Procedure for operating a converter and converter arrangement
The method and converter arrangement facilitate immediate data capture and analysis of desaturation errors in power semiconductor switches, simplifying fault diagnosis and reducing system downtime by using an expert system for precise error cause identification.
Patent Information
- Application Number
- DE102023203760
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-04-24
- Publication Date
- 2024-11-21
- Estimated Expiration
- 2043-04-24
AI Technical Summary
Existing systems face complex and costly root cause analysis for desaturation faults in power semiconductor switches, requiring immediate shutdown of electric motors/generators without precise identification of the fault cause.
A method and converter arrangement that records status data of converters in a non-volatile memory upon detecting desaturation errors and uses an expert system to determine the error cause, allowing for immediate data capture and analysis.
Simplifies error cause analysis by capturing and analyzing converter status data post-fault, enabling efficient identification of desaturation errors and reducing system downtime.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a method for operating a converter and a converter arrangement.
[0002] Converters, such as inverters, use power semiconductor switches, such as insulated-gate bipolar transistors (IGBTs) or MOSFETs, to convert high-voltage direct current electrical power into high-voltage alternating current power, which is transmitted to an electric motor / generator to generate torque for traction power in vehicles, for example.
[0003] Faults related to power semiconductor switches and electric motor / generator windings can result in excessive electrical current flow through the various components and increased voltage across a collector and emitter of one or more of the power semiconductor switches, which is called desaturation.
[0004] Common power semiconductor switches are controlled either in an ON state or an OFF state. In the ON state, with a properly functioning power semiconductor switch, the electrical current through the power semiconductor switch is saturated, meaning the entire available current flows through the switch, with a small portion being converted into heat energy due to the switch resistance. In the OFF state, the switch blocks the electrical current flow.
[0005] A desaturation fault in a power semiconductor switch can cause the electrical current to rise above a maximum operating limit, causing the power semiconductor switch to enter a linear mode, thereby desaturating the power semiconductor switch (DSAT or DESAT). A DSAT condition occurs when the voltage magnitude across the collector and emitter of a power semiconductor switch rises above a threshold, e.g., 1-2 volts when a gate-emitter voltage is boosted. Common responses to a DSAT fault include immediate shutdown of the electric motor / generator. Known systems are configured to monitor collector-emitter voltages of power semiconductor switches to detect DSAT faults and include control systems to cause immediate shutdown of the electric motor / generator upon detection of a DSAT fault.
[0006] To determine the specific cause of a DSAT error, an analysis at the system, hardware, and software levels is always required. Root cause analysis is usually performed using an Ishikawa diagram. These analyses are very complex and costly. Follow-up measurements are often necessary because the exact state of the system at the time of failure is unknown.
[0007] CN 1 01 793 928 A describes a method and system for real-time fault diagnosis for a phase-shifting full-bridge converter. Two lower bridge arms of a phase-shifting full-bridge converter are equipped with current sensors. In the positive half of a complete switching cycle of the phase-shifting full-bridge converter, the sensor of the front lower bridge arm receives a freewheeling current signal from an anti-parallel diode of the power tube. In the negative half of the switching cycle, the sensor of a rear lower bridge arm receives a freewheeling current signal from an anti-parallel diode of the power tube.The faults of the phase-shifting full-bridge converter are detected based on the logical relationship between the freewheeling current signal of the power tube's antiparallel diode and a power tube drive signal. The drive signal of the power tube is integrated with the freewheeling current signal obtained by sampling the positive or negative half. The invention allows for rapid and precise fault location and determination of fault types, creation of a fault database, and implementation of fault prediction.
[0008] The invention is based on the object of creating a method for operating a converter and a converter arrangement with which, in particular, a fault cause analysis after the occurrence of a DSAT fault can be simplified.
[0009] The object is achieved according to the invention by a method having the features of patent claim 1 or 2 and a converter arrangement having the features of patent claim 9 or 10. Advantageous embodiments of the invention emerge from the subclaims.
