Avoiding misdiagnosis for a control valve with an electropneumatic positioner

Abstract

To avoid misdiagnosis of a control valve with an electropneumatic positioner for regulating the position of a valve element, the end position of the valve element is recorded over time during operation. This allows the deviation of each end position from the expected end position to be determined. Furthermore, a histogram is recorded using the electropneumatic positioner during operation. The histogram is corrected using the determined deviation of each end position. Based on the corrected histogram, a diagnosis for the control valve is generated and displayed. If necessary, a maintenance measure for the control valve is then initiated. Due to the correction achieved by determining any changes in the end position of the valve element, the diagnoses are significantly more accurate.

Description

Field of invention

[0001] Electropneumatic positioners often have a problem with the control deviation histogram, which records the deviation between the setpoint and the actual position of the valve element. The positioner records the control deviation at regular intervals and stores it in the corresponding histogram bin as an incremented bin counter.

[0002] Even if the counter functions as intended, this recording does not differentiate between open and closed-loop control. Therefore, in addition to the actual setpoint deviation during operation, the histogram also includes all instances or times when the valve was completely closed (setpoint is zero). This is problematic when the closed position of the valve element is no longer exactly zero. In this case, the histogram will continuously increment the counter for a "non-zero control deviation," which actually represents an open-loop control situation where the positioner is not attempting to control and cannot compensate for the deviation between the zero setpoint and the measured position.

[0003] For valves that are rarely closed (low open-loop control time), the resulting distortion of the histogram should not pose a problem. However, for valves with a significant amount of open-loop control, this can quickly lead to a histogram that may result in incorrect conclusions. For example, if this behavior occurs, increased friction in various components of the control valve may be diagnosed.

[0004] Therefore, high histogram counts for large setpoint deviations when examining a valve that exhibits significant open-loop times and simultaneously experiences shifts in the lower end positions of the valve (e.g., due to contamination) falsely indicate a valve problem that does not actually exist.

[0005] Generally speaking, an end-position shift, i.e., the deviation of the actual valve position from the intended end position, influences the physical behavior of the control valve. In a positioner, various parameters and states are typically stored in the form of trend graphs and / or histograms. If the end position changes, this shift can alter the trend graphs and / or histograms, leading to misinterpretation.

[0006] German patent application DE 10 2020 113 437 B3 describes a method for detecting vibrations and / or shocks at a control valve. This method utilizes the position sensors already used to monitor the valve element position. The measured positions of the valve element are recorded and analyzed, with deviations from target positions indicating vibrations or shocks. This eliminates the need for additional, expensive vibration sensors.

[0007] German patent application DE 10 2018 116 048 A1 discloses a method for diagnosing the causes of changes in a control valve. The position of the actuator is recorded at least two different time intervals to determine an operating point or operating range for each interval. By comparing the operating points from the different time intervals, potential causes of changes, such as wear or process changes, can be diagnosed without the need for additional sensors.

[0008] German patent application DE 10 2008 062 292 presents a method for pressure-sensor-based diagnosis of the wear condition of a valve mechanism by determining static friction. For this purpose, the control pressure required to initiate the movement of the valve element after a standstill is measured. This pressure rise value is recorded multiple times, and the measured values ​​are stored and statistically analyzed to determine the likely influence of static friction as an indicator of the wear condition. Task

[0009] The object of the invention is to recognize such situations and to avoid resulting misdiagnoses. Solution

[0010] This problem is solved by the subject matter of the independent claim. Advantageous embodiments of the subject matter of the independent claim are identified in the dependent claims. The wording of all claims is hereby incorporated by reference into this description. The use of the singular is not intended to exclude the plural, and the reverse is also true unless otherwise disclosed.

[0011] The following section describes individual process steps in more detail. In a preferred embodiment of the invention, these steps are carried out in the specified order. However, the steps need not necessarily be performed in the specified order, and the process described may also include further, unmentioned steps.

[0012] To solve the problem, a method is proposed to avoid misdiagnosis for a control valve with an electropneumatic positioner for regulating the position of a valve element. The method comprises several steps.

[0013] First, the progression of the valve element's end position over time during operation is recorded. The end position of the valve element is typically defined as either the fully closed or the fully open state. Therefore, the end position cannot be recorded continuously, but only when the control valve is fully closed or fully open.

[0014] The deviation of the respective end position from an expected end position is then determined. An expected end position would be, for example, zero, 0%, full deflection, or 100% of the possible stroke of the valve element.

