Pipe support risk assessment method based on multi-type detection

Abstract

The application discloses a pipeline support danger assessment method based on multiple type detection, which comprises the following steps: numbering the pipeline support, using an ultrasonic flaw detector to perform ultrasonic detection on the non-destructive bolts; analyzing the stress of the pipeline support based on finite element simulation to obtain the maximum stress and the maximum displacement generated by the pipeline support; processing the pipeline support pictures based on image recognition to obtain the number of cracks of the pipeline support; performing non-destructive detection on the bolts using the ultrasonic flaw detector to obtain the number of damaged bolts; assessing the danger of each pipeline support, and assessing the danger of the entire pipeline support section according to the assessment results. The application can obtain the relevant data of the pipeline support by using multiple methods such as finite element simulation, image recognition and ultrasonic detection, comprehensively assess the danger of the pipeline support, improve the accuracy of the detection and assessment work, and eliminate the safety hazards.

Description

TECHNICAL FIELD

[0001] The present application relates to a pipeline support risk assessment method, and particularly relates to a pipeline support risk assessment method based on multi-type detection. BACKGROUND

[0002] In the oil and gas industry, pipeline supports serve as supporting components of pipelines and play a very important role in safe operation of the pipelines. Since the state of the pipeline supports is changing all the time after the pipeline supports are built, the state of the pipeline supports should be detected and evaluated in a timely manner. At present, detection of the pipeline supports depends on manual measurement, which requires a large amount of manpower and material resources, and not only the efficiency of detection and evaluation is low, but also the accuracy of the obtained data cannot be improved. SUMMARY

[0003] The present application aims at overcoming the deficiencies of the prior art and providing a pipeline support risk assessment method based on multi-type detection. The method can improve the accuracy of detection and evaluation and eliminate safety hazards.

[0004] To achieve the above object, the technical scheme adopted by the present application is as follows:

[0005] The pipeline support risk assessment method based on multi-type detection comprises the following steps:

[0006] Step 1: Number all the pipeline supports in a whole pipe section from 1 to p and count all the types of bolts used in the pipe section, use an ultrasonic detection probe of an industrial ultrasonic flaw detector to perform ultrasonic detection on all the types of non-destructive bolts, and obtain normal echo signals corresponding to the bolts of various types;

[0007] Step 2: Establish a record table of pipeline support information of data software, and record the data of the pipeline supports in the record table;

[0008] Step 3: Based on finite element simulation, analyze the load-bearing performance of each pipeline support on a pipe section respectively, and obtain the maximum stress and the maximum displacement generated by each pipeline support;

[0009] Step 4: Based on image recognition, use a camera to take a picture of each plane of each pipeline support respectively, then use data processing software to process the pictures of the pipeline supports, obtain the total number of cracks of each pipeline support, and record in the record table;

[0010] Step five, use the ultrasonic detection probe of the industrial ultrasonic flaw detector to nondestructively detect the bolts connected to each pipe support, the ultrasonic detection probe transmits the echo signal to the flaw detector, the echo signal is displayed on the oscilloscope screen of the flaw detector, by comparing the difference with the nondestructive bolt echo signal, judge whether the internal bolt of each pipe support is damaged, get the total number of damaged bolts of each pipe support, and record in the record table;

[0011] Step six, query the allowable stress [σ] corresponding to each pipe support q And allowable displacement [Δx] q , q=1, 2, …, p, q represents the number of pipe supports, p represents the total number of pipe supports, and record in the record table;

[0012] Step seven, establish a pipe support risk rating table in the data software, the pipe support risk rating table includes pipe support number, bearing capacity rating, surface quality rating, bolt damage rating, and support risk rating, then the risk of each pipe support is evaluated and recorded in the pipe support risk rating table;

[0013] Step eight, set a threshold, compare the total number of supports with different support risk rating in the pipe support risk rating table with the threshold, and evaluate the risk of the support section of all pipe supports constituting the entire pipe section.

[0014] The beneficial effects of the present application are: by using finite element simulation, image recognition and ultrasonic detection and other methods, the related data of the pipe support can be obtained, the risk of the pipe support can be comprehensively evaluated, the accuracy of the detection and evaluation work can be improved, and the safety hidden danger can be eliminated. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 A flow chart of a pipe support risk assessment method based on multiple type detection according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] The present application will be described in detail below in conjunction with the drawings and specific embodiments.

