DEFINING ABORT CRITERIA FOR A PARTICIPATION LIFT TEST ON A FLUIDLY OPERATED SAFETY VALVE AND DETERMINING THE FUNCTIONALITY OF A FLUIDLY OPERATED SAFETY VALVE

Description

Field of invention

[0001] The invention relates to a method for defining termination criteria for a partial-stroke test on a fluidically actuated safety valve. These termination criteria determine whether such a test is considered passed or failed and therefore have a decisive influence on the maintenance situation of fluidically actuated safety valves.

[0002] Furthermore, the invention relates to a method for determining the functionality of a fluidically driven safety valve.

[0003] If a valve element remains in one position for an extended period, deposits, corrosion, and / or adhesion can significantly increase static friction and thus the breakaway force required to move the valve element. The drive energy may then no longer be sufficient to move the valve element reliably.

[0004] For safety valves, it must be ensured that the drive energy is sufficient under all circumstances to move the valve element into a safety position. To guarantee that the valve element can be moved against increased frictional forces even after a long period of inactivity, a partial stroke test is therefore performed regularly. This test verifies the basic mobility of the valve element.

[0005] The typical course of a partial stroke test can be described using the following examples: Fig. 1 This diagram illustrates the relative stroke of the valve element and the pressure over time during a partial-stroke test on a safety valve. The safety valve has a single-acting pneumatic actuator with spring return. The safety position is normally closed (de-energized). The target value 210 shows the ideal, frictionless stroke profile when closing the valve during uniform venting. The pressure is reduced, and the spring forces of the actuator are released, moving the valve element towards the closed position. The actual pressure profile 220 initially shows a significant drop in pressure before the stroke (actual stroke) 230 changes. This shows the valve element breaking free from its open position, in which it may have already been stuck, i.e., subject to static friction. The difference between the initial maximum pressure and the pressure at the moment of breakaway is referred to as the breakaway pressure 240.The breakaway pressure is the pressure or force sufficient to overcome static friction and release the valve. The pressure in the actuator fluid at the moment of breakaway remains as a reserve (250) if it is greater than 0 bar. Otherwise, the safety valve can no longer function. After breakaway, the pressure is readjusted, i.e., increased again, to prevent or reduce overshoot. After overcoming static friction and the accelerated stroke movement, the actuator is vented smoothly (slow pressure reduction) until 90% of the stroke is reached. The actual stroke curve (230) runs approximately linearly and parallel to the target stroke curve (210) in the area of ​​sliding friction, shifted parallel by the amount of sliding friction. After reaching 90%, the pressure in the actuator is increased again so that the valve element fully opens against the spring forces. The partial stroke test is complete.

[0006] By eliminating time and applying pressure 220 over stroke 230, the stroke-pressure curve of the Fig. 2 The closing pressure reserve 320 is the pressure reserve that remains after the safety valve has closed. The closing pressure reserve 320, which is established under constant movement (sliding friction), can be determined by... Fig. 2 to be determined. For this purpose, the stroke-pressure curve is calculated according to Fig. 2 extrapolated further until the closed position (stroke = 0). State of the art

[0007] An actuating device for an on / off valve with pneumatic drive, in which a positioner is provided which includes a routine for performing a partial stroke test, is described in publication WO 2009 / 013205 A1.

[0008] Publication DE 10 2011 052 901 A1 also describes how the drive energy for carrying out a partial stroke test can be controlled in such a way as to prevent excessive overshoot after overcoming static friction.

[0009] During a partial stroke test, a stroke-pressure curve is recorded. This curve can be used, for example, to determine the pressure reserve of an actuator. Such a procedure is described in publication DE 10 2018 103 324 B3. The pressure reserve can indicate whether the valve will close reliably in a safety situation. Document US 2002 / 0145515 A1 discloses a system for manually performing an emergency shutdown test and collecting data in a process control environment.

[0010] To define the criteria for failing a partial stroke test, the following procedure is commonly used: A reference curve is recorded for a fully functional control valve. Experience has established the permissible deviations of a stroke-time curve or stroke-pressure curve that still ensure safe operation. These values ​​are permanently stored as termination limits and apply to the entire stroke-pressure curve and to all such valves.

