Device and method for determining the temperature of an electronic device

Abstract

An arrangement of a supply unit (2) comprising an industrial control system, wherein the supply unit (2) is assigned to a superior computing unit (3), wherein the supply unit (2) comprises at least one electronic component (100), wherein the at least one electronic component (100) has at least one electronic device (108, 109), wherein the supply unit (2) is configured to calculate a first temperature (T1) of the at least one electronic device (108, 109) by means of a first thermal model (M1), and the superior computing unit (3) is configured to calculate a second temperature (T2) of the at least one electronic device (108, 109) by means of a second thermal model (M2), wherein the supply unit (2) and the superior computing unit (3) jointly act such that at least the first temperature (T1) or the second temperature (T2) of the at least one electronic device (108, 109) is calculated.

Description

Technical Field

[0001] This invention relates to an apparatus comprising a supply unit of an industrial control system and a superior computing unit assigned to the supply unit. The supply unit has at least one electronic component, which includes at least one electronic device. The supply unit is configured and / or arranged to calculate a first temperature of the at least one electronic device using a first thermal model.

[0002] Furthermore, the present invention relates to a method for determining at least one temperature of at least one electronic device of at least one electronic component in a supply unit having a superior computing unit in an industrial control system. Here, the first temperature of the at least one electronic device can be calculated by the supply unit using a first thermal model.

[0003] The present invention also relates to a computer program, particularly a cloud application, the computer program including instructions that, when the program is executed by the aforementioned apparatus, cause the apparatus to perform the aforementioned method.

[0004] Furthermore, the present invention relates to a data carrier signal that transmits the aforementioned computer program, and to a machine-readable storage medium, such as a volatile or non-volatile medium, that includes the aforementioned computer program. Background Technology

[0005] The aforementioned types of apparatus and methods are known in the prior art. Calculating the temperature of electronic components, such as IGBTs (Insulated Gate Bipolar Transistors), is typically performed by means of software executable on the supply unit, particularly on the converter, and is usually based on sensor temperature, which measures the temperature of a cooling plate on which the IGBT module is applied. The device, such as the supply unit, can have a different number of temperature sensors. These sensors are then associated with a suitable chip. In converters configured as a combination of a control unit and power components, currently only the temperature on the power components is calculated, which in turn is provided to the control unit so that the control unit can display the temperature externally, for example, via an IBN tool.

[0006] For example, determining the temperature of an electronic device, such as an IGBT, can be used to protect the device from overheating. Here, if the calculated temperature value exceeds a preset value, the temperature of the corresponding electronic device can be reduced, for example, by cutting off the supply unit, such as a converter, or by a reaction in the supply unit, such as a reduction in current and / or a reduction in pulse frequency.

[0007] Furthermore, based on the temperature values ​​of the electronic devices and the power cycling profiles provided by the manufacturer, wear calculations can be performed on each device, such as each IGBT chip. The more accurate the chip temperature is determined, the more accurate the wear calculation can be performed.

[0008] Furthermore, it allows for better utilization of the devices / supply units that can determine chip temperature very precisely, since a lower safety margin must be considered when designing the devices / supply units.

[0009] Another problem is that if the power supply unit is not powered on, the temperature cannot be calculated. If the power supply unit is put back into operation, there is no information about the temperature of the electronics within the unit before it was shut down. This means that the temperature of the electronics must be estimated before the power supply unit is put back into operation. This estimation is unreliable and will not yield optimal results. Summary of the Invention

[0010] Based on the foregoing, the object of the present invention is to further improve the apparatus and method mentioned at the beginning, and thereby reliably and preferably achieve a more accurate determination of the temperature of the electronic devices in the supply unit, especially in the converter.

[0011] According to the present invention, the above objective is achieved by means of the aforementioned apparatus in the following manner: the upper-level computing unit is configured to calculate the second temperature of at least one electronic device by means of a second thermal model, wherein the supply unit and the upper-level computing unit (in operation) work together to calculate at least the first temperature or the second temperature of at least one electronic device.

