A storage tank intelligent detection data analysis method and detection system

Abstract

The invention discloses an intelligent detection data analysis method and a detection system for a storage tank, which relate to the technical field of storage tank detection, and comprise the following steps: based on an acoustic emission detection device, all acoustic emission information on the inner bottom of the storage tank within a first time is acquired, the acoustic emission information comprises an acoustic emission source and an acoustic emission signal waveform, which are respectively marked as a detection acoustic emission source and a detection waveform; acoustic emission signal waveforms refracted by a first number of welds are acquired, which are marked as historical waveforms, and a classification range is obtained based on historical waveform calculation; based on the classification range, it is judged whether the detection waveform is an acoustic emission signal waveform caused by a weld, and if so, the corresponding acoustic emission information is deleted; based on the remaining acoustic emission information, it is judged whether the inner bottom of the storage tank is corroded, and if so, a maintenance warning is issued; the invention is used to solve the problem of inaccurate judgment of corrosion defects at the bottom of the storage tank due to the influence of welds and the environment in the existing storage tank detection technology.

Description

Technical Field

[0001] The present invention relates to the technical field of storage tank detection, and in particular to a storage tank intelligent detection data analysis method and detection system. Background Art

[0002] Existing storage tanks are made of steel. After long-term use, the tank body will corrode, which will affect normal use and safety. Therefore, it is necessary to conduct corrosion detection on the tank. However, it is difficult and complicated to detect corrosion on the bottom of large storage tanks.

[0003] Existing corrosion detection technologies for the bottom of the inner side of storage tanks include ultrasonic detection, magnetic flux leakage detection, and acoustic emission detection. These detection technologies are not easy to determine whether it is a corrosion situation when a weld appears on the bottom of the tank or other environmental influences occur during corrosion detection. In particular, acoustic emission detection is used because the acoustic emission waves generated by corrosion will produce sound sources when passing through the weld during propagation. Therefore, it is easy to regard the weld as a corrosion point. For example, in the patent application with application publication number CN108844963A, an online monitoring system and method for corrosion defects of the bottom plate of a large storage tank are disclosed. This scheme uses acoustic emission for corrosion defect detection. However, if a weld appears on the bottom of the tank and the weld material is different or the weld is not smooth, bubbles will also appear when the ultrasonic guided wave emitted by the excitation device passes through the weld. Therefore, it is not easy to distinguish between the bubbles generated by the weld and the corrosion defect. In the existing tank detection technology, due to the influence of the weld and the environment, the judgment of the corrosion defect on the bottom of the inner side of the tank is inaccurate. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent, by obtaining all acoustic information of the inner bottom of the tank within the first time based on an acoustic emission detection device, judging whether the detection waveform is an acoustic emission signal waveform caused by a weld based on a classification range, and if so, deleting the corresponding acoustic information; judging whether it is corrosion of the inner bottom of the tank based on the remaining acoustic information, and if so, issuing a maintenance warning, so as to solve the problem of inaccurate judgment of corrosion defects on the inner bottom of the tank due to the influence of the weld and the environment in the existing tank detection technology.

[0005] To achieve the above objectives, in a first aspect, the present application provides a method for analyzing intelligent detection data of a storage tank, comprising the following steps:

[0006] Based on the acoustic emission detection device, all acoustic emission information on the bottom of the inner side of the storage tank is obtained within the first time, and the acoustic emission information includes the acoustic emission source and the acoustic emission signal waveform, which are marked as the detection acoustic emission source and the detection waveform respectively;

[0007] Acquire acoustic emission signal waveforms refracted by a first number of welds, mark them as historical waveforms, and calculate a classification range based on the historical waveforms;

[0008] Based on the classification range, determine whether the detected waveform is an acoustic emission signal waveform caused by a weld, and if so, delete the corresponding acoustic emission information;

[0009] Based on the remaining acoustic information, it is determined whether the bottom of the inner side of the tank is corroded. If so, a maintenance warning is issued.

[0010] Further, obtaining waveforms of acoustic emission signals refracted by a first number of welds, marking them as historical waveforms, and calculating and obtaining a classification range based on the historical waveforms includes the following sub-steps:

[0011] Acquire a first number of historical waveforms;

[0012] Obtain parameter data of each historical waveform, marked as historical wave parameters, where the historical wave parameters include: ring count, amplitude, duration and rise time;

[0013] The classification range method is used to obtain the range of each historical wave parameter, which is marked as the classification range.

[0014] Further, the classification range method includes:

[0015] Label the first quantity Dy;

[0016] Get any kind of historical wave parameter and mark it as processing parameter;

[0017] Sort the processing parameters from small to large, and set a sequence number corresponding to the processing parameter. The sequence number is an integer starting from 1.

