Fault prediction method and system for IGBT (Insulated Gate Bipolar Translator) module

Abstract

The invention discloses an IGBT module fault prediction method and system, the system comprises an IGBT module, the input end of the IGBT module is bidirectionally and electrically connected with an operation monitoring system, and the output end of the operation monitoring system is bidirectionally and electrically connected with a control circuit. According to the invention, internal data of an IGBT module is collected through a contact thermocouple, a thermistor, an infrared thermal imaging camera, a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short circuit test module and a Qg test module, is integrated through an integration collection module and then is transmitted to a control circuit; and the terminal processor processes the data and transmits the data to the IGBT module operation prediction system for prediction and judgment, so that the monitoring mode of the existing IGBT module can be improved, the existing and future states of the IGBT module can be monitored and predicted, and a user can timely master accidental dangerous conditions occurring in the operation process of the IGBT module. And the safety of the IGBT module is prevented from being influenced.

Description

Technical Field

[0001] The present invention relates to the technical field of IGBT module monitoring, and specifically to a fault prediction method and system for an IGBT module. Background Art

[0002] An IGBT (Insulated Gate Bipolar Transistor) is a composite fully-controlled voltage-driven power semiconductor device composed of a BJT and a MOS. It combines the advantages of both a MOSFET with a high input impedance and a GTR with a low conduction voltage drop. The GTR has a low saturation voltage drop and a large current density, but requires a large drive current. The MOSFET has a very small drive power and a fast switching speed, but a large conduction voltage drop and a small current density. The IGBT combines the advantages of the above two devices, with a small drive power and a low saturation voltage drop. The IGBT is very suitable for application in converter systems with a DC voltage of 600V or above, such as AC motors, frequency converters, switching power supplies, lighting circuits, traction drives, and other fields.

[0003] In order to improve the operating safety of the IGBT module, a monitoring system is required to monitor the IGBT module. For example, the patent number disclosed on the Chinese Patent Network is: 201610537013.1, and the patent name is: IGBT Module On-line Monitoring Method, which includes a to-be-tested IGBT module and at least one test IGBT module. The to-be-tested IGBT module and the test IGBT module form a bridge circuit, and the aging state of the to-be-tested IGBT module is tested according to the following method: input a short-circuit drive voltage to the to-be-tested IGBT module and control the to-be-tested IGBT module to work in the current saturation state, and input a working drive voltage to the test IGBT module to make the test IGBT module work normally. Detect the short-circuit current value of the to-be-tested IGBT module in the current saturation state, and judge the aging state of the to-be-tested IGBT module according to the short-circuit current value. It can accurately monitor the aging state of the IGBT module without stopping the machine or disassembling it, so as to accurately judge the aging state of the IGBT module and effectively avoid the economic losses caused by downtime detection and the damage caused by disassembling the IGBT module, which affects the measurement accuracy.

[0004] However, the existing monitoring methods for IGBT modules are relatively single, and can only monitor the existing state of the IGBT module. Users cannot timely master the unexpected dangerous situations that occur during the operation of the IGBT module, resulting in an impact on the safety of the IGBT module.

[0005] Therefore, it is necessary to design and transform the fault prediction method and system for the IGBT module. Summary of the Invention

[0006] To solve the problems raised in the above-mentioned background technology, the purpose of the present invention is to provide a fault prediction method and system for an IGBT module, which has the advantages of predicting and monitoring the IGBT module, and solves the problem that the existing monitoring methods for IGBT modules are relatively single, only able to monitor the existing state of the IGBT module, and users cannot timely grasp the unexpected dangerous situations that occur during the operation of the IGBT module, resulting in the safety of the IGBT module being affected.

