SEQUENTIAL FATIITATION TESTING METHOD

Abstract

Method for fatigue testing of a beam-shaped test specimen with a predetermined target load distribution, comprising at least a first test sequence (SQ1) of a first segment (SG1) of the test specimen, wherein one end of the first segment (SG1) is fixed in a test device (2) and is excited with a first predetermined load collective, comprising a load amplitude for each cross-section along the segment (SG1) of the test specimen, a load mean value for each cross-section along the segment (SG1) of the test specimen and a number of cycles, by an active load introduction means (4) at a first load frequency close to or equal to the system natural frequency of the test setup, and, subsequently, a second test sequence (SQ2) of a second segment (SG2) of the test specimen, which overlaps at least partially with the first segment (SG1) of the test specimen, wherein one end of the second segment (SG2) is fixed in a test device (2) and is excited with a second predetermined load collective by an active load introduction means (4) at a second load frequency close to or equal to the system natural frequency of the test setup, wherein the first predetermined load collective together with the second predetermined load collective accumulates at individual cross-sections along the test specimen and thereby achieves or exceeds the predetermined target load distribution, wherein the first load collective and the second load collective act together in the impact or pivoting direction or together in the impact and pivoting direction.

Description

[0001] The invention relates to a method for the sequential fatigue testing of beam-shaped test specimens. In particular, the method is suitable for testing rotor blades for wind turbines.

[0002] Rotor blades are subjected to high loads and wear during operation. They must therefore undergo thorough testing to ensure safe operation. Such testing is complex because newly developed rotor blade types have become increasingly longer over time and must be subjected to a predefined target load distribution along their entire length to simulate realistic loads. For this purpose, the rotor blades to be tested are clamped at one end in a test block, loaded by the targeted positioning of additional masses, decoupled masses, and / or spring elements, and harmonically excited to generate a test load distribution. This bending moment distribution is then introduced into the test specimen over a certain number of cycles. Such testing can be performed uniaxially, i.e., in either the flapping or pivoting direction, or biaxially, in both flapping and pivoting directions simultaneously.

[0003] One challenge of this testing procedure is generating the target load distribution along the entire test specimen without overloading any single section. If such a section is damaged to the point of failure during the test, the test cannot continue for the remaining length of the specimen without repairing the affected section. When testing rotor blades, the relatively short area near the blade root is particularly problematic. Due to the high forces this section is subjected to during operation, the target load at the root is also significantly higher. Since accelerations near the blade root are low, any additional masses applied in this area must be quite large to achieve the target load distribution.Further problems arise from the fact that adding mass significantly influences the mean bending moment distribution and lowers the system's natural frequency. The latter effect, in particular, necessitates longer test durations. To circumvent these problems, segmented testing can be performed. For this, the test specimen is divided into segments, which are then individually tested using the aforementioned method. However, this approach has the disadvantage that the area around the interface between the segments cannot be tested. An example of such a segmented testing procedure is shown in the publication "Development and feasibility study of segmented blade test methodology" (Ha, Kwangtae et al., 2020, Wind Energy Science, Vol. 5 No. 2, pp. 591-599). Publication DE 10 2018 218 515 A1 also describes a method for the biaxial testing of rotor blades.

[0004] The purpose of this patent application is therefore to propose a fatigue testing method that allows beam-shaped test specimens, such as rotor blades, to be tested efficiently and reliably along their entire length.

[0005] This problem is solved by a method according to independent claim 1. Advantageous embodiments will become apparent from the dependent claims, as well as from the description and the figures.

[0006] A method for fatigue testing a beam-shaped test specimen with a predetermined target load distribution comprises at least the steps of a first test sequence of a first segment of the test specimen, wherein one end of the first segment is fixed in a test fixture and is excited with a first predetermined load collective, comprising a load amplitude for each cross-section along the segment of the test specimen, a load mean value for each cross-section along the segment of the test specimen, and a number of cycles, by an active load introduction means at a first load frequency close to or equal to the system natural frequency of the test setup, and a second test sequence of a second segment of the test specimen, which overlaps at least partially with the first segment of the test specimen, wherein one end of the second segment is fixed in a test device and is excited with a second predetermined load collective by an active load introduction means at a second load frequency close to or equal to the system natural frequency of the test setup, wherein the first specified load collective together with the second specified load collective accumulates at individual cross-sections along the test specimen, thereby achieving or exceeding the specified target load distribution.

