Method for measuring of deflection or change the length of the product

Description of the invention

This invention relates to a method of measuring the deflection or change in length of a blade or rotor mast product in a wind power plant.

The present application relates to the German patent applications DE 38 21 642 and DE 37 12 780. From these applications it is known how changes in length or distance, which are caused by changes in physical parameters such as temperature or pressure, can be determined by measuring the delay electrical signals in a fiber optic cable exposed to physical impact. The signals are fed to the optical fiber using an optical multivibrator. In such a situation, the entire multi-pulse delay is determined with the high frequency counter. By comparing with the result of the reference counting, the deviation of the currently determined counting result from the reference counting result is determined, the length or distance difference is determined, and this length or distance difference is converted into a parameter of the physical interaction being investigated.

DE 37 12 780 describes a process and a device for accurately and quickly measuring the length of the line under test as well as the electrical discontinuity in the line. In this device, the pulse edge is introduced to the first end of the line with the pulse edge generator, the reflected edge of the pulse is detected, which returns from the discontinuity to the first end of the line, and after a certain time after the detection of the edge of the reflected pulse, the next pulse edge is triggered, so that the pulse edge generator repeatedly produces a pulse edge with a frequency associated with the propagation delay along the line length, and this frequency is measured. The publication DE 37 12 780 therefore describes how a discontinuity in a line can be detected and shows the possibility of using the reciprocal of this value, i.e. frequency more precisely, instead of a delay.

According to the publication DE 38 21 642, the delay in the line between the transmitter and the receiver is measured and the so-called stopwatch process causes the counting of clock pulses of much higher frequency to start with the emission of the pulse, the counting procedure being continued until the receiver receives the pulse. The counted value is a measure of the delay.

According to the invention, a method for measuring the deflection or change in length of a product, namely a rotor blade of a wind turbine, during its operation, wherein a line of conductors is connected to the rotor blade, is characterized in that when the rotor blade bends and / or changes in length due to increased load, a The length of the bent line of wires is also changed, the impulse signal is introduced into the bent line of the wires and the reference wire at their first end, generated by a signal generator, this pulse signal at the other end being received by a signal receiver which, upon receipt of the signal, triggers the triggering of a further pulse signal in the signal generator, and the device used counts the number of pulsed signals emitted in a predetermined unit time and the deflection and / or lengthening of the rotor blade is determined by comparing the measured number of pulsed signals emitted to the bent line of conductors over a given time unit with the count result from reference conductors.

Another embodiment of the invention is characterized in that, when the rotor blade is bent and / or the length changes due to its increased load, and the bent line of the lines also changes in length, a pulse signal is introduced into the bent line of the lines and the reference line at their first end. by means of a signal generator, and the reflector located at their other end reflects this pulse signal to their first end, where it is received by a signal receiver, by means of which the triggering of another pulse signal is then triggered in the signal generator, the signal generator and the signal receiver are connected to each other, and the time between the receipt of the pulse signal and the triggering of the next pulse signal by the signal generator is constant .

The above-mentioned line of conductors is either a line of electric conductors into which an impulse signal being an electrical signal is introduced, or the line of conductors is preferably a line of optical fiber conductors into which an impulse signal being an optical signal is inserted.

Preferably, the pulse signal is introduced into the length-varying bend line of conductors permanently connected, at least at its ends, to the rotor blade. More preferably, a line of conductors, at least in a predetermined area, is securely connected to the rotor blade,

And when the rotor blade is bent or lengthened, the conduit line is only stretched in this predetermined area.

As mentioned above, at least one electric conductor extends along the length of the rotor blade, said electric conductor starting at the rotor blade connection, running in the longitudinal direction of the rotor blade and back to the rotor blade connection, and a sensor is used which measures the resistance. wire and is connected to the resistance evaluation device.

Accordingly, the invention is based on the observation that the bending of the rotor blade always elongates the supporting structure, and this elongation, transferred to the electric conductor, causes a change in the resistance of the conductor, it may be called a bent conductor.

Since this change in resistance is proportional to the elongation of the conductor, it can be said that the change in resistance is proportional to the degree of bending of the rotor blade. In the simplest case, it is possible to determine the limit value of the change in resistance, and the fact of exceeding this limit value also means that the rotor blade is bent to the extent that the structure is damaged. Noticing such a situation therefore allows the rotor blade to be replaced or inspected in good time in order to decide whether or not the rotor blade is to be replaced.

