CN104034796B - Detection generating date device and method in a kind of pipe leakage - Google Patents

Abstract

A real-time processing device and method for pipeline magnetic flux leakage internal detection data, belonging to the technical field of pipeline detection, the device is installed on a pipeline internal detector, including: a magnetic flux leakage sensor unit, a signal conditioning module, an A/D conversion module and a central processing unit The magnetic flux leakage sensor unit includes a plurality of magnetic flux leakage sensors, which are evenly arranged on the detector in the pipeline along the circumferential direction of the cross section of the pipeline; the central processing unit includes a timing control module, a defect data discrimination module, and a defect data feature extraction module and data storage module; the method of the present invention can quickly identify and extract data features of abnormal data in defect detection, and only record the characteristics and related information of abnormal data, and can detect within 30 minutes after the internal detector completes the detection Analyzing the location of serious defects better prevents serious defects from leaking and causing catastrophic consequences.

Description

Device and method for real-time processing of pipeline magnetic flux leakage internal detection data

technical field

The invention belongs to the technical field of pipeline detection, and in particular relates to a device and method for real-time processing of pipeline magnetic flux leakage internal detection data.

Background technique

Pipeline transportation is an extremely important mode of transportation. For pipelines that have been in service for a long time, it is extremely necessary to analyze the location, shape and danger of defects by means of magnetic flux leakage testing. The judgment of defect shape and risk degree is mainly based on the data characteristics of the detection data at the defect. Therefore, the feature extraction of abnormal detection data is the most important link in the whole magnetic flux leakage detection process.

The internal detection of pipeline defects mainly uses the in-pipeline detector for magnetic flux leakage detection, and stores the detection data in the storage device of the in-pipeline detector, and then performs data processing after the detection is completed. The real-time and intelligence of data processing are poor. At the same time, adopting this data storage and processing method will require a storage device with a large amount of storage space. And requires a large power consumption. Because the oil pipeline, especially the submarine oil pipeline, generally has a long distance between nodes. The electric energy of the detector in the pipeline is relatively tight. Therefore, such a data storage and processing method is not very perfect. At present, there are relatively few studies on other data storage methods.

Contents of the invention

Aiming at the deficiencies in the prior art, the present invention provides a device and method for real-time processing of pipeline magnetic flux leakage internal detection data.

Technical scheme of the present invention:

A real-time processing device for pipeline magnetic flux leakage internal detection data is installed on a pipeline internal detector, including: a magnetic flux leakage sensor unit, a signal conditioning module, an A/D conversion module and a central processing unit;

The magnetic flux leakage sensor unit includes a plurality of magnetic flux leakage sensors, which are evenly arranged on the detector in the pipeline along the circumferential direction of the cross section of the pipeline. Each magnetic flux leakage sensor in the magnetic flux leakage sensor unit is used to detect the magnetic flux leakage signal of the pipeline and output an electrical Signal to signal conditioning module;

The signal conditioning module is used to filter and amplify the received electrical signal, and send the filtered and amplified electrical signal to the A/D conversion module;

The A/D conversion module is used to perform analog-to-digital conversion on the electrical signal received from the signal conditioning module and transmit the converted digital electrical signal to the central processing unit;

The central processing unit includes a timing control module, a defect data discrimination module, a defect data feature extraction module, and a data storage module;

The timing control module is used to control the conversion sequence of each path of the A/D conversion module;

The defect data discrimination module is used to receive the digital signal sent by the A/D conversion module, and discriminate the abnormal data and the validity of the abnormal data therein; determine the sorting number of each effective abnormal data in the received real-time data; Send the identified valid abnormal data to the defect data feature extraction module, and send the sorting number corresponding to the valid abnormal data to the data storage module;

The defect data feature extraction module is used to extract features from the received effective abnormal data, and send the extracted effective abnormal data features to the data storage module;

The data storage module is used for storing and outputting the received effective abnormal data features and their ranking numbers.

The method for real-time processing of pipeline magnetic flux leakage internal detection data by using the real-time processing device for pipeline magnetic flux leakage internal detection data includes the following steps:

Step 1: Obtain real-time pipeline internal detection magnetic flux leakage data and internal detection historical magnetic flux leakage data of normal pipe sections;

Taking the pipeline section between two girth welds as a unit, detect the magnetic flux leakage signal of multiple normal pipe sections, and establish the internal detection history magnetic flux leakage data record of the normal pipe section; the normal pipe section refers to the pipe section without defects;

Step 2: Determine the abnormal data threshold according to the internal inspection history magnetic flux leakage data of normal pipe sections;

