Considering the degree of loading of a power converter over time periods

Abstract

A computing device (6) repeatedly receives respective current data (D) from a control device (4) of a power converter (3) via an open network (7), said data describing a respective operating state of the power converter (3) and a reference time (t) for the respective operating state. It stores the respective received data (D) in addition to already stored data (D), which each describe a respective operating state of the power converter (3) and the respective reference time (t) for the respective operating state, so that it gradually builds up a history of the operating state of the power converter (3). The computing device (6) receives a start time (t1) and an end time (t2) from an operator (10), wherein the start time (t1) falls in the past and the end time (t2) falls after the start time (t1). It selects the stored data (D), wherein the respective reference time (t) thereof falls between the start time (t1) and the end time (t2). The computing device (6) determines a respective degree of loading (δV) of the power converter (3) for the reference times (t) of the selected data (D) on the basis of the operating states of the power converter (3) of the selected data (D) and outputs a time curve of the degree of loading (δV) and / or a variable derived from the degree of loading (δV), in particular the integral of the degree of loading (δV), and / or a frequency of the degree of loading (δV) as a function of the degree of loading (G) to the operator (10) in the form of a graphic (G).

Description

[0001] Description

[0002] Analysis of the load level of a power converter over time periods

[0003] The present invention relates to an operating method for a computing device,

[0004] - wherein the computing device repeatedly receives current data from a control unit of a power converter via an open network, describing a current operating state of the power converter and a reference point for the current operating state.

[0005] An open network is a network to which any components can be connected, allowing them to exchange information. Examples of such a network are the internet or a LAN (local area network).

[0006] The present invention further relates to a computer program, wherein the computer program comprises instructions which, when executed by a computing device, cause the computing device to execute such an operating procedure.

[0007] The present invention further relates to a computing device, wherein the computing device is programmed with such a computer program, so that the computing device performs such an operating procedure during operation.

[0008] The items mentioned are known.

[0009] Unplanned power converter failures are a significant source of disruption in factories and plants. Therefore, the remaining service life of power converter components is crucial for timely replacement. However, determining the remaining service life is difficult and involves considerable uncertainty. In particular, it is a problem that incorporates many different factors.

[0010] In the current state of the art, the manufacturer of the power inverter typically considers the so-called MTBF (mean time between failures) and performs corresponding calculations. These calculations are usually based on a typical operating cycle under assumed typical operating conditions such as inverter load, ambient temperature, humidity, air pressure, etc. This allows the MTBF to be estimated. The manufacturer provides the results of these considerations and calculations in the product documentation, thus making them available to the operator of the factory or plant.

[0011] The actual operation of the power inverter typically does not correspond, or at least not exactly correspond, to the assumptions on which the manufacturer based its determination of the MTBF (Mean Time Between Failures). The actual service life of the power inverter can therefore be greater or less than the service life specified in the product documentation.

[0012] In practice, the operator of a factory or plant is therefore faced with a dilemma: either replace the power converter prematurely or accept the risk of a sudden failure. Premature replacement incurs unnecessary costs. If the power converter fails suddenly, even greater damage is often to be expected. Therefore, efforts are underway to predict the actual remaining service life of a power converter more accurately.

[0013] Various methods for determining the remaining service life are known in the prior art.

[0014] For example, it is known to provide monitoring systems that use sensors to record and evaluate the actual operating status of the power inverter. The associated evaluation can be performed either inside or outside the control unit for the power inverter, as required.

[0015] Furthermore, data-based approaches are known that derive patterns from existing measurements and state variables. A deviation from a specific pattern can indicate an anomaly in the power converter.

[0016] Finally, models exist that estimate the remaining lifespan of a power converter online. These models are generally based on the power converter's power cycles, as known from its datasheets, and its estimated junction temperature. Changes in junction temperature during a power cycle contribute significantly to the wear of a power converter, such as its IGBT. Such models conveniently take actual load conditions into account. However, the accuracy with which the remaining lifespan is determined depends considerably on the thermal model of the power electronic component and the identification of the load cycles. Both the precise aging mechanism and the number and nature of the parameters are often difficult to ascertain.It can be modeled. Consequently, the prediction is inaccurate.

[0017] The object of the present invention is to create possibilities by which the operator is given a means to modify the operation of his factory or plant in such a targeted manner as to optimize the service life of the power converter.

[0018] The problem is solved by an operating method with the features of claim 1. Advantageous embodiments of the operating method are the subject of dependent claims 2 and 3.