[0011] If the position control of the control valve is a proportional control (also known as P-control), the control deviation (difference from the target position) and the pressure serve as termination criteria. During the partial-stroke test, the pressure chamber is slowly vented to determine whether the spring force is sufficient to release or move the valve element as intended. This requires a pressure sensor or pressure readings based on the IP converter signal. If, for example, the control deviation is greater than 3-5% or the pressure in the valve actuator is less than 2.5 bar, the test is considered failed. This means that the final test position must be achievable at a pressure greater than 2.5 bar to pass the test. Furthermore, the final position must not deviate by more than 3-5% from the final position in the reference curve. These fixed termination criteria are the same for all test phases and at all times.

[0012] These fixed abort limit criteria mean that, in the event of an abort, it is unclear how significant the change is compared to the previously measured, functionally safe state (during the last successful partial-stroke test). Changes in physical behavior caused by disturbances such as downtime, medium pressure, and / or medium temperature, but which do not fundamentally prevent the safety valve from functioning, are also not taken into account. Task

[0013] The object of the invention is to provide a method that provides better termination criteria for a partial stroke test, as well as a method by which the functionality of a fluidically driven safety valve can be determined more reliably. Solution

[0014] This problem is solved by the subject matter of independent claim 1. Advantageous embodiments of the subject matter of the independent claim are characterized in the dependent claims. The wording of all claims is hereby incorporated by reference into this description.

[0015] The use of the singular should not exclude the plural, and the same applies in reverse, unless otherwise stated.

[0016] The following section describes individual process steps in more detail. These steps do not necessarily have to be carried out in the order given, and the process described may also include further, unmentioned steps.

[0017] To solve this problem, a method is proposed for defining termination criteria for a partial-stroke test on a fluidically actuated safety valve with a valve element and a spring return. The valve element is to be moved into a safety position by the spring return when the pressure of the actuating fluid drops completely, with the pressure of the actuating fluid acting against the spring. The safety valve has means for determining the pressure of the actuating fluid and means for determining the position of the valve element. The method comprises the following steps: a) A first partial stroke test is performed on the safety valve by varying the pressure in the actuator fluid. This partial stroke test is carried out when the safety valve is operational, e.g., during commissioning. b) At least the position of the valve element and the pressure in the actuator fluid are recorded during the first partial stroke test. c) A first relationship is derived from the recordings, relating the position of the valve element and / or the time and / or the pressure of the actuator fluid and / or another parameter recorded during the first partial stroke test. This could, for example, be a stroke-pressure curve. d) This relationship is defined and stored as the reference curve for this safety valve in its operational state. e) At least one second relationship is defined, which has a predetermined distance from the reference curve.This at least one second relation can, for example, be a tolerance band around the stroke-pressure curve. f) The following is defined as a termination criterion for a subsequently performed, repeated partial stroke test, e.g., during future regular inspections of the same safety valve: If the repeated partial stroke test is performed on the same valve, the same data are recorded, and a third relation is derived from the data in the same way as for the reference curve, the repeated partial stroke test is considered failed if the third relation has a greater distance from the reference curve than the second relation at at least one point.

[0018] In this way, termination criteria for future partial-stroke tests can be defined for each specific safety valve, flexibly taking the valve's characteristics into account. Whether the valve passes the partial-stroke test or not therefore depends more on changes in the valve's maintenance condition and less on how much this particular valve differs from some ideal valve.

[0019] The safety valve has a position control system that is used in the partial stroke test. This position control system is a proportional control system with a control deviation. In the claimed embodiment of the method, the first relationship used is the sum of the target curve for the position control of the valve element and the control deviation.

[0020] A particularly simple design of the termination criteria for the partial stroke test is achieved if the second relation has a fixed distance of, for example, 2% from the reference curve. For the sake of simplicity, this distance can apply both upwards and downwards.

[0021] The termination criteria are more appropriate to the specific situation of the valve under consideration if the second relation has different distances to the reference curve for different positions of the valve element.