[0007] The method is suitable for fatigue testing a series of beam-shaped specimens and is not limited to the testing of rotor blades. "Beam-shaped" here refers to a body with a two-dimensional cross-section and a third-dimensional extension, which is large relative to the dimensions of the cross-section. This length of the beam defines a longitudinal direction. The cross-section can have any shape. In particular, the cross-section of the beam-shaped specimen can be rectangular, rounded, circular, and / or in the form of an aerodynamic profile. The cross-section can vary along the longitudinal axis of the specimen. Segments of the beam-shaped specimen are defined as sections of the specimen along its longitudinal direction that encompass the entire cross-section of the specimen.

[0008] The specified target load distribution corresponds to the fatigue load expected during real-world use of the test specimen. This target load distribution must be achieved at a minimum during testing. For example, in the case of a rotor blade for a wind turbine, the target load distribution corresponds to the expected load from the bending moments in the flapping and pivoting directions that act on the rotor blade during operation on a wind turbine.

[0009] In each test sequence, the respective segment of the test specimen is fixed in a test fixture of a test rig so that the longitudinal axis of the segment, and thus also of the test specimen, extends approximately horizontally from the fixing point. The segment therefore has two degrees of freedom for vibrations: parallel to the ground (pivot direction) and perpendicular to the ground (impact direction). The end of the segment can be fixed in various ways, for example, by securing it with bolts or screws, clamping it, or attaching it in a fixed-loose bearing. Depending on the type of fixing, the segment can also be rotated about its longitudinal axis to achieve a certain tilt angle to the ground. The segment is then excited to vibrate at a frequency close to or equal to the system's natural frequency by an active load application device. An active load application device could be, for example, an actuator or an eccentric exciter.The natural frequency of the system is determined by the segment, the test device, the actuator and any additional components.

[0010] During a test sequence, a segment is subjected to a load spectrum. This spectrum includes a load amplitude and mean load value for each cross-section along the segment, as well as the number of cycles over which the test sequence is performed. Bending moments and forces acting on each cross-section of the segment are thus defined. The duration over which the segments are subjected to these loads is defined by the number of cycles.

[0011] This procedure requires at least two test sequences on two segments of the test specimen that overlap at least partially in their longitudinal extent. The second segment can, for example, be a sub-segment of the first segment. Both segments are part of the same test specimen. The test specimen can be physically divided into several segments on which the test sequences are performed. However, it is also possible for the test specimen to remain intact, with the segments defined only as sections of the test specimen upon which specific load collectives are applied.

[0012] Over the at least two test sequences, the load amplitude and the load mean value accumulate over the cycles of the first and second load collectives for the individual cross-sections of the segments of the test specimen, so that the target load is reached or exceeded for each individual cross-section of the tested area of ​​the test specimen.

[0013] This allows for the achievement of target load distributions, particularly in areas where the tested segments overlap, which would not be possible through fatigue testing of the entire specimen or even individual segments. This is especially true for the interfaces where the segments are physically divided. Furthermore, adjusting the load collectives of the individual test sequences reduces the fatigue testing time. If one section of the specimen can only be tested under a low load, while a second section requires a high load, it is not necessary to test the entire specimen with a high number of test cycles under the low load.Instead, an initial segment encompassing both sections can be tested for a certain initial number of cycles under a low load and then divided so that the section to be tested under a high load is then tested for a second number of cycles under a higher load. The first section thus experiences only the lower load, while in the second section the loads from both test sequences accumulate, thereby achieving the necessary high load. This ensures that the specified target load is achieved for every cross-section along the test specimen without overloading any individual sections. Additionally, dividing the segment increases its natural frequency and thus its testing frequency, which in turn reduces the overall test duration.

[0014] In particular, the first segment tested in the first test sequence can correspond to the entire test specimen. In the first test sequence, the entire test specimen is thus tested with an initial load distribution, which can then be divided into a second segment and further segments. These overlap with the first segment. In this way, the entire test specimen can be tested with a predefined target load distribution.