[0017] As Figure 1 shown, a pipe support risk assessment method based on multiple type detection according to the present application includes the following steps:

[0018] Step one, number all the pipe supports in the entire pipe section from 1 to p, and count all the bolt types used in the pipe section, use the ultrasonic detection probe of the industrial ultrasonic flaw detector to ultrasonically detect all types of nondestructive bolts, and obtain the normal echo signals corresponding to various types of bolts;

[0019] Step two, establish a record table of data software pipeline support information, record the data of the pipeline support in the record table;

[0020] Step three, based on finite element simulation, the carrying capacity of each pipeline support on a pipe section is analyzed respectively, and the maximum stress and the maximum displacement of each pipeline support are obtained, the specific method is as follows:

[0021] First, install a force sensor on the No. 1 pipeline support and connect it to the computer, then use the force sensor to measure the load on the pipeline support for a period of time, the force sensor transmits the load data of the pipeline support to the computer, and selects the maximum load value F max Recorded, the period of time includes the force of the pipeline support under all working conditions; the all working conditions usually include pipeline normal smooth running, pipeline start-stop and other working conditions;

[0022] Second, create a three-dimensional pipeline support model in three-dimensional software, then import the three-dimensional pipeline support model into finite element analysis software, assume that the load on the pipeline support is F max , simulate and analyze the pipeline support to obtain the maximum stress value σ and the maximum displacement Δx of the pipeline support;

[0023] Third, repeat the first to third steps for all pipeline supports with other numbers in the entire pipe section to obtain the maximum stress value σ q and the maximum displacement Δx q of each pipeline support, and record them in the record table, q = 1, 2, …, p, q represents the number of the pipeline support, and p represents the total number of the pipeline support.

[0024] Step four, based on image recognition, use a camera to take a picture of each plane of each pipeline support, then use data processing software to process the pipeline support pictures to obtain the total number of cracks of each pipeline support, and record them in the record table, the specific method is as follows:

[0025] First, use a camera to take a picture of each plane of the No. 1 pipeline support, and mark the pictures from 1 to n in the order of shooting;

[0026] Second, import the first pipeline support picture into the data processing software, the R, G, B values of each pixel point in the picture are respectively recorded as R(x, y), G(x, y), and B(x, y), where x and y represent the horizontal and vertical coordinates of the pixel point in the picture respectively;

[0027] Third, the pipeline support picture is processed by gray scale, the gray value I(x, y) of each pixel point of the pipeline support picture is calculated as follows:

[0028] I(x, y) = 0.299R(x, y) + 0.587G(x, y) + 0.114B(x, y)

[0029] Let each pixel point R(x, y) = G(x, y) = B(x, y) = I(x, y) can be obtained pipe support gray picture;

[0030] Fourth, the gray processing of the pipe support picture is binarized, first, in the data processing software, the graythresh function is called to automatically calculate the best threshold I of the pipe support gray picture 最佳 ; Then, compare the relationship between each pixel point gray value I(x, y) and I 最佳 in the pipe support gray picture, if I(x, y) > I 最佳 , then I(x, y) is set to 1, otherwise, it is set to 0, that is, the binarized picture is obtained;

[0031] Fifth, the connected domain information in the binarized picture is extracted, first, in the data processing software, the bwlabel function is called, and the connected domain of the binarized picture is found in the 8-connected way, that is, the region whose I(x, y) = 1 is connected, and the number of connected domains k is obtained; Finally, the regionprops function is called to obtain the information of each connected domain, including the length l i , width b i and other information, wherein i represents the ith connected domain, i = 1, 2, …, k;

[0032] Sixth, the binarized picture is denoised, and other noise is filtered out. Since the crack shape is long and thin, the connected domain has the characteristics of large length and small width, so the minimum length-width ratio of the connected domain is set to T, and the non-crack connected domain is removed. The value of T can be adjusted according to the actual situation. The length-width ratio T i of each connected domain is calculated i / b i , wherein i represents the ith connected domain, i = 1, 2, …, k; compare the relationship between each connected domain T i and T, if T i > T, then the gray value I(x, y) of all pixel points in the ith connected domain is set to 1, otherwise, the gray value I(x, y) of all pixel points in the ith connected domain is set to 0, and finally the denoised crack binarized picture is obtained;