In a preferred embodiment of the invention, the load on the wind power plant may be determined on the basis of the rotor blade load, and the wind power plant may be shut down if a certain limit is exceeded. This avoids further and greater stresses.

In order to be able to compensate for the temperature dependent change in electrical conductor resistance and / or to obtain multiple measurement results, multiple electrical conductors can be used. These lines extend in the longitudinal direction of the rotor blade and are connected to the sensor. In such a case, multiple wires may be connected to the sensor or each wire may be connected to its own sensor. The temperature compensating conductor, which can be called the reference conductor, is positioned such that it is not subjected to bending load, but is only subject to a change in resistance with temperature. In this way, the change in resistance due to temperature is known and can be appropriately taken into account.

Most cables also allow redundant use. In the event of a problem with one cable, the change in resistance can still be measured using redundant cables. This eliminates the need for expensive repair caused by damage to the cable.

Moreover, the use of redundant conductors also allows the comparison of the change in resistance to be determined. In this way, each conductor can be checked by comparing the detected resistance change with the others.

According to a particularly advantageous feature, at least one of the bendable electrical lines extends a predetermined distance in a direction along the rotor blade, this distance however being shorter than the length of the rotor blade. The conduit does not extend to the top of the rotor blade, but terminates at a predetermined location on the rotor blade. In this way, individual bent conductors are subject to different effects depending on the bending of the rotor blade and the corresponding change in their resistance. Therefore, the exact shape of the bend can be determined from the different values ​​of the change in resistance.

In a preferred embodiment of the invention, the conduits which are shorter than the length of the rotor blade may also take the form of a branch of the conduit which extends over the entire length of the rotor blade. For this purpose, they are galvanically connected at certain points to a conduit that runs along the entire length of the rotor blade. In this way, the resolution level of the rotor blade bending can be varied depending on the number and spacing of the branches.

In order to avoid undesirable deformation of the bendable electric conductors, they are preferably connected to the rotor blade support structure and, according to a particularly advantageous feature, are included in the rotor blade support structure. In this case, according to a particularly advantageous feature, the bend lines, i.e. those which are to be stretched by bending the rotor blade, are inserted into the supporting structure. The conductors which are galvanically connected as branches to such conductors and which here act as return conductors can also be laid freely outside the support structure, e.g. in the form of a cable.

Accordingly, it is particularly advantageous to provide at least one electric conductor bent in each support structure in the longitudinal direction of the rotor blade. This solution also makes it possible to detect the direction of the rotor blade bending since one of the conductors

The electrical wiring is subjected to stretching, and therefore its resistance changes, more specifically increases, while at least one other wire is not stretched. With a suitable mounting structure, the conduit is subjected to a compressive load rather than a tensile load, and therefore its length is reduced rather than increased. Accordingly, its resistance changes at most in the opposite direction. This resistance therefore decreases.

According to a particularly advantageous feature, the electric conductors according to the invention contain at least one specific aluminum component or are made of aluminum. This already allows a significant change in the resistance in terms of the elastic strains of the conductor, this change in resistance being completely reversible and therefore reproducible. In a suitable manner, the bending of the rotor blade can be constantly monitored without the need to replace the electric wires of the bent or even the entire rotor blade once a bend has occurred. It should be noted that virtually any electrical cable is suitable as a sensor. However, such a conduit should have a surface with a certain roughness in order to obtain a good mechanical connection with the surrounding material.

In order not to have to replace the entire rotor blade in the event of a problem in the area of ​​the electric wires, e.g. after overstretching or due to material defects, the wires are preferably inserted into a member which is detachably attached to the support structure. Such an embodiment of the invention also allows it to be used in already manufactured rotor blades.