[0012] For example, a higher-level computing unit can take over temperature calculations—calculating only the second temperature when the supply unit is cut off. This is, for example, if the supply voltage is no longer available at the converter. From that moment on, information about the thermal path temperature is lost. The higher-level computing unit, such as a higher-level control device, edge device, or cloud server, can take over the temperature calculations and transmit the information to the converter when it is reconnected. In this way, it is ensured that whenever the industrial control system is running, a calculated, rather than estimated or assumed, temperature for at least one electronic device can be provided. In this way, the temperature of at least one electronic device can be reliably determined.

[0013] For example, as long as a data connection exists between the supply unit and the superior computing unit, it is possible to consider the following: the supply unit obtains all relevant data (for the operation of the supply unit) (e.g., temperature and preferred wear value) from the superior computing unit, which calculates a second temperature and preferably also calculates the wear value based on the second temperature. If the data connection fails, the supply unit can take over the calculation again.

[0014] Therefore, in one embodiment, it is advantageous to propose that the supply unit be configured as a converter, such as a converter for an industrial control system that operates automated facilities.

[0015] Furthermore, in one embodiment, it is advantageous to propose that the computing unit be configured as a higher-level control device, as an edge device, or as a cloud server.

[0016] In one implementation, it can be advantageously proposed that the corresponding thermal model is stored on the supply unit or the upper-level computing unit.

[0017] In one implementation, it is advantageous that the second thermal model is more detailed than the first thermal model, resulting in a more accurate calculation of the second temperature using the second thermal model compared to the first temperature calculated using the first thermal model. By improving the calculation of the temperature of at least one electronic component, wear calculations for at least one of the electronic devices can also be performed more effectively. Thus, better accuracy can be obtained.

[0018] In one implementation, it is advantageous to consider the current temperature of at least one electronic device when calculating the power loss using a second thermal model. For example, the above approach can be implemented as long as the conduction characteristics used for power loss calculation are related to the temperature calculated based on the current temperature (either in the upper-level computing unit or in the supply unit).

[0019] Therefore, the second thermal model can be calculated iteratively, where better results can be obtained with each iteration for determining the temperature of at least one electronic device. When calculating the power loss within the scope of the first thermal model, the conduction characteristic curve corresponding to the highest temperature of at least one electronic device is always used as the basis.

[0020] This allows for more accurate calculation of power loss. Excessive discrepancies between the power loss calculated using the first thermal model and the second thermal model indicate an incorrect design of the first thermal model, thus enabling the derivation of an improved, simplified version of the first thermal model.

[0021] In one implementation, it is advantageous to consider the nonlinear mapping of the conduction characteristic curve, for example via a lookup table, when calculating the second thermal model, and to take into account switching energy loss.

[0022] In one embodiment, it is advantageous to propose that at least one electronic component has two or more electronic devices, preferably two or more IGBT chips, wherein each electronic device, preferably each IGBT, is considered individually when calculating the second thermal model, such that the second temperature is calculated for each individual electronic component, preferably each individual IGBT chip.

[0023] In one embodiment, it is advantageous to propose that at least one electronic component has two or more electronic devices, preferably two or more IGBT chips, wherein thermal coupling between the individual electronic devices, preferably between the individual IGBTs, or between the IGBTs and diodes, etc., is taken into account when calculating the second thermal model.

[0024] In one embodiment, it is advantageous to propose that at least one electronic component has two or more electronic components, preferably two or more IGBT chips, wherein different thermal paths are considered for different electronic devices when determining the second temperature.

[0025] In one embodiment, it is advantageous to propose that the supply unit comprises two or more electronic components, preferably two or more IGBT modules, wherein thermal coupling between the individual electronic components, preferably between the individual IGBT modules, is taken into account when calculating the second temperature.

[0026] By significantly improving the determination of the temperature of each individual electronic component and optimizing the temperature of each individual IGBT chip, the wear of the IGBT module can be calculated more accurately, thereby leading to more precise conclusions about the current state of the IGBT.

[0027] In this implementation, it is advantageous that the electronic component includes at least one temperature sensor and that the second thermal model includes an option to determine the temperature of the temperature sensor—the so-called virtual sensor temperature.

[0028] The calculated virtual sensor temperature can be compared with the actual sensor temperature measured by the temperature sensor. Excessive deviation between the values ​​indicates a temperature sensor malfunction. An alarm can then be output, for example, indicating a temperature sensor malfunction. Here, the (first and / or second) temperature can be further calculated using the virtual sensor temperature.