[0022] It is particularly useful if the different distances of the second relation to the reference curve apply to different regions of the valve cone's movement during a partial-stroke test. Typically, the curve is therefore divided into four regions: the breakaway region, dominated by overcoming static friction; the region dominated by sliding friction, in which the valve element moves without further special features; the region of reversal of movement, when the valve element has reached 10% of its closing travel, which is determined by hysteresis; and the region, again dominated by sliding friction, of the return of the valve element to the fully open position.

[0023] This subdivision is useful because in certain areas, larger deviations from the ideal course of a partial-stroke test can occur without compromising the functionality of the safety valve, while in other areas, such large deviations must necessarily lead to termination and failure. For example, it is quite possible that a relatively large pressure drop is required during the breakaway phase, causing the valve element to lag significantly behind the reference curve. This is not critical if the valve element then approaches the reference curve more closely during the subsequent steps of the test. Therefore, a larger deviation can be permitted in a typical application during the breakaway phase.

[0024] For certain valve types or system configurations, it can also be advantageous to define the different distances and ranges in such a way as to account for stick-slip effects. These typically occur immediately after the start of movement and manifest themselves as a stroke over time that resembles a step function. When the valve element sticks, its position does not change. Then, there is a sudden correction of the position (slip). The magnitude of these steps can be used as a criterion for designing the at least one second relation. In the area where these steps occur, the distance of the at least one second relation from the reference curve should be made larger. In particular, it is conceivable that the tolerance range above the reference curve must be chosen to be larger than the tolerance range below the reference curve under these conditions.

[0025] Preferably, the second relationship exhibits different distances to the reference curve both above and below. This is particularly advantageous if the reference curve is a stroke-pressure curve. In this case, larger upward deviations than downward deviations are typically expected compared to the reference curve, since normal wear requires a greater reduction in the actuator pressure than a valve in ideal condition.

[0026] In a particularly preferred refinement of the method, the distances of the second relation to the reference curve are dynamically determined depending on the current conditions at the valve and / or the system at the time the partial-stroke test is performed. This allows for the consideration of influences such as downtime, temperature, and process pressure, which are unrelated or only minimally related to the fundamental functionality of the valve. For example, it is possible to allow larger deviations after a long period of inactivity of the valve and, correspondingly, to apply stricter criteria in cases where the valve was moved shortly before the partial-stroke test.

[0027] It is also advantageous if the partial stroke test is considered failed if the pressure in the drive fluid falls below a predetermined value. It is generally possible to specify limits for the breakaway pressure, the pressure at the reversal point of the partial stroke test movement, and / or the pressure during the entire test procedure.

[0028] To solve the problem, a method for determining the functionality of a fluidically actuated safety valve with a valve element and a spring return is proposed. The valve element is to be moved into a safety position by the spring return when the pressure of the actuating fluid drops completely; the pressure of the actuating fluid acts against the spring. The safety valve has means for determining the pressure of the actuating fluid and means for determining the position of the valve element. The method comprises the following steps: A partial stroke test is performed on the safety valve. A procedure as described above was previously performed on the same safety valve. Based on the termination criteria defined for this valve using the procedure described above, a decision is made as to whether the valve has passed the partial stroke test or not, and whether it can continue to be classified as functional.

[0029] This method allows for a more accurate determination of whether a safety valve is still functional. In particular, the result is more tailored to the specific situation of the valve under consideration, and the criterion violated in the event of failure may even provide clues as to the nature of the valve's impairment. Both the valve's maintenance cycles and the scope of maintenance work can thus be optimized.

[0030] The problem is further solved by the fact that, in a procedure like the one described above, the procedure steps are formulated as program code, which allows the procedure to run on at least one computer.

[0031] Furthermore, the problem is solved by a computer program which includes executable instructions which, when executed on a computing unit, a microcontroller, DSP, FPGA or computer or on a plurality of them in a network, executes the method according to one of the preceding method claims.

[0032] Furthermore, the problem is solved by a computer program with program code means to carry out the method according to the invention in one of its embodiments when the program is executed on a computing unit, a microcontroller, DSP, FPGA or computer, or on a plurality of these in a network. In particular, the program code means can be instructions stored on a computer-readable data carrier.