The subject matter of the invention in the examples of embodiments is explained by means of the drawing, in which: Fig. 1 shows a top view of a rotor blade according to a first embodiment of the invention in partial section, Fig. 2 shows a top view of a rotor blade according to a second embodiment of the invention in partial section, Fig. 3 shows a top view of a rotor blade according to a third embodiment of the invention in partial section, Fig. 4 is a simplified side view of a rotor blade which is bent in a first way, fig. 5 is a simplified side view of a rotor blade which is bent in a second way, fig. 6 is a simplified sectional view of a rotor blade, fig. 7 is a further simplified sectional view of a rotor blade. of the rotor, Fig. 8 shows the resistance curve of the conductor used according to the invention, Fig. 9 shows an embodiment of a rotor blade according to the invention together with measurement diagrams

b) - e) in the case of an analog phase shift, Fig. 10 is a schematic diagram of a measuring device according to the invention a) in the case of a rotor blade and the resulting measurement plots in the case of a digital phase shift (b) - (d)), Fig. 11 is a schematic diagram of a blade rotor with test lead, Fig. 12 shows the diagram of Fig. 11 with applied current or voltage, Fig. 13 is a schematic diagram of a rotor blade according to the invention with a test lead when inputting an input pulse and measuring an output pulse, as well as a corresponding timing diagram, Fig. 14 is a circuit diagram for a stopwatch process along with a timing diagram, and Fig. 15 is the circuit of Fig. 14, but using optical fibers, showing the corresponding measurement time diagram.

Fig. 1 is a plan view of the rotor blade 10. To clearly show the construction, the rotor blade 10 is shown in a partial section so that the arrangement of the two conduits 20, 22 can be seen.

The rotor blade 10 is mounted by means of a rotor blade cap 11 on the hub 12 of the wind power plant, which is only indicated for orientation here and in the other drawings. A bend conductor 20 and reference conductor 22 extend in the rotor blade from the root 11 of the rotor blade in the longitudinal direction of the rotor blade to the tip 13 of the rotor blade and back. Each of these flex lines 20 and reference 22 includes an input line 20a, 22a and a return line 20b, 22b. Both wires 20 and reference 22 are connected to a sensor 16 that determines their resistance.

In this case, the bend conduit 20 is shown to be straight. This is a symbolic representation of the fact that this conduit is connected to the rotor blade in such a way that stretching the rotor blade structure also stretches this flex conduit 20. The reference conductor 22 is shown non-linear in the area of ​​the rotor blade tip 13. This means that the conduit is not connected to the rotor blade 10 so as to be deformed with it.

PL 218 522 B1

Reference lead 22 should be free from distortion. Its resistance correspondingly varies primarily with temperature, so that the effect of temperature on the change in resistance is known and can be taken into account when detecting a change in resistance of the electric bend conductor 20 and does not lead to erroneous results.

The sensor 16 may be connected to the control system of the wind power plant, so that the deflection of the rotor blades may also be accounted for in the operation of the wind power plant.

Fig. 2 shows a construction similar to that already described from Fig. 1. It should be noted that four electric bend wires 20, 24, 25, 26 are shown. The temperature compensation reference wire is not shown for the sake of clarity, but of course it can. it can also be used in this embodiment.

In this drawing, all four electric bend conductors 20, 24, 25, 26 connected to the rotor blade 10 such that they follow the deflection of the rotor blade 10. However, since the conductors run at different lengths in the longitudinal direction of the rotor blade 10, it is possible to deduce the form of the overall deformation of the rotor blade 10 from the variation in the resistance of the individual conductors, if their length is known.

If there were only a first bend conduit 20 that extends up to the rotor blade tip 13, it would be possible to infer the deflection as such from the change in resistance. However, since the second bend conduit 25 does not extend as far as the rotor blade tip 13, it is possible to infer from the change in resistance of the first conduit 20 whether a deflection occurs at the rotor blade tip. If there is no change in resistance in the remaining bend conduits 24, 25, 26, i.e. changes in resistance due to deformation, then the recognizable deflection is limited only to the outer area of ​​the rotor blade 10 near the rotor blade tip.

If changes in resistance occur in bend conductors 20, 24 and 25 while the resistance of bend conductor 26 remains unchanged, this means that the rotor blade is bent approximately in the outer half in the longitudinal direction.

In this drawing, bend wires 20, 24, 25, 26 are also connected to a sensor 16 which in turn may be connected to a wind farm control system.

Figure 3 shows an alternative embodiment of the invention. The conductor 20 extends in the rotor blade 10 from the sensor 16 along the entire length of the rotor blade as far as the tip 13 of the rotor blade. The flexure branch lines 28 are connected to the first flex line 20 by a galvanic connection 29. The selection of the number and location of the flexure branch lines 28 may determine which areas of the rotor blade 10 are to be monitored with what accuracy. Or, in other words, with what accuracy and in which area of ​​the rotor blade 10 the deflection is to be detected.