[0015] The first segment of the test specimen can also correspond to the second segment. In this case, the test sequences are performed on the same segment. However, the test sequences can differ in their load collectives, so that different loads act on and accumulate on different cross-sections along the segment in the first and second test sequences.

[0016] The procedure requires at least two test sequences. However, it can be supplemented by further test sequences. These can be performed on the first segment, the second segment, or on a further segment of the test specimen. The segments tested in these further test sequences are each subjected to additional load collectives. These load collectives then accumulate to form the target load distribution. Each individual segment overlaps with at least one other segment. For example, the first segment can be the entire test specimen, which is then divided into two segments for a second and third test sequence. These segments each overlap with the first segment, but not with each other. By accumulating the first, second, and third load collectives, a target load distribution can be achieved for the entire length of the test specimen.The individual segments can be further subdivided for additional test sequences.

[0017] If further segments of the test specimen do not overlap, the respective test sequences can be performed simultaneously in a corresponding number of test fixtures. This reduces the test duration for the entire fatigue test.

[0018] The fatigue testing procedure can be performed uniaxially, both in the impact direction and in the pivot direction. The active load application device then excites the segments to vibrate in the respective loading direction. Different types of active load application devices are generally required for excitation in the impact and pivot directions.

[0019] The fatigue test can be performed for each segment under test, first in one loading direction and then in the other. In this case, different test setups are required for impact and pivoting directions.

[0020] Individual test sequences of the uniaxial fatigue test can be performed in the impact or pivoting direction, before, between, and / or after the individual test sequences in the opposite loading direction, or before and after an entire fatigue test in the opposite loading direction. The order of the individual test sequences is arbitrary. Thus, a fatigue test can be performed on a test specimen that is divided into segments in both the impact and pivoting directions.

[0021] The fatigue test can also be performed biaxially, i.e., simultaneously in the impact and rotation directions. For this purpose, the segments of the test specimen can be excited by at least one additional active load application device. In this embodiment, the first active load application device is designed to excite the segments in one loading direction, and the at least one additional load application device is designed to excite the segments in the other loading direction. The excitations in the respective loading directions can be performed at the same or different loading frequencies. In particular, the loading frequencies in the impact and rotation directions can correspond to the system natural frequencies of the test setup in the impact and rotation directions.

[0022] The loading frequencies of the first load application device and of at least one other active load application device can be selected to be identical or integer multiples of each other. In particular, the loading frequencies in the impact and pivoting directions can correspond to the system natural frequencies of the test setup in the impact and pivoting directions. Furthermore, the loading frequencies in the impact and pivoting directions can be arranged in relatively small integer ratios (1:1, 1:2, 2:1, etc.) to avoid complex movements, such as Lissajous figures, of the test specimen segments. For this purpose, the system natural frequencies of the test setup can be adjusted to the corresponding ratio, if necessary, by means of decoupled masses and / or elastic elements.

[0023] In fatigue testing, uniaxial and biaxial test sequences can be combined. The respective load collectives accumulate, ultimately resulting in a fatigue test in impact and pivot directions, as well as in other intermediate directions. This can be particularly useful when individual segments of the test specimen are susceptible to damage from loads in one direction or when testing is to be performed with a special focus on a particular direction.

[0024] Between the individual test sequences, the segments can also be tested using other test methods known from the prior art. It should be noted that if these test methods stress the segments, this stress will also accumulate in the overall stress distribution.

[0025] The load collectives of the individual test sequences can be set using passive load introduction devices. For example, fixed or decoupled masses can be attached to the test segments of the specimen. This allows the system's natural frequency to be changed. However, it should also be noted that these additional passive load introduction devices can exert an additional load on the test segments due to their weight. Another form of passive load introduction device can be an elastic element attached to the test segments. To set different load collectives for the at least two test sequences, different passive load introduction devices or different arrangements of passive load introduction devices can be provided for the different test sequences.

[0026] The test specimen can be, in particular, a rotor blade of a wind turbine. The segments of the individual test sequences can then be the entire rotor blade, the blade root, the tip, or a section of the rotor blade. Due to their shape, these sections of the rotor blade are subject to specific loads in real-world operation and thus have specific fatigue testing requirements that are difficult to reconcile when testing the entire rotor blade. A sequential fatigue test, as described here, allows the individual parts of the rotor blade to be subjected to specific loads without overloading individual sections or incurring additional testing time.