[0007] To achieve the above object, the present invention provides the following technical solutions: A fault prediction method and system for an IGBT module, including an IGBT module;

[0008] The input end of the IGBT module is bidirectionally electrically connected to an operation monitoring system, the output end of the operation monitoring system is bidirectionally electrically connected to a control circuit, the input end of the IGBT module is bidirectionally electrically connected to a thermal monitoring module, the output end of the thermal monitoring module is bidirectionally electrically connected to an integration and collection module, the output end of the integration and collection module is bidirectionally electrically connected to the input end of the control circuit, the thermal monitoring module is composed of a contact thermocouple, a thermistor and an infrared thermal imaging camera, the output end of the control circuit is bidirectionally electrically connected to a terminal processor, the output end of the terminal processor is bidirectionally electrically connected to an IGBT module operation prediction system, and the output end of the terminal processor is bidirectionally electrically connected to a classification storage module.

[0009] Preferably, the IGBT module operation prediction system includes a combined prediction model, the input end of the combined prediction model is bidirectionally electrically connected to the output end of the terminal processor, the output end of the combined prediction model is bidirectionally electrically connected to a comparison module, the input end of the comparison module is bidirectionally electrically connected to a long short-term memory network, and the input end of the long short-term memory network is bidirectionally electrically connected to the output end of the classification storage module.

[0010] Preferably, the output end of the comparison module is bidirectionally electrically connected to a prediction evaluation module, and the output end of the prediction evaluation module is bidirectionally electrically connected to a feedback module.

[0011] Preferably, the output end of the feedback module is electrically connected to the input end of the terminal processor, the output end of the feedback module is bidirectionally electrically connected to a wireless transceiver antenna, and the output end of the wireless transceiver antenna is remotely bidirectionally electrically connected to a remote monitoring platform.

[0012] Preferably, the combined prediction model is composed of a linear regression model, a grey prediction model and a BP neural network prediction model.

[0013] Preferably, an output terminal of the control circuit is electrically connected to a protection electrical appliance, an output terminal of the protection electrical appliance is electrically connected to an input terminal of the IGBT module, and the protection electrical appliance is composed of an intelligent fuse and a remote control circuit breaker.

[0014] Preferably, the operation monitoring system is composed of a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module, and a Qg test module.

[0015] A fault prediction method and system for an IGBT module include the following steps:

[0016] S1: During the operation of the IGBT module, a thermocouple, a thermistor, an infrared thermal imaging camera, a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module, and a Qg test module can collect internal data of the IGBT module and transfer it to the control circuit after integration by an integration collection module.

[0017] S2: The terminal processor processes the data and transfers it to the IGBT module operation prediction system. The linear regression model can use regression analysis in mathematical statistics to determine the quantitative relationship of interdependence between two or more variables. The grey prediction model can establish a mathematical model to make predictions based on a small amount of incomplete information collected by the IGBT module. Based on the past and present development laws of objective things, with the help of the data stored inside the long short-term memory network and scientific methods, the future development trend and situation of the IGBT module are described and analyzed, and scientific hypotheses and judgments are formed.

[0018] S3: The comparison module compares the prediction data. When there is an error in the data, the feedback module sends an alarm signal to the remote monitoring platform through a wireless transceiver antenna. At the same time, the feedback module can control the protection electrical appliance to turn on through the control circuit to prevent the IGBT module from being damaged.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0020] 1. In the present invention, a thermocouple, a thermistor, an infrared thermal imaging camera, a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module, and a Qg test module are used to collect internal data of the IGBT module and transfer it to the control circuit after integration by an integration collection module. The terminal processor processes the data and transfers it to the IGBT module operation prediction system for prediction and judgment, which can improve the existing monitoring method of the IGBT module, monitor and predict the current and future states of the IGBT module, and enable users to timely master the unexpected dangerous situations occurring during the operation of the IGBT module, avoiding the safety of the IGBT module from being affected.

[0021] 2. The present invention can predict the data state by setting up a combined prediction model and a comparison module, enabling users to monitor the operating conditions of the IGBT module and detect module failures earlier.

[0022] 3. The present invention can judge the prediction results by setting up a prediction evaluation module and a feedback module, avoid prediction errors, and improve the detection accuracy.

[0023] 4. The present invention can prompt remote monitoring users by setting up a wireless transceiver antenna and a remote monitoring platform, facilitating users to centrally master the operating states of multiple IGBT modules.

