Inertia-based method, system for detecting the bending strain of a pipeline, device and medium

Abstract

An inertial-based method and system for detecting bending strain in a pipeline, a device, and a medium. The method comprises: acquiring position information of a pipeline of a preset length, collected by an IMU (S1); performing a wavelet transformation on the position information to obtain an initial target position information (S2); performing sliding filtering on the initial target position information through a sliding window according to a preset sequence and window width, and determining a data window at each sliding of the sliding window on the initial target position information (S3); determining a fit coefficient by the minimum mean squared error depending on the respective data windows (S4).Using a numerical value at the center of each data window as a filter value, determining a filter sequence based on the respective filter values ​​and the fit coefficient, and using the filter sequence as the second target position information (S5); determining a pipe bending strain force based on the second target position information (S6); detecting the pipe bending strain based on the pipe bending strain force (S7). This method removes noise twice, thereby improving the accuracy of the position information calculated by an inertial measurement unit (IMU) and solving the problem of noise affecting the position information calculated by the IMU.

Description

Technical field

[0001] The present invention relates to the technical field of the inspection of oil and gas pipelines, in particular an inertia-based method and system for detecting a bending strain of a pipeline, a device and a medium. State of the art

[0002] Due to the specific characteristics of high pressure and long distances, pipelines can experience displacement and deformation caused by external forces as a result of geological disasters (such as earthquakes, landslides, thawing or heaving of permafrost) and other damage from third parties. These displacements and deformations subject the pipeline to both normal internal pressure loads and bending stresses. The presence of bending stresses seriously compromises the structural integrity and operational safety of the pipeline. In particular, severe defects involving bending stresses are more likely to lead to failures. Therefore, the inspection for bending stresses in long-distance pipelines has become a key focus for pipeline operators in recent years, as it is of paramount importance for accident prevention and ensuring pipeline safety.

[0003] The most important method for detecting pipeline defects is in-line inspection (IIT). Compared to external pipeline inspection, it offers advantages such as low cost, high efficiency, and reliable identification. An IIT robot uses the conveyed medium as its driving force and performs non-destructive testing of the pipeline for deformation, corrosion, cracks, and other defects. This provides a scientific basis for the operation, maintenance, and safety assessment of pipelines. The IIT robot uses an inertial measurement unit (IMU) based on strapdown inertial navigation technology to record centerline coordinates and calculate bending strain forces. However, the IIT robot's cuffs or support wheels collide with pipe walls, ring welds, spiral welds, and other features, resulting in slight oscillations and vibrations during the inspection process.The position information calculated by the IMU is affected by this noise. Disclosure of the invention

[0004] To overcome the problem that the position information calculated by the IMU is affected by noise, the present invention provides an inertia-based method and system for detecting a bending strain in a pipeline, a device, and a medium. In a first aspect, the present invention provides, to solve the aforementioned technical problem, an inertia-based method for detecting a bending strain in a pipeline, comprising the following steps: Acquiring position information of a pipeline with a preset length, collected by an inertial measurement unit (IMU); Performing a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the inertial measurement unit in order to obtain an initial target position information; Performing a sliding filter on the first target position information through a sliding window according to a preset sequence and a preset window width, and determining a data window on each sliding of the sliding window on the first target position information; Determining a fitting coefficient by the minimum mean squared error depending on the respective data windows; Using a numerical value at the center of each data window as a filter value, determining a filter sequence depending on the respective filter values ​​and the fitting coefficient, and using the filter sequence as the second target position information, wherein the second target position information is free from the influence of noise except for high-frequency noise from system noise and vibrations in the inertial measurement unit; Determining the bending strain force of the pipeline depending on the second target position information; Detecting the bending strain of the pipeline depending on the bending strain force of the pipeline.

[0005] In a second aspect, the present invention provides an inertia-based system for detecting a bending strain of a pipeline, comprising: a position information acquisition module configured to acquire position information of a pipeline of a preset length, collected by an inertial measurement unit; a module for determining initial target position information, configured to perform a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the inertial measurement unit in order to obtain initial target position information; a data window determination module configured to perform a sliding filtering of the first target position information by a sliding window according to a preset sequence and a preset window width, determining a data window on each slide of the sliding window over the first target position information; a fitting coefficient determination module configured to determine a fitting coefficient by the minimum mean squared error depending on the respective data windows; a fitting curve determination module configured to determine a fitting curve depending on each data window and fitting coefficient; a module for determining second target position information, configured to use a numerical value at the center of each data window as a filter value, determines a filter sequence depending on the respective filter values ​​and the fitting coefficient, and uses the filter sequence as second target position information, wherein the second target position information is free from the influence of noise except for high-frequency noise from system noise and vibrations in the inertial measurement unit; a pipeline bending strain force determination module configured to determine a bending strain force of the pipeline depending on the second target position information; and a detection module configured to detect bending strain of the pipeline depending on the bending strain force of the pipeline.

