the circuit topology of the portable IGBT module aging parameter detection device is as follows: an external 220V AC power source is connected, converted to 12V DC by an AC-DC switching power supply, and used to power the main circuit. When the rocker-shaped power switch S1 is closed, the 12V power source is connected to the main circuit. Pressing the output button S4 energizes the coil of the high-voltage relay K1, closing the corresponding contacts and closing K1.

[0057] like Figure 2 、 Figure 3 As shown, in this embodiment 2, the internal circuit of the above-mentioned device is encapsulated in the casing panel. Considering the portability and flexibility of the present device, the casing panel is fixed in a portable suitcase, and space is left on the left side inside the suitcase to store the power cord and alligator clips and their wiring.

[0058] For the IGBT module cut-off voltage parameter detection method, the gate and emitter of the IGBT module are short-circuited, a resistor is connected, and a negative voltage is connected. When the collector current reaches the specified breakdown current value, the corresponding voltage between the IGBT collector and emitter is the cut-off voltage V CES 、V CER 、V CEX .

[0059] In this embodiment 2, the operation method for testing the cut-off voltage of the IGBT module is as follows:

[0060] Return the potentiometer knob to zero, connect the AC220V power supply, and observe whether the voltage on the voltage digital display is 0. If it is not zero, press and hold the discharge button to discharge until the voltage value on the voltage digital display is 0.

[0061] Clamp two alligator clips on the collector and power emitter of the IGBT module, and plug the corresponding wires into the positive and negative poles of the DC output of the device described in this patent. Figure 4 shown.

[0062] Short-circuit the IGBT gate and the auxiliary emitter (connect a resistor and a negative voltage), turn on the power switch, and press the output button.

[0063] Adjust the potentiometer to increase the output voltage and observe the current value of the microampere current digital display. When the IGBT collector current reaches the specified breakdown current, stop applying pressure and record the data in the voltage digital display at this time. This data is the cut-off voltage (V CES 、V CER 、V CEX ).

[0064] After the test is complete, adjust the voltage knob to 0 and press the output button again to disconnect the output. Press and hold the discharge button to discharge. Wait until the voltage on the voltage display reaches 0, then disconnect the wiring. Turn off the power switch and unplug the AC220V power supply cable.

[0065] In this embodiment 2, for the IGBT module cut-off current parameter detection operation method, the gate and emitter of the IGBT module are short-circuited, a resistor is connected, and a negative voltage is connected. When the voltage between the collector and emitter reaches a specified voltage value, the corresponding IGBT collector current is the cut-off current I CES , I CER , I CEX .

[0066] The operation method for testing the cut-off current of the IGBT module is as follows:

[0067] Return the potentiometer knob (i.e., voltage adjustment knob) to zero, connect to AC220V power supply, and observe whether the voltage on the voltage digital display is 0. If it is not zero, press and hold the discharge button to discharge until the voltage value on the voltage digital display is 0.

[0068] Two alligator clips are respectively clamped on the collector electrode and power emitter electrode of the IGBT module, and the corresponding wiring is respectively inserted into the positive and negative poles of the DC output of the device described in this patent.

[0069] Short-circuit the IGBT gate and the auxiliary emitter (connect a resistor and a negative voltage), turn on the power switch, and press the output button.

[0070] Adjust the potentiometer to increase the output voltage of the device described in this patent, observe the voltage value of the voltage digital display, and when the voltage between the collector and emitter of the IGBT reaches the specified voltage value, stop applying the voltage and record the data in the microampere current digital display at this time. This data is the cut-off current (I CES , I CER , I CEX ).

[0071] After the test is completed, adjust the voltage knob to 0 and press the output button again to disconnect the output. Press and hold the discharge button to discharge. Wait until the voltage value on the voltage display reaches 0, then disconnect the wiring. Turn off the power switch and unplug the AC220V power supply cable.

[0072] After each test is completed, press the Save Data button and the data will be uploaded to the computer via Ethernet for storage and analysis. At the same time, if a USB flash drive is inserted, the test data will also be written to the specified file on the USB flash drive in a fixed format.