[0033] Furthermore, the problem is solved by a data carrier on which a data structure is stored which, after being loaded into a working and / or main memory of a computing unit, a microcontroller, DSPs, FPGAs or computer or a plurality of them in a network, can execute the inventive method in one of its embodiments.

[0034] The problem is also solved by a computer program product with program code stored on a machine-readable medium to carry out the method according to the invention in one of its embodiments when the program is executed on a computing unit, a microcontroller, DSP, FPGA, or computer, or on a plurality of these in a network. A computer program product is understood here to be the program as a marketable product. It can, in principle, exist in any form, for example, on paper or a computer-readable data carrier, and can, in particular, be distributed via a data transmission network.

[0035] Finally, the problem is solved by a modulated data signal which contains instructions executable by a computing unit, a microcontroller, DSP, FPGA or computer or by a plurality of them in a network to execute the method according to the invention in one of its embodiments.

[0036] The computer system for executing the method can be a single computer or microcontroller, DSPs, or FPGAs, as well as a network of microcontrollers, DSPs, FPGAs, or computers, for example, a closed, in-house network, or computers connected via the internet. Furthermore, the computer system can be implemented as a client-server configuration, with parts of the invention running on the server and others on a client.

[0037] 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. For example, range specifications always include all intermediate values ​​not explicitly stated and all conceivable sub-intervals.

[0038] Some 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 A stroke-time and the associated pressure-time curve during a partial stroke test on a typical safety valve; Fig. 2 A corresponding stroke-pressure curve; Fig. 3 Stroke-time curves for two possible profiles during a partial stroke test, including the associated setpoint curve; Fig. 4 Tolerance bands for the stroke-time curves from Fig. 3 ; Fig. 5 Range-dependent tolerance bands for the stroke-time curves from Fig. 3 ; and Fig. 6 an exemplary stroke-time curve from a partial stroke test on a control flap.

[0039] Fig. 3 This shows the stroke-time curves from typical partial-stroke tests of control valves. The target curve (dashed line) slowly reduces the valve stroke from 100% (fully open) to 90%, before fully opening the valve again. The return to the open position can be faster. Test 1 (dotted line) and Test 2 (solid line) represent possible real-world curves for such a partial-stroke test. In Test 2, the breakaway torque is higher, requiring the actuator to exert greater force. This causes the valve element to oscillate when it begins moving until the control system intervenes. Test 2 also shows a greater overall deviation from the target curve.

[0040] To solve the problem underlying the invention, the termination criteria for a partial stroke test must be adapted so that they fit the specific situation of, for example, Test 1 or Test 2. For this purpose, a reference curve is recorded in the new or technically sound condition of the valve, and the termination criteria for this valve are determined based on this reference curve. In the case of purely proportional control, as is generally the case with position controllers for valves, the control deviation (difference between the target and actual position of the valve element) can be considered. A tolerance (typically 2%) is then added to the maximum control deviation (e.g., 1% or 4%) that occurred during the reference measurement, resulting in a permissible maximum control deviation (e.g., 3% or 6%) as the termination criteria. These general conditions form tolerance bands around the target value curve in a stroke-time diagram.Such tolerance bands are exemplified in . Fig. 4 As shown. Based on test 2 (as a reference case), a wider tolerance band results than based on test 1, since test 2 already shows a larger control deviation, which represents the still functional normal case for the valve.

[0041] It is of course possible to design these criteria for upward and downward deviations differently (not shown).

[0042] A similar procedure can be applied to the pressure in the safety valve's actuator system. If a predetermined pressure value is undershot, a partial-stroke test is aborted. For example, a minimum actuator pressure of 1900 mbar could be reached in Test 1 and a minimum actuator pressure of 1500 mbar in Test 2. Assuming this corresponds to the fully functional state of the respective valve, a pressure cut-off limit of 1500 mbar can be defined for the valve in Test 1 and a pressure cut-off limit of 1100 mbar for the valve in Test 2, assuming a maximum permissible pressure deviation of 400 mbar from the normal state. The pressure condition is therefore significantly stricter for the valve in Test 1 than for the valve in Test 2.