[0029] Therefore, the second thermal model can have the option of calculating one (or more) virtual sensor temperatures, thereby enabling the verification of temperatures measured by temperature sensors using the calculated virtual sensor temperatures.

[0030] As already mentioned, in one embodiment, it is possible to propose that the supply unit comprises multiple (two, three, four, five, or six) electronic components, wherein each electronic component may have two or more electronic devices. Here, the second thermal model may have the option to take into account thermal coupling between the various electronic components and / or between the various electronic devices of the respective electronic components.

[0031] The above-mentioned objective of the present invention is also achieved in the method mentioned at the beginning of the present invention by means of a higher-level computing unit being able to calculate a second temperature of at least one electronic device by means of a second thermal model, wherein (if the industrial control system is in operation) the supply unit and the higher-level computing unit act to calculate at least one of the temperatures of at least one electronic device.

[0032] In one implementation, it would be advantageous to calculate the first temperature and the second temperature simultaneously.

[0033] It is possible to compare the chip temperature (first temperature) obtained in the supply unit, such as in the converter, with the temperature obtained in the upper-level computing unit, such as on a cloud server (second temperature). If the deviation is too large, it indicates that the (simplified) thermal model (first thermal model) in the converter is incorrectly designed, and therefore improvements to the first thermal model can be derived.

[0034] In one implementation, a further advantage is obtained if the (second) wear value of at least one electronic device is determined based on a second temperature. Similarly, the (first) wear value can be calculated based on a first temperature. If the calculated values ​​of the two wear values ​​deviate significantly from each other, this also indicates that the design of the first thermal model is incorrect. Improvements to the first thermal model can then be derived from this.

[0035] In one implementation, it would be advantageous to adapt the maintenance plan for the supply unit based on a wear value, which is determined according to a second temperature.

[0036] In one embodiment, the calculation of the first temperature or the second temperature can include at least one of the following sub-steps:

[0037] - Calculation of power loss

[0038] - Computational thermal path.

[0039] For example, it is possible to consider performing power loss calculations only on the converter (within the scope of the first thermal model) and transmitting the resulting values ​​to a higher-level computing unit, such as a higher-level control device. Therefore, in one embodiment, it is possible to propose performing only more complex thermal path calculations in the higher-level computing unit, such as in the higher-level control device (using the second thermal model).

[0040] Within the scope of the first thermal model, the losses of electronic devices, such as IGBTs and diodes, are calculated using the current operating point and data stored in the supply unit (Vcesat, switching loss characteristic curves, etc.).

[0041] Within the scope of the first thermal model, a first temperature (the travel temperature between the corresponding chip and its sensor) is calculated based on the loss obtained in the power loss calculation using data (thermal resistance and thermal time constant) stored in the supply unit. The term travel temperature is understood within the scope of this disclosure as a temperature difference calculated over a thermal path between the temperature at the chip (IGBT, diode, etc.) and a reference point, such as the temperature at a temperature sensor. The chip temperature is obtained by adding the corresponding travel temperature to the appropriate sensor temperature, since the aforementioned difference has been added to the corresponding sensor temperature.

[0042] In one implementation, it is advantageous to propose that the calculation of the second (but not the first) temperature includes one or more of the following sub-steps (which are not feasible when calculating the first temperature):

[0043] -Consider the current temperature in the power loss calculation.

[0044] -Consider the nonlinear mapping between the conduction characteristic curve and the switching energy loss.

[0045] - If at least one electronic component has two or more electronic devices, then each electronic device of the at least one electronic component is considered individually, such that a second temperature is calculated for each individual electronic device, and / or thermal coupling between the individual electronic devices is considered and / or different thermal paths are considered in different electronic devices.

[0046] - If the supply unit includes two or more electronic components, thermal coupling between the individual electronic components should be considered.

[0047] In one implementation, it would be advantageous for the electronic component to include at least one temperature sensor and for the temperature of the temperature sensor to be determined when calculating the second thermal model.

