A method and device for detecting defects in steel pipelines

Abstract

The present invention provides a method and device for detecting defects in steel pipelines. The method for detecting defects in steel pipelines comprises the following steps: S1: Detection using transient electromagnetic method: Transmitting transient electromagnetic excitation signals to the steel pipeline, after the surface of the steel pipeline is excited by the transient electromagnetic excitation signal, a transient electromagnetic attenuation signal is generated, and the transient electromagnetic attenuation signal generated by the surface of the steel pipeline is received; S2: Using weak magnetic method to detect the weak magnetic detection signal generated by the surface of the steel pipeline; S3: Fusion and enhancement processing of the transient electromagnetic attenuation signal and the weak magnetic detection signal. The device for detecting defects in steel pipelines comprises: a weak magnetic detection device, a transient electromagnetic detection device, and a data fusion host computer for fusing the weak magnetic detection signal and the transient electromagnetic signal. The present invention uses measurement of electrical variables or measurement of magnetic variables to achieve non-destructive detection.

Description

Technical Field

[0001] The present invention relates to the field of pipeline nondestructive testing, in particular to the field of measuring electrical variables or measuring magnetic variables, and specifically to a steel pipeline defect detection method and device. Background Art

[0002] According to the "Medium- and Long-Term Oil and Gas Pipeline Network Planning", it is estimated that the scale of China's pipeline network will reach 240,000 kilometers in 2025. Today, oil and gas pipeline transportation has become one of the important modes of transportation in China. However, with the rapid increase in the number of pipelines, pipeline accidents have also occurred frequently, and each pipeline accident has caused irreparable disasters to China's economy and security. The main cause of pipeline accidents is that some of the current pipelines are in disrepair, resulting in a large amount of corrosion and leakage. According to statistics, 20%~30% of the current pipelines have been in service for more than 20 years. The development of pipeline defect detection technology is one of the current research directions at home and abroad.

[0003] Pipelines are divided into buried pipelines and overhead pipelines. The main inspection methods for buried pipelines are divided into excavation inspection and trenchless inspection. The excavation inspection of buried pipelines adopts the "excavation-inspection-backfilling" method. This "excavation-inspection-backfilling" method is time-consuming and costly, and is not suitable for large-scale buried pipeline inspection.

[0004] In summary, the prior art has the following problems: the detection of buried pipelines is time-consuming and costly, and it is difficult to complete a large range of pipeline defect detection within a limited detection time. Summary of the invention

[0005] The present invention provides a steel pipeline defect detection method and device, which adopts the measurement of electrical variables or magnetic variables to solve the problems that the detection of buried pipelines is time-consuming, costly, and difficult to complete a large range of pipeline defect detection within a limited detection time.

[0006] To this end, the present invention proposes a steel pipeline defect detection method, which comprises the following steps:

[0007] S1: Detection by transient electromagnetic method: a transient electromagnetic excitation signal is emitted to the steel pipeline, and after the surface of the steel pipeline is excited by the transient electromagnetic excitation signal, a transient electromagnetic attenuation signal is generated, and the transient electromagnetic attenuation signal generated by the surface of the steel pipeline is received;

[0008] S2: Use weak magnetic method to detect the weak magnetic detection signal generated on the surface of the steel pipeline;

[0009] S3: The transient electromagnetic attenuation signal and the weak magnetic detection signal are fused and enhanced to obtain a fused steel pipeline defect image.

[0010] The transient electromagnetic attenuation signal obtains the transient electromagnetic detection image, the weak magnetic detection signal obtains the weak magnetic detection image, and the transient electromagnetic detection image and the weak magnetic detection image are fused to obtain a fused steel pipeline defect image, that is, a fused image of transient electromagnetic detection and weak magnetic detection.

[0011] Furthermore, the steel pipeline defect detection method further comprises:

[0012] Step S4: Evaluate the defect depth of the fused steel pipe defect image by establishing an artificial intelligence model or an AI model and determine the pipeline defect level based on the product standard of the steel pipe. The artificial intelligence model or the AI ​​model is, for example, a convolutional neural network model or a neural network model.

[0013] Furthermore, in step S3, the collected transient electromagnetic attenuation signal and weak magnetic detection signal are subjected to data alignment processing and normalization processing; wherein the collected transient electromagnetic attenuation signal is data A, and the weak magnetic detection signal is data B.