[0015] Regardless, at least one histogram and / or at least one trend graph is recorded during operation using the electropneumatic positioner. This histogram and / or trend graph represents the number of values ​​recorded over time during operation. The histogram and / or trend graph is selected from a group containing at least the following: ◯ a histogram for the control deviation of the actual value from the target value of the valve element position; a histogram for the position of the valve element; ◯ a trend graph for a stroke-pressure curve or pressure-stroke curve; ◯ a trend graph for the flow rate through the control valve as a function of the position of the valve element;

[0016] The trend graph for the stroke-pressure curve or pressure-stroke curve is usually represented in three dimensions.

[0017] The at least one recorded histogram and / or the at least one trend graph is corrected using the determined deviation of the respective end position.

[0018] Subsequently, a diagnosis for the control valve is created based on at least one corrected histogram and / or at least one trend graph.

[0019] This diagnosis will be output or saved.

[0020] If this is prompted by the issued diagnosis, at least one maintenance measure for the control valve will be initiated.

[0021] For position controllers that do not have enough memory to store all histograms and / or trend graphs with and without the influence of the end position shift, the correction of the histograms and / or trend graphs is performed afterwards.

[0022] The proposed method can prevent misdiagnoses in many cases. It allows for a determination of whether the shift in the control deviation histogram is due to the end position or a valve problem, such as high friction or pneumatic leakage. Correcting the valve position histogram for end position shifts allows for a better assessment of the valve design, as the end position shift no longer affects the valve position histogram.

[0023] The accuracy of the correction of the histogram and / or trend graph can be increased by taking into account the respective duration of time during which the valve element was in an end position or in a controlled position when correcting the at least one histogram and / or the at least one trend graph.

[0024] The end-position profile is acquired non-continuously, meaning the data for the end-position profile is not acquired continuously, but at irregular intervals, typically during the valve's operating cycles when it reaches its end positions and exhibits a change of at least ±0.25% from the last recorded end position. The graphs to be corrected are continuous histograms. The accuracy of the correction in this situation can be increased by considering temporal changes in the valve end positions, in addition to the recorded end positions, when correcting at least one histogram and / or at least one trend graph. This can be achieved, for example, by proportionally extrapolating the existing measured values ​​and their time intervals to the entire histogram recording period or by linear interpolation.By combining continuous and discontinuous data and considering the temporal changes in end positions, more accurate and reliable histograms can be generated. This methodology improves diagnostic capabilities and enables precise monitoring and maintenance of control valves.

[0025] To better inform the operators of process engineering plants or control valves, at least one additional indicator value can be output when correcting at least one histogram and / or at least one trend graph, which serves to check the validity of the correction and / or to warn of unlikely results.

[0026] The task is also performed by a computing unit that is suitable and configured to carry out the steps for correcting the at least one histogram and / or the at least one trend graph. These are the steps that begin with the step in which the at least one recorded histogram and / or the at least one trend graph is corrected using the determined deviation of the respective end position. Outsourcing these steps to an external computer, for example at the plant operator's site, on a server or edge device, or in the cloud, is sensible and advantageous if the computing power on the positioner is insufficient for the correction steps.

[0027] Furthermore, the computing unit may be suitable and configured to perform at least one of the additional steps of the procedure just described.

[0028] For the external calculation of the correction, an electropneumatic positioner for a control valve is required, which is suitable and configured to forward the data required to carry out the steps of the described procedure to the computing unit just described.

[0029] The problem is further solved by an electropneumatic positioner for a control valve, wherein the electropneumatic positioner comprises a computing unit and means that are suitable and equipped to perform the steps of the proposed method.

[0030] The task is also solved by a control valve with an electropneumatic positioner, as just described.

[0031] The task can also be solved by a process engineering plant with one of the described control valves.

[0032] Furthermore, the task can be solved by a computer program comprising commands that cause the described computing unit to perform the procedural steps for correcting the at least one histogram and / or the at least one trend graph.

[0033] Furthermore, the task can be solved by a computer program which includes commands that cause the described electropneumatic positioner and / or the described control valve to perform the described process steps.

[0034] Finally, a computer-readable medium can also solve the task, on which one of the computer programs just described is stored.

[0035] Further details and features will become apparent from the following description of preferred embodiments in conjunction with the figures. The respective features can be implemented individually or in combination. The possibilities for solving the problem are not limited to the embodiments shown.