[0048] In implementations without temperature sensors, it is possible to propose determining a reference temperature (e.g., always 25°C) for the (first and / or second) thermal model (fixed or always kept the same). Attached Figure Description

[0049] The present invention will now be described and explained in more detail with reference to the embodiments shown in the accompanying drawings. The drawings show:

[0050] Figure 1 This illustrates a portion of an industrial control system, which includes a converter with a higher-level control device.

[0051] Figure 2 The cross-section of the IGBT module is shown.

[0052] Figure 3A flowchart illustrating a method for parallel calculation of IGBT temperature, and

[0053] Figure 4 The device is shown connected to the cloud.

[0054] In the embodiments and drawings, elements that are the same or have the same function can be provided with the same reference numerals. Furthermore, the reference numerals in the claims and specification are only for better understanding of this application and should not in any way be considered as limiting the subject matter of the invention. Detailed Implementation

[0055] Figure 1 Device 1 is shown, corresponding to a device according to the invention. Device 1 includes a converter 2 having a superior control device 3. Converter 2 includes an IGBT module 100, which, for example, has an IGBT (bipolar transistor with insulated gate electrode) 108 and a diode 109.

[0056] Device 1 can, for example, be configured as part of an industrial control system (not shown) for process control in an automated facility. Here, converter 2 can be configured to supply current / voltage to (not shown) such as a rotating electric motor, especially an asynchronous motor.

[0057] Figure 2 A cross-sectional view through the IGBT module 100 is shown. Figure 2 It can also be deduced that: the IGBT module 100 includes a substrate 101. A thermal paste layer 103 is coated on one side of the substrate 101. A solder layer 105 is coated on the other side of the substrate 101 opposite to the side with thermal paste 102. The thermal paste layer 103 forms an intermediate layer between the substrate 101 and the coolant 102. The solder layer 105 forms an intermediate layer between the substrate 101 and a copper layer 104, wherein the copper layer 104 supports a ceramic substrate 106. The ceramic substrate 106 has copper layers 104 and 104' on both sides. An IGBT 108 and a diode 109 are disposed on a portion of another copper layer 104' away from the solder layer 105 and are fixed to the other copper layer 104' by means of another solder layer 107. The IGBT 108, the diode 109, and another portion of the other copper layer 104' away from the solder layer 105 are connected to a bonding wire 110. In addition, the IGBT module 100 has a temperature sensor 111. The heat flow through the IGBT module 100 (from the diode 109 and from the bipolar transistor 108 toward the coolant 102) is shown by arrow W.

[0058] The converter 2 may have software or computer program and mechanism for executing the software or computer program, wherein the software or program is able to calculate the first temperature T1 of the IGBT 108 by means of, for example, a first thermal model M1 stored in the converter 2.

[0059] The calculation using the first thermal model M1 can include, for example, two sub-steps TS1 and TS2, preferably consisting of these two sub-steps TS1 and TS2.

[0060] In the first sub-step TS1, the power loss can be calculated. Here, the losses of IGBT 109 (in the case of multiple bipolar transistors in all bipolar transistors) and diode 108 (in the case of multiple diodes in all diodes) can be calculated based on the current operating point of converter 2 and the data stored in converter 2 (Vcesat, switching loss characteristic curve, etc.).

[0061] For example, power loss can be calculated using a characteristic curve. Typically, this is based on the characteristic curve that results in the greatest power loss. For instance, the characteristic curve can be selected or determined based on a first or second calculated temperature. Therefore, power loss can be calculated based on the calculated chip temperature.

[0062] When temperature sensor 111 is present, the calculation of the first and / or second temperatures T1 and / or T2 can be performed using the sensor temperature measured by sensor 111. Here, temperature sensor 111 measures the temperature of cooling plate 102. It is conceivable that IGBT module 100 has multiple sensors 111. In multiple IGBT modules, different IGBT modules can have different numbers of temperature sensors. Each sensor is preferably associated with one or more chips, such as IGBT 108 or diode 109. Figure 2 (related to)

[0063] The thermal path can be calculated in the second sub-step TS2. Based on the loss calculated in the power loss calculation (sub-step TS1), the temperature (also called the travel temperature) between the chip, such as IGBT 108 or diode 109 and its sensor 111 can be calculated using data stored in converter 2 (e.g., thermal resistance and thermal time constant). In the case of a single chip, such as IGBT 108, this is the first temperature T1. In the case of multiple chips (IGBT 108, diode 109, etc.), the travel temperature between the respective chip and its sensor can be calculated for all chip-sensor pairs. As described above, the chip temperature can be obtained by adding the corresponding travel temperature to the adapted sensor temperature.