From the internal detection history magnetic flux leakage data of multiple normal pipe sections, respectively determine the maximum value of the magnetic flux leakage data of each normal pipe section, and obtain the average value of the maximum value of the magnetic flux leakage data of multiple normal pipe sections At the same time, obtain the average value of the magnetic flux leakage data of one section of normal pipe section; detect a section of defective pipe section, and obtain the average value of the magnetic flux leakage data of the defective pipe section; obtain the average value of the magnetic flux leakage data of the defective pipe section and the normal value of the section The ratio of the mean values of the pipe flux leakage data; that is,

$\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; no</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; li</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; =</span>\frac{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}$

In the formula: n is the number of normal pipe sections; X _max is the maximum value of magnetic flux leakage data for each normal pipe section; is the average value of the maximum value of magnetic flux leakage data of multiple normal pipe sections; is the average value of magnetic flux leakage data of a normal pipe section; X _i is the magnetic flux leakage data point of a normal pipe section; N ₁ is the number of magnetic flux leakage data points of a normal pipe section; is the average value of magnetic flux leakage data for a section of defective pipe section; X _1i is the data point of magnetic flux leakage for a section of defective pipe section; N ₂ is the number of data points for magnetic flux leakage of a section of defective pipe section; The ratio of the average value of the pipeline magnetic flux leakage data;

Then the abnormal data threshold is

Step 3: According to the abnormal data threshold, separate the abnormal data in the magnetic flux leakage detection data in the real-time pipeline, and record the abnormal data and its axial position; the axial direction refers to the length direction of the pipeline;

The method for separating the abnormal data is: the magnetic flux leakage data detected in the real-time pipeline greater than the abnormal data threshold K is regarded as abnormal data;

Step 4: Use the correlation analysis method to judge the validity of the abnormal data by comparing the abnormal data with its adjacent data; the adjacent data refers to the sensor detected at the same axial position as the sensor adjacent to the sensor that detected the abnormal data data;

The method of calculating the covariance between the abnormal data and its adjacent data is used to measure the correlation between the abnormal data and its adjacent data; because the distance between the adjacent two groups of sensors is relatively close, the adjacent two groups of signals have a certain similarity If the two sets of data reach a certain degree of positive correlation, it is considered that the data collection is correct, and the separated abnormal data is valid data; otherwise, the data collection is considered to be wrong, and the separated abnormal data is invalid data;

Step 5: Use the cubic spline interpolation method to perform cubic spline interpolation on the effective abnormal data measured by each sensor in real time, respectively, and obtain the cubic spline interpolation fitting curves of the effective abnormal data measured by each sensor in real time, referred to as each a curve;

The ordinate of each of the curves is the abnormal data value, and the abscissa is the axial detection position of the abnormal data;

Step 6: According to each curve, obtain the eigenvalues of the effective abnormal data measured by each sensor in real time;

Step 6.1: Obtain the axial eigenvalues of the effective abnormal data measured by each sensor in real time;

Step 6.1.1: Determine the peak value and valley value of each curve, and calculate the peak-valley value of each curve; the peak-valley value refers to the difference between the peak value and the valley value; the size of the peak-valley value is related to the depth of the pipeline defect; That is, determine the maximum value, minimum value, and difference between the maximum value and the minimum value of the effective abnormal data measured by each sensor in real time;

Step 6.1.2: Determine the abscissa value of the lowest valley point of each curve and the abscissa value of the second valley point, and calculate the valley value of each curve; the valley value refers to the abscissa value of the lowest valley point and The difference between the abscissa values of the sub-trough point; that is, determine the axial position of the minimum value and the axial position of the sub-minimum value of the effective abnormal data measured by each sensor in real time, and the axial position of the minimum value and the sub-minimum value The difference between the axial position of the value; the size of the valley value is related to the length of defect corrosion;

Step 6.1.3: Calculate the area of each curve respectively;

The formula for calculating the area of each curve is:

$\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span><span\; class="notranslate">\; \}</span>$

In the formula, S is the area of the curve, x _i (t) is the abnormal data point, min[ _xi (t)] is the minimum abnormal data;

Step 6.1.4: Calculate the energy of each curve;

The formula for calculating the energy of each curve is: Among them, _Se is the curve energy;

Step 6.1.5: Utilize the method of wavelet transform to obtain the inflection point spacing on each curve;

Method: Since the position of the inflection point required to be obtained is between the peak and the two troughs, only the part between the two troughs of the lowest valley point and the second lowest valley point of each curve is subjected to wavelet transformation, and the selected wavelet base is The second-order derivative of the smooth function; the zero point of the curve after continuous wavelet transformation is the inflection point of the original curve. The present invention selects the Mexican hat wavelet as the base wavelet, and performs abnormal data in the range of scale a=4 to a=32 Continuous wavelet transform. The wavelet scale is selected according to the transformation effect of the specific abnormal signal, and the effect is the best when the general scale is a=8, and the positions of the inflection points of each curve are obtained;