[0018] Determine whether u1*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u1*Dy as U1. If it is not an integer, find the average value of the processing parameters corresponding to the sequence numbers on both sides of u1*Dy and mark it as U1; where u1 is a constant in [0, 0.5];

[0019] Determine whether u2*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u2*Dy as U2. If it is not an integer, find the average value of the processing parameters corresponding to the sequence numbers on both sides of u2*Dy and mark it as U2; where u2=[(1 / u1)-1]*u1;

[0020] Set p1=U1-b*(U2-U1), p2=U2+b*(U2-U1); where p1 is the lower threshold, p2 is the upper threshold, and b is the range constant;

[0021] The setting range is: [p1, p2].

[0022] Furthermore, judging whether the detected waveform is an acoustic emission signal waveform caused by a weld based on the classification range also includes the following sub-steps:

[0023] Obtain parameter data of the detection waveform, mark it as detection parameter, and determine whether the detection parameter meets the corresponding classification range. If so, delete the sound information corresponding to the detection data.

[0024] Further, judging whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information includes the following sub-steps:

[0025] Establish a plane rectangular coordinate system, marked as the tank bottom coordinate system;

[0026] In the tank bottom coordinate system, starting from the X-axis, m straight lines are drawn in the positive direction of the Y-axis with an interval of n and parallel to the X-axis, marked as X-axis parallel lines. In the tank bottom coordinate system, starting from the Y-axis, m straight lines are drawn in the positive direction of the X-axis with an interval of n and parallel to the Y-axis, marked as Y-axis parallel lines. The X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, marked as detection squares.

[0027] The tank bottom contour and the detected acoustic emission source are drawn in the first quadrant of the tank bottom coordinate system; the acoustic emission source is the sound source position of the acoustic emission signal at the tank bottom.

[0028] Furthermore, judging whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information also includes the following sub-steps:

[0029] The abnormal area search method is used to obtain the abnormal areas of all detection grids and mark them as initial alarm areas.

[0030] Furthermore, the method for searching for abnormal areas includes:

[0031] Count the number of detected acoustic emission sources in each detection grid respectively, and mark it as the number of detections;

[0032] Mark the detection squares whose detection number is greater than 0 as sound-generating squares;

[0033] Get any sound square, mark it as the initial sound square, establish a 3x3 detection square area with the initial sound square as the center, mark it as the search area, and mark the eight detection squares in the search area except the initial sound square as neighboring squares;

[0034] Determine whether the neighboring square contains a sound-emitting square. If not, set the number of detections of the initial sound-emitting square to 0, and continue to select the next arbitrary sound-emitting square as the initial sound-emitting square. If it does, mark the sound-emitting square in the neighboring square as the execution sound-emitting square.

[0035] A new search area is established again with all the execution sound squares as the center of the search area, and the newly appeared sound squares are obtained from the new search area and marked as search sound squares. A new search area is established again with the search sound square as the center of the search area, and so on. New search areas are continuously established until there are no new sound squares in the new search area, and the sound squares contained in all search areas are taken as a whole and marked as abnormal areas;

[0036] Get any unsearched sound square, mark it as a newly selected sound square, start searching for a new abnormal area with the newly selected sound square, and repeatedly get new abnormal areas until all sound squares are searched.

[0037] Furthermore, judging whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information also includes the following sub-steps:

[0038] Count the number of squares in each initial alarm area, marked as Zk;

[0039] Count the total number of detections in each initial alarm area, marked as Jk;

[0040] Each detection ratio is calculated as: Ck=Jk / Zk; where Ck is the detection ratio;

[0041] Get the detection threshold;

[0042] Determine whether all Ck are greater than the detection threshold. If so, mark the initial alarm area corresponding to Ck greater than the detection threshold as the final alarm area, send the final alarm area information to the maintenance end, and issue a maintenance warning.

[0043] Furthermore, obtaining the detection threshold includes the following steps:

[0044] Obtain a second number of tank bottom corrosion areas, mark them as historical corrosion areas, obtain the number of acoustic emission sources in the historical corrosion areas within the first time, mark them as historical corrosion numbers;

[0045] Establish a plane rectangular coordinate system, marked as the corrosion coordinate system, so that the scale points on the X-axis and Y-axis of the corrosion coordinate system and the tank bottom coordinate system are the same;

[0046] In the tank bottom coordinate system, starting from the X-axis, m straight lines are drawn in the positive direction of the Y-axis with an interval of n and parallel to the X-axis, marked as X-axis parallel lines. In the tank bottom coordinate system, starting from the Y-axis, m straight lines are drawn in the positive direction of the X-axis with an interval of n and parallel to the Y-axis, marked as Y-axis parallel lines. The X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, marked as detection squares.

[0047] Plot the historical corrosion area in the first quadrant of the corrosion coordinate system;

[0048] The abnormal area of ​​the historical grid is obtained by using the abnormal area search method and marked as the historical abnormal area;

[0049] Get the number of squares in each historical anomaly area, marked as Zj;

[0050] Count the total number of detections in each historical anomaly area, marked as Jj;

[0051] Each historical ratio is calculated as: Cj=Jj / Zj; where Cj is the historical ratio;

[0052] Get the minimum value of the historical ratio and mark it as the detection threshold.