[0064] For example, the thermal path between the IGBT chip 108 and the sensor 111 can be described by three thermistors with three thermal time constants. Therefore, the current first temperature T1 of the IGBT 108 can be calculated using the current operating point (current, voltage, frequency, duty cycle) of the converter 2.

[0065] The upper-level control unit 3 is configured to calculate the second temperature T2 of the IGBT 108 using the second thermal model M2. For this purpose, the upper-level control unit 3 may also include software or a computer program and a mechanism for executing the software or computer program.

[0066] It should be understood that a converter can contain multiple IGBT modules. Here, for example, the temperatures T1 and T2 of the corresponding electronic components (IGBTs and / or diodes, etc.) in each individual IGBT module can be calculated using the aforementioned software program.

[0067] The converter 2 and the upper-level control device 3 work together to ensure that, when the automation facility and thus the industrial control system are running, at least one of temperatures T1 and T2 is always calculated, i.e., at every arbitrary point in time. For example, when the converter 2 is running, the first temperature T1 and the second temperature T2 can preferably be calculated simultaneously and, for example, compared with each other.

[0068] In addition, it is possible to consider calculating only the first or second temperature in order to save available computing resources.

[0069] Furthermore, it is advantageous that a second temperature T2 is calculated even when converter 2 is no longer powered on. Therefore, when converter 2 is put into operation, the calculated value of the second temperature T2 can be used as a starting value.

[0070] The first thermal model M1 can be constructed in a simplified manner, for example, to save computational resources of the converter 2. That is, the second thermal model M2 can be more detailed than the first thermal model M1.

[0071] For example, compared to the second thermal model M2, the first thermal model M1 can include one or more of the following simplifications:

[0072] - The current temperature T2 is not considered in the power loss calculation.

[0073] - A simplified, e.g., linearly-only description of the conduction characteristic curve and switching energy loss.

[0074] - The temperatures of the IGBT and diode are not considered separately (only the IGBT temperature is calculated in the first thermal model M1).

[0075] - A simplified description of thermal coupling between components, wherein, for example, thermal coupling between electronic devices (e.g., thermal coupling between IGBT 108 and diode 109 or between individual electronic components, such as IGBT module 100) is not considered.

[0076] - A simplified description of the thermal path of the device, where, for example, the same thermal path is considered for all IGBT chips 108.

[0077] Conversely, this means that the second thermal model M2 has one or more of the additional options listed above compared to the first thermal model M1.

[0078] By considering the current temperature, the temperature dependence of the conduction characteristic curve can be taken into account. This temperature dependence can be achieved, for example, in the form of feedback, where a second thermal model M2 is iteratively calculated, using the value calculated in the iteration for the second temperature T2 in the next iteration. For example, the second thermal model M2 can be calculated in the first iteration (e.g., when calculating power loss) using a characteristic curve corresponding to the highest temperature of the IGBT 108. Then, in the second iteration, the current temperature value of the IGBT 108, obtained in this way, can be used to perform power loss calculations starting from a different characteristic curve, where another characteristic curve is selected based on the current temperature value of the IGBT 108. Therefore, the second thermal model M2 can effectively consider the temperature dependence of the conduction characteristic curve.

[0079] In the first thermal model M1, for each chip (IGBT chip 108) or diode 109, the same thermal path between chips 108, 109 and temperature sensor 111 is used as the basis. In the thermal measurements of the converter, all paths between the corresponding chips (IGBT, diode) 108, 109 and the corresponding temperature sensor 111 can be measured and stored in the converter, where only the unique "worst-case" path can be considered and stored.

[0080] In the second thermal model M2, calculations can be performed without this assumption, and when calculating the second temperature T2, calculations can be based on different thermal paths for different chip-sensor pairs.