Step 6.1.6: Calculate the ratio of the peak-to-valley value to the valley-valley value, the ratio of the area to the peak-to-valley value, and the ratio of the area to the valley-to-valley value of each curve;

Step 6.2: Obtain the circumferential eigenvalues of the effective abnormal data measured by each sensor in real time; the circumferential direction refers to the circumferential direction of the pipe section;

Step 6.2.1: Determine the circumferential abnormal data threshold;

The method flow is: take the peak value of the effective abnormal data measured by the current sensor as the base point, and first judge the peak value of the effective abnormal data measured by the sensor immediately to the left or right of the current sensor near the axial position corresponding to the current sensor , if it is greater than or equal to the peak value of the effective abnormal data measured by the current sensor, then take this peak value as the maximum peak value, and continue to judge the peak value of the effective abnormal data measured by subsequent adjacent sensors in the same way along this direction , until the peak value of the effective abnormal data measured by the latter sensor is smaller than the peak value of the effective abnormal data measured by the previous sensor, the final maximum peak value is obtained, and the maximum peak value and the sensor corresponding to the maximum peak value are recorded; if If it is less than the peak value of the effective abnormal data measured by the current sensor, then judge the peak value of the effective abnormal data measured by the sensor immediately to the right or left of the current sensor, if it is greater than or equal to the peak value of the effective abnormal data measured by the current sensor , then take this peak value as the maximum peak value, and continue to judge the peak value of the effective abnormal data measured by the subsequent adjacent sensors in the same way in the same way until the peak value of the effective abnormal data measured by the subsequent sensor is smaller than the peak value of the previous sensor. Up to the peak value of the effective abnormal data measured by a sensor, the final maximum peak value is obtained, and the maximum peak value and the sensor corresponding to the maximum peak value are recorded; if the effective abnormality measured by the adjacent sensors in the left and right directions If the peak values of the data are smaller than the peak value of the effective abnormal data measured by the current sensor, the peak value of the effective abnormal data measured by the current sensor is taken as the final maximum peak value, and the maximum peak value and the current sensor are recorded; the final maximum peak value The half value of is used as the threshold of abnormal data in the circumferential direction;

Step 6.2.2: Determine the circumferential anomaly data;

The method is: if the peak value of the effective abnormal data measured by the sensors adjacent to the left and right directions of the sensor to which the maximum peak value belongs is greater than the threshold value of the circumferential abnormal data, it is regarded as the circumferential abnormal data and the corresponding sensor is regarded as being covered by defects in the pipe section. Within the range, the number of sensors covered by circumferential abnormal data and pipe section defects is obtained;

And record the axial position of the peak of the abnormal data measured by the sensor covered by the pipe defect. Based on the difference between the sensor position and the detection error, the axial position of each peak is different, and the least square method is used to determine the relative error. The position of , as the axial position of the peak value of the group of circumferential abnormal data;

Step 6.2.3: Use the cubic spline interpolation method to perform cubic spline interpolation on the circumferential abnormal data respectively to obtain the cubic spline interpolation fitting curve of the circumferential abnormal data, referred to as the circumferential curve;

Step 6.2.4: Calculate the area sum of the cubic spline interpolation fitting curve of the effective abnormal data measured in real time by multiple sensors covered by the pipe section defects;

Step 6.2.5: Take the same method as Step 6.1.1 to Step 6.1.6, and repeat Step 6.1.1 to Step 6.1.6 to obtain the characteristics of the circumferential curve, including: peak-valley value, valley-valley value, area , energy, distance between inflection points, ratio of peak to valley, area to peak to valley, area to valley to valley;

Step 7: output the eigenvalues of the effective abnormal data obtained in step 6;

Beneficial effect:

Utilizing the method of the present invention can quickly identify and extract data features of abnormal data in defect detection, and only record the characteristics and related information of abnormal data, and the amount of stored data is significantly higher than the traditional method of recording all data. significantly reduced. Due to the large amount of data recorded in the traditional data recording method, it generally takes several days or even ten days to obtain a complete data analysis report. However, the application of this method can analyze the location of serious defects within 30 minutes after the detection by the internal detector, thereby better preventing these serious defects from leaking and causing catastrophic consequences.