[0036] The exemplary embodiments are shown schematically in the figures. Identical reference numbers in the individual figures denote identical or functionally equivalent elements, or elements corresponding to each other in terms of their functions. Specifically, the figures show: Fig. 1 an uncorrected deviation histogram; Fig. 2 an example of recorded end positions, predominantly different from zero; Fig. 3 the weighting of the recorded final positions during correction; Fig. 4 a probability distribution for the final position of the valve element; Fig. 5A a correction histogram required for the correction of the control deviation histogram to take into account recorded end positions without a probability distribution for the end position; Fig. 5B a correction histogram required for the correction of the control deviation histogram to take into account recorded end positions with probability distribution for the end position; Fig. 6 an intermediate step in the correction of the control deviation histogram after subtraction of the values ​​from recorded end positions in an open control loop situation; Fig. 7 a further intermediate step in the correction of the control deviation histogram according to the distribution of negative values; Fig. 8. A further intermediate step in the correction of the control deviation histogram after removing the remaining negative values; and Fig. 9 the corrected control deviation histogram. I. Correction of the histogram “deviation from the rule” using the “end position curve”

[0037] Many position controllers continue to record values ​​for the control deviation histogram even when closing completely. However, in this operating state, the position controller is not in control mode, but rather in an open-loop control situation. This distorts the histogram. Using the end-position profile and the recorded operating time in a closed-loop control situation, the control deviation histogram can be corrected so that it only reflects the time spent in control mode. An example of correcting the control deviation histogram is described below.

[0038] First, the deviation histogram is needed. An example is in Fig.Figure 1 shows the following. In addition to the counters or frequencies or relative frequencies (bin counts) of the individual intervals, the interval boundaries are also needed for the histogram. In this case, the interval boundaries are in Fig. 1 recognizable: [-infinity, -5], [-5, -1], etc.

[0039] In addition, the deviations of the valve's lower end position from the expected zero end position over time are used. The lower end positions of the example valve over time are shown in Fig. Figure 2 shows the point at which the histogram is read. The dashed vertical line marks the histogram readout time. All points up to the time of histogram readout are used.

[0040] Finally, the recorded operating hours and the operating hours in the closed-loop control situation, together with the derived proportion in the open-loop control situation, are required. This is shown in the following table for the example under consideration. Table 1 Operating hours 3040 days and 14:00:00 h Operating hours in a closed-loop control situation 231 days and 21:00:00 h Operating hours in an open control loop situation 2808 days and 17:00:00 h Proportion of operating hours in the open control loop situation 92,4 %

[0041] The proposed solution corrects a histogram using the available data described above. Although the correction proposed here yields satisfactory results, an exact and completely accurate histogram correction is impossible due to the inherent incompleteness of the data. Assumptions and estimates are always used, which can provide a good approximation of actual past events. This is described in detail in the following sections.

[0042] One might assume that all the necessary information is already available at the lower end positions. This is not entirely true, as only values ​​that have a certain minimum distance to the previous value are recorded.

[0043] The resolution of the positions is therefore quite coarse, and it is unknown how long a lower end position lasted because it is not recorded when the valve's setpoint changes from closed to another position. For this reason, it is necessary to estimate which lower end positions were active and for how long, using assumptions to substitute for the missing data.

[0044] In a first step, the time differences between the measured lower end positions are calculated. Assuming that the actual closing times are evenly distributed over time, each lower endpoint can be weighted by the time delta to the next point. For the last point, the time difference until the histogram readout is used as the time difference.

[0045] In addition to these weights for each point, the total time period covered by the known lower end positions is also calculated. This is important because only the time intervals between the individual measurements of the lower end positions are known, but not how the open-loop control times should be distributed among them. For this purpose, the proportion of operating hours in the open-loop control situation calculated above is used.

[0046] In Fig. Figure 3 is a simple example with four lower end positions. In this example, the estimated time periods in the open-loop control situation (thick lines) are used as weights for their starting points. The total time of the open-loop control situation, including time before the first recorded point, is divided among the points according to these weights.

[0047] As a first indicator of the correction quality, the time intervals covered by the recorded points are summed and compared to the total operating hours. If the points cover only a small portion of the total operating hours, the correction quality may not be as good as if the points cover the entire recorded time. This information can be used as a warning to the user or as an error indication.

[0048] The histogram can now be corrected using the previously gathered information. For each lower end position, it must first be calculated which interval (bin) it belongs to. While this could easily be solved by simply using the bin limits as thresholds, this would not account for the fact that new values ​​are only recorded if they differ from the last recorded value by a minimum amount. In practice, a threshold of + / - 0.25% of the valve position is chosen.