[0073] In summary, in Example 2, the portable IGBT aging parameter detection and evaluation device described above can be used to detect parameters such as the cut-off voltage and cut-off current of the IGBT module, and can display kilovolt-level voltage values ​​and microampere-level current values. The test takes into account actual operating conditions and the actual IGBT situation. The measured data are static parameters of the IGBT module measured under actual application conditions, which are of guiding significance for the IGBT selection process and the identification of IGBT failure conditions and aging levels during on-site maintenance and operation. It is packaged in a suitcase, and the circuit topology used does not increase the volume too much while being able to complete the measurement and protection of the cut-off current and cut-off voltage of the IGBT module. It is portable and can be flexibly applied to measure the cut-off voltage and cut-off current of the IGBT module according to actual application scenarios. This device does not require complex operation and can support button control to upload each measured data to the computer host software via Ethernet. It also supports USB flash drives for data storage. It is very suitable for use by on-site maintenance and operation personnel.

[0074] In this embodiment 2, the structure design and operation method of the portable IGBT aging parameter detection and evaluation device are described. The structure is mainly aimed at the layout of the device circuit in the casing panel, as well as the layout of the digital display, buttons, voltage adjustment knobs, switches, and input and output interfaces on the panel; in addition, the structure also includes the suitcase-type design structure and layout described in this patent. The operation method mainly refers to using the device described in this patent to measure the cutoff voltage V of the urban rail traction-level IGBT module under the conditions of gate-emitter short circuit, connection resistance, and connection negative voltage. CES 、V CER 、V CEX and cut-off current I CES , I CER, I CEX operating procedures and methods.

[0075] Example 3

[0076] This embodiment 3 provides a transistor aging degree prediction system, including:

[0077] The transistor aging parameter detection device as described in Example 1 or Example 2, used to obtain the cut-off voltage and cut-off current of the transistor to be predicted;

[0078] A prediction module is used to process the obtained cutoff voltage and cutoff current using a pre-trained prediction model to obtain the aging status of the transistor to be predicted in the next time interval; wherein the prediction model is trained by a training set, and the training set includes multiple groups of data, wherein each group of data includes a set of cutoff voltage and cutoff current, and a label indicating the aging status of the transistor corresponding to the set of cutoff voltage and cutoff current.

[0079] In this embodiment 3, the transistor aging degree prediction system described above is used to implement a transistor aging degree prediction method, including:

[0080] Obtaining a cutoff voltage and a cutoff current of a transistor to be predicted;

[0081] The obtained cutoff voltage and cutoff current are processed using a pre-trained prediction model to obtain the aging status of the transistor to be predicted in the next time interval; wherein the prediction model is trained by a training set, and the training set includes multiple groups of data, wherein each group of data includes a set of cutoff voltage and cutoff current, and a label that marks the aging status of the transistor corresponding to the set of cutoff voltage and cutoff current.

[0082] Specifically, in this embodiment 3, the cutoff voltage and cutoff current of the IGBT at different aging levels throughout its life cycle are detected by the portable IGBT aging parameter detection and evaluation device, and the parameters detected each time are uploaded to the computer host via Ethernet. Data are selected according to a regular pattern of equal time intervals throughout the life cycle as a representation of different aging levels, 80% of the cutoff voltage and cutoff current data are used as a training set, and 20% of the data are used as a validation set. The cutoff voltage VCES and cutoff current ICES are used as inputs to the model, and the outputs are the cutoff voltage VCES and cutoff current ICES values ​​for the next time interval in the future. At the same time, the output is put into the classifier model for classification to determine whether the aging status of the IGBT module in the next time interval is qualified.

[0083] like Figure 5 As shown in the figure, the IGBT aging degree prediction and management method based on cut-off voltage and cut-off current is as follows:

[0084] Two independent LSTM time series prediction networks are used to predict the IGBT module cut-off voltage V for the next time interval. CES and cut-off current I CES Make predictions, such as Figure 6 As shown. Its input x1 t is the IGBT cut-off voltage V at different times CES Value, x2 t IGBT cut-off current I at different times CES Numeric value; the output is the prediction result, is the IGBT cut-off voltage V of the next time series CES Numerical value, IGBT cut-off current I CES The number of hidden neurons depends on the length N of the time series of the selected data. It needs to be determined according to the specific actual test situation. Here we only explain the method.