[0014] Furthermore, the one-dimensional data of transient electromagnetic detection and weak magnetic detection are converted into two-dimensional images through continuous wavelet transformation, that is, the transient electromagnetic attenuation signal (i.e. the one-dimensional data of transient electromagnetic detection, transient electromagnetic attenuation signal = one-dimensional data of transient electromagnetic detection) is transformed into a two-dimensional transient electromagnetic detection image through continuous wavelet transformation; the weak magnetic detection signal (one-dimensional data of weak magnetic detection) is transformed into a two-dimensional weak magnetic detection image through continuous wavelet transformation.

[0015] The coefficient of variation method is used to calculate the pixel weights of transient electromagnetic detection images and weak magnetic detection images, and the converted transient electromagnetic signals (two-dimensional transient electromagnetic detection images) and weak magnetic detection signals (two-dimensional weak magnetic detection images) are fused using the pixel weighting method according to the weight ratio obtained by the coefficient of variation method or other objective weighting methods. Among them, the coefficient of variation method can be implemented using the weight calculation method 3 on the CSDN blog: Coefficient of Variation method).

[0016] The present invention also proposes a steel pipeline defect detection device. The steel pipeline defect detection method of the present invention can be implemented by using the steel pipeline defect detection device. The steel pipeline defect detection device includes: a weak magnetic detection device, a transient electromagnetic detection device, and a data fusion host computer that fuses the weak magnetic detection signal and the transient electromagnetic signal; wherein the data fusion host computer is signal-connected to the weak magnetic detection device and the transient electromagnetic detection device, respectively.

[0017] The present invention is based on magnetic signals. By studying the relationship between transient electromagnetic signals and (i.e., weak magnetic detection signals) and steel pipeline defects, the transient electromagnetic signals and weak magnetic signals (i.e., weak magnetic detection signals) are enhanced by data fusion to complete the detection of steel pipeline defects. Then, by establishing an AI model (such as a CNN-BiGRU-Attention model), the defect depth of the fused steel pipeline defect image is estimated and the pipeline defect level is determined based on the ASME-B31G standard. This detection method has the advantages of simplicity, high efficiency, high precision, and small error. In addition, this method can also realize two-way detection of buried steel pipelines and aerial steel pipelines, solving the current problems of high cost, long time consumption, and inconvenience in steel pipeline detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a structural schematic diagram of a steel pipeline defect detection device according to an embodiment of the present invention, wherein the shielding plate is not pulled out;

[0019] Figure 2 It is a structural schematic diagram of a steel pipeline defect detection device according to an embodiment of the present invention, wherein the shielding plate has been pulled out;

[0020] Figure 3 It is a schematic structural diagram of the transient electromagnetic detection device of the present invention;

[0021] Figure 4A A schematic diagram of transient electromagnetic-weak magnetic detection data fusion enhancement according to an embodiment of the present invention;

[0022] Figure 4B It is another schematic diagram of transient electromagnetic-weak magnetic detection data fusion enhancement according to an embodiment of the present invention;

[0023] Figure 5 This is a schematic diagram of the CNN-BiGRU-Attention model of the present invention;

[0024] Figure 6 The schematic diagram of the ground-to-air bidirectional detection using transient electromagnetic method-weak magnetic fusion detection in the presence of an external magnetic field.

[0025] Explanation of the accompanying drawings: 11. Receiving coil; 12. Transmitting coil; 13. Transient electromagnetic instrument; 2. Data fusion host computer; 30. Weak magnetic detection signal acquisition device (weak magnetic detection device single-chip computer); 31. Weak magnetic detection sensor; 41. Upper magnetic shielding baffle; 42. Lower magnetic shielding baffle; 7. Slide rail; 81. Box body; 83. Box cover. DETAILED DESCRIPTION

[0026] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the present invention is now described.

[0027] like Figure 1 As shown, the present invention proposes a steel pipeline defect detection device, which can be used for buried steel pipeline defect detection and can also be used for aerial steel pipeline defect detection, including: three major parts, namely a weak magnetic detection device, a transient electromagnetic detection device and a data fusion host computer 2 that fuses weak magnetic detection signals and transient electromagnetic signals.