[0006] In a third aspect, the present invention further provides a computing device comprising a memory, a processor and a program stored in the memory and running on the processor, wherein the processor, when executing the program, implements the steps of the inertia-based method for detecting a bending strain of a pipeline as described above.

[0007] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, wherein the instructions, when executed on an end device, cause the end device to perform the steps of the inertia-based method for detecting a bending strain of a pipeline.

[0008] The inertia-based system provided by the present invention for detecting bending strain in a pipeline achieves the following advantageous effects: By performing wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the IMU, a first target position information is obtained; then, by sliding the sliding window on the first target position information, a data window is obtained, and by using the numerical value corresponding to the center point of each data window as a filter value in combination with the matching coefficient, a filter sequence is created, the filter sequence being used as the second target position information, thereby removing the influence of noise except for high-frequency noise from system noise and vibrations in the IMU;Finally, the second target position information determines the bending strain force of the pipeline, and the bending strain of the pipeline is detected depending on the bending strain force of the pipeline; the double noise removal improves the accuracy of the position information calculated by the IMU and solves the problem of the position information calculated by the IMU being affected by noise. Brief description of the drawings

[0009] To more clearly explain the technical solutions in the embodiments of the present invention or in the prior art, the present invention is further described below with reference to the attached drawings and embodiments. Fig. Figure 1 shows a schematic flowchart of an inertia-based method for detecting a bending strain of a pipeline according to an embodiment of the present invention; Fig. Figure 2 shows a schematic diagram of an SG filter principle; Fig. Figure 3 shows a diagram of the original position information acquired by an IIT robot before noise treatment; Fig. Figure 4 shows a diagram of the bending strain force calculated before noise treatment; Fig. Figure 5 shows a diagram of the target position information acquired by the IIT robot after noise removal; Fig. Figure 6 shows a diagram of the bending strain force calculated after noise treatment; Fig. Figure 7 shows a comparison diagram of the target positioning information for different filtering methods; Fig. Figure 8 shows a comparison diagram of the bending strain forces for different filter methods; Fig. Figure 9 shows a comparison diagram of the bending strain forces for directly lowered pipelines and pipelines lowered by other means before noise treatment using various filtering methods; Fig. Figure 10 shows a diagram of different shifts in the bending strain force during the method according to this embodiment; Fig. Figure 11 shows a comparison diagram of the absolute deviation of the bending strain forces for different filter methods; and Fig. Figure 12 shows a schematic representation of the structure of an inertia-based system for detecting a bending strain of a pipeline according to an embodiment of the present invention. Detailed descriptions

[0010] The following exemplary embodiments serve to further explain and supplement the present invention and do not constitute a limitation of the present invention. With reference to the accompanying figures, the inertia-based method and system for detecting a bending strain of a pipeline, the device, and the medium according to the exemplary embodiments of the present invention are described below.

[0011] As in Fig. Figure 1 shows an embodiment of the present invention providing an inertia-based method for detecting a bending strain of a pipeline, wherein the method comprises the following steps: S1: Acquiring position information of a pipeline of preset length, collected by an IMU.

[0012] The preset length can be adjusted depending on the actual situation.

[0013] S2: Performing a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the IMU in order to obtain an initial target position information.

[0014] The initial target position information is obtained via a Symlets wavelet transformation filter.

[0015] S3: Performing a sliding filter on the first target position information through a sliding window according to a preset sequence and a preset window width, and determining a data window on each sliding of the sliding window on the first target position information.

[0016] S4: Determining a fitting coefficient by the minimum mean squared error depending on the respective data windows.

[0017] S5: Using a numerical value at the center of each data window as a filter value, determining a filter sequence depending on the respective filter values ​​and the fitting coefficient, and using the filter sequence as the second target position information, wherein the second target position information is free from the influence of noise except for high-frequency noise from system noise and vibrations in the IMU.

[0018] The second target position information is obtained via an SG filter (Savitzky-Golay filter). Noise, excluding high-frequency noise from system noise and vibrations in the IMU, includes noise from ring welds and spiral welds, etc., which feeds into the position information of the ILI robot during detection over long distances and long periods.