[0081] Therefore, the first thermal model M1 within the converter software is designed solely to protect the IGBT chip 108 from overheating.

[0082] Furthermore, if the converter 2 includes multiple IGBT modules 100, the thermal coupling between the individual IGBT modules 100 can be additionally considered in the second thermal model M2. Therefore, reliable protection of the converter can be achieved while maintaining at least the converter performance unchanged. In particular, this prevents the converter from being dated too quickly.

[0083] If each chip (e.g., IGBT chip 108 or diode 109) has its own thermal path and each electronic device 108, 109 is thermally coupled to each other, then the wear of each chip 108, 109 can also be calculated, and the inaccuracy in calculating the wear of each chip 108, 109 can be reduced.

[0084] The calculation of the first temperature T1 in the software of converter 2 can be performed in parallel with the calculation of the second temperature T2 in the upper-level control device 3.

[0085] The upper-level control device 3 can be configured as an edge device or a cloud server. Here, the upper-level control device 3, such as an edge device, can have a significantly more complex second model M2 as described above for calculating power loss and thermal paths, and the second model M2 can also perform calculations efficiently (in terms of time and / or resources).

[0086] In the second thermal model M2, a nonlinear description of the conduction characteristic curve and switching loss energy can be considered (e.g., via a lookup table). This allows for a more accurate description of the conduction characteristic curve, for example. The conduction loss characteristic curve generally has the form of an e-function (exponential function). In the first thermal model M1, for simplicity, the e-function is described by a simple linear equation, inevitably leading to calculation errors. In the second thermal model M2, the characteristic curve can be accurately described (e.g., as a lookup table), thereby reducing errors in loss calculation.

[0087] For example, in the case of the second thermal model M2, each IGBT chip 108 can store its own thermal path.

[0088] Furthermore, the diode temperature in the second thermal model M2 can be calculated independently of the IGBT temperature.

[0089] Furthermore, due to the high computing power and virtually unlimited storage capacity at the edge, it is also possible to describe the very complex thermal coupling between the various components.

[0090] Figure 3 This illustrates an example of how converter 2 and the upstream control unit 3 work together. Arrows marked with "t" indicate the time sequence.

[0091] In step S10, converter 2 provides the upper-level control device 3 with relevant data for calculating the second thermal model M2. This data, for example, can be current, voltage, duty cycle, pulse frequency, etc.

[0092] There can be discrepancies between the data stored in the converter and the data required for calculation in the higher-level control device. For this reason, calculations using the first thermal model M1 and the second thermal model M2 can occur differently. For example, within the range of the first thermal model M1, calculations can be performed using the duty cycle, while within the range of the second thermal model M2, calculations can be performed using the modulation index. This allows for the transmission of different data.

[0093] In step S11, the first power loss can be calculated using converter 2, i.e., using the first thermal model M1. Preferably, in parallel with this, in step S20, the second (more complex) power loss can be calculated using the upper-level control device 3, i.e., using the second thermal model M2. When calculating the second power loss within the range of the second thermal model M2, one or more of the above-mentioned additional options can be selected. For example, the current second temperature T2 and / or the nonlinear description of the conduction characteristic curve and / or the switching energy loss can be considered here.

[0094] In the case of the second thermal model M2, it is also possible to calculate the temperature of temperature sensor 111, or, in the case of multiple temperature sensors, the temperature of each temperature sensor – step S201. The calculated temperature of the temperature sensor is also called the "virtual sensor temperature". The calculated virtual sensor temperature can then be compared with the actual measured temperature of temperature sensor 111 – step V1. Excessive deviation between the values ​​indicates a fault in the temperature sensor. Then, an alarm for a fault in temperature sensor 111 can be output. Subsequently, the calculation can be continued using the virtual sensor temperature with the help of the first thermal model M1 in converter 2 (and the second thermal model M2 in the higher-level control device 3).

[0095] In step S12, the first temperature T1 of the IGBT chip 108 in converter 2 can be calculated. For this purpose, the actual measured temperature or virtual sensor temperature of temperature sensor 111 can be used, as described above.