Description of drawings

Fig. 1 is a structural schematic diagram of a real-time processing device for pipeline magnetic flux leakage internal detection data according to an embodiment of the present invention;

Fig. 2 is the circuit principle diagram of the signal conditioning module of an embodiment of the present invention;

Fig. 3 is the interface circuit diagram of ADS7844 analog-to-digital converter and EP4CE15F17C8 FPGA of an embodiment of the present invention;

Fig. 4 is the EP4CE15F17C8 FPGA working flowchart of an embodiment of the present invention;

Fig. 5 is a working flow chart of the defective data discrimination module of an embodiment of the present invention;

Fig. 6 is a working flow chart of the defect data feature extraction module in an embodiment of the present invention;

Fig. 7 (a) is a schematic diagram of the curve peak-valley value of an embodiment of the present invention; (b) is a schematic diagram of the curve valley value of an embodiment of the present invention; (c) is a schematic diagram of the curve area of an embodiment of the present invention (d) is a schematic diagram of the curve inflection point spacing of an embodiment of the present invention;

FIG. 8 is a schematic diagram of the distance between inflection points of a curve after wavelet transformation according to an embodiment of the present invention.

detailed description

An embodiment of the present invention will be described in detail below in conjunction with the accompanying drawings.

The device for real-time processing of pipeline magnetic flux leakage internal detection data in this embodiment is installed on the pipeline internal detector, as shown in Figure 1, including: magnetic flux leakage sensor unit, signal conditioning module, A/D conversion module and central processing unit;

The magnetic flux leakage sensor unit in this embodiment includes 96 magnetic flux leakage sensors, which are evenly arranged on the detector in the pipeline along the circumferential direction of the pipeline section, and each magnetic flux leakage sensor in the magnetic flux leakage sensor unit is used to detect the pipeline magnetic flux leakage signal And output the electrical signal to the signal conditioning module;

The signal conditioning module of this embodiment is used to filter and amplify the received electrical signal, and send the filtered and amplified electrical signal to the A/D conversion module; the signal conditioning module of this embodiment is shown in Figure 2 As shown, the signal received from the Hall sensor is firstly filtered by the filter circuit, and then connected to the inverting input terminal 7 of the AD824 operational amplifier through the resistor R2 with a resistance value of 10K, and the non-inverting input terminal 8 is connected to a reference voltage of 2.5V , the output terminal 6 of the AD824 operational amplifier is connected to one end of a resistor R3 with a resistance value of 20K, one end of a resistor R1 with a resistance value of 20K, and one end of a 0.01pF capacitor C2, and the other end of the resistor R3 is connected to the output terminal of the signal conditioning module The input end of the A/D conversion chip, the other end of the resistor R1 is connected to the inverting input end of the operational amplifier, the other end of the capacitor C2 is grounded, the inverting input end 7 of the AD824 operational amplifier is also connected to one end of the 100pF capacitor C1, and the capacitor C1 The other end of the ground.

The A/D conversion module in this embodiment is used to perform analog-to-digital conversion on the pulse electrical signal received from the signal conditioning module and transmit the converted digital signal to the central processing unit under the control of the timing control module in the central processing unit. Processing unit; the A/D conversion module in this embodiment adopts an analog-to-digital converter whose model is ADS7844.

What the central processing unit in the present embodiment adopts is the FPGA that the model is EP4CE15F17C8, comprises sequence control module, defect data discriminating module, defect data characteristic extraction module and data storage module; The sequence control module in the present embodiment is used for controlling A The conversion order of each path of the /D conversion module; the data storage module in this embodiment is used to store the received abnormal data characteristics and their sorting numbers.

The interface circuit between EP4CE15F17C8FPGA and ADS7844 analog-to-digital converter in this embodiment, as shown in Figure 3, the ADS7844 analog-to-digital converter converts the voltage signal into a digital signal, and the five different output terminals of the ADS7844 analog-to-digital converter are respectively connected to the FPGA timing The custom I/O port of the control module, that is, the CS terminal of the ADS7844 analog-to-digital converter is connected to the I/O.71 terminal, the BUSY terminal of the ADS7844 analog-to-digital converter is connected to the I/O.72 terminal, and the DCLK of the ADS7844 analog-to-digital converter The terminal is connected to the I/O.73 terminal of the FPGA, the DIN terminal of the ADS7844 analog-to-digital converter is connected to the I/O.74 terminal, and the DOUT terminal of the ADS7844 analog-to-digital converter is connected to the I/O.75 terminal.

The workflow of the EP4CE15F17C8 FPGA in this embodiment, as shown in FIG. 4 , starts at step 401 .

In step 402, in this embodiment, the digital signal converted by the ADS7844 analog-to-digital converter is sent to the defect data discrimination module, and the defect data discrimination module discriminates the abnormal data and the validity of the abnormal data in the received real-time data, and the effective abnormal The data is the pipeline defect data; determine the sorting number of each valid abnormal data in the received real-time data; send the determined valid abnormal data to the defect data feature extraction module, and send the sorting number corresponding to the valid abnormal data to the data storage module ;

In step 403, the defect data feature extraction module in this embodiment extracts features from the received effective abnormal data, and sends the extracted feature values of the effective abnormal data to the data storage module;

In step 404, the data storage module in this embodiment outputs the sorting number corresponding to the stored valid abnormal data and the characteristic value of the valid abnormal data;

The workflow of the defect data discrimination module, as shown in FIG. 5 , starts at step 501 .