[0019] S6: Determining a bending strain force of the pipeline depending on the second target position information.

[0020] S7: Detecting the bending strain of the pipeline depending on the bending strain force of the pipeline.

[0021] In the present embodiment, a first target position information is obtained by performing the wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the IMU; then, a data window is obtained by sliding the sliding window on the first target position information, and a filter sequence is created by using the numerical value corresponding to the center point of each data window as the filter value in combination with the matching coefficient, with the filter sequence being used as the second target position information, thereby removing the influence of noise except for the high-frequency noise from system noise and vibrations in the IMU;Finally, the second target position information determines the bending strain force of the pipeline, and the bending strain of the pipeline is detected depending on the bending strain force of the pipeline; the double noise removal improves the accuracy of the position information calculated by the IMU and solves the problem of the position information calculated by the IMU being affected by noise.

[0022] Optionally, “performing a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the IMU to obtain an initial target position information” includes: performing a continuous wavelet transformation or a discrete wavelet transformation on the position information to obtain the initial target position information.

[0023] Both the continuous wavelet transform and the discrete wavelet transform are implemented using a Symlets wavelet transform filter. The wavelet transform is a time-frequency signal analysis technique that exhibits good localization properties in both the time and frequency domains. A wavelet consists of a set of wavelet basis functions that can describe the local properties of a signal in the time and frequency domains. It can perform signal analysis at various scales and analyze the signal in any time or space domain. The frequency resolution is higher in the high-frequency part of the signal, while the low-frequency part has higher time resolution.Therefore, the information loss during the processing of signals with small discontinuities is lower, which can reduce phase distortion during signal decomposition and reconstruction, thereby removing the influence of high-frequency noise from system noise and vibrations in the IMU.

[0024] Optionally, a continuous wavelet transformation is performed on the position information to obtain the first target position information according to the following formula: CWTx(a,b)=1|a|∫−∞∞x(t)ψ*(t−ba)dt; ψ*(x)=∑k=0N−1wkϕ(2t−k) W(z)=12(1+z−1)NQ(z) where CWT x (a, b) for the first target position information, x(t) for the position information, ψ* (x) for a parent wavelet function, a for a scaling factor, b for a shift factor, N for the order of a wavelet transformation filter, w kwhere W(z) represents a coefficient of the wavelet transformation filter, ϕ(t) represents a scaling function, and Q(z) represents a preset polynomial, with the result of W(z) being the value of w k is used.

[0025] Optionally, a discrete wavelet transformation is performed on the position information to obtain the first target position information according to the following formula: DWTx(j,k)=12j∫−∞∞x(t)ψ*(t−2jk2j)dt; ψ*(x)=∑k=0N−1wkϕ(2t−k) W(z)=12(1+z−1)NQ(z) where DWT x , (j, k) represents the first target position information, where j and k each represent an integer parameter, where x(t) represents the position information, ψ* (x) represents a parent wavelet function, a represents a scaling factor, b represents a shift factor, N represents the order of a wavelet transformation filter, w kwhere W(z) represents a coefficient of the wavelet transformation filter, ϕ(t) represents a scaling function, and Q(z) represents a preset polynomial, with the result of W(z) being the value of w k is used. Regardless of whether a continuous wavelet transform or a discrete wavelet transform is used, the parent wavelet function must satisfy the following conditions: (1) Limited support: ψ(t) = 0 if t<0 or t>1. (2) Orthogonality: ∫−∞∞ψ(t)ψ(t−k)dt=δ(k), where δ(k) represents the Dirac function. (3) Mean of zero: ∫−∞∞ψ(t)dt=0. (3) High-pass filtering: ψ̂(0) = 0 , where ψ̂(f) is the Fourier transform of ψ(t).

[0026] Furthermore, symlet wavelets are derived from Daubechies wavelets, all of which are generated by a low-pass filter W, where W can be decomposed as: W(z)=12(1+z−1)NQ(z) where N represents the order of the filter and Q(z) is a polynomial. For Daubechies wavelets, the square root of the minimum phase of Q(z) is chosen as the filter coefficient, while for Symlet wavelets, the square root of the approximately linear phase of Q(z) is chosen as the filter coefficient. In this way, Symlet wavelets have better symmetry and an approximately linear phase, while Daubechies wavelets have better frequency selectivity and a faster decay rate.