[0043] To obtain more situation-specific termination criteria for the partial stroke test, these criteria can also be designed to be area-dependent. This allows for statements about the precise test procedure and any changes to the fittings, enabling a differentiated diagnosis. For example, different termination criteria apply in the breakaway phase than in the continuous operation phase. A total of four or five areas can be defined: breakaway torque, sliding friction, hysteresis during direction reversal, sliding friction during retraction, and, if applicable, stick-slip effect. This is described in Fig. 5 The target curve is shown again as a dashed line. For clarity, only the lower tolerance bands of the termination criteria are shown. The limits for the valve in Test 1 are drawn as dotted lines (range of sliding friction, forward and backward movement) and as fine dashed lines (range of breakaway and hysteresis during reversal of motion), while the limits for the valve in Test 2 are drawn as coarse dashed lines (range of sliding friction, forward and backward movement) and solid lines (range of breakaway and hysteresis during reversal of motion).

[0044] Breakaway torque is defined as the point at which static friction is just overcome and the valve element begins to move. This marks the transition to sliding friction. In a typical valve, for example, deposits between the valve stem and the sealing seat, or increased friction within the valve assembly, can increase the breakaway torque over time. This can even lead to the partial stroke test being aborted if, for example, the control deviation tolerance is exceeded or the minimum pressure is undershot. A test abort in this range can therefore indicate a change in static friction due to permanent alterations in the mechanics of the assembly (e.g., altered packing friction or changes that only occur upon initial movement after a prolonged period of inactivity). This could, for example, be due to oil hardening. Appropriate maintenance measures can then be implemented.

[0045] Sliding friction refers to constant friction during sliding processes that is dependent on the direction of motion. A constant frictional force always occurs, which opposes the relative motion of two contacting surfaces and is independent of the speed of the relative motion. In a valve, packings (e.g., made of graphite or PTFE) used to protect against penetrating dirt or against the acting pressure of liquids or vapors result in relatively high sliding friction. This is because the packings are compressed by screws or springs and thus have a large contact area. A test interruption in this area can indicate changes in sliding friction or changes in the spring constant of the actuating springs.

[0046] Hysteresis here refers either to the difference in setpoint change (after a setpoint reversal) until a valve position changes, or to the pressure change required after a setpoint change until the valve moves. The primary cause of this hysteresis is mechanical (static) friction within the valve (especially in ball valves, for example) and / or the influence of flow forces (especially in butterfly valves, for example). If a test is aborted within the hysteresis range, it can be concluded that the static friction has been permanently altered due to mechanical changes in the valve structure (e.g., altered packing friction). By considering the changed friction in the breakaway torque range and within the hysteresis range, the torque that may have occurred, for example, during a period of inactivity, can be determined.

[0047] Sliding friction during return: Once the hysteresis is overcome, the valve element transitions from static friction to sliding friction. When returning to the starting position, the movement is usually faster to minimize the duration of the partial stroke test.

[0048] These areas are in Fig. 5 illustrated, in particular the different formation of the corresponding tolerance bands for the valves of test 1 and test 2.

[0049] When the static friction between two solids is greater than the sliding friction, the stick-slip effect can occur. Due to the seals on the piston rod and piston, pneumatic or hydraulic cylinders exhibit comparatively strong static and sliding friction. As soon as the piston starts moving, the air expands. Therefore, the effect of the stick-slip effect is particularly pronounced in slow movements that can only be achieved with the aid of small differential pressures. The stick-slip effect also occurs in dynamically loaded seals where high pressure loads, higher temperatures, and longer periods of inactivity make it difficult for the seal to detach from its seat. If a test is terminated based on the termination criterion of positional deviation or a pressure drop in this range, a change in the ratio of sliding to static friction can be inferred.The area where stick-slip effects occur is not shown here - typically these occur shortly after the valve element begins to move.

[0050] In Fig. 6 The typical stroke-time curve of a partial stroke test for a butterfly safety valve is shown (solid line), again together with the corresponding setpoint curve (dashed line). It can be seen that basically the same effects occur here as with a control valve, and that the procedure according to the invention is therefore also useful in this case.