[0028] Among them, Figure 2 and Figure 3 As shown, the transient electromagnetic detection device mainly includes a transient electromagnetic instrument 13, a transmitting coil 12 and a receiving coil 11.

[0029] like Figure 1 and Figure 2 As shown, the weak magnetic detection device mainly includes a weak magnetic detection signal acquisition device 30 and a weak magnetic detection sensor 31.

[0030] The transient electromagnetic detection device consists of a transient electromagnetic instrument, a transmitting coil, and a receiving coil. The three can be connected to each other through an aviation plug, or through other wired connections to form electrical connections and signal connections, for example, to form electrical connections and signal connections between the transient electromagnetic instrument and the transmitting coil and the receiving coil. The transient electromagnetic instrument transmits a bipolar rectangular pulse excitation signal to the transmitting coil, the transmitting coil generates a stable magnetic field in the entire space, and the receiving coil receives the attenuation signal generated by the eddy current on the surface of the steel pipeline. The transient electromagnetic instrument can use a mature transient electromagnetic detector sold on the market, such as the GBH-2 buried pipeline non-excavation corrosion detector based on TEM transient electromagnetic technology, TEM: Transient Electromagnetic Method. For example, the GBH-2 buried pipeline corrosion detector sold by Beijing Zhongcheng Jiayi Technology Co., Ltd. has the following main parameters: Pipe section length: The length of the pipe section covered by each measuring point is approximately equal to the sum of the side length of the sensor transmitting loop used and twice the center burial depth of the pipeline (L+2h), Detection accuracy: Under general interference conditions, the error can be controlled within 5%. (Weak magnetic range: ±400μT, sensitivity: 13nT / LSB)

[0031] The weak magnetic detection device includes a weak magnetic detection signal acquisition device 30 (i.e., a weak magnetic detection single chip computer) and a weak magnetic detection sensor 31, which are electrically and signal-connected to each other, and can perform weak magnetic defect detection on steel pipelines. The weak magnetic detection single chip computer is, for example, STM32. The weak magnetic detection sensor 31 performs defect detection on the steel pipeline to obtain detection data, and the weak magnetic detection single chip computer solves, stores and processes the data. The data fusion host computer 2 (also called transient electromagnetic-weak magnetic detection data fusion host computer) fuses and enhances the detection data of the transient electromagnetic detection device and the weak magnetic detection device, so that the detection data has higher accuracy and smaller errors. The weak magnetic detection device can be a fluxgate sensor. The weak magnetic detection sensor is, for example, the Witte Intelligent RM3100 three-axis electronic compass geomagnetic sensor (also called fluxgate sensor), with an operating current of 0.01A, an operating voltage of 5V, a range of ±400μT, and a sensitivity of 13nT / LSB. Figure 2 As shown, for example, four fluxgate sensors are provided, and the four fluxgate sensors are arranged opposite to each other in pairs, and the two opposite fluxgate sensors are spaced a certain distance apart and located on a straight line. The four fluxgate sensors form a cross-shaped matrix arrangement, so that weak magnetic detection of weak magnetic signals in all directions can be taken into account.

[0032] The weak magnetic detection device and the transient electromagnetic detection device are separated from each other and do not interfere with each other. For example, the transmitting coil and the receiving coil of the transient electromagnetic detection device, such as Figure 1 and Figure 2 As shown, there is no intersection or overlap with the horizontal projection of the weak magnetic field detection sensor 31 .

[0033] In addition, if Figure 1 and Figure 2 The steel pipeline defect detection device further includes: a shielding device for shielding the interference of external magnetic fields on transient electromagnetic signals and weak magnetic detection signals. The shielding device includes: two layers of shielding plates that can be pulled horizontally, namely an upper magnetic shielding baffle 41 and a lower magnetic shielding baffle 42, the upper magnetic shielding baffle 41 and the lower magnetic shielding baffle 42 form a shielding space; the steel pipeline defect detection device further includes: a box body 81 and a box cover 83 hinged to the box body, and a slide rail 7 for pulling the shielding plate is provided on the inner wall of the box body 81; when not in use, the box cover 83 is covered on the box body 81 to form a closed box that is easy to store and carry, and when in use, the box cover can be easily opened to expose the box body;

[0034] Among them, the transient electromagnetic instrument 13, the data fusion host computer 2 and the weak magnetic detection signal acquisition device 30 are located on the upper magnetic shielding baffle 41.