[0027] Optionally, "performing a sliding filter on the first target position information using a sliding window according to a preset sequence and a preset window width, and determining a data window on each sliding of the sliding window on the first target position information" is done according to the following formula: y(n)=∑k=0Daknk; where y(n) is an nth data window, D is a total number of sliding operations of the sliding window, n k for a polynomial of order n, and a k stands for a k-th glide.

[0028] The preset order and window width can be adjusted depending on the specific situation. In this embodiment, the preset window width is 2M+1, where M represents a preset value. Large, low-order sliding windows lead to signal distortion and simultaneously broaden the absorption line type, making it difficult to retain the desired information. Small, high-order sliding windows can preserve signal information better, but their noise filtering effect is weaker. Therefore, it is necessary to select an appropriate window width.

[0029] Optionally, this includes "determining a fitting coefficient by the minimum mean squared error depending on the respective data windows": εD=∑n=−MM(∑k=0Daknk−x[n])2; where ε D for a fitting coefficient, x[n] for an actual value of the nth data window, and M for the number of data windows.

[0030] Determining the adjustment coefficient using the minimum mean squared error aims to reduce the influence of errors and improve the accuracy of the adjustment curve when weighted adjustment of the filter values ​​corresponding to the respective data windows.

[0031] As in Fig. 2 shown, the one at the center (each black dot in) Fig. 1) corresponding numerical value y(i) = a for each data window i used as a filter value, where y(i) represents a filter value and a ifor an i-th window date, where the respective filter values ​​are adjusted by the adjustment coefficients to create a filter sequence y[n] (i.e., the adjustment curve in Fig. 1) to obtain, and where y[n] is used as the second target positioning information.

[0032] Optionally, the "determination of a bending strain force of the pipeline depending on the second target position information" is carried out according to the following formula: k=kv2+kh2 kv=ΔPΔs kh=−ΔAΔscos(P) ε=D2k εv=D2kv εh=D2kh; where ε represents a bending strain force of the pipeline, where ε v and ε h for a vertical or horizontal tensile force component of the pipeline, where D represents a nominal diameter of the pipeline, where k v and k hfor a vertical curvature or a horizontal curvature of the pipeline, where k represents a total curvature of the pipeline, P a pitch angle, s a kilometer marker, and A a course angle, where k v , k h and k all originate from the second target positioning information.

[0033] Since the bending strain force of the pipeline is proportional to the change in curvature in the elastic deformation range, the curvature of the pipeline must first be calculated to determine the bending strain force. When calculating the pipeline position, in addition to the pipeline coordinates in a UTM projection (Mercator Projection System Universal Test Message), the second target position information of the IIT robot is obtained in the local coordinate system. A pitch change of the IIT robot represents a change in the tilt angle of the pipeline relative to the horizontal plane in a fixed, observed pipeline, while the heading angle represents an angle between the direction along the pipeline and north. Therefore, the following relationships exist between the curvature of the pipeline and the second target position information: k=kv2+kh2 kv=ΔPΔs kh=−ΔAΔscos(P)

[0034] The bending strain of a pipeline is directly related to its curvature. It consists of two main components: circumferential strain and longitudinal strain, the latter being further subdivided into an axial parameter and a bending parameter. Assuming the pipeline's centerline is the neutral axis, the following relationships exist between the bending strain and the centerline curvature: ε=D2k εv=D2kv εh=D2kh.

[0035] If the pipeline is subjected to a true bending strain, the area of ​​influence of its bending should be large. Features such as ring welds would cause local strain changes. Therefore, as a criterion for adaptive filtering, strain changes caused by ring welds should be filtered out. If, upon obtaining preliminary results of the bending strain, the bending strain is greater than 0.125% within a 3 m range, the order of the SG filter is changed, with the adaptive SG filter processing noisy position information until the bending strain meets the requirements.

[0036] Optionally, a specific example is used to better illustrate this embodiment: Nine pipelines, each approximately 90 m long, were installed on five support pillars, with strain gauges positioned in the center of each pipeline. The maximum strain change is evident when comparing the bending strain forces detected by the IIT robot. The bending strain forces of straight pipelines and pipelines with varying displacements were analyzed. All tested pulling speeds were 1 m / s.