[0053] In a second aspect, the present application also provides an intelligent detection system for a storage tank, including a data acquisition module, a range acquisition module, a weld judgment module, and a corrosion judgment module;

[0054] The data acquisition module is used to acquire all acoustic emission information on the bottom of the inner side of the storage tank within the first time based on the acoustic emission detection device, and the acoustic emission information includes an acoustic emission source and an acoustic emission signal waveform, which are marked as a detection acoustic emission source and a detection waveform respectively;

[0055] The acquisition range module is used to acquire acoustic emission signal waveforms refracted by a first number of welds, mark them as historical waveforms, and calculate a classification range based on the historical waveforms;

[0056] The weld judgment module is used to judge whether the detected waveform is an acoustic emission signal waveform caused by a weld based on the classification range, and if so, delete the corresponding acoustic emission information;

[0057] The corrosion judgment module is used to judge whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information, and if so, issue a maintenance warning.

[0058] Beneficial effects of the present invention: The present invention obtains all acoustic information of the inner bottom of the storage tank in the first time based on the acoustic emission detection device, judges whether the detection waveform is the acoustic emission signal waveform caused by the weld based on the classification range, and if so, deletes the corresponding acoustic information; judges whether it is corrosion of the inner bottom of the storage tank based on the remaining acoustic information, and if so, issues a maintenance warning. The advantage of the present invention is that the influence of the weld and other data on the corrosion defect of the inner bottom of the storage tank is reduced, and the accuracy of the judgment and identification of the corrosion defect of the inner bottom of the storage tank is improved;

[0059] The present invention obtains the range of each historical wave parameter by using a classification range method. The advantage is that the numerical range of the wave refracted by the weld is obtained, which facilitates the distinction between the acoustic source refracted by the weld and the acoustic source generated by corrosion, thereby improving the accuracy of identifying corrosion defects on the bottom of the inner side of the storage tank.

[0060] The present invention obtains abnormal areas of all detection grids by utilizing the abnormal area search method, which has the advantage of eliminating the influence of non-corrosion generated waves and improving the accuracy of identifying corrosion defects on the bottom of the inner side of the storage tank. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a functional block diagram of the system of the present invention;

[0062] Figure 2 is a schematic diagram of the tank bottom coordinate system of the present invention;

[0063] Figure 3 is a schematic diagram of the search area of ​​the present invention;

[0064] Figure 4 The figure is a flow chart of the steps of the method of the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] Example 1, please refer to Figure 1 As shown, the present application provides an intelligent detection system for storage tanks, including a data acquisition module, a range acquisition module, a weld judgment module, and a corrosion judgment module;

[0067] The data acquisition module is used to acquire all acoustic information on the bottom of the inner side of the tank within the first time based on the acoustic emission detection device. The acoustic information includes the acoustic emission source and the acoustic emission signal waveform, which are marked as the detection acoustic emission source and the detection waveform respectively. The principle of the acoustic emission detection device is to evenly arrange the acoustic emission sensors around the tank wall, receive the acoustic emission signal generated by corrosion on the tank bottom within the first time, and comprehensively estimate the position of the sound source through the time difference when each acoustic emission sensor receives the acoustic emission signal, so as to understand the corrosion status of each area on the tank bottom. The acoustic emission source is the position of the sound source of the acoustic emission signal generated by corrosion. The first time is the time required for detection, which can be set according to the specific situation, such as 6 hours.

[0068] The acquisition range module is used to obtain the acoustic emission signal waveforms of the first number of weld refractions, mark them as historical waveforms, and calculate the classification range based on the historical waveforms;

[0069] The acquisition scope module is configured with the acquisition category scope strategy, which includes:

[0070] Acquire a first number of historical waveforms;

[0071] Obtain parameter data of each historical waveform, marked as historical wave parameters, wherein the historical wave parameters include: ring count, amplitude, duration and rise time; the historical wave parameters are obtained according to the historical waveform diagram;

[0072] The classification range method is used to obtain the range of each historical wave parameter, which is marked as the classification range.

[0073] The Get Range module is configured with the Classification Range method, which includes:

[0074] Label the first quantity Dy;

[0075] Get any kind of historical wave parameter and mark it as processing parameter;

[0076] Sort the processing parameters from small to large, and set a sequence number corresponding to the processing parameter. The sequence number is an integer starting from 1.

