Method for creating a quantity of data for examining a track bed using electromagnetic rays

Abstract

The invention relates to a method for creating a quantity of data for examining a track bed of a track using electromagnetic rays.

Description

[0001] Method for creating a data set for the investigation of a track bed using electromagnetic radiation

[0002] The invention disclosed here relates to a method according to the preamble of claim 1.

[0003] The invention relates to a method for creating a data set for examining a track bed using electromagnetic radiation.

[0004] The track bed comprises, on the one hand, sleepers, which are laid at a sleeper spacing. The sleeper spacing in the center of the track, for example, is standardized in Europe at 60.0 cm and is essentially constant in the track direction. A list of sleeper spacings can be found, for example, in the entry for "track class" on wikipedia.org. The disclosure of this invention is based on this standard. Any fluctuations in the sleeper spacing along a track axis have no influence on the method discussed here.

[0005] On the other hand, a track bed can also contain non-track objects such as contamination of the ballast, water in the substructure or even stones or bombs in the substructure.

[0006] According to current teaching, such non-track objects in the substructure can be detected using electromagnetic radiation. For this purpose, a transmitter at a measuring position emits a signal comprising electromagnetic radiation with an output frequency and an output amplitude at a specific output time into the track bed, in particular into the substructure. The signal is transmitted, absorbed, and reflected at a specific location in the substructure depending on its electrical properties (electrical reflection properties, dielectric permittivity, magnetic permeability, electrical conductivity). The reflected signal, comprising reflected electromagnetic radiation with a receiving frequency and a receiving amplitude, is received at a specific receiving time by a receiver arranged at the measuring position.

[0007] Preferably, an output signal is output and a reception signal is measured at a plurality or series of measuring positions, which measuring positions are spaced apart in the track direction by a measuring point distance.

[0008] A measurement pattern is created using conventional methods to describe the change in the amplitude of the reflected signal relative to the output signal, or the amplitude of the reflected signal and, in any case, the location (x, y, z) of the signal's reflection. The measurement pattern comprises a vector representation of the amplitude describing the location and can be created as a data set or as an image. The amplitude describes the change in the electrical properties between the materials arranged in the subsurface at the location of the signal's reflection. According to conventional methods, methods for creating such a measurement pattern are known.

[0009] In the prior art methods, the measurement point spacing of the measurement positions is selected exclusively such that a measurement position is selected essentially midway between two adjacent sleepers. This arrangement of the transmitter and receiver, as well as the exclusive measurement midway between two adjacent sleepers, ensures that a sleeper has no influence on the determined measurement pattern. Even a layperson recognizes that the exclusive selection of measurement positions at the same measurement point spacing as the sleeper spacing does not provide a satisfactory measurement. WO2023169870 describes a prior art for examining a track bed using electromagnetic radiation.In comparison to this prior art, WO2023169870 describes a new method for examining electromagnetic radiation, in which method the transmitter and the receiver for emitting or receiving electromagnetic radiation are spaced apart from each other and thus arranged adjacent to each other.

[0010] The method according to the invention has the task of reducing the distances between the measuring positions in the track direction and of reducing the influence of the sleepers as track-specific objects arranged at a regular distance on the measurement image via the substructure.

[0011] Zhang Xuebing, et al. Noise suppression of GPR data using Variational Mode Decomposition. 17th International Conference on Ground Penetrating Radar (GPR) 20180618 - 20180621;

[0012] Rapperswil, Switzerland does not describe the use of intrinsic functions via the measurement path.

[0013] Wanchao He, et al., VMD analysis of RF-tagged buried pipes for GPR applications. 2019 IEEE Asia-Pacific Microwave Conference (APMC). 20191210 - 20191213 does not describe the use of intrinsic functions over the measurement path.

[0014] Furthermore, it should be possible to select a measuring position independently of its distance from a sleeper, whereby the influence of the sleepers as regularly arranged track-specific objects on the measurement image via the substructure is reduced.

[0015] The tasks mentioned have the technical effect of obtaining a measurement image or a data set with a higher resolution, in which measurement image or in which data set small objects with a diameter of approximately 5.0 cm can be recognized.