[0035] The transmitting coil, the receiving coil and the weak magnetic detection sensor 31 are located between the upper magnetic shielding baffle 41 and the lower magnetic shielding baffle 42;

[0036] The present invention can detect buried steel pipelines or aerial steel pipelines in one direction or one side. Figure 6 As shown in the figure, in the presence of an external magnetic field, when using transient electromagnetic method-weak magnetic fusion detection for ground-air bidirectional detection, there will be external interference. For example, when detecting a buried steel pipeline on one side, the steel pipeline in the air will cause interference, and the steel pipeline in the air is the external interference at this time; similarly, when detecting an aerial steel pipeline on one side, the buried steel pipeline will cause interference, and the buried steel pipeline is the external interference at this time.

[0037] In order to avoid or reduce external interference, during one-way downward detection, that is, when detecting buried steel pipelines, the lower shielding plate is opened, that is, the lower magnetic shielding baffle 42 is opened, so that the transmitting coil, the receiving coil and the surface of the buried steel pipeline form an unobstructed transient electromagnetic field, and the weak magnetic detection sensor and the surface of the buried steel pipeline form an unobstructed weak magnetic detection magnetic field; finally, the transient electromagnetic attenuation signal and the weak magnetic detection signal of the buried steel pipeline are obtained; at the same time, the upper shielding plate is closed to shield the upper aerial pipeline from interference with the transmitting coil, the receiving coil, and the weak magnetic detection sensor.

[0038] Similarly, during one-way upward detection, that is, when conducting aerial steel pipeline detection, the upper shielding plate is opened, that is, the upper magnetic shielding baffle 41 is opened, so that the transmitting coil, the receiving coil and the surface of the aerial steel pipeline form an unobstructed transient electromagnetic field, and the weak magnetic detection sensor and the surface of the aerial steel pipeline form an unobstructed weak magnetic detection magnetic field; finally, the transient electromagnetic attenuation signal and the weak magnetic detection signal of the aerial steel pipeline are obtained; at the same time, the lower shielding plate is closed to shield the buried steel pipeline below from interfering with the transmitting coil, the receiving coil and the weak magnetic detection sensor.

[0039] The inventive principle of the present invention is:

[0040] Based on magnetic signals, by studying the relationship between transient electromagnetic signals and weak magnetic detection signals and steel pipeline defects, data fusion enhancement is performed on transient electromagnetic signals and weak magnetic detection signals to complete the detection of steel pipeline defects. Then, a CNN-BiGRU-Attention model is established to estimate the defect depth of the fused steel pipeline defect image and determine the pipeline defect level based on the ASME-B31G standard. This detection method has the advantages of simplicity, high efficiency, high accuracy and small error. In addition, this method can also realize two-way detection of buried steel pipelines and aerial steel pipelines, solving the current problems of high cost, long time consumption and inconvenience of steel pipeline detection.

[0041] The specific method is:

[0042] Transient electromagnetic-weak magnetic detection data fusion enhancement method:

[0043] This method fuses the data of two steel pipeline defect detection methods, transient electromagnetic detection and weak magnetic detection. In essence, it fuses the transient electromagnetic detection data and the weak magnetic detection data, collects the data of transient electromagnetic detection and weak magnetic detection, performs feature extraction transformation on the output data of transient electromagnetic detection and weak magnetic detection, extracts the feature vector representing the observed data, identifies the feature vector to complete the description of each sensor on the target, and then associates the description data of the target. Finally, a fusion algorithm is proposed to synthesize the transient electromagnetic detection data and the weak magnetic detection data.

[0044] After the steel pipeline is subjected to transient electromagnetic-weak magnetic detection fusion detection, the collected data is aligned and normalized. Let the transient electromagnetic detection data be A and the weak magnetic detection data be B.