[0037] As in Fig. 3 to Fig. As shown in Figure 4, the IIT robot is affected by various vibrations during an inspection process, such as those caused by features like ring welds and spiral welds. Positions like the tilt are disrupted by this noise, and the bending strain force is also inaccurate because it is calculated based on the position. During the inspection process, there are three types of noise in the IIT robot. High-frequency noise signals originate mainly from the robot's internal noise and vibrations that occur during the inspection. Small periodic peak signals are mainly due to a minor impact caused by the IIT robot as it traverses the spiral weld of the pipeline. There are also some large periodic peaks, mainly due to a significant impact caused by the IIT robot as it traverses the ring weld.Therefore, an effective method for noise suppression of the position information should be used.

[0038] If the method according to the present embodiment is used to process the original position information, as in Fig. 5 to Fig. 6 shown, is used, is from Fig. It is evident from Figure 5 that not only was the fixed-frequency noise caused by vibrations and shocks of the IIT robot due to the spiral weld during the inspection process eliminated, but also the positional changes caused by shocks as the IIT robot traversed the ring weld were reduced. The bending strain force calculated before and after filtering is shown in Figure 5. Fig. Figure 6 shows that this bending strain force eliminates the noise caused by vibrations and shocks due to weld seams in the pipeline, and the bending strain force is more accurate for straight pipelines.

[0039] To verify the proposed method, various filtering methods were compared. As in Fig. Figure 7 shows a comparison of the target position information obtained by denoising the original position information using only the wavelet transform filter and the method of the present embodiment. The comparison of the bending strain forces is shown in Fig. Figure 8 illustrates that the method of the present embodiment has a better noise reduction effect on the original position information and provides a more accurate bending strain force. Furthermore, the accuracy of the proposed method for measuring changes in bending strain force caused by flexural deformation of the pipeline is verified. Strain gauges were used to compare the detection results of an IIT robot installed in the center of a pipeline for a pull-through test. The pipeline for the pull-through test displaces different loads by 5, 10, 15, and 26.5 centimeters to simulate different degrees of flexural deformation. These different flexural deformations result in different bending strain forces in the pipeline.

[0040] Fig. Figure 9 shows a comparison of the bending strain forces for directly lowered pipelines and pipelines lowered by other means before noise reduction. Due to the noise, it is difficult to determine the true change in the bending strain force. The different shifts in the bending strain forces are shown in Fig. Figure 10 shows that the bending strain force changes accordingly when the pipeline exhibits different bending deformations.

[0041] To verify the effectiveness of the method in this embodiment, the cubic spline interpolation (CSP) filter algorithm, the symlets wavelet (SW) filter method, and the method of this embodiment were used to calculate and compare the data from the pull-through tests. The comparison of the different filter methods with respect to the bending strain force is shown in Table 1. According to Table 1, the change in strain gauge strain for a 5 cm drop in the pipe is 0.011%. Before data filtering, the strain change in bending strain force is calculated as 0.0007%, with an absolute deviation of 0.0040%. Using the three filter methods, the strain change in bending strain force is calculated as 0.015%, 0.014%, and 0.014%, respectively. The absolute deviations are 0.004%, 0.003% and 0.003% respectively.

[0042] For a pipe lowered by 10 cm, the change in strain gauge is 0.021%. Before data filtering, the strain change of the strain gauge is calculated as 0.0362%, with an absolute deviation of 0.0152%. Using the three filter methods, the strain change of the strain gauge is calculated as 0.0278%, 0.0263%, and 0.025%, respectively. The absolute deviations are 0.0068%, 0.0053%, and 0.004%, respectively.

[0043] For a pipe lowered by 15 cm, the strain gauge change is 0.032%. Before data filtering, the strain change from the pipe is calculated as 0.033%, with an absolute deviation of 0.001%. Using the three filter methods, the strain change from the pipe is calculated as 0.029%, 0.0342%, and 0.033%, respectively. The absolute deviations are 0.003%, 0.0022%, and 0.001%, respectively.

[0044] For a pipe lowered by 26.5 cm, the strain gauge reading changes by 0.055%. Before data filtering, the strain change is calculated as 0.0778%, with an absolute deviation of 0.0228%. Using the three filter methods, the strain change is calculated as 0.0712%, 0.0701%, and 0.068%, respectively. The absolute deviations are 0.0162%, 0.0151%, and 0.013%, respectively. Table 1: Comparison of different filter methods with regard to bending strain force PipelineSank Strainchanges forgaugemeasurement Before Filtered CSP SW W-SG Strainchanges AbsoluteDeviation Strainchanges AbsoluteDeviation Strainchanges AbsoluteDeviation Strainchanges AbsoluteDeviation 5 cm 0,0110% 0,0007% 0,0040% 0,0150% 0,0040% 0,0140% 0,0030% 0,0140% 0,0030% 10 cm 0,0210% 0,0362% 0,0152% 0,0278% 0,0068% 0,0263% 0,0053% 0,0250% 0,0040% 15 cm 0,0320% 0,0330% 0,0010% 0,0290% 0,0030% 0,0342% 0,0022% 0,0330% 0,0010% 26,5 cm 0,0550% 0,0778% 0,0228% 0,0712% 0,0162% 0,0701% 0,0151% 0,0680% 0,0130% AverageAbsoluteDeviation 0,0108% 0,0075% 0,0064% 0,0053%