[0033] Seventh, repeat the second to sixth steps, and in the data processing software, the other pictures of the pipe support numbered 1 are imported in turn to obtain the denoised crack binarized picture of the other pictures of the pipe support, and the number of cracks in each crack binarized picture is counted, which is denoted as N jLet j = 1, 2, ..., n, where j represents the image number of the pipe support and n represents the total number of images of the pipe support. Calculate the total number of cracks in the pipe support.

[0034] Step 8: Repeat steps 1 through 8 for all other pipe supports with different numbers throughout the entire pipe section to obtain the total number of cracks N for each pipe support. q q = 1, 2, ..., p, where q represents the pipe support number and p represents the total number of pipe supports, and these are recorded in the record table.

[0035] Step 5: Use the ultrasonic testing probe of an industrial ultrasonic flaw detector to perform non-destructive testing on the bolts connected to each pipe support. The ultrasonic testing probe transmits the echo signal to the flaw detector, and the echo signal is displayed on the oscilloscope screen of the flaw detector. By comparing the difference with the echo signal of the undamaged bolt, it is determined whether the bolts on each pipe support are damaged inside, and the total number of damaged bolts of each pipe support is obtained and recorded in the record table.

[0036] In this step, the ultrasonic testing probe acts as both a transmitter and a receiver, intermittently emitting ultrasonic waves in a pulsed manner. The industrial ultrasonic flaw detector can be the NDT610 model from Beijing Keda Instrument Technology Co., Ltd., which consists of an ultrasonic testing probe and a flaw detector. The specific method for this step is as follows:

[0037] Step 1: Number the bolts on pipe support No. 1 from 1 to m, and use an ultrasonic testing probe to test them in sequence. If the echo signal of a bolt is different from the normal signal, record the bolt number; otherwise, do not record it.

[0038] The second step is to count the number of damaged bolts, denoted as M.

[0039] Step 3: Repeat steps 2 and 3 for all other pipe supports with different numbers throughout the entire pipe section to obtain the total number of damaged bolts M for each pipe support. q q = 1, 2, ..., p, where q represents the pipe support number and p represents the total number of pipe supports, and these are recorded in the record table.

[0040] Step Six: Query (e.g., by consulting "Mechanical Design Handbook," Volume 1, edited by Cheng Daxian, 5th edition, Beijing: Chemical Industry Press, November 2007) the allowable stress [σ] for each pipe support. q Allowable displacement [Δx] q q = 1, 2, ..., p, where q represents the pipe support number and p represents the total number of pipe supports, and these are recorded in the record table.

[0041] Step seven, establish a pipeline support risk rating table in the data software, the pipeline support risk rating table includes pipeline support number, bearing capacity rating level, surface quality rating level, bolt damage rating level, support risk rating level, then the risk of each pipeline support is evaluated and recorded in the pipeline support risk rating table, the specific method of evaluation is as follows:

[0042] First step, pipeline support bearing capacity rating level: first, query the maximum stress value σ q , maximum displacement Δx q , allowable stress value [σ] q , allowable displacement [Δx] q of each pipeline support in the pipeline support information record table; second, judge the bearing capacity level of each support, if σ q <[σ] q and Δx q <[Δx] q , the pipeline support bearing capacity rating is a level, a level indicates that the bearing capacity is sufficient; otherwise, the pipeline bearing capacity rating is c level, c level indicates that the bearing capacity is insufficient; finally, record the bearing capacity rating level of each support in the pipeline support risk rating table;

[0043] Second step, pipeline support surface quality rating level: first, query the number of pipeline support cracks N q of each pipeline support in the record table; second, judge the surface quality rating level of each support, if N q <1, the surface quality rating is a level, a level indicates that the support surface has no cracks; if 1≤N q <3, the surface quality rating is b level, b level indicates that the support surface has a small number of cracks; if N q ≥3, the rating is c level, c level indicates that the support surface has multiple cracks; finally, record the surface quality rating level of each support in the pipeline support risk rating table;