[0010] In particular, a method for operating a converter is provided, wherein power semiconductor switches of the converter are monitored for the presence of desaturation errors, wherein upon detection of a desaturation error in at least one of the power semiconductor switches, acquired status data of the converter, in particular before measures are triggered, are stored in a non-volatile memory, wherein the acquired status data are fed to an expert system which is configured to determine at least one cause of the error based on the acquired status data, wherein a result of the expert system is stored in the non-volatile memory.
[0011] Furthermore, in particular, a converter arrangement is created, comprising a converter, and a control device, wherein the control device is configured to monitor power semiconductor switches of the converter for the presence of desaturation errors and, upon detection of a desaturation error in at least one of the power semiconductor switches, to store acquired status data of the converter, in particular before measures are triggered, in a non-volatile memory, furthermore to provide an expert system which is configured to determine at least one cause of the error based on the acquired status data, to feed the acquired status data to the expert system, and to store a result of the expert system in the non-volatile memory.
[0012] The method and the converter arrangement make it possible, when a desaturation fault occurs, to firstly store recorded converter status data in a non-volatile memory for later fault cause analysis and, secondly, to perform an initial fault cause analysis and also to store the results obtained in a non-volatile memory. This simplifies fault cause analysis because the status data are recorded when the desaturation fault occurs and, moreover, already known fault causes have been checked. For this purpose, it is provided that when a desaturation fault is detected in at least one of the power semiconductor switches, recorded converter status data is stored in a non-volatile memory before measures are triggered (such as transferring the converter to a safe state, for example by switching it off). Furthermore, the recorded status data is fed to an expert system.The expert system is designed to determine at least one cause of the error based on the recorded status data, with a result of the expert system also being stored in the non-volatile memory.
[0013] One advantage of the system is that the process steps can be performed immediately after the desaturation fault is detected. Should a desaturation fault have such serious consequences that part of the converter's power electronics is destroyed, data, possibly incomplete, can still be provided to support the root cause analysis.
[0014] The non-volatile memory is in particular a flash memory (especially flash EEPROM) or a non-volatile RAM (non-volatile random-access memory, NVRAM).
[0015] The acquired status data may include at least one of the following: phase voltages, phase currents, a high-voltage DC voltage, a low-voltage DC voltage, an intermediate circuit voltage, a gate driver supply voltage, a control supply voltage, a board temperature, a gate driver IC temperature, values of readable gate driver IC registers, a power module temperature, thermal model temperatures, an insulation monitor status, a software version, a parameter value version.
[0016] The expert system is stored in particular in a memory, for example, a main memory or non-volatile memory. The expert system particularly comprises an assignment of converter states, which are represented by the state data, to one or more fault causes. The expert system can be designed, for example, as a control system and / or mathematical function and / or mapping. The expert system can be generated, for example, based on empirical test series, field tests, and / or data collected from the field during normal operation. In principle, simulations can also be used additionally or alternatively to generate the expert system. Furthermore, the expert system can also use artificial intelligence methods, in particular machine learning methods.
[0017] Parts of the converter arrangement, in particular the control device, can be implemented individually or collectively as a combination of hardware and software, for example as program code executed on a microcontroller or microprocessor. However, it can also be provided that parts are implemented individually or collectively as an application-specific integrated circuit (ASIC) and / or a field-programmable gate array (FPGA).
[0018] The term converter includes in particular inverters, rectifiers, DC-DC converters and AC-DC converters.
[0019] The process and converter arrangement can be used in all areas where converters are used. For example, in electric or hybrid vehicles, in connection with solar or wind power systems, and in the field of industrial converters.
[0020] In one embodiment, values of at least some of the acquired status data are latched for storage immediately after the desaturation fault is detected. This prevents modified values from being saved if the status data of the converter is changed by triggering and initiating measures, such as shutting down the converter and / or an electrical machine powered by it. However, it is possible to ensure that the status data that was present when the desaturation fault occurred is stored in the non-volatile memory.
[0021] In one embodiment, the acquired status data is stored in the non-volatile memory with the highest priority. This ensures, in particular, that the acquired status data is saved before the converter and other components are potentially destroyed, and saving is then no longer possible. Even if the desaturation fault causes significant damage, at least as much status data as possible can still be stored in the non-volatile memory, thus still supporting a root cause analysis.