[0019] According to the invention, an operating method of the type mentioned above is designed by:

[0020] - that the computing device stores the received data in addition to already stored data, which each describe a respective operating state of the power converter and the respective reference time for the respective operating state, so that the computing device gradually builds up a history of the operating state of the power converter,

[0021] - that the computing device receives a start time and an end time from an operator, where the start time is in the past and the end time is after the start time,

[0022] - that the computing device selects the stored data whose respective reference time lies between the start time and the end time, and

[0023] - that the computing device determines a respective load level of the power converter for the reference times of the selected data based on the operating states of the power converter of the selected data and outputs a time course of the load level and / or a quantity derived from the load level, in particular the integral of the load level, and / or a frequency of the load level as a function of the load level as a graphic to the operator.

[0024] By building up a history, it is possible to retrospectively analyze the operating states of specific time periods. These analyses, which are based on predefined start and end times, can be performed concurrently with ongoing data acquisition. The operator can specify the start and end times of the time period to the computer. Depending on the operating procedure, in some cases only predetermined, fixed time periods can be specified. In other cases, the operator can specify any desired time period.

[0025] The starting point must be in the past, and the ending point must also be after the starting point. However, the time interval, i.e., the time difference between the starting and ending points, can vary. The ending point can also be the current point in time. It is even possible to define the ending point in such a way that the computer continuously or periodically updates it as the historical data grows.

[0026] In many cases, it will be useful to display a graph of the load level over time to the operator. In other cases, it may be possible to display a graph of a value derived from the load level. A suitable derived value could be the integral of the load level. In other cases, it may be useful to determine the frequencies of specific load levels (histograms) and display them to the operator. In both cases—whether displaying a time-based graph or a histogram—the operator of the factory or plant can, at least in some cases, adjust the operation of the factory or plant to reduce wear on the power converter.

[0027] The load level can be determined as required. For example, it can be the junction temperature or the percentage of the switching cycle during which the inverter's semiconductor switches are open. Preferably, however, the load level increases with the degree of wear on the power inverter. In particular, the load level can be proportional to the wear.

[0028] Preferably, the computing device color-codes areas of the output graphic depending on the load level. This makes it particularly easy for the operator to evaluate the output graphic.

[0029] The problem is further solved by a computer program with the features of claim 4. According to the invention, the commands cause the computing device to execute an operating method according to the invention. The problem is further solved by a computing device with the features of claim 5. According to the invention, the computing device is programmed with a computer program according to the invention, such that the computing device executes an operating method according to the invention during operation.

[0030] The load level is advantageously understood as the indication of the wear of the IGBT or the diodes. This wear is advantageously determined or calculated within a drive, particularly within the inverter.

[0031] The determination or calculation is achieved in particular based on the current operating point of the inverter (especially based on chip temperatures).

[0032] Wear can therefore be determined using the characteristic curves of the number of cycles provided by semiconductor manufacturers.

[0033] Preferably, chip temperatures, heat sink temperatures, rotational speeds, currents, overload counters and / or operating states are also recorded.

[0034] A comparison of temperatures (especially coolant temperatures, ambient temperatures and / or heat sink temperatures), currents and / or DC voltages against reference curves and a subsequent assessment of whether the operation is OK is fundamentally possible.

[0035] Advantageously, within the scope of the present invention, a comparison of the wear values ​​against a reference curve is performed. The wear values ​​can be used as the sole value for comparison.

[0036] Additionally, a fuzzy logic factor is advantageous for the evaluation, as wear values ​​are also subject to inaccuracies (for example, inaccuracies in the calculation of chip temperatures).

[0037] This uncertainty factor is preferably configured such that values ​​in a first range of the reference curve are considered acceptable, values ​​in a second range are considered problematic, and values ​​in a third range are considered very problematic. In other words, this uncertainty factor is preferably configured such that values ​​up to, for example, 50% above the reference curve are considered acceptable (green range, in accordance with the abbreviations in the description of FIG 11: "gr", "yl", and "rd" for green, yellow, and red), values ​​above 50% are considered problematic (yellow range), and values ​​above 75% are considered very problematic (red range).

[0038] The rendering can be performed on an external edge and / or cloud.

[0039] The wear counter is preferably calculated on the drive, especially on the inverter, but it is equally conceivable that it is also calculated on the edge / cloud.

[0040] The properties, features, and advantages of this invention described above, as well as the manner in which they are achieved, will become clearer and more readily understandable in connection with the following description of the exemplary embodiments, which are explained in more detail in conjunction with the drawings. These drawings show, in schematic representation:

[0041] FIG 1 a block diagram,

[0042] FIGS. 2 to 4 are a flowchart.

[0043] FIG 5 a flowchart,

[0044] FIG 6 a graphic,

[0045] FIG 7 a flowchart,

[0046] FIG 8 a graphic,

[0047] FIG 9 a flowchart,

[0048] FIG 10 a graphic,

[0049] FIGS 11 and 12 are modifications of FIGS 6 and 10.