[0085] The data selected at a certain time interval are sorted according to the time series and normalized. The normalization formula used in this method is:

[0086]

[0087] Where min is the minimum value in the data, and max is the maximum value in the data.

[0088] The output prediction result restoration formula is:

[0089]

[0090] The LSTM network calculation process for predicting the IGBT cut-off voltage and cut-off current is as follows:

[0091] First, set an initial hidden state h10, h20 and cell state C10, C20, and use the value of the random generator function, x1 t and x2 t are the cut-off voltage and cut-off current values ​​of the IGBT in the t-th time series respectively; the outputs f11 and f21 of the forget gate in the cut-off voltage and cut-off current networks are solved respectively. The activation function of the solution process is selected as the sigmoid function, so the expressions of f11 and f21 are shown in Equations (5) and (6).

[0092] f11=σ(W1 f ×h10+U1 f ×x11+b1 f ) (5)

[0093] f21=σ(W2 f×h20+U2 f ×x21+b2 f ) (6)

[0094] Among them W1 f , W2 f , U1 f , U2 f , b1 f , b2 f are the coefficients and bias of the linear relationship, and σ is the sigmoid activation function.

[0095] The input gate consists of two parts. The outputs of the first part are i11 and i21, and the outputs of the second part are a11 and a21. The activation function used in the first part is the sigmoid function, and the activation function used in the second part is the tanh function. The output expressions of the two parts of the input gate are shown in Equations (7)-(10).

[0096] i11=σ(W1 i ×h10+U1 i ×x11+b1 i ) (7)

[0097] i21=σ(W2 i ×h20+U2 i ×x21+b2 i ) (8)

[0098] a11=tanh(W1 a ×h10+U1 a ×x11+b1 a ) (9)

[0099] a21=tanh(W2 a ×h20+U2 a ×x21+b2 a ) (10)

[0100] Among them W1 i , W2 i , U1 i , U2 i , b1 i , b2 i , W1 a , W2a, U1 a , U2 a , b1 a , b2 a are the coefficients and bias of the linear relationship, and σ is the sigmoid activation function.

[0101] The output results of the forget gate and input gate will act on the cell states C1 and C2. The cell state expression of the first time data is shown in formula (11) (12).

[0102] C11=C10⊙f11+i11⊙a11 (11)

[0103] C21=C20⊙f21+i21⊙a21 (12)

[0104] where ⊙ is the Hadamard product.

[0105] The hidden states h11 and h21 consist of two parts. The first part is the output gate values ​​o11 and o21, which are determined by the hidden state at the previous time and the IGBT cut-off voltage and cut-off current at the current time. Its expressions are shown in Equations (13) and (14). The second part consists of the cell state and the tanh activation function.

[0106] o11=σ(W1 o ×h10+U1 o ×x11+b1 o ) (13)

[0107] o21=σ(W2 o ×h20+U2 o ×x21+b2 o ) (14)

[0108] Among them W1 o , W2 o , U1 o , U2 o , b1 o , b2 o , W1 a , W2a, U1 a , U2 a , b1 a , b2 a are the coefficients and bias of the linear relationship, and σ is the sigmoid activation function.

[0109] Therefore, the expressions of hidden states h11 and h21 can be obtained as shown in Equations (15) and (16).

[0110] h11=o11⊙tanh(C11) (15)

[0111] h21=o21⊙tanh(C21) (16)

[0112] After calculating the hidden state and cell state of the first time, the IGBT cut-off voltage and cut-off current network of each other time series are calculated according to the above calculation process. After N times of calculation, the IGBT cut-off voltage V under the current time series t can be obtained. CES and cut-off current I CES The next time series normalized forecast value of .