[0077] Determine whether u1*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u1*Dy as U1. If it is not an integer, find the average value of the processing parameters corresponding to the sequence numbers on both sides of u1*Dy and mark it as U1; where u1 is a constant in [0, 0.5]; u1 is best set to 0.25;

[0078] Determine whether u2*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u2*Dy as U2. If it is not an integer, find the average value of the processing parameters corresponding to the sequence numbers on both sides of u2*Dy and mark it as U2; where u2=[(1 / u1)-1]*u1;

[0079] Set p1=U1-b*(U2-U1), p2=U2+b*(U2-U1); p1 is the lower threshold, p2 is the upper threshold, and b is the range constant; b is set according to u1 and the amount of filtered data; for example, if u1 is set to 0.25, b is set to 0.75;

[0080] The setting range is: [p1, p2];

[0081] The principle of setting the range is: because the same type of processing parameters are used for range acquisition, and the same type of processing parameters are distributed more concentratedly, such as u1 is set to 0.25, and U2-U1 is the difference in data distribution in the middle 1 / 2 part, so p1 can be set to obtain a smaller threshold based on U1 data value minus b*(U2-U1), so p3 can be set to obtain a larger threshold based on U3 data value plus b*(U2-U1); and thus the range of the processing parameters can be obtained;

[0082] In practical applications, the first quantity is marked as Dy=1000. If the processing parameter is the rise time, when u1 is set to 0.25, it is determined whether 1000*0.25=250 is an integer. 250 is an integer, and the processing parameter 20us with a sequence number of 250 is marked as U1. u2=[(1 / u1)-1]*u1=0.75 is determined whether 1000*0.75=750 is an integer. 750 is an integer, and the processing parameter 30us with a sequence number of 750 is marked as U2. b is set to 0.75, and p1=U1-b*(U2-U1)=12.5 and p2=U2+b*(U2-U1)=37.5 are set. The setting range is [12.5, 37.5]. Therefore, the classification range of the rise time is [12.5us, 37.5us]. The classification range of ringing count, energy, amplitude, and duration is obtained by this method.

[0083] The weld judgment module is used to judge whether the detected waveform is the acoustic emission signal waveform caused by the weld based on the classification range. If so, the corresponding acoustic emission information is deleted; if there is a weld on the bottom of the tank and the weld material is different or the weld is not smooth, the acoustic emission source generated by corrosion will also appear when passing through the weld, so it is necessary to distinguish the acoustic emission sources generated by the weld and the corrosion defects;

[0084] The weld judgment module is configured with a classification method strategy, and the classification method strategy includes:

[0085] Obtain parameter data of the detection waveform, mark it as detection parameter, determine whether the detection parameter meets the corresponding classification range, if so, delete the sound information corresponding to the detection data; if so, it can be determined that the detection waveform is a detection waveform generated by the weld, which is interference information, so delete the sound information corresponding to the detection data;

[0086] In actual applications, the detection waveform parameters are ring count = 9, amplitude = 25mv, duration = 300us and rise time = 30us, respectively. This is used to determine whether they are within the corresponding classification range. For example, the classification range of the rise time is [12.5us, 37.5us], and the rise time = 30us is within [12.5us, 37.5us], which means that it meets the range. This is used to determine whether all detection waveform parameters are met. If they meet the range, the sound information corresponding to the detection data is deleted.

[0087] The corrosion judgment module is used to judge whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information. If so, a maintenance warning is issued;

[0088] The corrosion judgment module is configured with a detection grid drawing strategy, and the classification method strategy includes:

[0089] See also Figure 2 As shown, a plane rectangular coordinate system is established, marked as the tank bottom coordinate system;

[0090] In the tank bottom coordinate system, take the X-axis as the starting line and draw m straight lines parallel to the X-axis in the positive direction of the Y-axis with an interval of n, which are marked as X-axis parallel lines. In the tank bottom coordinate system, take the Y-axis as the starting line and draw m straight lines parallel to the Y-axis in the positive direction of the X-axis with an interval of n, which are marked as Y-axis parallel lines. The X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, which are marked as detection squares. The drawing of the detection square is established as an example to facilitate the analysis and judgment of the distribution of acoustic emission sources within the tank bottom contour. The settings of n and m can be set according to the size of the tank bottom, and the specifications should not be too large. For example, if the tank bottom is a circle with a diameter of 5m, n=0.4cm and m=1500 respectively, which can be adjusted according to the specific application.

[0091] Draw the tank bottom contour and the detected acoustic emission source in the first quadrant of the tank bottom coordinate system; the acoustic emission source is the sound source position of the acoustic emission signal at the bottom of the tank; the placement should satisfy the coverage range of all tank bottom contours and acoustic emission sources in the detection grid.

[0092] The corrosion judgment module is configured with a strategy for obtaining an initial alarm area, which includes: using a method for searching abnormal areas to obtain abnormal areas of all detection grids, and marking them as initial alarm areas.