[0045] Data alignment processing:

[0046] (1)

[0047] For example, x represents the horizontal coordinate of the detection data (the unit of pipeline length is m), y represents the vertical coordinate of the detection data (weak magnetic field is magnetic induction intensity: T, transient electromagnetic is induced electromotive force: V; Xi+1 represents the i+1th signal data collected by the sensor, that is, the horizontal coordinate of the i+1th position on the pipeline length collected by the sensor, the unit is m, Xi represents the i-th signal data collected by the sensor, that is, the horizontal coordinate of the i-th position on the pipeline length, Yi+1 represents the vertical coordinate of the i+1th detection data (; weak magnetic field is magnetic induction intensity: T, transient electromagnetic is induced electromotive force: V; A*, B* are the process or results after A and B interpolation,

[0048] Among them, interpolation uses mean interpolation

[0049] Normalization:

[0050] (2)

[0051] x represents the horizontal coordinate of the test data (the unit of pipeline length is m), new represents the new x value, xmin represents the minimum x value, xmax represents the maximum x value, A', B' represents the process or result of A*, B* normalization,

[0052] The one-dimensional data of transient electromagnetic method detection and weak magnetic detection are converted into two-dimensional images through continuous wavelet transformation. If the following three conditions are met, then it is called is the basic wavelet, for any signal , f(t) is the continuous wavelet function at time t, and its continuous wavelet change formula is as follows:

[0053] Define the conditions:

[0054] (3)

[0055] t represents time, the unit is s, is the basic wavelet, unitless, C is the normalization constant, unitless, ω is the frequency variable, unit is rad / s, represents its complex conjugate, unitless, f(t) is the time domain signal to be analyzed, L² (R) represents the space of square integrable functions and has no units.

[0056] Continuous wavelet transformation formula:

[0057] (4)

[0058] WTψ(a,b) is the result of continuous wavelet transform, which is a function of scale parameter a and position parameter b. a is the scale parameter, which determines the scaling of the wavelet function and has no unit. b is the translation parameter, which determines the position of the wavelet function on the time axis and has no unit. f(t) is the time domain signal to be analyzed, t is the time in seconds.

[0059] (t) represents the complex conjugate function of the wavelet function; stands for complex conjugation;

[0060] The pixel weights of transient electromagnetic detection images and weak magnetic detection images are calculated using the coefficient of variation method, and the converted transient electromagnetic signals and weak magnetic detection signals are fused using the pixel weighting method according to the weight ratio obtained using the coefficient of variation method.

[0061] Data collection after data normalization:

[0062] (5)

[0063] X11, X12, Xn1…Xn2, represent the signal data collected by the sensor; no unit, X represents the parameter matrix

[0064] Calculate the mean of each indicator:

[0065] (6)

[0066] Calculate the parameter average, n represents the total number, Aj represents the average value of each column, no unit; X ij Represents the parameter of the i-th row and j-th column in the matrix

[0067] Calculate the standard deviation (mean square error) for each metric:

[0068] (7)

[0069] Sj represents the mean square error of each column, unitless

[0070] Calculate the coefficient of variation for each indicator:

[0071] (8)

[0072] V j represents the coefficient of variation and has no unit;

[0073] Calculate weights:

[0074] (9)

[0075] represent V j The average value of , Wj represents the percentage weight value, no unit;

[0076] For example, in a specific detection, the weak magnetic detection sensor is, for example, the Witt Intelligent RM3100 three-axis electronic compass geomagnetic sensor (also known as fluxgate sensor, range: ±400μT, sensitivity: 13nT / LSB), and the transient electromagnetic instrument is the GBH-2 buried pipeline corrosion detector. Through the above formula, it can be concluded that when these two devices are used, the difference in detection data shows that the weight values ​​of the transient electromagnetic detection image and the weak magnetic detection image are 0.44 and 0.56 respectively. The weight is assigned to the pixel weighting method, and the transient electromagnetic detection image and the weak magnetic detection image are image fused. Figure 4A This is a schematic diagram of transient electromagnetic-weak magnetic detection data fusion enhancement according to an embodiment of the present invention. Figure 4B It is another schematic diagram of transient electromagnetic-weak magnetic detection data fusion enhancement according to an embodiment of the present invention; Figure 4A and Figure 4B They are the color and black and white representations of the same transient electromagnetic-weak magnetic detection data fusion enhancement schematic diagram. Figure 4A and Figure 4B It can be seen that the diagram after the fusion of transient electromagnetic and weak magnetic detection data has a more obvious fusion enhancement effect than a single transient electromagnetic detection diagram or a single weak magnetic detection image.