[0024] 5. The present invention is composed of a linear regression model, a grey prediction model, and a BP neural network prediction model. It can use regression analysis in mathematical statistics to determine the variable statistical data of the quantitative relationship of interdependence between two or more variables. The grey prediction model can establish a mathematical model to make predictions based on the small amount of incomplete information collected by the IGBT module. Based on the past and present development laws of objective things, with the help of the data stored inside the long short-term memory network and scientific methods, it describes and analyzes the future development trends and conditions of the IGBT module, and forms scientific hypotheses and judgments.

[0025] 6. The present invention can protect the IGBT module by setting up a protection electrical appliance, preventing the IGBT module from being damaged accidentally.

[0026] 7. The present invention can monitor the IGBT module by setting up a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module, and a Qg test module, improve the data acquisition range, judge whether the tested FWD chip is normal, judge the short-circuit states of CE, GE, and GC. At the same time, the carrier concentration, mobility, and diffusion coefficient of the IGBT power module are all affected by temperature. The contact thermocouple, thermistor, and infrared thermal imaging camera can collect and judge the static parameters in the on or off state and the dynamic parameters at the moment of on or off. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0029] As Figure 1 shown, a fault prediction method and system for an IGBT module provided by the present invention includes an IGBT module;

[0030] The input end of the IGBT module is bidirectionally electrically connected to an operation monitoring system, the output end of the operation monitoring system is bidirectionally electrically connected to a control circuit, the input end of the IGBT module is bidirectionally electrically connected to a thermal monitoring module, the output end of the thermal monitoring module is bidirectionally electrically connected to an integration collection module, the output end of the integration collection module is bidirectionally electrically connected to the input end of the control circuit. The thermal monitoring module is composed of a contact thermocouple, a thermistor and an infrared thermal imaging camera. The output end of the control circuit is bidirectionally electrically connected to a terminal processor, the output end of the terminal processor is bidirectionally electrically connected to an IGBT module operation prediction system, and the output end of the terminal processor is bidirectionally electrically connected to a classification storage module.

[0031] Referring to Figure 1 , the IGBT module operation prediction system includes a combined prediction model. The input end of the combined prediction model is bidirectionally electrically connected to the output end of the terminal processor, the output end of the combined prediction model is bidirectionally electrically connected to a comparison module, the input end of the comparison module is bidirectionally electrically connected to a long short-term memory network, and the input end of the long short-term memory network is bidirectionally electrically connected to the output end of the classification storage module.

[0032] As a technical optimization scheme of the present invention, by setting the combined prediction model and the comparison module, the data state can be predicted, enabling users to monitor the operation status of the IGBT module and detect the failure of the module earlier.

[0033] Referring to Figure 1 , the output end of the comparison module is bidirectionally electrically connected to a prediction evaluation module, and the output end of the prediction evaluation module is bidirectionally electrically connected to a feedback module.

[0034] As a technical optimization scheme of the present invention, by setting the prediction evaluation module and the feedback module, the prediction result can be judged, avoiding prediction errors and improving the detection accuracy.

[0035] Referring to Figure 1 , the output end of the feedback module is electrically connected to the input end of the terminal processor, the output end of the feedback module is bidirectionally electrically connected to a wireless transceiver antenna, and the output end of the wireless transceiver antenna is remotely bidirectionally electrically connected to a remote monitoring platform.

[0036] As a technical optimization solution of the present invention, by setting up a wireless transceiver antenna and a remote monitoring platform, it is possible to prompt remote monitoring users, facilitating users to centrally master the operating status of multiple IGBT modules.

[0037] Reference Figure 1 , the combined prediction model is composed of a linear regression model, a grey prediction model, and a BP neural network prediction model.

[0038] As a technical optimization solution of the present invention, by setting up a combined model composed of a linear regression model, a grey prediction model, and a BP neural network prediction model, it is possible to use regression analysis in mathematical statistics to determine the variable statistical data of the quantitative relationship of interdependence between two or more variables. The grey prediction model can establish a mathematical model to make predictions for the small amount of incomplete information collected by the IGBT module, and based on the development laws of objective things in the past and present, with the help of the data stored inside the long short-term memory network and scientific methods, describe and analyze the future development trends and conditions of the IGBT module, and form scientific hypotheses and judgments.