[0016] According to the invention this is achieved by claim 1. An embodiment of the method according to the invention is characterized in that at least one function f describing the measurement pattern for a plurality of locations on the track bed is created, which function f is described by n intrinsic functions fintrN (23) each with a wave number wl equal to n times the inverse sleeper spacing s with n=1,2,3... and a residual function frest, for which residual function n J frest = f - fintrN 1 is valid, which intrinsic functions and which residual function are created by means of a VMD algorithm, which intrinsic functions and which residual function are functions over the measurement path, the dataset being created as the sum of the n intrinsic functions provided with a factor and the residual function provided with a residual factor.

[0017] The teaching of the VMD algorithm (VMD Variational Mode Decomposition) can be applied to solve the problem of the invention. In particular, the teaching of the VMD algorithm is adapted to the technical problem at hand.

[0018] The following generally applies to the number of waves wl where s is the threshold distance and applies for n=l,2,3...

[0019] The plurality of locations can be defined by a curve or a line. In a graphic representation of the plurality of locations, the locations are arranged on this curve or line. For example, the standardized sleeper spacing at track center in Europe is 60.0 cm. For example, "Railway Geotechnics" mentions a sleeper spacing of 50.0 cm (19.5 inches) for wooden sleepers or a sleeper spacing of 60.0 cm (24.0 inches) for concrete sleepers. The function is defined with at least one intrinsic function with a wave number equal to the inverse sleeper spacing and thus with a wave number of 1 / 60.0 cm. -1 or by n intrinsic functions, each with a wave number equal to an inverse n-fold threshold distance and thus 1 / (n*60, 0) cm -1with n=l,2,3... and in any case the residual function is described (or mathematically approximated). A Fast Fourier Transformation (FFT) can be used to investigate which of the wavenumber n*l / 60, 0 cm -1 with n=1, 2, 3... are suitable. Previous unpublished experiments have shown that three intrinsic functions with a wavenumber n*1 / 60, 0 cm-1 with n=3 provide sufficiently good results.

[0020] In a curved track, the sleeper spacing in the inner area of ​​the track is smaller than in the outer area. The user can take this circumstance into account, particularly in a three-dimensional measurement of the substructure, by considering the smaller sleeper spacing for a geometric section in the inner area of ​​the curve and the larger clamp spacing for a geometric section in the outer area of ​​the curve when selecting the number of corrugations mentioned here. The user therefore selects the number of corrugations mentioned here depending on the radial position of the geometric section and thus the actual sleeper spacing, which is also a question of the accuracy of the implementation of the method according to the invention. In all cases, a number of corrugations equal to n times the inverse sleeper spacing is selected.By creating a data set as the sum of the intrinsic function with a factor or intrinsic functions with n factors and the residual function with a residual factor, the influence of the thresholds on the data set can be controlled by the choice of the mentioned factors.

[0021] The factors for the intrinsic functions can be chosen by a user for each intrinsic function and thus can be different.

[0022] By definition, a measurement pattern is determined at a location (x,y,z).

[0023] The measurement pattern can be a two-dimensional measurement pattern or a three-dimensional pattern, where, for example, a coordinate of x, y, z is constant.

[0024] The determined intrinsic functions and the residual function can be two-dimensional functions or three-dimensional functions. The disclosure of the invention focuses on two-dimensional intrinsic functions and a two-dimensional residual function.

[0025] The disclosure of the invention does not exclude three-dimensional intrinsic functions and a likewise three-dimensional residual function. The person skilled in the art is able to modify the method steps directed to two-dimensional intrinsic functions and a two-dimensional residual function in such a way that three-dimensionality is taken into account. The method according to the invention can be characterized in that the function f is described by k further intrinsic functions fintrK, each with a further wave number w2 not equal to k times the inverse threshold distance s with k=1, 2, 3..., wherein the data set is defined as a sum of the n intrinsic functions fintrN, each provided with a factor, the k further intrinsic functions fintrK, each provided with a further factor, and the further provided with a further residual factor.

[0026] Residual function is created, whereby for the further residual function applies .