[0093] The corrosion judgment module is configured with a method for searching abnormal areas, which includes:

[0094] Count the number of detected acoustic emission sources in each detection grid respectively, and mark it as the number of detections;

[0095] Mark the detection squares whose detection number is greater than 0 as sound-generating squares;

[0096] See also Figure 3 As shown, obtain any sound square, mark it as the initial sound square, establish a 3x3 detection square area with the initial sound square as the center, mark it as the search area, and mark the eight detection squares in the search area except the initial sound square as neighboring squares;

[0097] Determine whether the neighboring grid contains a sound-generating grid. If not, set the number of detections of the initial sound-generating grid to 0, and continue to select the next arbitrary sound-generating grid as the initial sound-generating grid. If it does, mark the sound-generating grid in the neighboring grid as the executing sound-generating grid. The detection grid to be established is small in size, and the individual sound source is generated by noise, so delete the number of detections at this point.

[0098] A new search area is established again with all the execution sound squares as the center of the search area, and the newly appeared sound squares are obtained from the new search area and marked as search sound squares. A new search area is established again with the search sound square as the center of the search area, and so on. New search areas are continuously established until there are no new sound squares in the new search area, and the sound squares contained in all search areas are taken as a whole and marked as abnormal areas;

[0099] Get any unsearched sound grid, mark it as the newly selected sound grid, start searching for a new abnormal area with the newly selected sound grid, and repeatedly get new abnormal areas until all sound grids are searched. This abnormal area method classifies adjacent sound points into one area and deletes isolated noise sources at the same time, so that the data is more accurate when making regional judgments based on abnormal areas.

[0100] The corrosion judgment module is configured with a strategy for obtaining the final alarm area, which includes: counting the number of squares in each initial alarm area, marked as Zk;

[0101] Count the total number of detections in each initial alarm area, marked as Jk;

[0102] Each detection ratio is calculated as: Ck=Jk / Zk; where Ck is the detection ratio; the detection ratio is the average number of detections in each detection square in the abnormal area;

[0103] Get the detection threshold;

[0104] Determine whether all Ck are greater than the detection threshold. If so, mark the initial alarm area corresponding to Ck greater than the detection threshold as the final alarm area, send the final alarm area information to the maintenance end, and issue a maintenance warning; if the detection ratio is too large, that is, the number of sound sources in this area is too many, it is determined that corrosion occurs in this area;

[0105] In practical applications, taking the calculation of a detection ratio as an example, the number of squares in the initial alarm area Zk is counted as 10, the total number of detections in the initial alarm area Jk is counted as 500, and the detection ratio Ck=Jk / Zk=50 is calculated; Ck=50 is greater than the detection threshold = 20, the initial alarm area is marked as the final alarm area, the final alarm area information is given to the maintenance end, and a maintenance warning is issued. In the specific implementation, each detection ratio must be calculated, and then it is determined whether all Ck are greater than the detection threshold.

[0106] The corrosion judgment module is configured with a detection threshold acquisition strategy, which includes:

[0107] Obtain a second number of tank bottom corrosion areas, mark them as historical corrosion areas, obtain the number of acoustic emission sources in the historical corrosion areas within the first time, mark them as historical corrosion numbers;

[0108] Establish a plane rectangular coordinate system, marked as the corrosion coordinate system, so that the scale points on the X-axis and Y-axis of the corrosion coordinate system and the tank bottom coordinate system are the same;

[0109] In the tank bottom coordinate system, starting from the X-axis, m straight lines are drawn in the positive direction of the Y-axis with an interval of n and parallel to the X-axis, marked as X-axis parallel lines. In the tank bottom coordinate system, starting from the Y-axis, m straight lines are drawn in the positive direction of the X-axis with an interval of n and parallel to the Y-axis, marked as Y-axis parallel lines. The X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, marked as detection squares.

[0110] Plot the historical corrosion area in the first quadrant of the corrosion coordinate system;

[0111] The abnormal area of ​​the historical grid is obtained by using the abnormal area search method and marked as the historical abnormal area;

[0112] Get the number of squares in each historical anomaly area, marked as Zj;

[0113] Count the total number of detections in each historical anomaly area, marked as Jj;

[0114] Each historical ratio is calculated as: Cj=Jj / Zj; where Cj is the historical ratio;

[0115] Get the minimum value of the historical ratio and mark it as the detection threshold. If corrosion occurs, the number of detections in the corrosion area will be large. Therefore, the minimum historical ratio of the corrosion area can be set as the detection threshold. If it is less than the detection threshold, it is not a corrosion area, and if it is greater than the detection threshold, it is not a corrosion area.

[0116] In practical applications, taking a historical ratio as an example, the number of squares in the historical abnormal area Zk is 60, the total number of detections in the historical abnormal area Jk is 1200, and the detection ratio Ck=Jk / Zk=20 is calculated; then each historical ratio is calculated as 20, 22, 40 and 82, etc., and the minimum value of the historical ratio is 20, that is, the detection threshold is 20.