[0045] Here, "Pipeline Sank" stands for pipeline sinking, "Strain changes for gauge measurement" for the changes in the strain gauge, "Strain changes" for the strain change in the bending strain force, "Absolute Deviation" for the absolute deviation, CSP for the cubic spline interpolation filter algorithm, SW for the Symlets-Wavelet (SW) filter method, and W-SG for the method of the present embodiment. A comparison of the absolute deviation of the bending strain force for different filter methods is described in Fig. Figure 11 shows the average absolute deviation before data filtering, where the average absolute deviation is 0.0108%. When using the three filtering methods, the average absolute deviations of the bending strain force are 0.0075%, 0.0064%, and 0.0053%, respectively, with the average relative deviations being 26.9%, 21.7%, and 18.2%, respectively. Clearly, the method of the present embodiment exhibits higher performance and provides a more accurate bending strain force. Based on pull-through tests and the results, the method proposed in the present invention can achieve good computational performance for bending strain forces.

[0046] As in Fig. As shown in Figure 12, the embodiment of the present invention further provides an inertia-based system for detecting a bending strain of a pipeline, comprising: a position information acquisition module 101 configured to acquire position information of a pipeline of preset length, collected by an inertial measurement unit; a module for determining initial target position information 102, which is configured to perform a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the inertial measurement unit in order to obtain initial target position information; a data window determination module 103, which is configured to perform a sliding filtering of the first target position information by a sliding window according to a preset sequence and a preset window width, and determines a data window on each sliding of the sliding window on the first target position information; a fitting coefficient determination module 104, which is configured to determine a fitting coefficient by the minimum mean squared error depending on the respective data windows; a module for determining second target position information 105, which is configured to use a numerical value at the center of each data window as a filter value, determines a filter sequence depending on the respective filter values ​​and the fitting coefficient, and uses the filter sequence as second target position information, wherein the second target position information is free from the influence of noise except for high-frequency noise from system noise and vibrations in the inertial measurement unit; a pipeline bending strain force determination module 106, configured to determine a bending strain force of the pipeline depending on the second target position information; and a detection module 107 that is configured to detect a bending strain of the pipeline depending on the bending strain force of the pipeline.

[0047] Optionally, the module for determining initial target position information 102 is specifically configured to, to perform a continuous wavelet transformation or a discrete wavelet transformation on the position information to obtain initial target position information. Optionally, the module for determining initial target position information 102 is specifically configured to to perform a continuous wavelet transformation on the position information to obtain an initial target position information, as shown in the following formula: CWTx(a,b)=1|a|∫−∞∞x(t)ψ*(t−ba)dt; ψ*(x)=∑k=0N−1wkϕ(2t−k) W(z)=12(1+z−1)NQ(z) where CWT x(a, b) for the first target position information, x(t) for the position information, ψ* (x) for a parent wavelet function, a for a scaling factor, b for a shift factor, N for the order of a wavelet transformation filter, w k for a coefficient of the wavelet transformation filter, ϕ(t) for a scaling function, and Q(z) represents a predefined polynomial, where the result of W(z) is the value of w k is used.

[0048] Optionally, the module for determining initial target position information 102 is specifically configured to, to perform a discrete wavelet transformation on the position information to obtain an initial target position information, as shown in the following formula: CWTx(j,k)=12j∫−∞∞x(t)ψ*(t−2jk2j)dt; ψ*(x)=∑k=0N−1wkϕ(2t−k) W(z)=12(1+z−1)NQ(z) where DWT x, (j, k) represents the first target position information, where j and k each represent an integer parameter, where x(t) represents the position information, ψ* (x) represents a parent wavelet function, a represents a scaling factor, b represents a shift factor, N represents the order of a wavelet transformation filter, w k where W(z) represents a coefficient of the wavelet transformation filter, ϕ(t) represents a scaling function, and Q(z) represents a preset polynomial, with the result of W(z) being the value of w k is used.