[0050] According to FIG. 1, an energy flow between an electrical supply network 1 and an electrical unit 2 (usually an electrical load, but sometimes also an electrical energy source such as a generator) is controlled via a power converter 3. For this purpose, the power converter 3 is supplied with control commands C by means of an associated control unit 4.

[0051] The control unit 4 detects operating parameters B of the power converter 3, in particular the voltages and currents flowing across the various semiconductor switches of the power converter 3. Furthermore, the control unit 4 may also be aware of other operating parameters B of the power converter 3, such as the switching frequency, pulse width, and duty cycle of the various semiconductor switches. The control unit 4 may also receive information I from the environment of the power converter 3 via sensors 5, such as the ambient temperature, humidity, air pressure, and other parameters.

[0052] The control unit 4 executes an operating procedure which is explained below in conjunction with FIG. 2. The procedure is only briefly described because it is of minor importance within the scope of the present invention.

[0053] According to FIG. 2, the control unit 4 determines the control commands C for the power inverter 3 in step S1 and controls the power inverter 3 in step S2 according to the control commands C. The determination of the control commands C and the corresponding control of the power inverter 3 are not, as such, the subject of the present invention.

[0054] In step S3, the control unit 4 receives the operating parameters B and the information I. In step S4, the control unit 4 assigns the corresponding acquisition time to the operating parameters B and the information I and stores it internally.

[0055] In step S5, the control unit 4 checks whether it should transmit data D to a computer 6 with which it is connected via an open network 7 (for example, the internet or a LAN). If not, the control unit 4 returns directly to step S1. If, however, it does, in step S6 the control unit 4 processes the operating variables B and information I that it has stored during a specific period. This period typically extends from the time of the last transmission of data D to the computer 6 to the current time. By processing the operating variables B and information I, the control unit 4 determines the data D. The data D describes a current operating state of the power converter 3.In the simplest case, the processing is trivial, so that the data D directly correspond to the operating variables B and information I. However, more complex processing is also possible, especially statistical analyses. In this case, the data D corresponds to the statistical analyses. For example, the control unit 4 can determine the mean, median, minimum, maximum, dispersion, variance, and other parameters. Furthermore, in step S6, the control unit 4 assigns a reference time t to the determined data D. This combination—that is, the data D determined through analysis and the reference time t—is transmitted by the control unit 4 to the computer 6 via network 7 in step S7.

[0056] The computing device 6 is programmed with a computer program 8. The computer program 8 comprises instructions 9 that can be executed directly by the computing device 6. When executed by the computing device 9, the instructions 9 cause the computing device 6 to perform an operating procedure. Part of this operating procedure is explained in more detail below in conjunction with FIG. 3, and another part in conjunction with FIG. 4. The procedures of FIG. 3 and 4 are executed in parallel.

[0057] According to FIG. 3, in step S11, the computer 6 receives the data D and its corresponding reference time t. The data D and the reference time t are transmitted from the control unit 4 to the computer 6 via the network 7. In step S12, the computer 6 stores the data D received in step S11, including the corresponding reference time t. The data in step S12 is stored in such a way that similar data D and their corresponding reference times t, which the computer 6 has already received and stored at earlier times, are not overwritten. The data in step S12 is therefore stored in addition to this data D and its corresponding reference times t. The computer 6 then returns to step S11, so that steps S11 and S12 are executed repeatedly.As a result, the computing unit 6 gradually builds up a history of the operating state of the power converter 3.

[0058] According to FIG. 4, in step S21, the computing unit 6 checks whether an operator 10 (see FIG. 1) has submitted a request for evaluation. If so, the computing unit 6 executes steps S22 to S26. Otherwise, the computing unit 6 returns directly to step S21.

[0059] In step S22, the computer 6 receives a start time t1 and an end time t2 from the operator 10. The start time t1 is in the past. In principle, it can be any point in the past. The end time t2 is after the start time t1. The end time t2 can also be in the past. However, it can also be in the present. Examples of the evaluation period defined by the start time t1 and the end time t2 are today, yesterday, last week, last month, last year, and the entire service life of the power converter 3 to date. In step S23, the computer 6 selects the stored data D whose respective reference time t lies between the start time t1 and the end time t2.The data D, whose respective reference time t corresponds exactly to the start time t1 or exactly to the end time t2, can also be selected or excluded as needed. The question of how these two singular cases are handled is of secondary importance.

[0060] In step S24, the computing unit 6 determines a respective load factor ÖV of the power converter 3 for the reference times t of the selected data D, based on the operating states of the power converter 3. The load factor ÖV preferably increases with the extent to which the power converter 3 is subject to wear. In particular, the load factor ÖV can be proportional to the wear. The load factor ÖV can be normalized, for example, to the expected service life of the power converter 3.