In particular, in the embodiment shown in this drawing, it is evident that it is advantageous to provide a plug connection in the area of ​​the rotor blade root 11 (although not shown), since in the manufacture of the rotor blade 10 the bend lines 20, 28 may already be connected to the plug connection. and you can check the connections. After assembling the rotor blade on site, it is then possible to use cables that have already been made for connection to the sensor. This ensures a simple assembly with a low probability of making a mistake.

Fig. 4 is a schematic side view of a curved rotor blade 10. The drawing also shows the area of ​​the blade root 11, the marked hub 12 and the sensor 16. In this case, the distance of the bend conduits 20 from the surface of the rotor blade 10 should in no way be interpreted as conserving scale. When measuring the degree of deflection in this way, it is more accurate the closer the bend conduits 20 run to the surface of the rotor blade 10. For the sake of clarity of the drawing, a distinction is made between the surface of the rotor blade 10 and the bend conduits 20. The rotor blade 10 is bent downward in this drawing. The side of the rotor blade 10 which faces the deflection direction is in the drawing from below, while the opposite side is from above.

The drawing shows that the first folding conduit 20 is positioned at the upper side of the rotor blade 10 and the second folding conduit 21 is positioned at the lower side. With the shown rotor blade deflection, the first conduit 20 at the upper side of the rotor blade is stretched and exhibits a marked change in resistance that can be picked up by the sensor 16. With this rotor blade bending, the second bend conduit 21 at the lower side of the rotor blade 10 is not stretched, but is compressed. It certainly does not cause any increase in the resistance of this conductor. Consequently, the direction of the rotor blade bending can be deduced from the change in the resistance of the first bend conductor 20 at the upper side of the rotor blade 10.

PL 218 522 B1

Fig. 5 shows a special case of a bent rotor blade 10 which, however, is not unusual in practice. In this case, the rotor blade is bent in its central region in the direction of arrow A (downward), but in its outer region close to the rotor blade tip 13, it is bent in the direction of arrow B, i.e. towards the upper side of the rotor blade. Equipping the rotor blade 10 with bent conduits 20, 21. both extend up to the tip 13 of the rotor blade. will stretch both of these wires.

If the failure situation is rejected, it is already possible to deduce a dangerous bending of the rotor blade 10 on this basis and control the power plant accordingly. for example, turn it off. However, it should be noted that the actual shape of the rotor blade bending cannot yet be deduced from this. If further bend leads 23, 25 are provided that do not extend to the tip 13 of the rotor blade. then, in the situation of the illustrated bending, the bent conduit 25 is also stretched and therefore its resistance increases accordingly. Accordingly, the actual bending of the rotor blade 10 can be deduced from the measurement of the resistance or the variation in the resistance of the bend leads 20, 21, 23, 25 by the sensor 16. In this connection, it should be emphasized once again. that the bend conduits 20, 21, 23, 25 extend very close to each other and as close as possible to the respective surfaces of the rotor blade 10. so that the resulting extension of the second bend conductor 23 on the lower side of the rotor blade does not actually occur.

As an alternative to this embodiment with a plurality of bend lines 20, 21, 23, 25 of different lengths. which may be in the form of wire loops here. the embodiment of the invention shown in Fig. 1 can of course also be used on the upper side and / or on the lower side of the rotor blade 10. Of course, the advantages described here also follow. in particular, the possibility of achieving a certain degree of accuracy in detecting rotor blade deflection by selecting the number and spacing of the branches.

Figures 6 and 7 show simplified cross sections of a rotor blade according to the invention. In the rotor blade shown in Fig. 6, supporting structures 34, 36 extending in the longitudinal direction are shown. These support structures 34, 36 can be, for example, strips of fibers. that is, load-bearing structures made of tufts of glass fibers and epoxy resin. which extend substantially over the entire length of the rotor blade.

Bend and reference electrical lines 20, 21, 22, 23 are embedded in the support structures 34, 36. The input and return lines are labeled a and b respectively for reference. that each arrangement relates to a conduit extending from the base of the rotor blade in the longitudinal direction of the rotor blade and back.

The use of bent and reference conductors 20, 21, 22, 23 in load-bearing structures 34, 36 means. that their course can be very accurately determined. Due to the fact that they run as close as possible to the respective rotor blade surface, it is possible to draw conclusions from changes in resistance with an appropriate degree of certainty.