[0027] In addition to the n intrinsic functions with a wave number l / (n*60, 0) cm -1 the function is described (or approximated) by a further intrinsic function or k further intrinsic functions, which intrinsic function or k intrinsic functions have different further wave numbers w2 for the wave numbers l / (n*60, 0) with n=l,2,3... The user can choose the further wave numbers according to his experience. Previous, unpublished experiments have shown that further wave numbers in a range greater than zero and less than 2*1 / 60, 0 cm -1 deliver good results. The user selects a low wavenumber, preferably close to zero, to detect non-track objects such as aerial bombs.

[0028] The remaining function can then be displayed on a monitor.

[0029] In analogy to the above description of the comparison of the residual function with reference functions, the further

[0030] The residual function can be compared with other reference functions to determine another similarity value. Analogous to the above description, another state and / or another property of the subsurface can be determined.

[0031] The further condition may be the presence of an object such as the above-mentioned aerial bomb in the underground or in a part of the underground.

[0032] The user can specify an additional wavenumber and a further wavenumber range, as well as the number of additional intrinsic functions with the additional wavenumbers in the specified additional wavenumber range. The user can select the additional wavenumber range based on their experience and / or the properties of an off-track object to be detected. The user selects the number of additional intrinsic functions depending on the additional wavenumber range. The user can select a large number of additional intrinsic functions for a large additional wavenumber range. The user can select a small number of additional intrinsic functions for a small additional wavenumber range. For example, a number of 5 to 10 additional intrinsic functions can be selected for a wavenumber range of 100 cm. -1 be considered sensible.

[0033] For the further wave number w2 the following applies in general

[0034] W2 = — and 0 < w2 < — , k*sk*s where s is the threshold distance and applies for k=l,2,3...

[0035] The above description mentions that the mentioned intrinsic functions are assigned a factor, and the other intrinsic functions are assigned a further factor. The mentioned other intrinsic functions and other intrinsic functions can be reduced or increased by the additional factor, respectively, whereby the proportion of the mentioned intrinsic functions or other intrinsic functions in the data set can be controlled as the sum of the functions. This has the further technical effect of reducing the influence of the thresholds in the data set.

[0036] The number n of intrinsic functions can be determined in a first iterative calculation procedure.

[0037] In this first iterative calculation procedure, the number n is increased until the change in the surface integral of the residual function, specifically the two-dimensional residual function, falls below a first limit specified by the user when the number n increases to n+1. For a three-dimensional residual function, a volume integral of the residual function is determined, and the number n is increased from n to n+1 until the volume integral falls below a limit specified by the user.

[0038] It is also possible to determine the change in the area integral or volume integral when the number n of the intrinsic function is changed to n+1, and increase the number from n to n+1 until the change in the area integral or volume integral falls below a user-specified limit. In other words, the number n is changed to n+1 until the area integral or volume integral approaches an area value or volume value, respectively.

[0039] The number k of further intrinsic functions can be determined in a second iterative calculation method, whereby the number n of intrinsic functions is determined from the first iterative calculation method. The number k is then increased until the change in the area integral of the further residual function when increasing from k to k+1 falls below a second limit specified by a user. In the above-mentioned iterative calculation methods, the question of the required number n of intrinsic functions and the required number k of further intrinsic functions is solved in such a way that by increasing the number n or k, the area integral or the volume integral of the residual function or the further residual function is approximated to a mathematical limit. A continuous increase in the number n or k brings this mathematical limit closer and closer to it.In order to maintain efficiency, in particular when carrying out the method according to the invention as a computer-implemented method, the first limit value and, if appropriate, the second limit value are introduced.

[0040] As already mentioned, the method according to the invention or parts of the method according to the invention can be carried out as a computer-implemented method.

[0041] The method according to the invention can be characterized in that at least one function f describing the measurement pattern for a plurality of locations of a single elevation of the track bed is created.

[0042] The locations of a single altitude can be defined via the coordinates of the location, whereby a range is specified for the z-coordinate.

[0043] The measurement pattern can be displayed on a screen and thus structured line by line or in contour lines. The area-by-area description of the measurement pattern can advantageously be adapted to this structure using the function. The function can describe the amplitudes of a line of the measurement pattern.