[0077] The CNN-BiGRU-Attention model is used to estimate or evaluate the defect depth of the fused pipeline defect image, such as Figure 5As shown in the figure, the model consists of four main components: 2DCNN module, BiGRU module, attention mechanism and fully connected (FC) module. The 2DCNN module has three branches, which are composed of three different sizes of convolution kernels. The kernel sizes are 3*3, 5*5 and 7*7, respectively, which can extract detailed features, local features and global features. The BiGRU module includes two BiGRU layers, which can realize nonlinear fitting of sample features. The BiGRU module is optimized by using the attention mechanism, and the weight of the hidden state of the BiGRU layer can be determined. In addition, other AI models or artificial intelligence models, such as other convolutional neural network models, such as BP, can also be used to estimate or evaluate the depth of defects. The CNN-BiGRU-Attention model combines the advantages of convolutional neural network (CNN) in local feature extraction, the ability of bidirectional gated recurrent unit (BiGRU) to capture contextual information when processing sequence data, and the focus of attention mechanism (Attention) on key information, so that the model can more effectively extract key features in text when processing natural language processing tasks, while taking into account contextual information, thereby significantly improving the accuracy of understanding and prediction, with an accuracy of up to 95%.

[0078] Based on the ASME-B31G standard, five defect states are divided according to the defect depth of the pipeline, as shown in Table 1. The prediction results of the CNN-BiGRU-Attention model are used to determine the level of the pipeline defect state, with an accuracy of 95%.

[0079] Table 1. Defect status identification table

[0080]

[0081] Bidirectional detection of steel pipelines on the ground and in the air:

[0082] The bidirectional detection of steel pipelines on the ground and in the air is based on the transient electromagnetic-weak magnetic detection fusion enhancement method. When the fusion sensor transmits a signal, it transmits a signal to the ground and the air at the same time. The transmitted signal causes the steel pipelines on the ground and in the air to generate a stable magnetic field. The receiving sensor receives the magnetic signal and expresses it in the form of voltage, thereby realizing the bidirectional detection of steel pipelines on the ground and in the air.

[0083] When the steel pipelines on the ground and in the air appear separately, according to the half-space theory, the medium on the other side is regarded as the background environment and will not affect the transient electromagnetic method-weak magnetic fusion detection results. However, when the steel pipelines appear on the ground and in the air at the same time, the transient electromagnetic-weak magnetic detection fusion enhancement device (the transient electromagnetic-weak magnetic fusion detection sensor includes a transmitting coil, a receiving coil and a weak magnetic detection sensor) will receive two different magnetic signals, resulting in errors in the detection results. In order to avoid errors in the detection results when the bidirectional steel pipelines on the ground and in the air appear at the same time, the magnetic shielding plate and the slide rail are combined to isolate the electromagnetic signal on one side, thereby realizing the detection of the ground and air steel pipelines that appear simultaneously in a certain place.

[0084] Beneficial effects:

[0085] 1. Compared with the existing single-sensor steel pipeline defect detection, the present invention adopts the pixel weighting method to fuse the transient electromagnetic detection steel pipeline data and the weak magnetic detection steel pipeline data, and adopts the coefficient of variation method to perform pixel weight analysis on the transient electromagnetic-weak magnetic detection data, and then establishes a CNN model (convolutional neural network), such as a CNN-BiGRU-Attention model, to estimate the defect depth of the fused steel pipeline defect image and determine the pipeline defect level based on the ASME-B31G

[32] standard. This makes the data fusion detection method more accurate and less error-prone.

[0086] 2. Compared with the existing buried steel pipeline detection and aerial steel pipeline detection, the present invention can realize the detection of ground steel pipelines and aerial steel pipelines separately based on the half-space theory without changing the detection method and device, making the detection process more convenient and efficient.