[0022] In one embodiment, the result of the expert system is encoded in the non-volatile memory in the form of a bit field. This allows information about possible error causes to be provided in a particularly simple and clear manner. The error causes can then be encoded as numbers using the bit field, so that error codes can be provided that correspond to one or more error causes.
[0023] In one embodiment, a further development provides that each bit in the bit field is assigned a cause of error. In this simple case, each bit in the bit field corresponds to a cause of error. The value of the respective bit in the expert system's result then encodes the presence (e.g., "1") or absence (e.g., "0") of the associated cause of error.
[0024] In one alternative, a start test is performed when starting the converter. The start test comprises at least checking a supply voltage and / or checking voltage signals at the converter's gate drivers and / or checking switching operations of the converter's gates and / or checking a resulting phase voltage, with the test result being stored in the non-volatile memory. This can increase the reliability of the converter. Furthermore, it can provide additional data for later fault root cause analysis.
[0025] In one embodiment, the expert system stores the acquired status data and / or determines the cause of the fault using a computing device reserved at least for this time. This ensures that the acquired status data can be stored and / or the cause of the fault can be determined before operation of the converter or converter arrangement is no longer possible.If the control device of the converter arrangement is, for example, part of the drive train of an electric vehicle and has multiple computing devices, one of the computing devices can be used exclusively for storing the acquired status data and / or determining the cause of the fault using the expert system, while one or more other computing devices of the control device can be used for initiating and / or controlling measures that transfer the converter and / or the electric vehicle to a safe state. The separate computing device ensures that as much of the acquired status data as possible can still be stored and / or that the result of the expert system can still be determined and saved.
[0026] Another alternative, in addition or alternatively, provides for at least one counter to be incremented or decremented depending on the error based on the expert system's result. This allows for a gradation of errors based on their frequency to be taken into account.
[0027] In one embodiment, at least one error correction measure is initiated based on the result of the expert system and / or a counter reading of the at least one counter. This allows less critical errors to be ignored for the time being and / or measures to be selected and / or triggered depending on a counter reading of the at least one counter, for example, to attempt to correct the error or the cause of the error. If, for example, overheating of a power module of the converter is determined to be the cause of the error, the counter is incremented by one (or decremented, depending on the specific design of the counter).
[0028] The converter can then be switched off. For example, if this is possible at the time, the ignition of an electric vehicle can be switched off and then switched back on after a short wait time for the power module to cool down. The process can then be repeated. If the counter has reached a specified counter value for the aforementioned example fault cause, the converter is permanently switched off, for example, and restarting is blocked.
[0029] Further features of the converter configuration are described in the various embodiments of the method. The advantages of the converter configuration are the same as those of the various embodiments of the method.
[0030] Furthermore, in particular, a vehicle is also provided, comprising at least one converter arrangement according to one of the described embodiments. A vehicle is, in particular, a motor vehicle, in particular an electric vehicle or a hybrid vehicle. In principle, however, the vehicle can also be another land, rail, water, air, or space vehicle, for example, an air taxi or a drone.
[0031] The invention will be explained in more detail below using preferred embodiments with reference to the figure. Herein: Fig. 1 a schematic representation of embodiments of the converter arrangement.
[0032] The Fig.1 shows a schematic representation of embodiments of the converter assembly 1. The converter assembly 1 is arranged in particular in a vehicle 50, in particular an electric vehicle, where it forms, for example, part of a traction train. The converter assembly 1 is configured to carry out the method described in this disclosure. The method described in this disclosure is explained in more detail below using the converter assembly 1.
[0033] The converter assembly 1 comprises a converter 2 and a control device 3. The control device 3 can, in principle, also be part of the converter 2 or can control and / or regulate the converter 2. The converter 2 is, for example, a pulse-controlled inverter that serves to power a multi-phase electrical machine of the vehicle 50. The control device 3 comprises, for example, a computing device 4 and a memory 5.
[0034] Furthermore, the control device 3 comprises a non-volatile memory 6, for example a flash memory or a non-volatile RAM memory (NVRAM).