[0044] The following description of the definition of the mentioned

[0045] Factors is based on the basic idea of ​​reducing the influence of the thresholds on the amount of data formed by the sum of the functions.

[0046] The method according to the invention can be characterized in that the n factors for the n intrinsic functions are either equal to zero or greater than zero and less than one, the k further factors for the k further intrinsic functions are equal to one and the residual factor for the residual function is equal to one.

[0047] This definition of the aforementioned factors creates a dataset with a reduced influence of the thresholds by reducing the contribution of the intrinsic function to the dataset via the factor. In general, the n factors for the intrinsic functions are smaller than the k additional factors for the additional function and the residual factor for the residual function.

[0048] By setting a reducing factor for the intrinsic functions equal to zero, a data set is created that does not include the n intrinsic functions with n = 1, 2, 3, etc. By setting the additional factors and the residual factor unequal to zero, preferably equal to or greater than one, a data set is created that is formed by the additional intrinsic functions and the residual functions.

[0049] A mitigating factor can be defined for each intrinsic function.

[0050] The mitigating factor can be defined by the user. The user can choose the mitigating factor based on their experience.

[0051] The method according to the invention can be characterized in that the n factors for the n intrinsic functions are equal to one, the k further factors for the k further intrinsic functions are greater than one and the residual factor for the residual function is greater than one.

[0052] This definition of the mentioned factors creates a dataset with a reduced influence of the thresholds by increasing the share of the other intrinsic functions over the other factors and the share of the residual function over the residual factor in the dataset.

[0053] The other factors and the residual factor can be the same.

[0054] The method according to the invention can be characterized in that the measuring point distance between the measuring positions is less than half the threshold distance.

[0055] For example, a measuring point spacing of 12.0 cm, preferably 4.0 cm, can be selected. The measuring point spacing can be determined using a state-of-the-art displacement measuring system.

[0056] The transmitter and receiver can be arranged on a vehicle, in particular a rail vehicle, which is moving on the track in one direction of travel. The transmitter can be controlled by a control device such that the transmitter only emits a signal after the vehicle has moved on the track in the direction of travel by a measuring distance. The receiver can be controlled by the control device such that the receiver only receives a signal after the vehicle has moved by a measuring distance. The transmitter and receiver can be controlled such that the transmitter and receiver only emit or receive a signal at the measuring positions offset by the measuring distance.The method according to the invention can be characterized in that the location described by the measurement pattern comprises a three-dimensional location specification and a three-dimensional data set is created or the location described by the measurement pattern comprises a two-dimensional location specification and a two-dimensional data set is created as a geometric intersection.

[0057] The method according to the invention can be characterized in that the amount of data is displayed on a screen.

[0058] Furthermore, the aforementioned factors can be displayed. The user can preferably change the aforementioned factors independently of one another, with a change in at least one of the aforementioned factors being indicated by the data set displayed on the screen.

[0059] The user can thus control the reduction or increase of the aforementioned intrinsic functions using the factors mentioned above. As explained above, the data set can also be formed by at least one additional intrinsic function, whereby the user can define the additional wave numbers of the additional intrinsic functions. The user can control the definition of the additional wave numbers by displaying the data set, which data set is also formed by the additional intrinsic functions.

[0060] The method according to the invention can be characterized in that the measurement pattern is displayed on the screen.

[0061] The above description of the invention mentions the user's ability to control the generated data set. This controllability can be achieved, in particular, by displaying the measurement pattern and the data set. The user can thus, in particular, control the effect of the method according to the invention. In particular, the user can change the at least one reducing factor or the at least one additional wave number.

[0062] The invention disclosed here also relates to a computer program product for carrying out the method described above. The invention disclosed here also relates to a storage medium on which the computer program product for carrying out the method according to the invention is stored.

[0063] The invention disclosed here also relates to a computer program product for implementing the method described above. The signal emitted by the transmitter and the signal received by the receiver are input values ​​for the method implemented with the computer program product. Furthermore, the additional wave numbers can be input values ​​of the computer program.

[0064] The set of data created from the input values ​​is the output value of the computer program product.