[0051] Furthermore, it is advantageous to dynamically adjust the abort criteria based on the current process conditions. For example, the current downtime, medium temperature, or process pressure can be factored into the abort criteria for a partial stroke test. For instance, the criteria for breaking the test after a longer downtime can be more lenient, while they must be stricter if the valve was only recently moved, as a deviation in this case could more readily indicate a valve defect, which would necessitate appropriate maintenance. glossary Fluidic actuator of a valve

[0052] A valve is referred to as a fluidically driven valve when the valve's actuator rod is moved by a diaphragm that is pressurized and thereby positioned by a fluid, typically compressed air. Stroke-pressure curve

[0053] The stroke-pressure curve of a valve or a partial stroke test indicates the position of the valve element, in other words: the stroke, as a function of the pressure in the drive fluid of the fluidically driven valve. Partial Stroke Test: See partial stroke test PST: see partial stroke test relation

[0054] Here, a relation is understood as a mathematical relationship that, unlike a function (or mapping), is not bijective. Of particular interest are relations that are left-total. This means that in a relation between sets A and B, for every element in A, there exists at least one element in B, but not necessarily exactly one; there may well be several. Safety valve

[0055] Safety valves, as used here, are actuators with an on / off function and safety-relevant applications. Actuators consist of an actuator – typically fluidic – and a movable valve element and serve to regulate fluid flow. The valves can be either rotary or poppet valves. Single-acting pneumatic actuators are generally used in safety-relevant valves. These actuators, which are spring-loaded on one side, move independently to a safe position when the actuator is vented, i.e., when the compressed air escapes from the actuator chamber. This occurs, for example, when a flow-pressure (I / P) converter or a solenoid valve is no longer energized.

[0056] In safety valves, the safety valve is often open during normal operation, and closes automatically in the event of a fault (e.g., power failure). The compressed air always acts against the spring force. If the actuator is then vented, the valve begins to close because the spring forces are released. In the stroke-pressure curve, this configuration causes the valve to break free against static friction when the pressure in the actuator is reduced.

[0057] The safety position can also be open when de-energized (drive vented) and closed when energized (drive ventilated). Stick-slip effects

[0058] The stick-slip effect (from the English words "stick" and "slip"), also known as the static friction effect or (self-excited) frictional vibration, describes the jerky sliding motion of solid bodies moving against each other, i.e., the periodically alternating sticking and sliding. Well-known examples include chalk on a blackboard, creaking doors, squeaking brakes, rattling windshield wipers, and a wet fingertip stuttering across a latex balloon or setting the rim of a drinking glass into vibration. Partial stroke test ( Partial Stroke Test, PST)

[0059] To ensure the safe operation of a valve, the actuator's movement is tested regularly or cyclically. During these tests, the valve should not move completely into its safety position to avoid disrupting ongoing operations. In a partial-stroke test, the actuator is moved only as far as necessary to ensure it travels a portion of its range without significantly impacting the system's processes. This test also verifies whether the actuator can break free from its position. After the partial-stroke test, the actuator returns to its initial position. This test verifies the actuator's basic mobility. Valve member

[0060] The valve element is the component that closes the valve when it is pressed onto the valve seat. Reference sign

[0061] 210 Target value of stroke 220 Actual value of pressure 230 Actual value of stroke 240 Breakaway pressure 250 Breakaway pressure reserve 310 Breakaway reserve 320 Closing pressure reserve 330 Two points for determining the interpolation line 340 Interpolation line 350 Pressure value at stroke = 0 360 90% of stroke, end of PST cited literature cited patent literature