[0049] Optionally, the data window determination module 103 is specifically configured to, to perform a sliding filter on the first target position information using a sliding window according to a preset sequence and a preset window width, and to determine a data window on each sliding of the sliding window on the first target position information, as shown in the following formula: y(n)=∑k=0Daknk; where y(n) is an nth data window, D is a total number of sliding operations of the sliding window, n k for a polynomial of order n, and a k stands for a k-th glide.

[0050] Optionally, the adjustment coefficient determination module 104 is specifically configured to determine an adjustment coefficient by the minimum mean squared error depending on the respective data windows, which includes: εD=∑n=−MM(∑k=0Daknk−x[n])2; where ε Dfor a fitting coefficient, x[n] for an actual value of the nth data window, and M for the number of data windows.

[0051] Optionally, the pipe bending strain force determination module 106 is specifically configured to, to determine a bending strain force of the pipeline depending on the second target position information, as shown in the following formula: k=kv2+kh2 kv=ΔPΔs kh=−ΔAΔscos(P) ε=D2k εv=D2kv εh=D2kh; where ε represents a bending strain force of the pipeline, where ε v and ε h for a vertical or horizontal tensile force component of the pipeline, where D represents a nominal diameter of the pipeline, where k v and k h for a vertical curvature or a horizontal curvature of the pipeline, where k represents a total curvature of the pipeline, P a pitch angle, s a kilometer marker, and A a course angle, where k v , k h and k all originate from the second target positioning information.

[0052] In an embodiment of the present invention, a computing device is further provided comprising a memory, a processor and a program that is stored in the memory and runs on the processor, wherein, when executing the program, the processor implements part or all of the steps of the inertia-based method for detecting a bending strain of a pipeline as described above.

[0053] A computer can be chosen as the computing device; accordingly, its program is computer software. Furthermore, the individual parameters and steps in the computing device according to the present invention can be related to the individual parameters and steps in the exemplary embodiments of the inertia-based method for detecting a bending strain of a pipeline as described above, which will not be repeated here. It is known to those skilled in the art that the present invention can be implemented as a system, method, or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms: pure hardware form, pure software form (including firmware, resident software, microcode, etc.), or a combination of hardware and software, which are generally referred to here as a "circuit," "module," or "system."Furthermore, in some embodiments, the present invention can also be implemented in the form of a computer program product on one or more computer-readable media, wherein the computer-readable medium contains computer-readable program code. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or means, or any combination thereof.

[0054] In this description, references to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., mean that the specific features, structures, materials, or special characteristics described in connection with that embodiment or example are included in at least one embodiment or example of the present invention. In this description, a schematic formulation of the above terms does not necessarily refer to the same embodiment or example. Furthermore, the described specific features, structures, materials, or special characteristics can be suitably combined in one or more embodiments or examples. Moreover, the person skilled in the art can distinguish the various embodiments or examples described in this description.Combine examples and their characteristics, as long as they do not contradict each other.

[0055] Although the embodiments of the present invention have been shown and described above, it should be understood that the embodiments mentioned above are exemplary and should not be construed as limiting the present invention. The person skilled in the art can make changes, modifications, substitutions, and alterations to the embodiments mentioned above within the scope of the present invention.