[0044] Third step, pipeline support bolt damage rating level: first, query the number of bolt damage M q of each pipeline support in the record table; second, judge the bolt damage rating level of each support, if M q <1, the bolt damage rating is a level, a level indicates that the support bolt has no damage; if 1≤M q <3, the bolt damage rating is b level, b level indicates that a small number of bolts have internal damage; if M q ≥3, the bolt damage rating is c level, c level indicates that multiple bolts have internal damage; finally, record the bolt damage rating level of each support in the pipeline support risk rating table;

[0045] Fourth step, the risk assessment level of the pipe support: for any pipe support, if the bearing capacity assessment level, surface quality assessment level, bolt damage assessment level are the same, then the rating is the risk assessment level of the pipe support; if the bearing capacity assessment level, surface quality assessment level, bolt damage assessment level are different, then the highest risk rating among them is taken as the risk assessment level of the pipe support, the risk from high to low is divided into c, b, a; use this method to assess the risk level of all pipe supports, and record it in the pipe support risk assessment level table;

[0046] Fifth step, if the pipe support risk assessment is a level, it means that the support bearing capacity is sufficient, there is no damage, and no further inspection is needed; if the pipe support risk assessment is b level, it means that the support has some problems, and further inspection of the pipe support is needed before determining whether to take measures to maintain the support; if the pipe support risk assessment is c level, it means that the support has multiple problems, and measures need to be taken to maintain the support.

[0047] Step eight, set a threshold, compare the total number of supports with different support risk assessment levels in the pipe support risk assessment table with the threshold, and assess the risk of the support section of all pipe supports in the entire pipe section, the specific method is as follows:

[0048] First step, count the number of a level pipe supports in the pipe support risk assessment table, denoted as X, the number of b level pipe supports, denoted as Y, and the number of c level pipe supports, denoted as Z;

[0049] Second step, if Z = 0 and then the support section is rated A; if Z = 0 and then the support section is rated B; if Z ≠ 0 and then the support section is rated B, and other conditions are rated C, p represents the total number of pipe supports;

[0050] Third step, if the support section is rated A, the overall performance of the support section is good, and individual supports may need to take appropriate measures; B level: slightly lower than the normal use requirement, part of the support should take measures; C level: measures should be taken to maintain the entire support section.

Claims

1. A method for assessing the risk of a pipe support based on multi-type detection, characterized by The method comprises the following steps: Step one, number all pipe supports from 1 to p in the whole pipe section, and count all bolt types used in the pipe section, use the ultrasonic detection probe of the industrial ultrasonic flaw detector to perform ultrasonic detection on the non-destructive bolts of all types, and obtain the normal echo signals corresponding to the bolts of various types; Step two, establish a record table of pipe support information in the data software, and record the data of the pipe support in the record table; Step three, based on finite element simulation, analyze the bearing performance of each pipe support on a pipe section respectively, and obtain the maximum stress and the maximum displacement generated by each pipe support; Step four, based on image recognition, use a camera to take a picture of each plane of each pipe support respectively, then use data processing software to process the pipe support pictures, obtain the total number of cracks of each pipe support, and record them in the record table; Step five, use the ultrasonic detection probe of the industrial ultrasonic flaw detector to perform non-destructive detection on the bolts connected to each pipe support, the ultrasonic detection probe transmits the echo signal to the flaw detector, and the echo signal is displayed on the oscilloscope screen of the flaw detector, by comparing the difference between the echo signal and the normal signal, it is judged whether the internal bolt of each pipe support is damaged, and the total number of damaged bolts of each pipe support is obtained and recorded in the record table; Step six, query each pipe support corresponding allowable stress [σ] q and allowable displacement [Δx] q , q = 1, 2, …, p, q represents the number of pipe supports, p represents the total number of pipe supports, and is recorded in the record table; Step seven, establish a pipe support risk rating table in the data software, the pipe support risk rating table includes pipe support number, bearing capacity rating, surface quality rating, bolt damage rating, and support risk rating, then the risk of each pipe support is evaluated and recorded in the pipe support risk rating table; Step eight, set a threshold, compare the total number of supports with different support risk rating levels in the pipe support risk rating table with the threshold respectively, and evaluate the risk of the support section of all pipe supports constituting the whole pipe section.