[0113] Assuming the current time is t, the predicted output value expressions are shown in equations (17) and (18).

[0114]

[0115]

[0116] in is the IGBT cut-off voltage V at the current time t CES and cut-off current I CES The next time series normalized forecast value, V1, V2, b1, b2 are the coefficients of the linear relationship. σ is the sigmoid activation function.

[0117] The loss function is obtained by comparing the output prediction function with the actual value. The gradient descent method is used on the loss function to obtain the gradient calculation formula of each parameter. The parameters are continuously updated and iterated until the error requirements are met. Finally, the IGBT cut-off voltage V at the same time interval can be obtained. CES and cut-off current I CES The LSTM prediction model uses a time series as input and outputs the normalized predicted values ​​of the IGBT cut-off voltage and cut-off current in the next time series. Because the loss function expression needs to be selected based on actual conditions, this method only describes the backpropagation process.

[0118] The normalized predicted IGBT cutoff voltage and current values ​​are restored using Equation (4). Combined with the parameter judgment criteria, the restored predicted values ​​are classified and evaluated using a supervised learning classification method. The specific method needs to be determined based on the prediction data and the comparison of classification results. This embodiment does not provide a detailed description of the classification method. Finally, the host computer output displays the probability of whether the next time series IGBT aging degree meets the standard.

[0119] In summary, the IGBT aging degree prediction and management method based on cut-off voltage and cut-off current described in Example 3 can use a series of time series consisting of the measured IGBT cut-off voltage and cut-off current to predict the IGBT cut-off voltage and cut-off current of the next time series using a two-layer LSTM network method, and classify and evaluate them, thereby predicting and evaluating the probability of IGBT aging degree in the next time series.

[0120] Example 4

[0121] Embodiment 4 of the present invention provides a non-transitory computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, a method for predicting transistor aging degree is implemented. The method includes:

[0122] Obtaining a cutoff voltage and a cutoff current of a transistor to be predicted;

[0123] The obtained cutoff voltage and cutoff current are processed using a pre-trained prediction model to obtain the aging status of the transistor to be predicted in the next time interval; wherein the prediction model is trained by a training set, and the training set includes multiple groups of data, wherein each group of data includes a set of cutoff voltage and cutoff current, and a label that marks the aging status of the transistor corresponding to the set of cutoff voltage and cutoff current.

[0124] Example 5

[0125] Embodiment 5 of the present invention provides a computer program (product), including a computer program, wherein when the computer program is run on one or more processors, the computer program is used to implement a method for predicting transistor aging, the method comprising:

[0126] Obtaining a cutoff voltage and a cutoff current of a transistor to be predicted;

[0127] The obtained cutoff voltage and cutoff current are processed using a pre-trained prediction model to obtain the aging status of the transistor to be predicted in the next time interval; wherein the prediction model is trained by a training set, and the training set includes multiple groups of data, wherein each group of data includes a set of cutoff voltage and cutoff current, and a label that marks the aging status of the transistor corresponding to the set of cutoff voltage and cutoff current.

[0128] Example 6

[0129] Embodiment 6 of the present invention provides an electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory to cause the electronic device to execute instructions for implementing a method for predicting transistor aging, the method comprising:

[0130] Obtaining a cutoff voltage and a cutoff current of a transistor to be predicted;

[0131] The obtained cutoff voltage and cutoff current are processed using a pre-trained prediction model to obtain the aging status of the transistor to be predicted in the next time interval; wherein the prediction model is trained by a training set, and the training set includes multiple groups of data, wherein each group of data includes a set of cutoff voltage and cutoff current, and a label that marks the aging status of the transistor corresponding to the set of cutoff voltage and cutoff current.