Fig. 7 also shows the supporting structures 34, 36. In this case, however, it should be noted. that the bend and reference lines 20, 21, 22, 23 are not located in the support structures 34, 36 themselves but in the support members 38. These support members 38 can be of the same structure as the support structures 34, 36 so that the interaction of the support members 38 with bent and reference conductors 20, 21, 22, 23 exactly corresponded to the cooperation of the conductor with the supporting structures 34, 36.

In this case, the support members 38 may be stationary. but detachably attached to the load-bearing structures 34, 36. Should a cable need to be replaced due to material defects or other damage. this does not necessarily result in the loss of the entire rotor blade or a very costly repair. but the respective support element 38 detaches from the support structure 34, 36 and is replaced with a new one.

Such an embodiment of the invention. thanks to the possibility of an appropriate choice. As regards the connection between the support structures 34, 36 and the support elements 38 or also the connection between the surface of the rotor blade 10 (obviously the inner one) and the support elements 38, it is possible to apply the invention to rotor blades. which have already been executed.

Fig. 8 shows the course of the experimentally tested conductor resistance as a function of the tensile stress. The left range 40 follows a straight line. in the middle range 42 the curve rises considerably, while in the right range 44 the curve initially runs in a straight line. before there is again a sharp increase in resistance with a subsequent decrease in resistance and a final increase in resistance. The range of the 44 curve on the right is characteristic of

Tearing the electric cable when the tensile stress is too high. On the other hand, the change in resistance in the middle range 42 of the curve course is within the range of elastic deformation of the electric conductor. In a series of measurements, when determining this curve, the range of elastic deformation of the electric conductor was found with a stretching in the longitudinal direction of less than 1% of the conductor length, and in the case of aluminum in particular in the range of 0.3%. Stretching the aluminum conductor in the longitudinal direction by 0.3% is therefore an elastic deformation which, however, causes a significant and detectable change in resistance. This has been found in a number of measurements up to 25 ΜΩ.

Since the deformation is elastic, the electrical conductor is not damaged by it and the change in resistance is reproducible. Rotor blade bending can be repeatedly detected using the same electrical cables.

Figures 9 and 10 show the complement and alternative of the above-described process or method, respectively. The method can be carried out with analog and / or digital signals. In both the solutions shown in Figures 9 and 10, it is common that the delay of the signal in the circuit does not affect the delay detection procedure. This enables the actual delay in the wireline to be marked.

The analog and digital solutions in Figures 9 and 10 are substantially comparable. In both cases, there are two lines between the transmitter and the receiver, namely the line of reference conductors 51, the length of which does not vary, and a measuring line of bent conductors 50 parallel to it, the extension of which enables the measurement of the bending of e.g. In addition to the fact that here the delays between the reference line 51 and the measuring line of the bend conductors 50 can be compared, the two lines are also subject to the same thermal effects so that their influence is compensated.

In an analog circuit (first alternative) at rest, the reference signal (analog electrical signal) and the measurement signal are in phase. There is therefore a summed signal of the same frequency but with a greater amplitude.

If a phase shift occurs due to the stretching of the bend measuring line, the summed signal obviously also changes. On the one hand, the peak-to-peak value is lower than for in-phase signals, and there is also a change in the envelope of the summed signal.

The method for detecting such changes is well known in the art. You can see that the amplitude goes down until a phase shift of 180 °. Outside of this range, up to the full period, signs must also be taken into account in order to obtain correct information regarding the phase position.

In the case of a digital solution with phase-compatible input of signals to the receiver, the lowest arithmetic mean is obtained (also, of course, depending on the duty cycle of the pulse waveform). Assuming that the duty cycle is still the same, however, the arithmetic mean increases with increasing phase shift between the reference signal and the measurement signal. It is therefore a measure of the phase shift of the signals at the receiver.

The above-described process can be carried out with electrical, optical and, in principle, also acoustic signals. Basically, the situation is that for a slight stretch it is preferable to choose a high frequency (i.e. a frequency greater than 1 kHz, preferably several MHz), while for a large stretch it is preferable to choose a low frequency to obtain the appropriate phase shift within one period.