[0096] The second temperature T2 of the IGBT chip 108 can preferably be calculated in parallel with step S12 in the higher-level control device 3. When calculating the second temperature T2 within the range of the second thermal model M2, one or more of the above-mentioned additional options can be selected. Here, for example, thermal coupling between different IGBT chips and / or between IGBT chips and diodes (and / or other electronic devices not shown here that may include the IGBT module 100) can be considered, and / or different thermal paths can be used for different IGBTs and / or diodes (and / or other electronic devices not shown here that may include the IGBT module 100), etc.

[0097] In step V2, the first temperature T1 of the IGBT 108 can be compared with the second temperature T2.

[0098] In step S13, the first wear of the IGBT 108 can be calculated based on the first temperature T1 calculated in the converter 2. Preferably, in parallel with this, in step S22, the second wear of the IGBT 108 can be calculated on the upper-level control device 3 based on the calculated second temperature T2.

[0099] In step V3, the first wear can be compared with the second wear.

[0100] Based on the above, the wear of IGBT 108 can be accurately calculated using the second temperature T2 obtained through the second thermal model M2. Therefore, a more accurate description of the condition of the IGBT module 100 can be obtained, thereby enabling better maintenance services to be provided to customers. Furthermore, it allows for improved robustness for customers.

[0101] Furthermore, it is conceivable that, as long as a data connection exists between converter 2 and the upper-level control device 3, converter 2 obtains all operational data (temperature and wear values) from the upper-level control device 3 during operation; the upper-level control device could, for example, be located in the cloud. In this case, it is unnecessary to calculate the first thermal model M1. This conserves the resources of converter 2.

[0102] In the event of a data connection failure, converter 2 can again take over the calculation of the first temperature T1 using its own value.

[0103] It is also possible to consider that as long as converter 2 is turned off, the upper-level control device 3, such as an edge device, takes over the calculation of the second temperature T2. Therefore, the temperature can still be determined even when converter 2 has no supply voltage and can no longer determine the first temperature T1. This is advantageous because when converter 2 is turned off, information about the temperature of IGBT 108, and most importantly, information about the thermal path, is lost. If the temperature is not calculated on the upper-level control device 3 during the period when converter 2 is off, this information will no longer be provided when converter 2 is turned back on. That is, when converter 2 is off (no supply voltage), the upper-level control device 3 can continue to calculate the temperature and inform converter 2 of the temperature when it is turned back on.

[0104] If a significant difference is found when comparing values ​​performed in one or more of steps V1, V2, and V3 above, this indicates that the design of the first thermal model M1 is incorrect. From this, improvements to the development of the first thermal model M1 on converter 2 can be derived. Therefore, the development of the first thermal model M1 can continue. Furthermore, this can assist in the further development of the runtime software on the converter.

[0105] Figure 4 An exemplary device 1 connected to cloud 4 is shown. As already explained, the first temperature T1 of electronic devices 108, 109 is calculated in the software of converter 2 using a first thermal model M1, wherein the first thermal model M1 can include two of the aforementioned group steps TS1 and TS2. Here, the upper-level control device 3 can include multiple components. For example, the upper-level control device 3 includes a local (not located in the cloud) computing unit 30, which is designed to communicate with the cloud server 31. A computer program 32 is stored executablely on the computing unit 30, for example, on a hard disk not shown here. When the computer program 32 is executed, it causes device 1 to perform one or more of the above-described method steps. Here, the computer program 32 can cause the computing unit 30 to: calculate a second thermal model M2, and access the computing resources of the cloud server 31 when needed, in order to determine, for example, the second temperature T2 based on one or more of the aforementioned options of the second thermal model M2. In particular, considering that all thermal couplings would be computationally intensive, the computing resources of cloud server 31 would be very useful in this case.

[0106] Although the invention has been illustrated and described in detail by way of examples, the invention is not limited to the disclosed examples. Variations thereof can be deduced by those skilled in the art without departing from the scope of the invention as defined by the appended claims. In particular, the features described in conjunction with the method can also be used in or as a supplement to the device, and vice versa.