[0065] The dataset generated by the method according to the invention can form the basis for an analysis of the track bed, in particular the substructure. For example, the track bed, in particular the substructure, can be analyzed based on the dataset generated by the method according to the invention using artificial intelligence.

[0066] A sample of the data set can be compared with a reference sample. Discontinuities in the data set can be identified by comparing areas of the data set.

[0067] The dataset created using the method according to the invention can form the basis for controlling an automatic or semi-automatic control system for a machine that treats the track bed. Semi-automatic control of a machine is understood to mean a control system in which a person can intervene.

[0068] The invention is further explained with reference to the following embodiments shown in the figures:

[0069] Fig. 1 shows a track vehicle with the necessary devices for an advantageous implementation of the method according to the invention,

[0070] Fig. 2 shows a measurement pattern of a track,

[0071] Fig. 3 shows the mathematical functions used to create the data set shown in Fig. 4,

[0072] Fig. 4 shows a data set created from a measurement pattern of a track,

[0073] Fig. 5 shows a measurement image according to the state of the art,

[0074] Fig. 6 shows a measurement image,

[0075] Fig. 7 shows a data set created according to the method according to the invention,

[0076] Fig. 8 illustrates the application of the method according to the invention for a line from Figure 6.

[0077] The embodiments shown in the figures merely illustrate possible embodiments. It should be noted at this point that the invention is not limited to these specifically illustrated embodiments. Combinations of the individual embodiments with one another and a combination of an embodiment with the general description above are also possible. These further possible combinations do not need to be explicitly mentioned, since these further possible combinations are within the skill of the person skilled in this technical field based on the teaching of technical action based on the present invention.

[0078] The scope of protection is determined by the claims. However, the description and drawings must be used to interpret the claims. Individual features or combinations of features from the various embodiments shown and described may represent independent inventive solutions. The problem underlying these independent inventive solutions can be derived from the description.

[0079] In the figures, the following elements are identified by the preceding reference numerals:

[0080] 1 substructure

[0081] 2-5 sleepers

[0082] 6 rail

[0083] 7 Radar device

[0084] 8 Direction of travel

[0085] 9 output signal

[0086] 10 reflected signal

[0087] 11-13 Measuring position

[0088] 14-16 additional measuring positions

[0089] 17 horizontal axis

[0090] 18 vertical axis

[0091] 19 upper area of ​​the measurement image

[0092] 20 lines

[0093] 21 Function

[0094] 22 Diagram of function

[0095] 23 (first) intrinsic function

[0096] 24 Diagram of (first) intrinsic function

[0097] 25 further first intrinsic function

[0098] 26 Diagram of further first intrinsic function

[0099] 27 further second intrinsic function

[0100] 28 Diagram of further second intrinsic function 29 Residual function

[0101] 30 diagram of residual function

[0102] 31 upper range of the data set

[0103] 32 (free)

[0104] 33 second intrinsic function

[0105] 34 Diagram of second intrinsic function

[0106] 35 third additional intrinsic function

[0107] 36 Diagram of the third intrinsic function

[0108] 37 fourth additional intrinsic function

[0109] 38 Diagram of fourth additional intrinsic function

[0110] Figure 1 shows a sectional view through a track bed. The track bed comprises a substructure 1 to be examined, sleepers 2-5, and a (visible) rail 6. A radar device 7 comprising a transmitter for emitting a signal comprising electromagnetic radiation into the substructure 1 is moved relative to the substructure 1 in a direction of travel 8. The emitted signal 9 penetrates a partial area of ​​the substructure 1 and is reflected by partial areas of the substructure 1. The radar device 7 comprises a receiver, which receives a portion of the reflected signal 10. Using current teaching, it is possible to create a measurement pattern across the substructure 1 from the signals 9, 10. Such a measurement pattern is shown, for example, in Figure 5.

[0111] According to the prior art, as shown in Figure 5 and thus deviating from Figures 2 and 6, a signal 9 is emitted and a signal 10 is received at the measuring positions 11, 12, 13 using the radar device 7. In a prior art method, the measuring positions 11, 12, 13 are located exclusively in the middle between the thresholds 2-5; for example, the measuring position 11 is in the middle between the thresholds 2, 3. In a prior art method, the measuring positions 11, 12, 13 are selected such that the thresholds 2-5 have no influence on the emitted signal 9 or the reflected signal 10. However, even a layperson recognizes that a measurement exclusively at the measuring positions 11, 12, 13 cannot provide a satisfactory measurement above the substructure 1.