2. The multi-type detection based pipe support hazard assessment method of claim 1, wherein: The specific method of step three is as follows: Firstly, install the force sensor on the pipe support numbered 1 and connect it with the computer, then use the force sensor to measure the load of the pipe support in a period of time, the force sensor transmits the load data of the pipe support to the computer, and select the maximum load value F from the measured load data max Record the load of the pipe support in a period of time, which includes the force of the pipe support in all working conditions, including the normal and smooth running of the pipe, the starting and stopping of the pipe Second step, create a three-dimensional pipe support model in three-dimensional software, then import the three-dimensional pipe support model into the finite element analysis software, assuming that the load on the pipe support is F max The maximum stress value σ and the maximum displacement Δx of the pipe support are obtained by simulating and analyzing the pipe support. Third step, repeat the first step to the third step for all pipe supports of other numbers in the whole pipe section, to get the maximum stress value σ of each pipe support q and the maximum displacement Δx q and record in the record table, q = 1, 2, …, p, q represents the number of pipe supports, and p represents the total number of pipe supports.

3. The multi-type detection based pipe support hazard assessment method of claim 1, wherein: The specific method of step four is as follows: First step, use a camera to take a picture of each plane of the pipe support numbered 1, and mark the pictures from 1 to n in the order of shooting; Second step, import the first pipe support picture into the data processing software, and record the R, G, and B values of each pixel point in the picture as R(x, y), G(x, y), and B(x, y) respectively, wherein x and y represent the horizontal and vertical coordinates of the pixel point in the picture respectively; Third step, perform grayscale processing on the pipe support picture, and the grayscale value I(x, y) of each pixel point in the pipe support picture is calculated according to the following formula: I(x, y) = 0.299R(x, y) + 0.587G(x, y) + 0.114B(x, y) Let R(x, y) = G(x, y) = B(x, y) = I(x, y) for each pixel point, and the grayscale picture of the pipe support can be obtained; Fourthly, the gray-scale image of the pipe support is binarized. Firstly, the graythresh function is called in the data processing software to automatically calculate the optimal threshold I 最佳 of the gray-scale image of the pipe support. Then, the relationship between the gray-scale value I(x, y) of each pixel in the gray-scale image of the pipe support and I 最佳 is compared. If I(x, y) > I 最佳 , I(x, y) is set as 1, otherwise, it is set as 0, and the binarized image is obtained. In the fifth step, the information of the connected domain in the binary image is extracted. First, in the data processing software, the bwlabel function is called to find the connected domain of the binary image in the 8-connected mode, that is, to find the region with I(x, y) = 1 and connected, and the number k of the connected domains is obtained. Finally, the regionprops function is called to obtain the information of each connected domain, and the information includes the length l and the width b of the circumscribed rectangle i i of the connected domain, where i represents the i-th connected domain, i = 1, 2, …, k.​ Step 6, set the minimum aspect ratio of the connected domain as T, remove the non-crack connected domain, and calculate the aspect ratio T of each connected domain i = l i / b i , wherein i represents the i-th connected domain, i = 1, 2, …, k; compare the relationship between each connected domain T i and T, if T i > T, set the gray value I(x, y) of all pixel points in the i-th connected domain to 1, otherwise, set the gray value I(x, y) of all pixel points in the i-th connected domain to 0, and finally obtain the denoised crack binary picture; Step 7, repeat steps 2 to 6, import other pictures of the No. 1 pipeline support into the data processing software in sequence to obtain the crack binaryzation pictures of the other pictures of the pipeline support after noise reduction, and count the number of cracks in each crack binaryzation picture, denoted as N j , j = 1, 2, …, n, j represents the picture number of the pipeline support, and n represents the total number of pictures of the pipeline support, and the total number of cracks of the pipeline support is calculated Step 8, repeat the first step to the eighth step for all the other numbered pipe supports of the whole pipe section to get the total number of cracks N of each pipe support q , q = 1, 2, …, p, q represents the number of pipe supports, p represents the total number of pipe supports, and is recorded in the record table.