[0062] WO 2009 / 013205 A1 DE 10 2011 052 901 A1 DE 10 2018 103 324 B3 US2002 / 0145515

Claims

1. Method for establishing termination criteria for a partial stroke test of a fluidically driven safety valve having a valve member and a spring return; wherein the valve member is intended to be moved into a safety position by the spring return in case of a complete loss of pressure of the drive fluid; wherein the safety valve has means for determining the pressure of the drive fluid; wherein the pressure of the drive fluid acts against the spring; wherein the safety valve has means for determining the position of the valve member; wherein the valve has a position control which is used during the partial stroke test, and wherein said position control is a proportional control with a control deviation, wherein the method comprises the following steps: a first partial stroke test of the safety valve is performed by variation of the pressure in the drive fluid; wherein the first partial stroke test is performed at a point in time at which the safety valve is operable; at least the position of the valve member is recorded during the first partial stroke test; a first relation is derived from the recorded data, which puts the position of the valve member and / or the time and / or the pressure of the drive fluid and / or a further parameter recorded during the first partial stroke test in relation to one another; wherein the sum of the target curve for the position control of the valve member and the control deviation is used as first relation. this relation is defined as reference curve for this safety valve while operable and saved; at least one second relation having a predetermined distance from the reference curve is defined; the following is established as termination criterion for a repeated partial stroke test of the same valve to be performed at a later time: if the repeated partial stroke test is performed at the same valve; and if the same data is recorded thereby; and if a third relation is determined from said data in the same manner as the reference curve, then the repeated partial stroke test is considered failed if the third relation has a greater distance to the reference curve than the second relation at at least one point.

2. Method according to the preceding claim, characterized in that the at least one second relation has a distance of 2% to the reference curve; wherein this distance applies upwards and downwards.

3. Method according to claim 1, characterized in that the at least one second relation has different distances to the reference curve for different positions of the valve member.

4. Method according to the immediately preceding claim, characterized in that the different distances of the at least one second relation to the reference curve respectively apply to different ranges of the movement of the valve member during a partial stroke test.

5. Method according to the immediately preceding claim, characterized in that the different distances and ranges are determined such that stick-slip effects are taken into account.

6. Method according to claim 1, characterized in that the at least one second relation has different respective upward and downward distances to the reference curve.

7. Method according to any one of the preceding claims, characterized in that the distances of the second relation to the reference curve are dynamically determined depending on current conditions at the valve and / or the plant at the time at which the partial stroke test is performed.

8. Method according to any one of the preceding claims, characterized in that the partial stroke test is additionally considered failed if the pressure in the drive fluid falls below a predetermined value.

9. Method for determining the operability of a fluidically driven safety valve having a valve member and a spring return; wherein the valve member is intended to be moved into a safety position by the spring return in case of a complete loss of pressure of the drive fluid; wherein the safety valve has means for determining the pressure of the drive fluid; wherein the pressure of the drive fluid acts against the spring; wherein the safety valve has means for determining the position of the valve member; wherein the method comprises the following steps: a partial stroke test is performed at the safety valve; a method according to one of the preceding claims was performed on the same safety valve at a previous point in time; based on the termination criteria defined for this valve by this method according to any one of the preceding claims, it is decided whether or not the valve has passed the partial stroke test and whether or not it is still to be considered operable.

10. Method according to any one of the preceding claims, wherein the method steps are formulated as program code, with which the method is performable on at least one computer.

11. Computer program, characterized in that it comprises executable instructions which perform the method according to any one of the preceding method claims 1 to 10 when executed on a processing unit, a microcontroller, DSP, FPGA or computer or a plurality thereof in a network.

12. Computer program having program code means for performing a method according to any one of the preceding method claims 1 to 10 if the computer program is executed on a processing unit, a microcontroller, DSP, FPGA or computer or a plurality thereof in a network.

13. Data storage medium, on which a data structure is stored that performs the method according to any one of the preceding method claims 1 to 10 after being loaded into a working and / or main memory of a processing unit, a microcontroller, DSP, FPGA or computer or a plurality thereof in a network.

14. Computer program product having program code means stored on a machine-readable storage medium for performing all steps according to any one of the preceding method claims 1 to 10 if the program is executed on a processing unit, a microcontroller, DSP, FPGA or computer or a plurality thereof in a network.

15. Modulated data signal containing instructions executable by a processing unit, a microcontroller, DSP, FPGA or computer or a plurality thereof in a network for performing a method according to any one of the preceding method claims 1 to 10.