In step 502, obtain the real-time detection magnetic flux leakage data in the pipeline and the internal detection historical magnetic flux leakage data of the normal pipe section;

Taking the pipeline section between two girth welds as a unit, detect the magnetic flux leakage signal of multiple normal pipe sections, and establish the internal detection history magnetic flux leakage data record of the normal pipe section; the normal pipe section refers to the pipe section without defects;

In step 503, an abnormal data threshold is determined according to the internal detection history magnetic flux leakage data of the normal pipe section;

From the internal detection history magnetic flux leakage data of multiple normal pipe sections, respectively determine the maximum value of the magnetic flux leakage data of each normal pipe section, and obtain the average value of the maximum value of the magnetic flux leakage data of multiple normal pipe sections At the same time, obtain the average value of the magnetic flux leakage data of one section of normal pipe section; detect a section of natural corrosion pipe section, and obtain the average value of the magnetic flux leakage data of the natural corrosion pipe section; obtain the average value of the magnetic flux leakage data of the natural corrosion pipe section and the obtained The ratio of the average value of the magnetic flux leakage data of a normal pipe section; that is,

$\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; no</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; li</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; =</span>\frac{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}$

In the formula: n is the number of normal pipe sections; X _max is the maximum value of magnetic flux leakage data for each normal pipe section; is the average value of the maximum value of magnetic flux leakage data of multiple normal pipe sections; is the average value of magnetic flux leakage data of a normal pipe section; X _i is the magnetic flux leakage data point of a normal pipe section; N ₁ is the number of magnetic flux leakage data points of a normal pipe section; is the average value of the magnetic flux leakage data of the naturally corroded pipe section; X _1i is the magnetic flux leakage data point of the naturally corroded pipe section; N ₂ is the number of data points of the magnetic flux leakage data of the naturally corroded pipe section; The ratio of the means of the data;

Then the abnormal data threshold is

In step 504: according to the abnormal data threshold, the abnormal data in the magnetic flux leakage detection data in the real-time pipeline is separated, and the abnormal data and its axial position are recorded; the axial direction refers to the length direction of the pipeline;

The method for separating the abnormal data is: the magnetic flux leakage data detected in the real-time pipeline greater than the abnormal data threshold K is regarded as abnormal data;

In step 505: Using the correlation analysis method, the validity of the abnormal data is judged by comparing the abnormal data with its adjacent data; adjacent data refers to the sensor detected at the same axial position as the sensor adjacent to the sensor that detected the abnormal data The data;

The method of calculating the covariance between the abnormal data and its adjacent data is used to measure the correlation between the abnormal data and its adjacent data; because the distance between the adjacent two groups of sensors is relatively close, the adjacent two groups of signals have a certain similarity If the two sets of data reach a certain degree of positive correlation, it is considered that the data collection is correct, and the separated abnormal data is valid data; otherwise, the data collection is considered to be wrong, and the separated abnormal data is invalid data;

COV(x,y)=E[(x _i -E(x _i ))(y _i -E(y _i ))] (i=1,2,...,n)

In the formula, COV(x, y) is the covariance of the two sets of data of x and y; x _i is the abnormal data point; y _i is the adjacent data point of x _i ; n ₁ is the number of abnormal data points; E(x) is the expectation of abnormal data points; E(y) is the expectation of adjacent data points of x _i ;

Covariance represents the correlation between two variables. If the trend of the two variables is consistent, that is, if one of them is greater than its own expected value, and the other is also greater than its own expected value, then the covariance between the two variables is Positive value. If two variables have opposite trends, that is, one of them is greater than its expected value and the other is less than its own expected value, then the covariance between the two variables is negative. If x and y are statistically independent, then the covariance between the two is 0; the validity of the abnormal data is judged by the threshold value of the covariance coefficient r, and the threshold value of the covariance coefficient r determined by experiments in this implementation method is 0.75 , the abnormal data corresponding to the covariance coefficient greater than the threshold 0.75 is effective abnormal data;

$\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; )</span>}{\sqrt{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\sqrt{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>\sqrt{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}$

In step 506, using the cubic spline interpolation method, cubic spline interpolation is performed on the effective abnormal data measured by each sensor in real time, respectively, and the cubic spline interpolation fitting curves of the effective abnormal data measured by each sensor in real time are respectively obtained, referred to as various curves;

The ordinate of each of the curves is the abnormal data value, and the abscissa is the axial detection position of the abnormal data;

The workflow of the defect data feature extraction module, as shown in FIG. 6 , starts at step 601 .