4. The multi-type detection based pipe rack hazard assessment method of claim 1, wherein: The specific method of step five is as follows: First step, number the bolts on the pipe support numbered 1 from 1 to m, and use the ultrasonic detection probe to detect them in sequence, if the echo signal of the bolt is different from the normal signal, record the number of the bolt, otherwise, do not record; Second step, count the number of damaged bolts, recorded as M; Third step, repeat the second step to the third step to all pipe supports of other numbers of the whole pipe section, get the total number of damaged bolts M of each pipe support q , q = 1, 2, …, p, q represents the number of pipe supports, p represents the total number of pipe supports, and is recorded in the record table.

5. The multi-type detection based pipe rack hazard assessment method according to claim 1, wherein: The specific method of step seven is as follows: First, in the pipe support information record table to query each pipe support maximum stress value σ q , the maximum displacement Δx q , allowable stress value [σ] q , allowable displacement [Δx] q ; second, to determine the carrying capacity of each support level, if σ q <[σ] q and Δx q <[Δx] q , the pipe support carrying capacity rating a level, a level indicates that the carrying capacity is sufficient; otherwise the pipe carrying capacity rating for c level, c level indicates that the carrying capacity is insufficient; finally, the carrying capacity of each support rating level recorded in the pipe support risk rating table; Second step, surface quality rating of pipe support: first, in the record table to query each pipe support pipe support crack number N q ; second, to determine the surface quality rating of each support, if N q <1, the surface quality rating is a, a indicates that the support surface has no cracks; if 1≤N q <3, the surface quality rating is b, b indicates that the support surface has a small number of cracks; if N q ≥3, the rating is c, c indicates that the support surface has multiple cracks; finally, the surface quality rating of each support is recorded in the pipe support risk rating table; Third step, the bolt damage assessment level of the pipe support: first, in the record table to query the number of bolt damage M of each pipe support q ; second, to determine the bolt damage assessment level of each support, if M q <1, the bolt damage assessment is a level, a level indicates that the support bolt has no damage; if 1≤M q <3, the bolt damage assessment is b level, b level indicates that a small amount of bolt has internal damage; if M q ≥3, the bolt damage assessment is c level, c level indicates that multiple bolts have internal damage; finally, the bolt damage assessment level of each support is recorded in the pipe support risk rating table; Fourth step, evaluate the risk level of the pipeline support: for any pipeline support, if the bearing capacity evaluation level, surface quality evaluation level, and bolt damage evaluation level are the same, the evaluation level is the risk evaluation level of the pipeline support; if the bearing capacity evaluation level, surface quality evaluation level, and bolt damage evaluation level are different, the highest risk evaluation level among them is taken as the risk evaluation level of the pipeline support, and the risk is divided into c, b, and a from high to low; use this method to evaluate the risk level of all pipeline supports, and record it in the pipeline support risk evaluation level table; Fifth step, if the pipeline support risk evaluation is a level, it means that the support bearing capacity is sufficient, there is no damage, and no further inspection is needed; if the pipeline support risk evaluation is b level, it means that the support has some problems, and further inspection of the pipeline support is needed before determining whether to take measures to maintain the support; if the pipeline support risk evaluation is c level, it means that the support has multiple problems, and measures need to be taken to maintain the support.

6. The multi-type detection based pipe rack hazard assessment method according to claim 5, characterized in that: The specific method of step eight is as follows: First step, count the number of pipeline supports with a level risk evaluation in the pipeline support risk evaluation level table, recorded as X, the number of pipeline supports with b level risk evaluation, recorded as Y, and the number of pipeline supports with c level risk evaluation, recorded as Z; Second step, if Z = 0 and then the bracket section is rated as A level; if Z = 0 and then the bracket section is rated as B level; if Z ≠ 0 and then the bracket section is rated as B level, and the other cases are rated as C level, and p represents the total number of pipe supports. Third step, if the support section is evaluated as A level, the overall performance of the support section is good, and appropriate measures may be taken for individual supports; B level: slightly lower than the normal use requirement, and some supports should take measures; C level: measures should be taken to maintain the entire support section.

Images