Claims

1. An apparatus comprising a supply unit (2) for an industrial control system, wherein, A higher-level computing unit (3) is assigned to the supply unit (2), wherein the supply unit (2) includes at least one electronic component (100), wherein at least one electronic component (100) has at least one electronic device (108, 109), wherein the supply unit (2) is configured to calculate a first temperature (T1) of at least one electronic device (108, 109) by means of a first thermal model (M1), and the higher-level computing unit (3) is configured to calculate the first temperature (T1) of at least one electronic device (108, 109) by means of a second thermal model (M2). The second temperature (T2) of 108, 109) is calculated, wherein the supply unit (2) and the upper-level calculation unit (3) work together to calculate at least one of the electronic devices (108, 109) at the first temperature (T1) or the second temperature (T2), wherein the supply unit (2) is designed to provide the upper-level calculation unit (3) with data related to the calculation of the second thermal model (M2), and the upper-level calculation unit (3) is designed to calculate the second temperature (T2) when the supply unit (2) is not powered on.

2. The apparatus according to claim 1, wherein, The supply unit is configured as a converter (2).

3. The apparatus according to claim 1 or 2, wherein, The computing unit is configured as a higher-level control device (3), an edge device, or a cloud server.

4. The apparatus according to claim 1 or 2, wherein, The second thermal model (M2) is more detailed than the first thermal model (M1), so that the second temperature calculated by the second thermal model is more accurate than the first temperature calculated by the first thermal model.

5. The apparatus according to claim 1 or 2, wherein, The electronic component (100) includes at least one temperature sensor (111), and the second thermal model (M2) includes an option to determine the temperature of the temperature sensor (111).

6. The apparatus according to claim 5, wherein, The first temperature (T1) is determined based on the temperature obtained by the temperature sensor (111) using the second thermal model (M2).

7. The apparatus according to claim 1 or 2, wherein, The electronic device is constructed as a semiconductor device.

8. The apparatus according to claim 7, wherein, The electronic device is configured as an IGBT (108) or as a diode (109).

9. The apparatus according to claim 7, wherein, The electronic component (100) is preferably constructed as an IGBT module.

10. A method for determining at least one temperature (T1, T2) of at least one electronic device (108, 109) of at least one electronic component (100) of a supply unit (2) of an industrial control system, wherein, The supply unit (2) is provided with a superior calculation unit (3), wherein the supply unit (2) is capable of calculating a first temperature (T1) of at least one of the electronic devices (108, 109) by means of a first thermal model (M1), wherein the superior calculation unit (3) is capable of calculating a second temperature (T2) of at least one of the electronic devices (108, 109) by means of a second thermal model (M2), wherein the supply unit (2) and the superior calculation unit (3) work together to calculate at least one of the temperatures (T1, T2) of at least one of the electronic devices (108, 109), wherein the supply unit (2) provides the superior calculation unit (3) with data related to the calculation of the second thermal model (M2), and the superior calculation unit (3) calculates the second temperature (T2) when the supply unit (2) is not powered on.

11. The method according to claim 10, wherein, The wear value of at least one of the electronic devices (108, 109) is determined based on the second temperature (T2).

12. The method according to claim 11, wherein, The maintenance plan for the supply unit (2) is adapted based on the wear value.

13. The method according to claim 10, 11 or 12, wherein, Simultaneously calculate the first temperature (T1) and the second temperature (T2).

14. The method according to any one of claims 10 to 12, wherein, The calculation of the first temperature or the second temperature includes at least one of the following sub-steps: - Calculate the power loss. - Computational thermal path.

15. The method according to any one of claims 10 to 12, wherein, The second thermal model (M2) is more detailed than the first thermal model (M1), making the second temperature calculated by the second thermal model more accurate than the first temperature calculated by the first thermal model, and the calculation of the first temperature (T1) is simpler than the calculation of the second temperature (T2).

16. The method according to any one of claims 10 to 12, wherein, The electronic component (100) includes at least one temperature sensor (111), and the temperature of the temperature sensor (111) is determined when calculating the second thermal model (M2).

17. The method according to claim 16, wherein, The first temperature (T1) is determined based on the temperature obtained by the temperature sensor (111) using the second thermal model (M2).

18. A computer program including instructions, particularly a cloud application, wherein when executed by a device according to any one of claims 1 to 9, the instructions cause the device to perform the method according to any one of claims 10 to 17.

19. A machine-readable storage medium comprising the computer program of claim 18.

Images