[0132] In summary, the insulated gate bipolar transistor aging parameter detection device, aging degree prediction method and system described in the embodiments of the present invention can realize the insulated gate bipolar transistor (IGBT) at the cut-off voltage (V CES 、V CER 、V CEX ) and cutoff current (I CES , I CER , I CEX ) Portable parameter detection device and method. The device can detect static parameters such as IGBT cutoff voltage and cutoff current. It utilizes a suitcase-style package, boasting a compact size, ease of portability, and simple operation. This device is of great significance to field personnel in static parameter calibration and aging failure assessment of IGBT devices and modules, and is an essential component of device and module applications. The power module in the device can achieve a continuously adjustable output voltage of 0 to 5000V by adjusting the potentiometer knob, meeting the voltage requirements for testing static parameters such as cutoff voltage and cutoff current of common high-voltage, high-power IGBT devices and modules. Regarding data measurement, the device uses an external voltage divider to reduce the high voltage across the voltmeter, resolving the issue of insufficient range due to excessive test voltage. A microampere-level digital current meter is also used to meet the requirements for cutoff current parameter measurement. Furthermore, the device supports both Ethernet and USB interfaces for data transmission and storage. The method described in this patent is used on a host computer to perform data analysis and evaluate and predict IGBT aging. Due to its portability and multi-interface capabilities, it can flexibly perform static parameter testing and data upload on IGBT modules in different application scenarios, and has the ability to assess and predict the degree of IGBT module aging. It can assist and guide on-site maintenance personnel in decision-making and management, meet actual on-site testing needs, reduce maintenance costs, and improve system reliability.

[0133] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0134] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0135] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0136] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide the functions for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0137] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solutions disclosed in the present invention without the need for creative work should be included in the scope of protection of the present invention.

Claims

1. A transistor aging degree prediction system, characterized in that: include: A transistor aging parameter detection device, used to obtain the cut-off voltage and cut-off current of the transistor to be predicted; A prediction module is used to process the obtained cut-off voltage and cut-off current using a pre-trained prediction model to obtain the aging state of the transistor to be predicted in the next time interval; wherein the prediction model includes an LSTM network and a classifier model; the LSTM network is used to take the cut-off voltage and cut-off current time series at the same time interval as input, and obtain the normalized prediction values ​​of the cut-off voltage and cut-off current of the next time series as output; the classifier model is used to restore the normalized cut-off voltage and cut-off current prediction values, and classify and evaluate the restored prediction values ​​using a supervised learning classification method in combination with parameter judgment criteria to obtain the probability of whether the aging degree of the transistor in the next time series meets the standard; wherein the prediction model is obtained by training a training set, and the training set includes multiple groups of data, wherein each group of data includes a group of cut-off voltage and cut-off current, and a label indicating the aging state of the transistor corresponding to the group of cut-off voltage and cut-off current; wherein the training process of the LSTM network includes: First, set an initial hidden state h10, h20 and cell state C10, C20, and use the value of the random generator function, x1 t and x2 t are the IGBT cut-off voltage and cut-off current values ​​of the t-th time series respectively; the outputs f11 and f21 of the forget gate in the cut-off voltage and cut-off current networks are solved respectively; The input gate consists of two parts. The output of the first part is i11, i21, and the output of the second part is a11, a21. The output results of the forget gate and the input gate will act on the cell states C1 and C2. The hidden states h11 and h21 consist of two parts. The first part is the output gate value o11 and o21, which is determined by the hidden state at the previous time and the IGBT cut-off voltage and cut-off current at the current time. The expressions of the hidden states h11 and h21 can be obtained. After calculating the hidden state and cell state of the first time, calculate the IGBT cut-off voltage and cut-off current of each other time series in turn. After N calculations, the IGBT cut-off voltage V under the current time series t is finally obtained. CES and cut-off current I CES The next time series normalized forecast value of ; The loss function is obtained by using the output prediction function and the actual value. The gradient descent method is used on the loss function to obtain the gradient calculation formula of each parameter. The parameters are continuously updated and iterated until the error requirements are met. Finally, the IGBT cut-off voltage V is obtained at the same time interval. CES and cut-off current I CES An LSTM network that takes the time series as input and the normalized predicted values ​​of the IGBT cut-off voltage and cut-off current of the next time series as output.