[0049] Therefore, for example, a triangular probability distribution is assumed for the final position (the integral of the overall distribution is 1), where the maximum probability lies at the recorded value. At the limits of the threshold, the probability is zero. This results in a triangular probability distribution that is used to assign the individual points to the intervals / bins.

[0050] Alternatively, one could assume a different probability distribution, such as a normal distribution.

[0051] An example is in Fig. Figure 4 shows the point in the middle representing the recorded lower end position, the dashed black line representing the threshold range, and the thick lines representing the probability distribution.

[0052] In Fig. 5A shows the histogram for the point to which no distribution is applied, while in Fig.Figure 5B shows the histogram for the same point to which the triangular distribution described above is applied. For clarity, the values ​​have been shifted to negative values.

[0053] This process is repeated for each available point, and the resulting bins are weighted and summed to obtain the overall correction histogram.

[0054] The correction histogram for the final position obtained in this way can be applied to the original input histogram by subtracting the correction histogram from the input histogram or adding it in the case of negative values.

[0055] In Fig.Figure 6 shows the resulting histogram. In this example, it can be seen that while a significant portion of the range from -1 to -0.5 was removed by the correction, some of it remains, and instead, a significant negative range between -0.5 and -0.2 is now present. Therefore, the following further steps are necessary.

[0056] With the preceding correction, there is no guarantee that all bin numbers are still positive; in fact, it is very likely that some bin numbers are now negative. To correct this inconsistent result, there are two options. The first option would be to discard the negative bin numbers and thus ignore them. The second option would be to redistribute the negative values ​​to other bins. While redistribution is permissible since it is not known exactly where correction is needed, any redistribution must be carried out with caution, as otherwise relevant information might be lost from the histogram.

[0057] The proposed solution is a weighted redistribution of all bin values. The largest negative field is iteratively redistributed to its neighboring fields. It is suggested to use one or two higher and one or two lower bins, if available. The redistribution can only be performed on bins with positive values, and the distribution between the bins should be proportional to the positive values ​​of the bins and inversely proportional to the distance from the bin. The inversely proportional weighting of the distance is based on the decreasing probability of a misassignment with increasing bin distance. The proportionality to the amplitude accounts for the more frequent occurrence of the more heavily weighted intervals and also avoids erroneous corrections of intervals into negative values.

[0058] There will be situations where the negative bins have no matching positive counterpart in the adjacent bins, and large numbers of bins are moved without leading to a helpful solution. To handle such situations, it is proposed to count all moved bin numbers and compare them to a threshold value at the end. If the number of bins is greater than the threshold, an error is reported, and the correction is considered invalid and / or a warning is issued to the user.

[0059] To return to the example described, in Fig. Figure 7 shows the result of the redistribution. The large negative area was redistributed to its neighbors, leaving a small negative area in the middle.

[0060] All remaining negative bin numbers could not be redistributed and are discarded in this step. In practice, the negative bins can simply be set to zero.

[0061] Here too, the sum of all discarded values ​​is compared to a threshold. If the threshold is exceeded, the correction is considered invalid because too large a portion of the histogram was discarded when separating negative values. A corresponding message can be displayed.

[0062] In Fig. Figure 8 shows the histogram after truncation. In this case, the discarded negative values ​​are small, so the correction can be considered valid.

[0063] After removing part of the original histogram and even discarding some counts entirely, the total histogram value may deviate significantly from the desired 100%. To compensate for this, the histogram is scaled to the intended value. It is suggested that the histogram always be scaled to 100%, disregarding the total value of the original histogram before the correction.

[0064] In the special case of a correction that removes almost 100% of the original histogram, which can only occur with valves with a very high idle time, rescaling would either be impossible (division by zero) or would lead to an increase in numerical noise (division by a small number). Therefore, the total value of the histogram before rescaling must be higher than a certain threshold. A small value such as 10 is suggested. -6This threshold can be used to set all values ​​to exactly zero if the sum of the histogram falls below this threshold. In such a case, an error message or warning can be issued.

[0065] In Fig. Figure 9 shows the successful scaling for the example valve presented here.

[0066] The proposed procedure returns the corrected histogram and five validity checks. While a corrected histogram will always be generated, its validity depends on these checks and whether they meet their criteria. The checks and their proposed values ​​are listed in Table 2.

[0067] An uncorrected histogram could incorrectly indicate friction or pneumatic leakage of the control valve. I.1 Alternative Approaches I.1.1 Determination of the assumed lower end positions and their duration

[0068] There are many possible assumptions that could be made to get from the lower end positions to weighted time durations for each point.