In step 602, obtain the axial eigenvalues of the effective abnormal data measured by each sensor in real time;

A. Determine the peak value and the valley value of each curve, and calculate the peak-valley value Y _pp of each curve; As shown in Figure 7 (a), the peak-valley value Y _pp refers to the difference between the peak value and the valley value; That is, Determine the maximum value, minimum value and the difference between the maximum value and the minimum value of the effective abnormal data measured by each sensor in real time; the size of the peak and valley value Y _pp is related to the depth of the pipeline defect;

B. determine the abscissa value of the lowest valley point of each curve and the abscissa value of the second valley point, and calculate the valley value X _pp of each curve; As shown in Figure 7 (b), the valley value X _pp refers to is the difference between the abscissa value of the lowest valley point and the abscissa value of the sub-valley point; that is, to determine the axial position of the minimum value and the axial position of the second minimum value of the effective abnormal data measured by each sensor in real time, and the The difference between the axial position of the minimum value and the axial position of the second minimum value; the size of the valley value X _pp is related to the length of defect corrosion;

C. Calculate the area of each curve separately;

As shown in Figure 7(c), the formula for calculating the area of each curve is:

$\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span><span\; class="notranslate">\; \}</span>$

In the formula, S is the area of the curve, x _i (t) is the abnormal data point, and min[ _xi (t)] is the minimum abnormal data; the size of the curve area is related to the comprehensive situation of the length and depth of defect corrosion;

D. Calculate the energy of each curve;

The formula for calculating the energy of each curve is: Among them, _Se is the curve energy; the size of the curve energy is related to the comprehensive situation of the length and depth of defect corrosion;

E. Utilize the wavelet transform method to find the inflection point spacing on each curve;

Method: As shown in Figure 7(d), X _kk represents the inflection point spacing, and the size of the inflection point spacing X _kk is related to the length of defect corrosion; since the required inflection point positions are respectively between the peak and the two troughs, only Carry out wavelet transformation to the part between two troughs of each curve, the selected wavelet basis is the second order derivative of smooth function; The zero point of the curve after continuous wavelet transformation is the inflection point of the original curve, and the present invention selects the Mexican hat The wavelet is used as the basic wavelet, and the continuous wavelet transform is performed on the abnormal data in the range of scale a=4 to a=32. The wavelet scale is selected according to the transformation effect of the specific abnormal signal, and the effect is the best when the general scale is a=8, and the positions of the inflection points of each curve are obtained, as shown in Figure 8;

F. Find the ratio of the peak-valley value Y _pp to the valley value X _pp of each curve, the ratio of the area S to the peak-valley value Y _pp , the ratio of the area S to the valley value X _pp ;

In step 603, the circumferential characteristic value of the effective abnormal data measured by each sensor in real time is obtained; the circumferential direction refers to the circumferential direction of the pipeline section;

H. Determine the circumferential abnormal data threshold;

The method flow is: take the peak value of the effective abnormal data measured by the current sensor as the base point, and first judge the peak value of the effective abnormal data measured by the sensor immediately to the left or right of the current sensor near the axial position corresponding to the current sensor , if it is greater than or equal to the peak value of the effective abnormal data measured by the current sensor, then take this peak value as the maximum peak value, and continue to judge the peak value of the effective abnormal data measured by subsequent adjacent sensors in the same way along this direction , until the peak value of the effective abnormal data measured by the latter sensor is smaller than the peak value of the effective abnormal data measured by the previous sensor, the final maximum peak value is obtained, and the maximum peak value and the sensor corresponding to the maximum peak value are recorded; if If it is less than the peak value of the effective abnormal data measured by the current sensor, then judge the peak value of the effective abnormal data measured by the sensor immediately to the right or left of the current sensor, if it is greater than or equal to the peak value of the effective abnormal data measured by the current sensor , then take this peak value as the maximum peak value, and continue to judge the peak value of the effective abnormal data measured by the subsequent adjacent sensors in the same way, until the peak value of the effective abnormal data measured by the latter sensor is smaller than the previous one. Up to the peak value of the effective abnormal data measured by a sensor, the final maximum peak value is obtained, and the maximum peak value and the sensor corresponding to the maximum peak value are recorded; if the effective abnormality measured by the adjacent sensors in the left and right directions If the peak values of the data are smaller than the peak value of the effective abnormal data measured by the current sensor, the peak value of the effective abnormal data measured by the current sensor is taken as the final maximum peak value, and the maximum peak value and the current sensor are recorded; the final maximum peak value The half value of is used as the threshold of abnormal data in the circumferential direction;

I. Determine the circumferential abnormal data;

The method is: if the peak value of the effective abnormal data measured by the sensors adjacent to the left and right directions of the sensor to which the maximum peak value belongs is greater than the threshold value of the circumferential abnormal data, it is regarded as the circumferential abnormal data and the corresponding sensor is regarded as being covered by defects in the pipe section. Within the range, the number of sensors covered by circumferential abnormal data and pipe section defects is obtained;