[0039] Reference Figure 1 , the output end of the control circuit is electrically connected to a protection electrical appliance, and the output end of the protection electrical appliance is electrically connected to the input end of the IGBT module. The protection electrical appliance is composed of an intelligent fuse and a remote control circuit breaker.

[0040] As a technical optimization solution of the present invention, by setting up the protection electrical appliance, it is possible to protect the IGBT module and prevent the IGBT module from being damaged due to accidents.

[0041] Reference Figure 1 , the operation monitoring system is composed of a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module, and a Qg test module.

[0042] As a technical optimization solution of the present invention, by setting up a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module, and a Qg test module, it is possible to monitor the IGBT module, improve the data acquisition range, judge whether the test FWD chip is normal, judge the short-circuit states of CE, GE, and GC. At the same time, the carrier concentration, mobility, and diffusion coefficient of the IGBT power module are all affected by temperature. The contact thermocouple, thermistor, and infrared thermal imaging camera can collect and judge the static parameters in the on or off state and the dynamic parameters at the moment of on or off.

[0043] Reference Figure 1 , a fault prediction method and system for an IGBT module, comprising the following steps:

[0044] S1: During the operation of the IGBT module, the thermocouple, thermistor, infrared thermal imaging camera, diffusion coefficient monitoring module, mobility monitoring module, carrier concentration monitoring module, short-circuit test module, and Qg test module can collect the internal data of the IGBT module, and after integration using the integration collection module, transfer it to the control circuit;

[0045] S2: The terminal processor processes the data and transfers it to the IGBT module operation prediction system. The linear regression model can use regression analysis in mathematical statistics to determine the variable statistical data of the quantitative relationship of interdependence between two or more variables. The grey prediction model can establish a mathematical model to make predictions for the small amount of incomplete information collected by the IGBT module. Based on the past and present development laws of objective things, with the help of the data stored in the long short-term memory network and scientific methods, describe and analyze the future development trends and conditions of the IGBT module, and form scientific hypotheses and judgments;

[0046] S3: The comparison module compares the predicted data. When there is an error in the data, the feedback module sends an alarm signal to the remote monitoring platform through the wireless transceiver antenna. At the same time, the feedback module can use the control circuit to control the protection electrical appliance to turn on, preventing the IGBT module from being damaged.

[0047] The working principle and usage process of the present invention: During the operation of the IGBT module, the thermocouple, thermistor, infrared thermal imaging camera, diffusion coefficient monitoring module, mobility monitoring module, carrier concentration monitoring module, short-circuit test module, and Qg test module can collect the internal data of the IGBT module, and after integration using the integration collection module, transfer it to the control circuit. The terminal processor processes the data and transfers it to the IGBT module operation prediction system. The linear regression model can use regression analysis in mathematical statistics to determine the variable statistical data of the quantitative relationship of interdependence between two or more variables. The grey prediction model can establish a mathematical model to make predictions for the small amount of incomplete information collected by the IGBT module. Based on the past and present development laws of objective things, with the help of the data stored in the long short-term memory network and scientific methods, describe and analyze the future development trends and conditions of the IGBT module, and form scientific hypotheses and judgments. The comparison module compares the predicted data. When there is an error in the data, the feedback module sends an alarm signal to the remote monitoring platform through the wireless transceiver antenna. At the same time, the feedback module can use the control circuit to control the protection electrical appliance to turn on, preventing the IGBT module from being damaged.

[0048] In summary, for the fault prediction method and system of the IGBT module, the internal data of the IGBT module is collected through a contact thermocouple, a thermistor, an infrared thermal imaging camera, a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module, and a Qg test module, and is integrated by an integration collection module and then transmitted to a control circuit. The terminal processor processes the data and transmits it to the IGBT module operation prediction system for prediction and judgment. It can improve the existing monitoring method of the IGBT module, monitor and predict the current and future states of the IGBT module, enabling users to timely grasp unexpected dangerous situations that occur during the operation of the IGBT module and avoid affecting the safety of the IGBT module.