[0069] One assumption could be that the lower end position can be modeled as a constant. Based on this assumption, an average of all lower end positions was calculated (mean, median, and time-weighted average were tested). As expected, this works well for a small number of points, but not for valves where the lower end positions vary over time or exhibit a large spread across multiple bins.

[0070] Another assumption could be that the points are treated individually, but all take the same amount of time. In this case, the total time of the open-loop control situation is divided by the number of recorded lower end positions, which gives the duration for each point. As expected, this works well for evenly distributed lower end positions, but misrepresents valves where the lower end positions are concentrated at certain times.

[0071] As a variation of the proposed method, it was also tested to not weight each point by the time until the next recorded lower end position, but rather to distribute each time interval between the two points in the vicinity evenly. The advantage could be that the points are weighted more evenly, but this introduces a substantive error into the calculation. The results were therefore equally good for some systems, but in many cases this method led to worse results than the preferred method.

[0072] Finally, a more complex approach was tested, intended to approximate an optimal distribution algorithm. First, the values ​​that could potentially be counted for each bin are summed. Then, the number of available correction possibilities for each bin is divided by each bin value. The result is a measure of how much potential correction is available for each bin. The bin with the lowest possible correction ratio is then selected and corrected first. Ideally, this ensures that even bin values ​​only partially reached by a given point can be fully corrected if necessary. The results for this method were good, but overall not significantly better and in some cases even worse than the proposed method. Therefore, the increased effort and complexity are not justified. I.1.2 Redistribution of negative bin values

[0073] There are several ways to redistribute negative bin values. The first parameter that can be set is the number of adjacent bins across which the values ​​could be distributed. The proposed method uses one or two adjacent bins in each direction, but any other number is conceivable, including all bins. In testing, using only one bin proved most effective in distributing negative bins nearby while preventing the data from being carried too far across the histogram.

[0074] Initial tests also evaluated simpler methods. One method, for example, could involve distributing each negative field evenly across the two adjacent fields and repeating this process iteratively until no more values ​​can be shifted. II. Correction of the valve position histogram using the end position profile

[0075] Another possibility would be to check whether the valve is leaving its control range due to an end-position shift. For example, many positioners can only reach setpoints greater than 1%. If there is an end-position shift of, say, 5%, this corresponds in reality to a valve opening of 6%.

[0076] This is also useful because the K v -value or the K v The curve of a valve or a valve seat-valve cone combination is defined as a function of the stroke. In the case of an end-position shift, the valve position and the current K are correct. v The value no longer matches. After correction, the new histogram includes the valve position, which is again based on the K v -curve is aligned. This can be used to assess the interpretation.

[0077] The valve position can be used to infer the design of the histogram. A significant negative end-position shift means the valve needs to open less to achieve the desired flow rate than when new, potentially leading to a misinterpretation of the design due to wear. III. Correction of the pressure-stroke curve using the end position profile

[0078] The pressure-stroke curve can be used, for example, to detect spring breakage or pneumatic leakage. A disturbance variable in this curve is the pressure difference Δp of the medium, which drops between the seat and the cone. If the end position changes, the pressure difference for a given valve position changes under constant process conditions. By taking the end position shift into account, this effect can be factored out. IV. Correction of the control deviation histogram and the valve position using the end position profile

[0079] If the control deviation histogram is correlated with the valve position and the end position profile in several readouts, a change in friction can be inferred if only the control deviation histogram changes.

[0080] In a continuously operating system, if the valve position remains unchanged or only slightly altered, it can be assumed that the process has not changed or has changed very little, thus ruling out an influence of the pressure differential Δp of the medium on the control deviation. If the control deviation now changes, this can be attributed to a change in the control valve, such as a change in the friction conditions at the valve. Glossary: ​​Pressure-stroke curve or stroke-pressure curve

[0081] The stroke-pressure curve of a control valve with an electropneumatic positioner for regulating the position of a valve element represents the stroke – i.e., the position of the valve element – ​​as a function of the pressure in the pneumatic actuator of the control valve. A pressure-stroke curve represents the opposite. electropneumatic positioner

[0082] An electropneumatic positioner is a positioner for a pneumatic actuator that receives an electrical signal as its input and converts it into compressed air with a corresponding, regulated pressure at the output. The position of a component of the device driven by the actuator, e.g., a valve element, is taken into account as the actual value during control. closed control loop situation