[0112] Figure 2 shows a measurement pattern across substructure 1, which measurement pattern is created by arranging the radar devices at the measurement positions 11, 12, 13 shown in Figure 1 and at additional measurement positions 14-16. The additional measurement positions 14-16 are shown in Figure 1 as examples to illustrate that the measurement positions 11-16 for the measurement pattern shown in Figure 2 have a smaller distance than the threshold distance. The additional measurement positions 14-16 are selected independently of the distance of the additional measurement positions 14-16 from the thresholds 3-5.

[0113] Figure 2 shows a measurement pattern created by positioning the radar device 7 at multiple measurement positions other than the exemplary measurement positions 11-16. The radar device 7 can be positioned at a series of measurement positions with a spacing of 4.0 cm to 12.0 cm.

[0114] In Figure 2, the measurement positions 11-16 are plotted along the horizontal axis 17 of the measurement image. The path in the direction of travel 8 is plotted along the horizontal axis 17 of the measurement image. The measurement image shown in Figure 2 shows the amplitudes of the reflected rays for a traveled distance of approximately 120.0 meters.

[0115] The time span of the reflected signal from the time of output to the time of reflection is plotted on the vertical axis 18. The time span plotted on the vertical axis 18 is equivalent to the depth in the subsurface (z<0). The creation of a measurement image as shown in Figure 2 is generally known in the art. As explained above, a measurement image is created according to the state of the art by exclusively positioning the radar device 7 at measurement positions between the thresholds. In the measurement image created in Figure 2, the radar device 7 was positioned at measurement positions with a distance of less than 60.0 cm, such as a distance of 4.0 to 12.0 cm. The influence of the thresholds on the measurement image can be seen from the approximately regular pattern of amplitudes in the upper region 19 of the measurement image.This influence is also present in the areas other than the upper area 19, but is not so easily visible to the naked eye.

[0116] The measurement image shown in Figure 2 shows the amplitude of the reflected signal as a function of location (x, z). For clarity, only a two-dimensional measurement image is discussed here.

[0117] At least one function f is created that describes the measurement pattern for a multitude of locations on the track bed. The measurement image shown in Figure 2 illustrates the special case of a line-by-line setup. It is obvious that the description of the measurement pattern is adapted to this line-by-line setup, thus creating a mathematical function that describes the measurement pattern, in particular the amplitude or the change in amplitude in a line. A line of the measurement pattern encompasses a multitude of locations.

[0118] Preferably, a series of functions is created, with one function describing one line of the measurement pattern, such as line 20. However, in the following description of the figures, only the creation of one function and the processing of this function will be discussed in order to keep the description concise. Figure 3 shows an example of a function 21, which function describes the measurement image, in particular the (change in) amplitude in line 20 (see Figure 2). It is of no particular importance for the disclosure of the method according to the invention that line 20 is part of the above-mentioned upper region 19, since the method according to the invention comprises the creation of functions across all lines. Only the effects of the method according to the invention are clearly visible in this line 20, which is why the method according to the invention is discussed using line 20 as an example.

[0119] In diagram 22, the path of the reflected signal is shown on the horizontal axis and the amplitude on the vertical axis. Function 21 is a mathematical function with a multitude of wavelengths. Function 21 has different wavelengths in subranges.

[0120] Instead of Function 21, the individual measured values ​​can also be considered. The subsequent procedural steps are to be applied to the measured values ​​instead of Function 21.

[0121] Applying the theory of the VMD algorithm (VMD Variational Mode Decomposition), the function is described by an intrinsic function 23 with a wave number of 1 / 60.0. The theory of the VMD algorithm is adapted to the present problem of reducing the influence of the sleepers on the measurement image by choosing an intrinsic function with a wave number of 1 / 60.0. The peculiarity of the track bed, that the sleepers have a standardized sleeper spacing of, for example, 60.0 cm, is taken into account.