[0117] Example 2, please refer to Figure 4 As shown, the present application provides a method for analyzing intelligent detection data of a storage tank, comprising the following steps:

[0118] Step S1, based on the acoustic emission detection device, all acoustic emission information on the inner bottom of the storage tank within the first time is obtained, and the acoustic emission information includes an acoustic emission source and an acoustic emission signal waveform, which are marked as a detection acoustic emission source and a detection waveform respectively.

[0119] Step S2, obtaining waveforms of acoustic emission signals refracted by a first number of welds, marking them as historical waveforms, and obtaining a classification range based on the historical waveforms; Step S2 includes the following sub-steps:

[0120] Step S201, obtaining a first number of historical waveforms;

[0121] Step S202, obtaining parameter data of each historical waveform, marked as historical wave parameters, wherein the historical wave parameters include: ring count, amplitude, duration and rise time.

[0122] Step S203, using the classification range method to obtain the range of each historical wave parameter, marked as the classification range; Step S203 includes the following sub-steps:

[0123] Step S20301, marking the first quantity as Dy;

[0124] Step S20302, obtaining any kind of historical wave parameter and marking it as a processing parameter; sorting the processing parameters from small to large, and setting a sequence number corresponding to the processing parameter, where the sequence number is an integer starting from 1;

[0125] Step S20303, determine whether u1*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u1*Dy as U1. If it is not an integer, calculate the average value of the processing parameters corresponding to the sequence numbers on both sides of u1*Dy and mark it as U1; where u1 is a constant in [0, 0.5].

[0126] Step S20304, determine whether u2*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u2*Dy as U2. If it is not an integer, calculate the average value of the processing parameters corresponding to the sequence numbers on both sides of u2*Dy and mark it as U2; where u2=[(1 / u1)-1]*u1;

[0127] Step S20305, set p1=U1-b*(U2-U1), p2=U2+b*(U2-U1); where p1 is the lower threshold, p2 is the upper threshold, and b is the range constant;

[0128] Step S20306, set the range to: [p1, p2].

[0129] Step S3, judging whether the detected waveform is an acoustic emission signal waveform caused by a weld based on the classification range, and if so, deleting the corresponding acoustic emission information; Step S3 includes the following sub-steps:

[0130] Step S301, obtaining parameter data of the detection waveform, marking it as detection parameters, and determining whether the detection parameters meet the corresponding classification range. If so, deleting the sound information corresponding to the detection data.

[0131] Step S4, based on the remaining acoustic information, determine whether the bottom of the inner side of the tank is corroded, and if so, issue a maintenance warning; Step S4 includes the following sub-steps:

[0132] Step S401, obtaining parameter data of the detection waveform, marking it as a detection parameter, and determining whether the detection parameter satisfies the corresponding classification range. If so, deleting the sound information corresponding to the detection data;

[0133] Step S402, establishing a plane rectangular coordinate system, marked as the tank bottom coordinate system;

[0134] Step S403: In the tank bottom coordinate system, m straight lines are drawn with an interval of n and parallel to the X-axis in the positive direction of the Y-axis starting from the X-axis, and are marked as X-axis parallel lines. In the tank bottom coordinate system, m straight lines are drawn with an interval of n and parallel to the Y-axis in the positive direction of the X-axis starting from the Y-axis, and are marked as Y-axis parallel lines. The X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, which are marked as detection squares.

[0135] Step S404, drawing the tank bottom contour and the detected acoustic emission source in the first quadrant of the tank bottom coordinate system; wherein the acoustic emission source is the sound source position of the acoustic emission signal at the tank bottom;

[0136] Step S405, using the abnormal area search method to obtain the abnormal areas of all detection grids, and marking them as initial alarm areas; Step S405 includes the following sub-steps:

[0137] Step S40501, respectively counting the number of detected acoustic emission sources in each detection grid, and marking it as the detection number;

[0138] Step S40502, marking the detection squares with a detection number greater than 0 as sound-generating squares;

[0139] Step S40503, obtaining any sound emitting square, marking it as the initial sound emitting square, establishing a 3x3 detection square area with the initial sound emitting square as the center, marking it as the search area, and marking the eight detection squares in the search area except the initial sound emitting square as neighboring squares;

[0140] Step S40504, determining whether the neighboring grid contains a vocalization grid, if not, setting the detection number of the initial vocalization grid to 0, and continuing to select the next arbitrary vocalization grid as the initial vocalization grid, if it does, marking the vocalization grid in the neighboring grid as the execution vocalization grid;

[0141] Step S40505, taking all the executed vocalization squares as the center of the search area, respectively, to establish a new search area again, obtain the newly appeared vocalization square from the new search area, mark it as the search vocalization square, take the search vocalization square as the center of the search area to establish a new search area again, and so on, continue to establish new search areas until there are no new vocalization squares in the new search area, and take the vocalization squares contained in all the search areas as a whole and mark them as abnormal areas;

[0142] Step S40506, obtain any unsearched sound square, mark it as a newly selected sound square, start searching for a new abnormal area with the newly selected sound square, and repeatedly obtain new abnormal areas until all sound squares are searched.