[0049] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0050] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A fault prediction system for an IGBT module, comprising the IGBT module; Characterized in that: The input end of the IGBT module is bidirectionally electrically connected to an operation monitoring system, the output end of the operation monitoring system is bidirectionally electrically connected to a control circuit, the input end of the IGBT module is bidirectionally electrically connected to a thermal monitoring module, the output end of the thermal monitoring module is bidirectionally electrically connected to an integration and collection module, the output end of the integration and collection module is bidirectionally electrically connected to the input end of the control circuit, the thermal monitoring module consists of a contact thermocouple, a thermistor and an infrared thermal imaging camera, the output end of the control circuit is bidirectionally electrically connected to a terminal processor, the output end of the terminal processor is bidirectionally electrically connected to an IGBT module operation prediction system, and the output end of the terminal processor is bidirectionally electrically connected to a classification storage module.

2. The fault prediction system for an IGBT module according to claim 1, Characterized in that: The IGBT module operation prediction system includes a combined prediction model, the input end of the combined prediction model is bidirectionally electrically connected to the output end of the terminal processor, the output end of the combined prediction model is bidirectionally electrically connected to a comparison module, the input end of the comparison module is bidirectionally electrically connected to a long short-term memory network, and the input end of the long short-term memory network is bidirectionally electrically connected to the output end of the classification storage module.

3. The fault prediction system for an IGBT module according to claim 2, Characterized in that: The output end of the comparison module is bidirectionally electrically connected to a prediction evaluation module, and the output end of the prediction evaluation module is bidirectionally electrically connected to a feedback module.

4. The fault prediction system for an IGBT module according to claim 3, Characterized in that: The output end of the feedback module is electrically connected to the input end of the terminal processor, the output end of the feedback module is bidirectionally electrically connected to a wireless transceiver antenna, and the output end of the wireless transceiver antenna is remotely bidirectionally electrically connected to a remote monitoring platform.

5. The fault prediction system for an IGBT module according to claim 2, Characterized in that: The combined prediction model consists of a linear regression model, a grey prediction model and a BP neural network prediction model.

6. The fault prediction system for an IGBT module according to claim 1, Characterized in that: The output end of the control circuit is electrically connected to a protection electrical appliance, the output end of the protection electrical appliance is electrically connected to the input end of the IGBT module, and the protection electrical appliance consists of an intelligent fuse and a remote control circuit breaker.

7. The fault prediction system for an IGBT module according to claim 1, Characterized in that: The operation monitoring system consists of a diffusion coefficient monitoring module, a mobility monitoring module, a carrier concentration monitoring module, a short-circuit test module and a Qg test module.

8. The fault prediction method for an IGBT module according to any one of the above claims, Characterized in that: Comprising the following steps: S1: During the operation of the IGBT module, the thermocouple, thermistor, infrared thermal imaging camera, diffusion coefficient monitoring module, mobility monitoring module, carrier concentration monitoring module, short - circuit test module and Qg test module can collect the internal data of the IGBT module and transfer it to the control circuit after integration by the integration collection module; S2: The terminal processor processes the data and transfers it to the IGBT module operation prediction system. The linear regression model can use regression analysis in mathematical statistics to determine the variable statistical data of the quantitative relationship of interdependence between two or more variables. The grey prediction model can establish a mathematical model to make predictions for the small and incomplete information collected by the IGBT module. Based on the past and present development laws of objective things, with the help of the data stored inside the long - short - term memory network and scientific methods, it describes and analyzes the future development trend and situation of the IGBT module, and forms scientific hypotheses and judgments; S3: The comparison module compares the predicted data. When there are errors in the data, the feedback module sends an alarm signal to the remote monitoring platform through the wireless transceiver antenna. At the same time, the feedback module can control the protection electrical appliance to turn on by using the control circuit to prevent the IGBT module from being damaged.

Images