And record the axial position of the peak of the abnormal data measured by the sensor covered by the pipe defect. Based on the difference between the sensor position and the detection error, the axial position of each peak is different, and the least square method is used to determine the relative error. The position of , as the axial position of the peak value of the group of circumferential abnormal data;

J. Use the cubic spline interpolation method to perform cubic spline interpolation on the circumferential abnormal data respectively, and obtain the cubic spline interpolation fitting curve of the circumferential abnormal data, referred to as the circumferential curve;

K. Calculate the area sum of the cubic spline interpolation fitting curve of the effective abnormal data measured in real time by multiple sensors covered by the pipe section defects;

L. Take the same method as A to F, and repeat A to F to obtain the characteristics of the circumferential curve, including: peak-to-valley value, valley-to-valley value, area, energy, distance between inflection points, ratio of peak-to-valley value to valley-to-valley value , the ratio of the area to the peak-to-valley value, the ratio of the area to the valley-to-valley value;

Although the specific embodiments of the present invention have been described above, those skilled in the art should understand that these are only examples, and various changes or modifications can be made to these embodiments without departing from the principles and principles of the present invention. substance. The scope of the invention is limited only by the appended claims.

Claims (3)

1. a method for detection generating date in pipe leakage, uses detection generating date device in pipe leakage Realizing, this device is arranged on in-pipeline detector, comprising: leakage field sensor unit, Signal-regulated kinase, A/D modulus of conversion Block and CPU；

Described leakage field sensor unit includes multiple leakage field sensor, and the plurality of leakage field sensor is along pipeline section circumferencial direction Being evenly arranged on in-pipeline detector, each leakage field sensor in described leakage field sensor unit is used to detect pipeline leakage Magnetic signal also exports the signal of telecommunication to Signal-regulated kinase；

Described Signal-regulated kinase is for being filtered and processing and amplifying the signal of telecommunication received, and will filter and processing and amplifying After the signal of telecommunication and deliver to A/D modular converter；

Described A/D modular converter for carrying out analog digital conversion and by after conversion to the signal of telecommunication received from Signal-regulated kinase Digital signal is sent to CPU；

Described CPU, including time-sequence control mode, defective data discrimination module, defective data characteristic extracting module and Data memory module；

Described time-sequence control mode is for controlling the change over order of each path of A/D modular converter；

Described defective data discrimination module is used for receiving the digital signal that A/D modular converter transmits, and to abnormal data therein And the effectiveness of abnormal data differentiates；Determine each effective anomaly data sequence number in the real time data received；Will The effective anomaly data determined send to defective data characteristic extracting module, are sent by sequence number corresponding for effective anomaly data To data memory module；

Described defective data characteristic extracting module is for extracting feature to the effective anomaly data received and effective by extract The feature of abnormal data sends to data memory module；

Described data memory module is for storing the effective anomaly data characteristics received and sequence number thereof and export；

It is characterized in that: comprise the following steps:

Step 1: obtain the interior detection history magnetic flux leakage data of real-time pipeline detection magnetic flux leakage data and normal pipeline section；

The magnetic leakage signal of the detection normal pipeline section of multistage, and set up the interior detection history magnetic flux leakage data record of normal pipeline section；Normal pipe Section refers to the pipeline section not having defect to occur；

Step 2: according to the interior detection history magnetic flux leakage data of normal pipeline section, determine abnormal data threshold value；

Step 3: according to abnormal data threshold value, isolate the abnormal data in real-time pipeline detection magnetic flux leakage data, and to exception Data and axial location thereof carry out record；Axially refer to pipe lengths；

Step 4: utilize correlation analysis, is adjacent the comparison of data by abnormal data, it determines having of abnormal data Effect property；Adjacent data refers to the number that the sensor adjacent with sensor abnormal data being detected detects in same axial position According to；

Use the method calculating the covariance that abnormal data is adjacent data, weigh abnormal data and be adjacent between data Dependency；The computing formula of covariance coefficient r is as follows, then corresponding to the covariance coefficient of the threshold value being more than covariance coefficient r Abnormal data is effective abnormal data；The threshold value of covariance coefficient r is determined by experiment；

$r=\frac{\underset{i=1}{\overset{n_1}{\&Sigma;}}(x_i-\stackrel{\&OverBar;}{x})(y_i-\stackrel{\&OverBar;}{y})}{\sqrt{\underset{i=1}{\overset{n_1}{\&Sigma;}}{(x_i-\stackrel{\&OverBar;}{x})}^2{(y_i-\stackrel{\&OverBar;}{y})}^2}}=\frac{n_1\underset{i=1}{\overset{n_1}{\&Sigma;}}{x}_i{y}_i-\underset{i=1}{\overset{n_1}{\&Sigma;}}{x}_i\&CenterDot;\underset{i=1}{\overset{n_1}{\&Sigma;}}{y}_i}{\sqrt{n_1\underset{i=1}{\overset{n_1}{\&Sigma;}}{x_i}^2-{\left(\underset{i=1}{\overset{n_1}{\&Sigma;}}{x}_i\right)}^2}\&CenterDot;\sqrt{n_1\underset{i=1}{\overset{n_1}{\&Sigma;}}{y_i}^2-{\left(\underset{i=1}{\overset{n_1}{\&Sigma;}}{y}_i\right)}^2}},i=1,2,...,n_1$