[0143] Step S406, counting the number of squares in each initial alarm area, marked as Zk; counting the total number of detections in each initial alarm area, marked as Jk;

[0144] Step S407, calculating each detection ratio as: Ck=Jk / Zk; wherein Ck is the detection ratio;

[0145] Step S408, obtaining a detection threshold; Step S408 includes the following sub-steps:

[0146] Step S40801, obtaining a second number of corrosion areas on the bottom of the storage tank, marking them as historical corrosion areas, and obtaining the number of acoustic emission sources in the historical corrosion areas within a first period of time, marking them as historical corrosion numbers;

[0147] Step S40802, establish a plane rectangular coordinate system, mark it as the corrosion coordinate system, and make the scale points on the X-axis and Y-axis of the corrosion coordinate system and the tank bottom coordinate system the same;

[0148] Step S40803, in the tank bottom coordinate system, with the X axis as the starting line, m straight lines are drawn in the positive direction of the Y axis with an interval of n and parallel to the X axis, marked as X-axis parallel lines; in the tank bottom coordinate system, with the Y axis as the starting line, m straight lines are drawn in the positive direction of the X axis with an interval of n and parallel to the Y axis, marked as Y-axis parallel lines; the X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, marked as detection squares;

[0149] Step S40804, drawing the historical corrosion area in the first quadrant of the corrosion coordinate system;

[0150] Step S40805, using the abnormal area search method to obtain the abnormal area of ​​the historical grid, and mark it as the historical abnormal area;

[0151] Step S40806, obtaining the number of squares in each historical abnormal area, marked as Zj; counting the total number of detections in each historical abnormal area, marked as Jj;

[0152] Step S40807, calculate each historical ratio as: Cj=Jj / Zj; where Cj is the historical ratio;

[0153] Step S40808, obtain the minimum value of the historical ratio and mark it as the detection threshold.

[0154] Step S409, determine whether all Ck are greater than the detection threshold. If so, mark the initial alarm area corresponding to Ck greater than the detection threshold as the final alarm area, send the final alarm area information to the maintenance end, and issue a maintenance warning.

[0155] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0156] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

Claims

1. A method for analyzing intelligent detection data of storage tanks, characterized in that: The steps include: Based on the acoustic emission detection device, all acoustic emission information on the bottom of the inner side of the storage tank is obtained within the first time, and the acoustic emission information includes the acoustic emission source and the acoustic emission signal waveform, which are marked as the detection acoustic emission source and the detection waveform respectively; Acquire acoustic emission signal waveforms refracted by a first number of welds, mark them as historical waveforms, and calculate a classification range based on the historical waveforms; Based on the classification range, determine whether the detected waveform is an acoustic emission signal waveform caused by a weld, and if so, delete the corresponding acoustic emission information; Based on the remaining acoustic information, determine whether the bottom of the tank is corroded. If so, issue a maintenance warning. Acquiring the acoustic emission signal waveforms of the first number of weld refractions, marking them as historical waveforms, and calculating and obtaining the classification range based on the historical waveforms includes the following sub-steps: Acquire a first number of historical waveforms; Obtain parameter data of each historical waveform, marked as historical wave parameters, where the historical wave parameters include: ring count, amplitude, duration and rise time; The classification range method is used to obtain the range of each historical wave parameter, which is marked as the classification range; Classification scope methods include: Label the first quantity Dy; Get any kind of historical wave parameter and mark it as processing parameter; Sort the processing parameters from small to large, and set a sequence number corresponding to the processing parameter. The sequence number is an integer starting from 1. Determine whether u1*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u1*Dy as U1. If it is not an integer, find the average value of the processing parameters corresponding to the sequence numbers on both sides of u1*Dy and mark it as U1; where u1 is a constant in [0, 0.5]; Determine whether u2*Dy is an integer. If it is an integer, mark the processing parameter with the sequence number u2*Dy as U2. If it is not an integer, find the average value of the processing parameters corresponding to the sequence numbers on both sides of u2*Dy and mark it as U2; where u2=[(1 / u1)-1]*u1; Set p1=U1-b*(U2-U1), p2=U2+b*(U2-U1); where p1 is the lower threshold, p2 is the upper threshold, and b is the range constant; Set the classification range to: [p1, p2]; Judging whether the detected waveform is an acoustic emission signal waveform caused by a weld based on the classification range also includes the following sub-steps: Obtain parameter data of the detection waveform, mark it as a detection parameter, and determine whether the detection parameter meets the corresponding classification range. If so, delete the sound information corresponding to the detection data; Determining whether the bottom of the tank is corroded based on the remaining acoustic information includes the following sub-steps: Establish a plane rectangular coordinate system, marked as the tank bottom coordinate system; In the tank bottom coordinate system, starting from the X-axis, m straight lines are drawn in the positive direction of the Y-axis with an interval of n and parallel to the X-axis, marked as X-axis parallel lines. In the tank bottom coordinate system, starting from the Y-axis, m straight lines are drawn in the positive direction of the X-axis with an interval of n and parallel to the Y-axis, marked as Y-axis parallel lines. The X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, marked as detection squares. Draw the tank bottom contour and the detected acoustic emission source in the first quadrant of the tank bottom coordinate system; the acoustic emission source is the sound source position of the acoustic emission signal at the tank bottom; Judging whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information also includes the following sub-steps: The abnormal area search method is used to obtain the abnormal area of ​​all detection grids and mark them as the initial alarm area; Methods for searching for abnormal areas include: Count the number of detected acoustic emission sources in each detection grid respectively, and mark it as the number of detections; Mark the detection squares whose detection number is greater than 0 as sound-generating squares; Get any sound square, mark it as the initial sound square, establish a 3x3 detection square area with the initial sound square as the center, mark it as the search area, and mark the eight detection squares in the search area except the initial sound square as neighboring squares; Determine whether the neighboring square contains a sound-emitting square. If not, set the number of detections of the initial sound-emitting square to 0, and continue to select the next arbitrary sound-emitting square as the initial sound-emitting square. If it does, mark the sound-emitting square in the neighboring square as the execution sound-emitting square. A new search area is established again with all the execution sound squares as the center of the search area, and the newly appeared sound squares are obtained from the new search area and marked as search sound squares. A new search area is established again with the search sound square as the center of the search area, and so on. New search areas are continuously established until there are no new sound squares in the new search area, and the sound squares contained in all search areas are taken as a whole and marked as abnormal areas; Obtain any unsearched sound square, mark it as a newly selected sound square, start searching for a new abnormal area with the newly selected sound square, and repeatedly obtain new abnormal areas until all sound squares are searched; Judging whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information also includes the following sub-steps: Count the number of squares in each initial alarm area, marked as Zk; Count the total number of detections in each initial alarm area, marked as Jk; Each detection ratio is calculated as: Ck=Jk / Zk; where Ck is the detection ratio; Get the detection threshold; Determine whether all Ck are greater than the detection threshold. If so, mark the initial alarm area corresponding to Ck greater than the detection threshold as the final alarm area, send the final alarm area information to the maintenance end, and issue a maintenance warning.