2. The transistor aging prediction system according to claim 1, wherein: The transistor aging parameter detection device comprises: A switching power supply for converting an external AC power supply into a DC power supply, an adjustable power module being connected to the switching power supply, and a power switch being provided between the switching power supply and the adjustable power module; The adjustable power supply module is connected to a capacitor, a potentiometer, a voltage digital display, a voltage divider and a current digital display; an output button is provided between the switching power supply and the adjustable power supply module, and the output button is connected to a first relay.

3. The transistor aging prediction system according to claim 2, wherein: An input-side fuse is provided between the power switch and the adjustable power module, and the adjustable power module is connected to an output-side ceramic fuse.

4. The transistor aging degree prediction system according to claim 3, characterized in that: The adjustable power supply module is connected to a discharge resistor, a discharge button is connected between the power switch and the switching power supply, and the discharge button is connected to a second relay.

5. The transistor aging prediction system according to claim 4, wherein: It also includes a casing, which is provided with a power input socket connected to the switching power supply for connecting to an external AC power supply, a voltage adjustment knob connected to the voltage divider, the output button and the discharge button are both provided on the casing, and the casing is also provided with an output socket connected to the adjustable power supply module.

6. The transistor aging prediction system according to claim 5, characterized in that: The housing is provided with a USB interface, an Ethernet interface and a data saving and uploading button.

7. A method for predicting transistor aging degree, characterized in that: include: Obtaining a cutoff voltage and a cutoff current of a transistor to be predicted; The obtained cut-off voltage and cut-off current are processed using a pre-trained prediction model to obtain the aging state of the transistor to be predicted in the next time interval; wherein the prediction model includes an LSTM network and a classifier model; the LSTM network is used to take the cut-off voltage and cut-off current time series at the same time interval as input, and obtain the normalized prediction values ​​of the cut-off voltage and cut-off current of the next time series as output; the classifier model is used to restore the normalized cut-off voltage and cut-off current prediction values, and classify and evaluate the restored prediction values ​​using a supervised learning classification method in combination with a parameter judgment standard to obtain the probability of whether the aging degree of the transistor in the next time series meets the standard; wherein the prediction model is obtained by training a training set, and the training set includes multiple groups of data, wherein each group of data includes a group of cut-off voltage and cut-off current, and a label indicating the aging state of the transistor corresponding to the group of cut-off voltage and cut-off current; wherein the training process of the LSTM network includes: First, set an initial hidden state h10, h20 and cell state C10, C20, and use the value of the random generator function, x1 t and x2 t are the IGBT cut-off voltage and cut-off current values ​​of the t-th time series respectively; the outputs f11 and f21 of the forget gate in the cut-off voltage and cut-off current networks are solved respectively; The input gate consists of two parts. The output of the first part is i11, i21, and the output of the second part is a11, a21. The output results of the forget gate and the input gate will act on the cell states C1 and C2. The hidden states h11 and h21 consist of two parts. The first part is the output gate value o11 and o21, which is determined by the hidden state at the previous time and the IGBT cut-off voltage and cut-off current at the current time. The expressions of the hidden states h11 and h21 can be obtained. After calculating the hidden state and cell state of the first time, calculate the IGBT cut-off voltage and cut-off current of each other time series in turn. After N calculations, the IGBT cut-off voltage V under the current time series t is finally obtained. CES and cut-off current I CES The next time series normalized forecast value of ; The loss function is obtained by using the output prediction function and the actual value. The gradient descent method is used on the loss function to obtain the gradient calculation formula of each parameter. The parameters are continuously updated and iterated until the error requirements are met. Finally, the IGBT cut-off voltage V is obtained at the same time interval. CES and cut-off current I CES An LSTM network that takes the time series as input and the normalized predicted values ​​of the IGBT cut-off voltage and cut-off current of the next time series as output.

8. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by the processor, the transistor aging degree prediction method according to claim 7 is implemented.

9. A computer program product, characterized in that The invention comprises a computer program, which is used to implement the transistor aging degree prediction method according to claim 7 when the computer program is run on one or more processors.

10. An electronic device, characterized in that: include: A processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to execute instructions for implementing the transistor aging degree prediction method as described in claim 7.

Images