Claims

[1] Inertia-based method for detecting bending strain in a pipeline, characterized by the following steps: Capturing position information of a pipeline with a preset length, collected by an inertial measurement unit; Performing a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the inertial measurement unit in order to obtain an initial target position information; Performing a sliding filter on the first target position information through a sliding window according to a preset sequence and a preset window width, and Determine a data window on each slide of the sliding window on the first target position information; Determining a fitting coefficient by the minimum mean squared error depending on the respective data windows; Using a numeric value in the center of each data window as a filter value, Determining a filter sequence depending on the respective filter values ​​and the matching coefficient, and using the filter sequence as the second target position information, wherein the second target position information is free from the influence of noise except for high-frequency noise from system noise and vibrations in the inertial measurement unit; Determining the bending strain force of the pipeline depending on the second target position information; Detecting the bending strain of the pipeline depending on the bending strain force of the pipeline. [2] Method according to claim 1, characterized by, that includes “performing a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the inertial measurement unit in order to obtain an initial target position information”: Performing a continuous wavelet transformation or a discrete wavelet transformation on the position information to obtain the first target position information. [3] Method according to claim 2, characterized by , that the “performing a continuous wavelet transformation on the position information to obtain the first target position information” is done according to the following formula: CWTx(a,b)=1|a|∫−∞∞x(t)ψ*(t−ba)dt; ψ*(x)=∑k=0N−1wkϕ(2t−k) W(z)=12(1+z−1)NQ(z) where CWT x(a, b) for the first target position information, x(t) for the position information, ψ* (x) for a parent wavelet function, a for a scaling factor, b for a shift factor, N for the order of a wavelet transformation filter, w k where W(z) represents a coefficient of the wavelet transformation filter, ϕ(t) represents a scaling function, and Q(z) represents a preset polynomial, with the result of W(z) being the value of w k is used. [4] Method according to claim 2, characterized by , that the “performance of a discrete wavelet transformation on the position information to obtain the first target position information” is carried out according to the following formula: DWTx(j,k)=12j∫−∞∞x(t)ψ*(t−2jk2j)dt; ψ*(x)=∑k=0N−1wkϕ(2t−k) W(z)=12(1+z−1)NQ(z) where DWT x(j, k) represents the first target position information, where j and k each represent an integer parameter, where x(t) represents the position information, ψ* (x) represents a parent wavelet function, a represents a scaling factor, b represents a shift factor, N represents the order of a wavelet transformation filter, w k where W(z) represents a coefficient of the wavelet transformation filter, ϕ(t) represents a scaling function, and Q(z) represents a preset polynomial, with the result of W(z) being the value of w k is used. [5] Method according to claim 1, characterized by , that “performing a sliding filter on the first target position information by means of a sliding window according to a preset sequence and a preset window width, and determining a data window on each sliding of the sliding window on the first target position information” is done according to the following formula: y(n)=∑k=0Daknk; where y(n) is an nth data window, D is a total number of sliding operations of the sliding window, n k for a polynomial of order n, and a k stands for a k-th glide. [6] Method according to claim 5, characterized by , that includes “determining a fitting coefficient by the minimum mean squared error depending on the respective data windows”: εD=∑n=−MM(∑k=0Daknk−x[n])2; where ε D for a fitting coefficient, x[n] for an actual value of the nth data window, and M for the number of data windows. [7] Method according to any one of claims 1 to 5, characterized by , that the “determination of a bending strain force of the pipeline depending on the second target position information” is carried out according to the following formula: k=kv2+kh2 kv=ΔPΔs kh=−ΔAΔscos(P) ε=D2k εv=D2kv εh=D2kh; where ε represents a bending strain force of the pipeline, where ε v and ε h for a vertical or horizontal tensile force component of the pipeline, where D represents a nominal diameter of the pipeline, where k v and k h for a vertical curvature or a horizontal curvature of the pipeline, where k represents the total curvature of the pipeline, P a pitch angle, s a kilometer marker, and A a course angle. where k v , k h and k all originate from the second target positioning information. [8] Inertia-based system for detecting bending strain of a pipeline, characterized by , that it includes: a position information acquisition module configured to acquire position information of a pipeline of a preset length, collected by an inertial measurement unit; a module for determining initial target position information, configured to perform a wavelet transformation on the position information to remove the influence of high-frequency noise from system noise and vibrations in the inertial measurement unit in order to obtain initial target position information; a data window determination module configured to perform a sliding filtering of the first target position information by a sliding window according to a preset sequence and a preset window width, determining a data window on each slide of the sliding window over the first target position information; a fitting coefficient determination module configured to determine a fitting coefficient by the minimum mean squared error depending on the respective data windows; a module for determining second target position information, configured to use a numerical value at the center of each data window as a filter value, determines a filter sequence depending on the respective filter values ​​and the fitting coefficient, and uses the filter sequence as second target position information, wherein the second target position information is free from the influence of noise except for high-frequency noise from system noise and vibrations in the inertial measurement unit; a pipeline bending strain force determination module configured to determine a bending strain force of the pipeline depending on the second target position information; and a detection module configured to detect bending strain of the pipeline depending on the bending strain force of the pipeline. [9] Computing device comprising a memory, a processor and a program stored in the memory and running on the processor, characterized by , that the processor, when executing the program, implements the steps of the inertia-based method for detecting a bending strain of a pipeline according to one of claims 1 to 7. [10] Computer-readable storage medium, characterized by that instructions are stored in the computer-readable storage medium, wherein, when executed on an end device, the instructions cause the end device to perform the steps of the inertia-based method for detecting a bending strain of a pipeline according to any one of claims 1 to 7.

Images