[0083] A closed-loop control situation is one in which the control loop actually regulates. In such a situation, it attempts, for example, to move a valve element to the position corresponding to the setpoint. In a closed-loop control situation, continuous correction is used to try to bring the actual value and the setpoint into alignment. IP converter

[0084] An IP converter converts an electrical signal (I) into pneumatic pressure. K v -Value

[0085] The K v The K-value of a valve is a measure of the flow of fluids through the valve and is used to describe its flow capacity. Technically, the K-value indicates v This value indicates the amount of water, in cubic meters, that can flow through the valve at a pressure loss of 1 bar per hour. It is expressed in m 3 / h measured. open control loop situation

[0086] An open-loop control situation occurs when the control process transitions into direct control, for example, when a boundary value is approached. Thus, if, for instance, a valve element is moved to its fully open or fully closed position, control ceases. Only the extreme position is being controlled. Control deviation, target value deviation

[0087] In control engineering, this refers to the deviation between the target and actual values. Table 2 Description and definition Acceptance threshold Percentage of total operating hours that must be covered by the recorded lower end positions for the result to be considered valid. > 60% Percentage of negative bins that are not redistributed (and must be discarded) for the correction to still be considered valid < 20 % Percentage of redistributed (shifted across the bin boundaries) negative values ​​so that the correction is still considered valid < 40 % Ratio of the open control loop situation (hours in the open control loop / total operating hours) so that the correction is still considered valid < 80 % Number of recorded lower end positions for correction to still be considered valid > 8

Claims

[1] Method for avoiding misdiagnosis for a control valve with an electropneumatic positioner for controlling the position of a valve element, comprising the following steps: 1.1 Recording the progression of an end position of the valve element over time during operation; 1.2 Determining the deviation of the respective final position from an expected final position; 1.3 Acquisition of at least one histogram and / or at least one trend graph using the electropneumatic positioner during operation, wherein the at least one histogram represents the number of values ​​determined over time that were recorded during operation, and wherein the at least one histogram and / or the at least one trend graph is selected from a group that includes at least the following histograms and / or trend graphs: 1.3.1 a histogram for the control deviation of the actual value from the target value of the position of the valve element; 1.3.2 a histogram for the position of the valve element; 1.3.3 a trend graph for a stroke-pressure curve; 1.3.4 a trend graph for the flow through the control valve as a function of the position of the valve element; 1.4 Correction of at least one histogram and / or at least one trend graph using the determined deviation of the respective end position; 1.5 Creating a diagnosis for the control valve based on at least one corrected histogram and / or at least one corrected trend graph; 1.6 Output and / or save the diagnosis; and 1.7 Initiate at least one maintenance measure for the control valve if this is prompted by the diagnosis. [2] Method according to the preceding claim, characterized by, that when correcting the at least one histogram and / or the at least one trend graph, the respective duration of time during which the valve element was in an end position or in a regulated position is taken into account. [3] Method according to any one of the preceding claims, characterized by , that when correcting the at least one histogram and / or the at least one trend graph, temporal changes in the end positions of the valve element are taken into account in addition to the recorded end positions of the valve element. [4] Method according to any one of the preceding claims, characterized by , that when correcting at least one histogram and / or at least one trend graph, at least one additional indicator value is output, which serves to check the validity of the correction and / or to warn of unlikely results. [5] Computing unit suitable and configured to perform steps 1.4 to 1.7 according to claim 1. [6] Computing unit according to the immediately preceding claim, characterized by that the computing unit is suitable and equipped to perform at least one of the additional steps of process claims 2 to 4. [7] Electropneumatic positioner for a control valve, which is suitable and configured to forward the data required to carry out the steps of the method according to any one of claims 1 to 4 to the computing unit according to claim 5. [8] Electropneumatic positioner for a control valve, wherein the electropneumatic positioner comprises a computing unit and means which are suitable and configured to perform the steps of the method according to any of the preceding method claims. [9] Control valve with an electropneumatic positioner according to claim 7 or 8. [10] Process engineering plant with a control valve according to the immediately preceding claim. [11] Computer program comprising instructions that cause the computing unit according to claim 5 to perform the process steps 1.4 to 1.7 according to claim 1. [12] Computer program comprising commands that cause the electropneumatic positioner according to claim 8 and / or the control valve according to claim 9 to perform the method steps according to any one of the preceding method claims 1 to 4. [13] Computer-readable medium on which the computer program according to claim 11 or 12 is stored.

Images