[0027] By avoiding overloading individual sections, the method offers a further advantage, as it allows for a reduction in the number of cycles. A load spectrum can be varied in such a way that the same load is achieved with a slight increase in load amplitude and a simultaneous significant reduction in the number of cycles. If the overload, which is primarily caused by excessively high load amplitudes, is reduced by the proposed method, the load amplitude can be increased across the entire segment to further decrease the number of cycles.

[0028] The described embodiments of the subject matter of the present application can be used individually as well as in combination to achieve additional effects and to provide a reliable and efficient method for the sequential fatigue testing of beam-shaped test specimens, in particular rotor blades for wind turbines.

[0029] The aforementioned aspects, as well as further aspects of the invention, will become apparent from the detailed description of the exemplary embodiments, which is given with the aid of the following drawings, of which: Fig. 1 schematically represents three general uniaxial test sequences in the direction of impact, Fig. 2 schematically represents three general uniaxial test sequences in the direction of rotation, Fig. 3 schematically represents three general biaxial test sequences, Fig. 4 describes the process of a sequential fatigue test with three test sequences, Fig. 5. The relative load distributions along a relative specimen length according to a method as described in Fig. 4 shows, Fig. 6 describes the process of a sequential fatigue test with two test sequences, Fig. 7 represents the process of a further sequential fatigue test with two test sequences, wherein the tested segment is the entire test specimen in each case, Fig. 8 describes the process of a sequential fatigue test with four test sequences, Fig. 9a describes the procedure of a uniaxial sequential fatigue test with two test sequences, each in the direction of impact, Fig. 9b describes the process of a uniaxial sequential fatigue test with two test sequences, each in the direction of rotation. Fig. 10 describes the process of a sequential fatigue test in which uniaxial test sequences are combined in the impact and pivoting directions, Fig. 11 describes the process of a biaxial sequential fatigue test with three test sequences, and Fig. 12 describes the process of a sequential fatigue test in which uniaxial and biaxial test sequences are combined.

[0030] The claimed method will now be explained in more detail with reference to the accompanying drawings. Reference numerals refer to the same elements. Rotor blades are shown in the drawings as an example of a test specimen. However, other beam-shaped test specimens can also be tested using this method.

[0031] In Fig. 1, Fig. 2 and Fig. Figure 3 schematically depicts test setups that can be used for the inventive process. These setups correspond to setups that can also be used for other processes known from the prior art.

[0032] Fig. Figure 1 shows a test setup A1 for uniaxial testing in the direction of impact. The test specimen 1, depicted as a rotor blade, is fixed in a test fixture 2. A rotor blade can, for example, be attached to the test fixture 2 via its connection points at the blade root. Additional masses 3 are attached to the test specimen as passive load introduction elements. An additional passive load introduction element, designed as an elastic element 5, is also shown. The active load introduction element 4 is an actuator designed to excite the test specimen 1 to vibrations in the direction of impact. The active load introduction element 4 specifies a load amplitude and excites the test specimen 1 for a predetermined number of cycles at a frequency that corresponds to or is close to the system's natural frequency.The entire test setup has a system natural frequency, which results from the properties of the test specimen and the passive load application means. In test setup A2, a segment 1, corresponding to the root of a rotor blade, is tested in the same way. In test setup A3, a segment 1, corresponding to the tip of a rotor blade, is tested, with a fixed-loose mounting used for fixation in the test fixture 2. The individual segments do not necessarily have to correspond to the blade roots and tips shown. A segment could just as easily correspond to a different section of a rotor blade or a segment of a completely different test specimen.

[0033] Fig. Figure 2 shows a test setup B1 for uniaxial testing in the rotation direction, analogous to the test setup shown in Figure 1. In this setup, the active load application means is designed to excite the test specimen 1 in the rotation direction, and the passive load application means are also configured for this loading direction. Test setups B2 and B3, in turn, represent the testing of segments 1.