[0087] The above description is only an illustrative embodiment of the present invention and is not intended to limit the scope of the present invention. The components of the present invention can be combined with each other without conflict, and any equivalent changes and modifications made by any technician in the field without departing from the concept and principle of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for detecting defects in steel pipelines, characterized in that: The steel pipeline defect detection method comprises the following steps: S1: Detection by transient electromagnetic method: a transient electromagnetic excitation signal is emitted to the steel pipeline, and after the surface of the steel pipeline is excited by the transient electromagnetic excitation signal, a transient electromagnetic attenuation signal is generated, and the transient electromagnetic attenuation signal generated by the surface of the steel pipeline is received; S2: Use weak magnetic method to detect the weak magnetic detection signal generated on the surface of the steel pipeline; S3: performing fusion and enhancement processing on the transient electromagnetic attenuation signal and the weak magnetic detection signal to obtain a fused steel pipeline defect image; The one-dimensional data of transient electromagnetic detection and weak magnetic detection are converted into two-dimensional images through continuous wavelet transformation. The pixel weights of transient electromagnetic detection images and weak magnetic detection images are calculated using the coefficient of variation method. The converted transient electromagnetic signals and weak magnetic detection signals are fused using the pixel weighting method according to the weight ratio obtained by the coefficient of variation method.

2. The steel pipeline defect detection method according to claim 1, characterized in that: The steel pipeline defect detection method further comprises: Step S4: An artificial intelligence model is established to evaluate the defect depth of the fused steel pipe defect image and to determine the pipe defect grade based on the product standard of the steel pipe.

3. The steel pipeline defect detection method according to claim 1, characterized in that: The steel pipeline defect detection method is a unidirectional detection, that is, only the unidirectional downward detection of the buried steel pipeline is detected or only the unidirectional upward detection of the aerial steel pipeline is detected. When the unidirectional detection is performed under the condition that the buried steel pipeline and the aerial steel pipeline exist at the same time, in step S1, the interference of the buried steel pipeline on the transient electromagnetic detection of the aerial steel pipeline is shielded, or the interference of the aerial steel pipeline on the transient electromagnetic detection of the buried steel pipeline is shielded.

4. The steel pipeline defect detection method according to claim 3, characterized in that: In step S1, a bipolar rectangular pulse transient electromagnetic excitation signal is transmitted to the transmitting coil through the transient electromagnetic instrument, the transmitting coil generates a magnetic field in the entire space, and the transient electromagnetic attenuation signal generated by the eddy current on the surface of the steel pipeline is received by the receiving coil. The transient electromagnetic instrument receives the transient electromagnetic attenuation signal to generate the defect information of the buried pipeline; the transient electromagnetic instrument is electrically connected to the transmitting coil and the receiving coil respectively; The weak magnetic detection signal generated on the surface of the steel pipeline is detected by a weak magnetic detection sensor, and the weak magnetic detection signal is solved, stored and processed by a weak magnetic detection single chip microcomputer. The weak magnetic detection sensor and the weak magnetic detection single chip microcomputer are electrically connected and signal connected. An upper and lower layer of shielding plates that can be pulled horizontally are arranged to form a shielding space, and the transmitting coil, the receiving coil and the weak magnetic detection sensor are located in the shielding space.

5. The steel pipeline defect detection method according to claim 4, characterized in that: During one-way downward detection, the lower shielding plate is pulled open to form an unobstructed transient electromagnetic field between the transmitting coil, the receiving coil and the surface of the buried steel pipeline, and an unobstructed weak magnetic detection magnetic field between the weak magnetic detection sensor and the surface of the buried steel pipeline; finally, the transient electromagnetic attenuation signal and the weak magnetic detection signal of the buried steel pipeline are obtained; at the same time, the upper shielding plate is closed to shield the upper aerial pipeline from interfering with the transmitting coil, the receiving coil, and the weak magnetic detection sensor.

6. The steel pipeline defect detection method according to claim 4, characterized in that: During one-way upward detection, the upper shielding plate is pulled open to form an unobstructed transient electromagnetic field between the transmitting coil, the receiving coil and the surface of the aerial steel pipeline, and an unobstructed weak magnetic detection magnetic field between the weak magnetic detection sensor and the surface of the aerial steel pipeline; finally, the transient electromagnetic attenuation signal and the weak magnetic detection signal of the aerial steel pipeline are obtained; at the same time, the lower shielding plate is closed to shield the buried steel pipeline below from interfering with the transmitting coil, the receiving coil and the weak magnetic detection sensor.

7. The steel pipeline defect detection method according to claim 1, characterized in that: In step S3, the collected transient electromagnetic attenuation signal and weak magnetic detection signal are subjected to data alignment and normalization processing.

Images