[0035] The control device 3 is configured to monitor power semiconductor switches (e.g. IGBTs or MOSFETs) of the converter 2 for the presence of desaturation errors and, if a desaturation error is detected in at least one of the power semiconductor switches, to store acquired status data 10 of the converter 2 in the non-volatile memory 4, in particular before measures are triggered. The power semiconductor switches are monitored in a manner known per se. Furthermore, the control device 3 is configured to provide an expert system 20. For example, the expert system 20 is stored in the form of description data in the memory 5 and can be retrieved and applied from there by the computing device 4. The expert system 20 is configured to determine at least one fault cause 21-x based on the acquired status data 10. The acquired status data 10 is fed to the expert system 20 for this purpose.A result 21 of the expert system 20 is stored by the control device 3 in the non-volatile memory 6.
[0036] In particular, it is provided that values of at least a portion of the acquired status data 10 are latched for storage immediately after the desaturation fault is detected. In particular, the values of the acquired status data 10 that existed when the desaturation fault was detected are stored so that these values are available for a fault root cause analysis. It may also be provided to access historical values and store one or more historical values of the acquired status data 10, for example, values that were acquired immediately before and upon the occurrence of the desaturation fault. This can provide even better support for the fault root cause analysis.
[0037] In particular, it is provided that the recorded status data 10 is stored in the non-volatile memory 6 with the highest priority. In particular, the computing operations ("tasks") provided for this purpose are processed with the highest priority in the computing device 4.
[0038] It can be provided that the result 21 of the expert system 20 is encoded in the non-volatile memory 6 in the form of a bit field.
[0039] In a further development, each bit in the bit field can be assigned a 21-x error cause. For example, the bit field can have a size of 32 or 64 bits, allowing a total of 32 or 64 error causes to be directly specified. In principle, another coding is also possible, in which the bits do not correspond to a single 21-x error cause, but rather in which numbers encoded by the bits represent individual 21-x error causes or combinations of 21-x error causes.
[0040] In one alternative, a start test is performed when starting the converter 2, wherein the start test comprises at least checking a supply voltage and / or checking voltage signals at gate drivers of the converter 2 and / or checking switching operations of gates of the converter 2 and / or checking a respective resulting phase voltage, wherein a check result 22 is stored in the non-volatile memory 6. The voltages can be checked, for example, by comparing them with minimum and / or maximum threshold values. The voltage signals and / or the switching operations can be checked, for example, by checking a predetermined or expected correspondence between the voltage signals and the respective switching states of the gate drivers and gates.The phase voltages can, for example, be compared with values derived from the switching states of the gates, whereby any deviation that occurs can be checked using predefined threshold values.
[0041] It can be provided that the storage of the recorded status data 10 and / or the determination of the cause of the error 21-x by means of the expert system 20 is carried out by means of a computing device 7 reserved at least for this time.
[0042] In another alternative, it is additionally or alternatively provided that, based on the result 21 of the expert system 20, at least one counter 23 is incremented or decremented depending on the error. For example, one or more counters 23 can be provided for one or more error causes 21-x.
[0043] In particular, it can be provided that, based on the result 21 of the expert system 20 and / or a counter reading of the at least one counter 23, at least one action 30 for troubleshooting is initiated. For this purpose, in particular, an action signal 31 can be generated and provided, which is fed, for example, to a vehicle control system 51. For example, it can be provided that a counter 23 is assigned to a fault cause 21-x of the type "overheating of the power module."