[0034] Fig. Figure 3 shows a test setup C1 for biaxial testing. The test setup includes two active load application elements 4, one of which excites the test specimen 1 in the impact direction and the other of which excites the test specimen 1 in the pivoting direction. The passive load application elements are also designed to withstand loading in both directions. In addition to the additional masses 3 and the elastic elements 5, test setup C1 includes decoupled masses 6. These are not limited to biaxial testing but could also be used in all other test setups. It should be noted that test setup C1 can have different system natural frequencies in the two loading directions, and the active load application elements 4 must excite the test specimen 1 to vibrations at frequencies matched to these frequencies.It is advantageous if the excitation in the impact and pivot directions occurs at frequencies that are identical or in a small integer ratio to each other. This allows the movement of the test specimen 1 to be restricted to elliptical or figure-eight paths. To adjust the system natural frequencies independently of each other, the passive load introduction devices can be attached to the test specimen 1 in the corresponding direction. Test setups C2 and C3, in turn, represent the biaxial testing of segments 1. Test setups that include the one described in... Fig. 1 to Fig. The three shown correspond to those used to perform the sequential fatigue test. The following description and flowcharts illustrate this. Fig. 4 and Fig. 6 to Fig. 12 therefore refers to these examination structures.

[0035] Fig. Figure 4 shows the procedure of the sequential fatigue test according to the invention with three test sequences. For a test specimen, in this example a rotor blade, a target load distribution D is defined. target The test sequence SQ1 begins with the initial test of a first segment SG1 of the test specimen, which in this case corresponds to the entire test specimen, in the rotational direction. A test setup B1 is used for this purpose. Actuator 4 excites the first segment SG1 with a predefined initial load spectrum, comprising a load amplitude for each cross-section along the first segment SG1, a load mean value for each cross-section along segment SG1 of the test specimen, and a specified number of cycles. This results in a test load distribution D. testThe first test sequence SQ1 is completed. After completion, the first segment SG1 is divided into a second segment SG2 and a third segment SG3. In this example, the second segment SG2 is the root of the rotor blade and the third segment SG3 is the tip of the rotor blade. These segments therefore naturally overlap with the first segment SG1. The following are used for testing: Fig. Two known test setups, B2 and B3. The second and third test sequences, SQ2 and SQ3, can be performed simultaneously. The second and third segments, SG2 and SG3, of the test specimen are each excited with a second and third load collective, respectively. The test load distributions D obtained in the individual test sequences test The loads of the individual segments accumulate to form an overall load distribution, which is the target load distribution D. target achieved or exceeded.

[0036] The results of the fatigue test from Fig. 4 are shown in the diagram in Fig. Figure 5 shows the relative specimen length z / L on the x-axis of the diagram, and the test load distribution relative to the target load distribution D on the y-axis. target Area E shows the entire area of ​​the test specimen to be tested. The first test sequence SQ1 is performed on the entire test specimen, resulting in load curve A. As can be seen, this remains below the target load in areas F and H, i.e., D. test / D target < 1. In area G, the necessary load is already reached or exceeded in this sequence SQ1, D test / D target1. For the second and third test sequences, the test specimen is divided. In the second sequence, SQ2, only the root segment SG2 is loaded, generating load curve B. To achieve the target load in region F, region I is loaded again. In the third sequence, SQ3, only the tip segment SG3 is loaded, generating load curve C. Region J is loaded again to achieve the target load in region H. The load curves of the three test sequences are added together to form the accumulated total load curve D. For this, D applies. test / D target > 1 for the entire area to be tested. The target load distribution D target This was therefore achieved or exceeded for every cross-section along the test specimen.

[0037] Fig. Figure 6 shows an implementation of the procedure in which only two test sequences are performed. In the first sequence, SQ1, an entire test specimen is excited as the first segment, SG1, with a first load collective in the direction of impact, using a test setup corresponding to A1. In the second test sequence, SQ2, a second segment, SG2, is excited with a second load collective in a test corresponding to A2. The resulting test loads are added together in the area of ​​the second segment, SG2, thus enabling, for example, improved testing of a rotor blade root.

[0038] In Fig. In section 7, the first segment SG1 tested in the first test sequence SQ1 corresponds to the segment tested in the second test sequence SQ2. Both segments represent the entire test specimen, which is tested in a test setup according to A1. The only difference between the first test sequence SQ1 and the second test sequence SQ2 lies in the load collectives used. For example, cross-sections of the test specimen near the fixed end can be subjected to higher loads by the first load collective than cross-sections near the loose end, while the opposite is true for the second load collective. Together, these load collectives accumulate to at least the specified target load distribution.