[0122] As explained above, the number of waves can be adjusted to the threshold spacing, which according to the standard is 60.0 cm. It is also possible to select a number of waves different from 1 / 60.0 if the threshold spacing differs from the standardized distance mentioned as an example.

[0123] The first intrinsic function 23 with a wave number 1 / 60,0 created using the VMD algorithm is shown in diagram 24.

[0124] Referring to the above description, it is possible to further describe function 21 by n intrinsic functions, each with a wavelength of l / (n*60, 0). The creation of these n intrinsic functions is optional and is therefore not discussed further in the figure description.

[0125] Figure 3 shows the special case where additional intrinsic functions 25, 27 are created with additional wave numbers that differ from the wave number. The creation of the additional intrinsic functions is optional but advantageous, which is why the additional intrinsic functions are shown in Figure 2, and the creation of the additional intrinsic functions is discussed in this figure description.

[0126] Using the VMD algorithm, the function 21 is described by a first further intrinsic function 25 with a first further wave number not equal to l / (n*60, 0cm) with n=l, 2, 3... The first further intrinsic function 25 is shown in diagram 26.

[0127] Using the VMD algorithm, function 21 is described by a second intrinsic function 27 with a second wave number not equal to 1 / (n*60, 0cm) with n=1, 2, 3... The first intrinsic function 27 is shown in diagram 28.

[0128] The first additional wave number and the second additional wave number are entered by the user, subject to the conditions specified above. The description section above contains a description of how to create the additional intrinsic functions. This description should be applied accordingly.

[0129] A residual function 29 is created, for which the condition defined in the claims applies. The creation of the residual function is known from the teachings of the VMD algorithm.

[0130] A data set is created for row 20 as the sum of the intrinsic function 23 with a factor, the further intrinsic functions 25, 27 with further factors and the residual function 29 with a residual factor.

[0131] As mentioned above, a function can be created for each row as a region of the measurement image. A data set can be created for each row as described above.

[0132] Figure 4 shows a dataset created using the method described above, applying the VMD algorithm. It is clearly visible to the naked eye that the influence of thresholds 2-5 is minimized, particularly in the upper region 31 of the dataset.

[0133] The data set is created by reducing the intrinsic function by a factor equal to zero. It is also conceivable to apply a factor other than zero, as mentioned in the general description. The additional factor and the residual factor are set equal to one.

[0134] Figures 2, 3, and 4 relate to a representation of signals across the track bed in a longitudinal section of the track bed. Figures 2, 3, and 4 illustrate a two-dimensional application. The method according to the invention can be applied to a three-dimensional analysis and representation of the signals. Instead of the two-dimensional function used in the presented example (intrinsic function, further intrinsic function, and residual function), three-dimensional mathematical functions are to be used.

[0135] Figure 5 shows a measurement image (or radargram) created using a state-of-the-art method. The measurement positions used to create this measurement image are spaced approximately 60 cm apart (corresponding to the threshold spacing).

[0136] Figure 6 shows a measurement image created using measured values ​​from measurement positions spaced 6.0 cm apart. The influence of the thresholds on the measurement result is clearly visible, particularly in the upper area of ​​the measurement image.

[0137] Figure 7 shows a dataset created from the measurement image shown in Figure 6 using the method according to the invention. No influence of the thresholds on the dataset is evident.

[0138] The method according to the invention is applied to all lines defining the regions of the measurement image of Figure 6; Figure 8 illustrates only the application of the method to line 20.

[0139] The function 21 (input signal) describing the amplitudes is shown in diagram 22. This function 21 (input signal) is

[0140] - a first intrinsic function 23 (IMF3) with a first wavelength 1 / (1*60, 0) (n=1, with reference to the above description),

[0141] - a second intrinsic function 33 (IMF1) with a second wavelength 1 / (2*60, 0) (n=2, with reference to the above description). The function 21 (input signal) is further described with

[0142] - another first intrinsic function 25 ( IMF2 ) with another first wavelength 2, 520m -1 and

[0143] - another second intrinsic function 27 (IMF4) with another second wavelength 1.077 m -1 ,

[0144] - a further third intrinsic function 35 (IMF5) with a further third wavelength 0.585 m -1 ,

[0145] - a further fourth intrinsic function 37 (IMF6) with a further fourth wavelength 0.155 m -1 , - a residual function 29 (residual signal). The n

[0146] Factors for the intrinsic functions 23 and 33 are set to zero to create the dataset. The k additional factors for the other intrinsic functions are set to one to create the dataset.