[0044] If this fault cause 21-x occurs, for example, the counter 23 is incremented by one (or, conversely, decremented by one, depending on the specific design). After this fault cause 21-x occurs, an attempt is made to correct the fault using a measure 30. For example, in an electric vehicle, the ignition is turned off (after the electric vehicle has been transferred to a safe state) so that the overheated power module can cool down. After a specified waiting period, the ignition is turned back on, and continued operation of the electric vehicle is possible. If a desaturation fault with the same fault cause 21-x then occurs again, the associated counter 23 is incremented (or decremented) by one again, and measure 30 is executed again, and so on.However, if the counter reading on counter 23 reaches a threshold, the final measure 30 is, for example, a complete shutdown (after the electric vehicle has been brought into a safe state), which means that it can no longer be restarted and a visit to the workshop is necessary. List of reference symbols 1 Converter arrangement 2 inverters 3 Control device 4 Computing device 5 storage 6 non-volatile memory 7 reserved computing device 10 recorded status data 20 Expert system 21 results 21-x Error cause 22 Review result 23 points 30 Measure 31 action signal 50 vehicles 51 Vehicle control
Claims
[1] Method for operating a converter (2), wherein power semiconductors of the converter (2) are monitored for the presence of desaturation errors, wherein, upon detection of a desaturation fault in at least one of the power semiconductors, recorded status data (10) of the converter (2) are stored in a non-volatile memory (6), wherein the recorded status data (10) are fed to an expert system (20) which is designed to determine at least one cause of the error (21-x) based on the recorded status data (10), wherein a result (21) of the expert system (20) is stored in the non-volatile memory (6), wherein a start test is carried out when starting the converter (2), wherein the start test comprises at least a check of a supply voltage and / or a check of voltage signals at gate drivers of the converter (2) and / or a check of switching operations of gates of the converter (2) and / or a check of a respectively resulting phase voltage, wherein a check result (22) is stored in the non-volatile memory (6). [2] Method for operating a converter (2), wherein power semiconductors of the converter (2) are monitored for the presence of desaturation errors, wherein, upon detection of a desaturation fault in at least one of the power semiconductors, recorded status data (10) of the converter (2) are stored in a non-volatile memory (6), wherein the recorded status data (10) are fed to an expert system (20) which is designed to determine at least one cause of the error (21-x) based on the recorded status data (10), wherein a result (21) of the expert system (20) is stored in the non-volatile memory (6), wherein, based on the result (21) of the expert system (20), at least one counter (23) is increased or decreased depending on the error. [3] Method according to claim 1 or 2, characterized by that values of at least part of the acquired state data (10) are locked for storage immediately after the desaturation error is detected. [4] Method according to one of the preceding claims, characterized by that the storage of the recorded status data (10) in the non-volatile memory (6) takes place with the highest priority. [5] Method according to one of the preceding claims, characterized bythat the result (21) of the expert system (20) is encoded in the non-volatile memory (6) in the form of a bit field. [6] Method according to claim 5, characterized by that each bit in the bit field is assigned an error cause (21-x). [7] Method according to one of the preceding claims, characterized by that the storage of the recorded status data (10) and / or the determination of the cause of the error (21-x) by means of the expert system (20) is carried out by means of a computing device (7) reserved at least for this time. [8] Method according to one of the preceding claims, characterized by that, based on the result (21) of the expert system (20) and / or a counter reading of the at least one counter (23), at least one error correction measure is initiated. [9] Converter arrangement (1), comprising: an inverter (2), and a control device (3), wherein the control device (3) is designed to monitor power semiconductors of the converter (2) for the presence of desaturation errors, and when a desaturation fault is detected in at least one of the power semiconductors, storing the status data (10) of the converter (2) in a non-volatile memory (5), further providing an expert system (20) which is designed to determine at least one cause of error (21-x) based on the recorded status data (10), to feed the recorded status data (10) to the expert system (20), and storing a result (21) of the expert system (20) in the non-volatile memory (6), and to carry out a start test when starting the converter (2), wherein the start test comprises at least a check of a supply voltage and / or a check of voltage signals at gate drivers of the converter (2) and / or a check of switching operations of gates of the converter (2) and / or a check of a respectively resulting phase voltage, wherein a check result (22) is stored in the non-volatile memory (6). [10] Converter arrangement (1), comprising: an inverter (2), and a control device (3), wherein the control device (3) is designed to monitor power semiconductors of the converter (2) for the presence of desaturation errors, and when a desaturation fault is detected in at least one of the power semiconductors, storing the status data (10) of the converter (2) in a non-volatile memory (5), further providing an expert system (20) which is designed to determine at least one cause of error (21-x) based on the recorded status data (10), to feed the recorded status data (10) to the expert system (20), and storing a result (21) of the expert system (20) in the non-volatile memory (6), and based on the result (21) of the expert system (20), to increase or decrease at least one counter (23) depending on the error.
Citation Information
Patent Citations
Phase-shifting full-bridge convertor real-time fault diagnosis method and system
CN101793928A
CN000101793928A