Fig. 11 (similar to Figs. 9 and 10) shows a rotor blade 10 with the measurement line of conductors 50 disposed therein, preferably in the surface of the rotor blade in the form of a measurement cable (or optical fiber). Mechanical loads (caused by wind) bend the rotor blade and the cable is stretched or reduced in length. This change in load is therefore proportional to the change in length:

AF ~ ΔΙ

Expect a length change in the range of 0.0% - 0.2%, which corresponds to a load change of 0.0 - 100%. The task is therefore to determine the change in length with the highest possible level of resolution.

In the first solution, the resistance of the conductor can be taken as proportional to the length, and therefore also to the load.

PL 218 522 B1

AR - ΔΙ ~ AF

A current is applied to the conductor and the voltage drop is measured over it, as shown in Fig. 2.

Tests have shown that this principle is applicable in practice.

There are some problems, however, as a high measurement accuracy (<0.002%) is required as the signal is 0.2% absolute and must also be at least 100 degrees. Moreover, the resistance of the conductor varies greatly with the temperature of the conductor. Noise is superimposed on the signal, which may be generated by electric and magnetic fields. This is especially noticeable during a thunderstorm. A cable with an attached electronic circuit can be damaged by a direct lightning strike.

An alternative solution is shown in Fig. 13. In this case, the length of the bent wire is determined by the pulse delay. The speed is 2/3 the speed of light, or about 200,000 km / s.

As it can be seen from Fig. 13:

At ~ ΔΙ - AF and the transition time change is a measure of the load.

With a line length of 40 m assumed, this gives t = 200 ns with an Δt overlay of 0-400 ps.

Since such time is not easy to measure, an inverse is formed, more precisely defined as:

f = -

AND

So it is now about frequency.

The frequency values ​​are very easily determined with the final desired accuracy (by adjusting the gating time of the frequency counter).

The frequency is now determined from the signal delay by a procedure whereby the incoming pulse immediately sends a fresh pulse to the line (stopwatch process). The number of pulses emitted in one second determines the frequency.

Fig. 14 is a schematic diagram of the corresponding circuit along with the corresponding timing diagram.

The pulse is replaced by a change in level and the frequency is as expected:

£ 1 X 1 f = -, where t = ^{2tp yy 2}(F)

With a cable length of 40 m, this gives:

AND

Due to the influence of the load, the frequencies are in the range of 2.5-2.505 MHz, i.e. a change in value by 5000 Hz.

A frequency counter with a gating time of 20 ms would give 50 values ​​per second with a 1% load resolution. These values ​​then already contain the mean values ​​of the 50 individual length measurements.

The advantage is then that no sensitive analog sensing system is needed as there is a high signal-to-noise ratio (0V or 10V) and there are no troublesome delay variations due to temperature fluctuations.

This takes place in particular according to the embodiment of Figs. 9 and 10, where there is also a reference line in addition to the measurement line.

The above-described solution can also be realized optically. In this case, the cable is replaced by an optical fiber, and feedback is provided by the optical fiber transmitter and receiver, as shown in Fig. 15.

In this case, the advantages are in particular that no sensitive analog sensor system is needed because there is a high signal-to-noise ratio (light on or off), no troublesome delay variations due to temperature fluctuations, no noise interference on the fiber due to electric or magnetic fields and no impacts from lightning are to be expected.

PL 218 522 B1

It should be expected that in practice the blade deflection will already be measured with a length change of less than 1 mm. As the present application states that the invention is to be applied to the variation in the length of a rotor blade, it should be noted that it is also possible to measure the twist of a rotor blade if the measuring line is suitably arranged in a spiral shape on the surface of the rotor blade such that the twist of the rotor blade is the rotor also automatically changes the length of the coiled cable line.

In particular, the measuring method according to the invention can also be used to monitor the rotor blade parts with regard to their load as well as their longitudinal extension by the measuring procedure, which can be very helpful especially in the area of ​​the blade tip when gusts of wind are present in order to detect better than hitherto associated hazards. in a specific individual case with the given blade loads.

It should be clearly emphasized that what is shown in figures 11-14 can of course also be easily combined with what is shown in the other drawings. Regarding the stopwatch process, it is evident here that this means, inter alia, that the stopwatch is stopped upon receipt of the emitted pulse and restarted each time and accordingly determines the frequency of stop-start cycles in a given time range, e.g. 1 s.

Accordingly, it is possible to use any shape of the pulse signal, inter alia also a narrowed pulse signal according to the sin x / x function.