[0039] Fig. Figure 8 represents a sequential fatigue test with four test sequences. The first two sequences, SQ1 and SQ2, correspond to the sequences from [reference missing]. Fig. 7. The test specimen is then divided into two segments, SG2 and SG3. Since these segments are no longer connected, test sequences SQ3 and SQ4 can be performed simultaneously, using test setups A2 and A3 respectively. The four load collectives accumulate and reach or exceed the target load distribution for each cross-section of the test specimen.

[0040] In Fig. Figure 9a shows such a sequential fatigue test again in the direction of impact. Fig. Figure 9b, however, shows the fatigue test with two test sequences in the direction of rotation. Fig. 9a First, a test sequence SQ1 according to test setup A1 is performed on a first segment SG, which corresponds to the entire test specimen. Then, in a second sequence SQ2 according to test setup A3, a second segment SG2 is tested. Fig. 9b. First, a test sequence SQ1 according to test setup B1 is performed on a first segment SG1, which corresponds to the entire test specimen. Then, in a second sequence SQ2 according to test setup B2, a second segment SG2 is tested. Both tests incorporate passive load introduction devices in the form of fixed masses 3 and elastic elements 5. These must be adapted to the loading direction. As can be seen, the uniaxial fatigue tests in one loading direction are otherwise analogous. The load collectives in the respective loading direction accumulate and reach or exceed the target load distribution for that loading direction for each cross-section of the test specimen.

[0041] The in Fig. The 10 illustrated methods combine several uniaxial test sequences. In a first sequence SQ1, the first segment SG1 corresponds to the entire test specimen. This is excited uniaxially in the rotation direction with a first load collective (see Fig. 2 B1). In the second sequence SQ2, the entire test specimen is also tested, but it is excited uniaxially in the direction of impact with a second load collective (see Fig. 1 A1). Only then is the test specimen divided into two segments SG2 and SG3, corresponding to the root and tip regions of a rotor blade. In the third test sequence SQ3, the second segment SG2 is first subjected to a third load collective in the direction of rotation (see Fig. 2 B2) and then in the fourth test sequence SQ4 with a fourth load collective in the direction of impact (see Fig. 1 A2). The third segment SG3 is excited during the fifth test sequence SQ5 with a fifth load collective in the direction of impact (see Fig. 1 A3). Further sequences in the pivoting direction would also be conceivable for this segment SG3; however, this is omitted because the first and third load collectives in the pivoting direction already accumulate to the target load distribution for each cross-section along the longitudinal axis of the test specimen. Similarly, the second, fourth, and fifth load collectives in the impact direction accumulate to the target load distribution for each cross-section along the longitudinal axis of the test specimen. Thus, the sequential fatigue test can be performed uniaxially in both the impact and pivoting directions, and a test can still be carried out in both loading directions.

[0042] In Fig. Figure 11 shows a biaxial fatigue test according to the procedure. Test setups are used as described in Fig. 3 shows the use of a system that simultaneously excites the segments to be tested in both the impact and rotation directions. In a first sequence SQ1, the entire test specimen is biaxially excited as the first segment SG1 (see Fig. 3 C1). Then the test specimen is divided and the segments SG2 and SG3 are processed in a second sequence SQ2 (see Fig. 3 C2) and third sequence SQ3 (see Fig. 3 C3) tested. This allows the test to be carried out simultaneously in both directions of loading.

[0043] In Fig. 12. Uniaxial and biaxial test sequences are combined to achieve or exceed the target load distribution for each cross-section of the test specimen in both loading directions. The procedure corresponds to that described in Fig. 10 shown, however, the excitation in the first test sequence SQ1 and third test sequence SQ3 is biaxial (see Fig. 3 C1 and C2). Additionally, the first segment SG1 is excited uniaxially in the direction of rotation in the second test sequence SQ2 and the second segment SG2 in a fifth sequence SQ5 (see Fig. 2 B1 and B2) and the third segment SG3 in a fourth sequence SQ4 in the direction of the stroke (see Fig. 1 A3).