2. A method for analyzing intelligent detection data of a storage tank according to claim 1, characterized in that: Obtaining the detection threshold includes the following steps: Obtain a second number of tank bottom corrosion areas, mark them as historical corrosion areas, obtain the number of acoustic emission sources in the historical corrosion areas within the first time, mark them as historical corrosion numbers; Establish a plane rectangular coordinate system, marked as the corrosion coordinate system, so that the scale points on the X-axis and Y-axis of the corrosion coordinate system and the tank bottom coordinate system are the same; In the tank bottom coordinate system, starting from the X-axis, m straight lines are drawn in the positive direction of the Y-axis with an interval of n and parallel to the X-axis, marked as X-axis parallel lines. In the tank bottom coordinate system, starting from the Y-axis, m straight lines are drawn in the positive direction of the X-axis with an interval of n and parallel to the Y-axis, marked as Y-axis parallel lines. The X-axis parallel lines and the Y-axis parallel lines form (m-1)*(m-1) squares, marked as detection squares. Plot the historical corrosion area in the first quadrant of the corrosion coordinate system; The abnormal area of ​​the historical grid is obtained by using the abnormal area search method and marked as the historical abnormal area; Get the number of squares in each historical anomaly area, marked as Zj; Count the total number of detections in each historical anomaly area, marked as Jj; Each historical ratio is calculated as: Cj=Jj / Zj; where Cj is the historical ratio; Get the minimum value of the historical ratio and mark it as the detection threshold.

3. A storage tank intelligent detection system, used to implement a storage tank intelligent detection data analysis method according to claim 1 or 2, characterized in that: It includes a data acquisition module, a range acquisition module, a weld judgment module and a corrosion judgment module; The data acquisition module is used to acquire all acoustic emission information on the bottom of the inner side of the storage tank within the first time based on the acoustic emission detection device, and the acoustic emission information includes an acoustic emission source and an acoustic emission signal waveform, which are marked as a detection acoustic emission source and a detection waveform respectively; The acquisition range module is used to acquire acoustic emission signal waveforms refracted by a first number of welds, mark them as historical waveforms, and calculate a classification range based on the historical waveforms; The weld judgment module is used to judge whether the detected waveform is an acoustic emission signal waveform caused by a weld based on the classification range, and if so, delete the corresponding acoustic emission information; The corrosion judgment module is used to judge whether the bottom of the inner side of the tank is corroded based on the remaining acoustic information, and if so, issue a maintenance warning.

Images