Claims

Patent claims 1. A method for creating a data set for examining a track bed of a track by means of electromagnetic radiation, which track bed comprises sleepers (2, 3, 4, 5) laid with a sleeper spacing s and a substructure (1) with one or more layers of ballast, which substructure (1) may include objects foreign to the track bed such as dirt or boulders or other foreign bodies, wherein at a measuring position (11-16) a signal (9) comprising electromagnetic radiation with an output frequency and an output amplitude is output into the track bed by means of a transmitter at an output time, which signal is reflected, absorbed and transmitted at a location in the track bed,wherein a reflected signal (10) comprising reflected electromagnetic rays having a reception frequency and a reception amplitude is received by means of a receiver arranged at or adjacent to the measuring position (11-16) at a reception time, wherein an output signal is output and a reception signal is measured at a series of measuring positions (11-16), which measuring positions (11-16) are spaced apart in the track direction by a measuring point distance, wherein a measuring pattern describing the change in the amplitude of the reflected signal (10) to the output signal (9) over a location (x, y, z) or the amplitude of the reflected signal (9) over a location (x, y, z) in the track bed is created, wherein the y-axis is transverse to a track direction of the track, the z-axis describes elevations of the track bed and the x-axis is oriented in the direction of the track axis, characterized in that at least one function f (21) describing the measurement pattern for a plurality of locations of the track bed is created, which function f (21) is described by n intrinsic functions fintrN (23) each with a wave number wl at the level of the n-fold inverse sleeper spacing s with n=1,2,3... and a residual function frest (29), for which residual function (29) which intrinsic functions and which residual function are created by means of a VMD algorithm, which intrinsic functions and which residual function are functions over the measurement path, whereby the data set is created as the sum of the n intrinsic functions (23) provided with a factor and the residual function (29) provided with a residual factor.

2. Method according to claim 1, characterized in that the function f is described by k further intrinsic functions fintrK (25, 27) each with a further wave number w2 not equal to the k-fold inverse threshold distance s with k=l,2,3..., which intrinsic functions and which residual function are created by means of a VMD algorithm, which intrinsic functions and which residual function are functions over the measurement path, wherein the data set is a sum of the n intrinsic functions fintrN (23) each provided with a factor, the k further intrinsic functions fintrK (25, 27) each provided with a further factor and the further residual function (29) provided with a further residual factor, whereby for the further residual function applies .

3. Method according to one of claims 1 to 2, characterized in that at least one function f (21) describing the measurement pattern for a plurality of locations of a single elevation of the track bed is created.

4. Method according to one of claims 1 to 3, characterized in that the factors for the intrinsic functions (23) are either equal to zero or greater than zero and less than one, the further factors for the further intrinsic functions (25, 27) are equal to one and the residual factor for the residual function (29) or the further residual factor for the further residual functions (29) is equal to one.

5. Method according to one of claims 1 to 3, characterized in that the n factors for the n intrinsic functions (23) are equal to one, the k further factors for the k further intrinsic functions (25, 27) are greater than one and the residual factor for the residual function (29) or the further residual factor for the further residual function (29) is greater than one.

6. Method according to one of claims 1 to 5, characterized in that the measuring point distance between the measuring positions (11-16) is less than half the threshold distance s.

7. Method according to one of claims 1 to 6, characterized in that the location described by the measurement pattern comprises a three-dimensional location specification and a three-dimensional data set is created or the location described by the measurement pattern comprises a two-dimensional location specification and a two-dimensional data set is created, a geometric intersection.

8. Method according to one of claims 1 to 7, characterized in that the data set is displayed on a screen.

9. Method according to claim 8, characterized in that the measurement pattern is displayed on the screen.

10. Computer program product for carrying out the method according to claims 1 to 9.

Images