[0044] This makes it possible to achieve or exceed the target load distribution for each cross-section of the test specimen even in areas where further biaxial test sequences would be time-consuming or risk damaging the segments too much.

[0045] The embodiments shown here are not limiting. In particular, the features of these embodiments can be combined to achieve additional effects. It is obvious to the person skilled in the art that modifications can be made to these embodiments without departing from the fundamental principles of the subject matter of this patent application, the scope of which is defined in the claims.

Claims

[1] Method for fatigue testing of a beam-shaped test specimen with a predetermined target load distribution, comprising at least a first test sequence (SQ1) of a first segment (SG1) of the test specimen, wherein one end of the first segment (SG1) is fixed in a test device (2) and is excited with a first predetermined load collective, comprising a load amplitude for each cross-section along the segment (SG1) of the test specimen, a load mean value for each cross-section along the segment (SG1) of the test specimen and a number of cycles, by an active load introduction means (4) at a first load frequency close to or equal to the system natural frequency of the test setup, and, subsequently, a second test sequence (SQ2) of a second segment (SG2) of the test specimen, which overlaps at least partially with the first segment (SG1) of the test specimen, wherein one end of the second segment (SG2) is fixed in a test device (2) and is excited with a second predetermined load collective by an active load introduction means (4) at a second load frequency close to or equal to the system natural frequency of the test setup, wherein the first predetermined load collective together with the second predetermined load collective accumulates at individual cross-sections along the test specimen and thereby achieves or exceeds the predetermined target load distribution, wherein the first load collective and the second load collective act together in the impact or pivoting direction or together in the impact and pivoting direction. [2] Method according to claim 1, characterized by, that the first segment (SG1) of the test specimen corresponds to the entire test specimen. [3] Method according to any one of the preceding claims, characterized by , that the first segment (SG1) of the test specimen corresponds to the second segment (SG2) of the test specimen. [4] Method according to any one of the preceding claims, characterized by , that the procedure includes further test sequences (SQ3, SQ4, SQ5) in which the first segment (SG1), the second segment (SG2), and other segments different from these are tested with further predefined load collectives, whereby the first predefined load collective together with the second predefined load collective and further predefined load collectives accumulates to the predefined target load distribution, whereby individual segments each overlap at least partially with at least one other segment. [5] Method according to claim 4, characterized by, that test sequences on segments of the test specimen that do not overlap are performed simultaneously. [6] Method according to any one of the preceding claims, characterized by that the fatigue test is performed uniaxially, in the direction of impact or pivoting. [7] Method according to any one of claims 1 to 5, characterized by , that the fatigue test is carried out uniaxially, successively in the impact and pivoting directions, whereby the order of individual test sequences is arbitrary. [8] Method according to claim 7, characterized by , that individual test sequences of the uniaxial fatigue test in the impact or pivot direction are carried out before, between and / or after the individual test sequences in the respective other loading direction. [9] Method according to any one of claims 1 to 5, characterized by, that the fatigue test is carried out biaxially, simultaneously in the impact and pivoting directions, wherein the test specimen is additionally excited by at least one further active load introduction means (4) with the same or different loading frequency as by the first active load introduction means (4). [10] Method according to claim 9, characterized by , that the loading frequencies of the first load introduction means (4) and of the at least one further active load introduction means (4) are set such that they are identical to each other or integer multiples of each other, wherein the loading frequencies correspond in particular to the system natural frequencies of the test setup for a vibration in the respective loading direction. [11] Method according to any one of claims 6 to 10, characterized by, that fatigue test sequences are combined uniaxially in impact and / or pivot direction and biaxially in impact and pivot direction, and that the respective load collectives accumulate. [12] Method according to any one of the preceding claims, characterized by that further testing procedures are applied to the segments of the test specimen between the individual test sequences. [13] Method according to any one of the preceding claims, characterized by , that the load collectives are set by attaching at least one passive load introduction device in the form of a fixed mass (3) or a decoupled mass (6) to the tested segment. [14] Method according to any one of the preceding claims, characterized by , that the load collectives are modified such that at least one passive load introduction device in the form of an elastic element (5) is attached to the tested segment. [15] Method according to any one of the preceding claims, characterized by that the test specimen is a rotor blade of a wind turbine.

Images