[0061] In step S25, the computer unit 6 generates a graph G. The generated graph G is based on the public transport load levels determined in step S24. In step S26, the computer unit 6 outputs the generated graph G to the operator 10. The computer unit 6 can then, for example, return to step S21.

[0062] Possible types of graphic G are explained in more detail below.

[0063] For example, step S25 can be implemented as explained in more detail below in conjunction with FIG. 5. According to FIG. 5, step S25 is designed such that the computing unit 6 determines a time-dependent profile of the load level ÖV as graph G. This representation makes it particularly easy for the operator 10 to see at which reference times t a particularly high wear level ÖV occurred. In some cases, by analyzing the operational process in which the power converter 3 is used, the cause can be identified. Sometimes the operational process can even be modified so that the particularly high wear levels ÖV can be avoided or at least reduced. FIG. 6 shows a possible time-dependent profile of the load level ÖV.

[0064] Alternatively or additionally, step S25 according to FIG. 7 can be implemented such that the computing unit 6 in step S25 determines a time course of a quantity derived from the public transport load level as graph G. The derived quantity can, in particular, be the integral of the public transport load level. This representation makes it particularly easy for the operator 10 to see the total wear and tear that has occurred within the evaluation period. FIG. 8 shows a possible time course of the derived quantity.

[0065] Alternatively or additionally, step S25 according to FIG. 9 can be implemented such that the computing unit 6 determines a frequency H of the public transport load level as a graph G in step S25. This representation makes it particularly easy for the operator 10 to see with what frequency H certain public transport load levels occurred within the evaluation period. FIG. 10 shows a possible frequency distribution.

[0066] Preferably, the computing device color-codes six areas of the output graphic G depending on the public transport occupancy level. This is illustrated below in FIGS. 11 and 12. FIGS. 11 and 12 are modifications of FIGS. 6 and 10. Unlike FIGS. 6 and 10, however, FIGS. 11 and 12 include areas that are displayed in different colors. For example, the abbreviations "gr", "yl", and "rd" can stand for green, yellow, and red, respectively.

[0067] In addition to graph G, other values ​​can be output, such as the maximum public transport load level that occurred within the evaluation period, a maximum junction temperature of the semiconductor switches, and others.

[0068] The present invention offers many advantages. In particular, it does not merely consider the current state or integrate accumulated wear in isolation, but rather enables analyses of temporal trends and frequencies. This allows the operator 10 to perform more in-depth analyses that offer potential for future optimizations. The evaluations can also be linked with results available from other sources. The computing device 6 can be a standalone computing device located locally at the operator's location. However, it can also be a component of a so-called cloud. The same applies to the memory in which the history is stored.

[0069] Although the invention has been further illustrated and described in detail by the preferred embodiments, the invention is not limited by the disclosed examples and other variations can be derived from them by the person skilled in the art without leaving the scope of protection of the invention.

Claims

Patent claims 1. Operating procedures for a computing facility (6), - wherein the computing device (6) repeatedly receives current data (D) from a control device (4) of a power converter (3) via an open network (7), which describes a respective operating state of the power converter (3) and a reference time (t) for the respective operating state, - wherein the computing device (6) stores the received data (D) in addition to already stored data (D), which each describe a respective operating state of the power converter (3) and the respective reference time (t) for the respective operating state, so that the computing device (6) gradually builds up a history of the operating state of the power converter (3), - wherein the computing device (6) receives a start time (t1) and an end time (t2) from an operator (10), wherein the start time (t1) is in the past and the end time (t2) is after the start time (t1), - wherein the computing device (6) selects the stored data (D) whose respective reference time (t) lies between the start time (t1) and the end time (t2), and - wherein the computing device (6) determines a respective load level (ÖV) of the power converter (3) for the reference times (t) of the selected data (D) based on the operating states of the power converter (3) of the selected data (D) and outputs a time course of the load level (ÖV) and / or a quantity derived from the load level (ÖV), in particular the integral of the load level (ÖV), and / or a frequency (H) of the load level (ÖV) as a function of the load level (ÖV) as a graph (G) to the operator (10).

2. Operating method according to claim 1, characterized in that the load level (ÖV) increases with the extent to which the power converter (3) is subject to wear, in particular is proportional to the wear.

3. Operating method according to claim 1 or 2, characterized in that the computing device (6) color-codes areas of the output graphic (G) depending on the load level (ÖV).

4. Computer program, wherein the computer program comprises instructions (9) which, when executed by a computing device (6), cause the computing device (6) to execute an operating procedure according to any one of claims 1 to 3.

5. Computing device, wherein the computing device is programmed with a computer program (8) according to claim 4, such that the computing device executes an operating procedure according to any one of claims 1 to 3 during operation.

Images