In formula, x_iFor effective anomaly data point；For x_iMeansigma methods；y_iFor x_iConsecutive number strong point；For y_iMeansigma methods；n₁For Exceptional data point quantity；

Step 5: utilize cubic spline interpolation, the effective abnormal data recorded each sensor in real time respectively carries out three samples Bar interpolation, respectively obtains the cubic spline interpolation matched curve of the effective anomaly data that each sensor records in real time, is called for short each bar Curve；

The vertical coordinate of described each bar curve is abnormal data value, abscissa be abnormal data axially detect position；

Step 6: respectively according to each bar curve, ask for the eigenvalue of the effective anomaly data that each sensor records in real time；

Step 6.1: ask for the axial eigenvalue of effective anomaly data；

Step 6.1.1: determine peak value and the valley of each bar curve, and calculate the peak-to-valley value of each bar curve；Peak-to-valley value refers to peak Value and the difference of valley；That is, determine the maximum of effective anomaly data, minima and maximum that each sensor records in real time with The difference of little value；The size of peak-to-valley value is relevant with the degree of depth of defect of pipeline；

Step 6.1.2: determine abscissa value and the abscissa value of time low valley point of the minimum valley point of each bar curve, and calculate each bar The paddy valley of curve；Paddy valley refers to the abscissa value of minimum valley point and the difference of the abscissa value of time low valley point；That is, determine respectively The axial location of the minima of the effective anomaly data that sensor records in real time and the axial location of secondary minima, and described minimum The axial location of value and the difference of the axial location of time minima；The size of paddy valley is relevant with the length of defect etching；

Step 6.1.3: calculate the area of each bar curve respectively；

The computing formula of each bar area under the curve is:

$S=\underset{i=1}{\overset{N}{\&Sigma;}}\{x_i\left(t\right)-min\&lsqb;x_i\left(t\right)\&rsqb;\}$

In formula, S is area under the curve, x_iT () is exceptional data point, min [x_i(t)] it is minimum abnormal data；N represents each bar curve Exceptional data point quantity；

Step 6.1.4: calculate each bar curve energy；

The computing formula of each bar curve energy is:Wherein, S_eFor curve energy；

Step 6.1.5: the method utilizing wavelet transformation, asks for the flex point spacing on each bar curve；

Method: due to the required corner position taken respectively between crest and two troughs, therefore only respectively to each bar curve Minimum valley point and secondary low valley point between part carry out wavelet transformation, the wavelet basis chosen is the second dervative of smooth function； The zero point of the curve after continuous wavelet transform, is the flex point of virgin curve；

Step 6.1.6: ask for peak-to-valley value and the ratio of paddy valley, area and the ratio of peak-to-valley value, area and the paddy of each bar curve The ratio of valley；

Step 6.2: ask for the circumferential eigenvalue of effective anomaly data；Circumference refers to pipeline section circumferencial direction；

Step 6.2.1: determine circumference abnormal data threshold value；

Method flow is: the peak value of the effective anomaly data recorded with current sensor is as basic point, in corresponding axial of this peak value Near position, first the peak value of the effective anomaly data that the sensor of on the left of current sensor or right side next-door neighbour records is sentenced Disconnected, if the peak value of its effective anomaly data recorded more than or equal to current sensor, then using this peak value as peak-peak, press According to same method, the peak value continuing on the effective anomaly data that follow-up adjacent sensors is recorded by the direction successively is sentenced Disconnected, until the peak of effective anomaly data that the peak value of effective anomaly data that a rear sensor records records less than previous sensor Till value, obtain final peak-peak, and the sensor corresponding to this peak-peak and this peak-peak is carried out record；If The peak value of its effective anomaly data recorded less than current sensor, the then sensor on the right side of current sensor or left side next-door neighbour The peak value of the effective anomaly data recorded judges, if its effective anomaly data recorded more than or equal to current sensor Peak value, then using this peak value as peak-peak, after the same method, continue on the direction successively to follow-up adjacent sensors The peak value of the effective anomaly data recorded judges, until the peak value of effective anomaly data that a rear sensor records is less than front Till the peak value of the effective anomaly data that one sensor records, obtain final peak-peak, and to this peak-peak and this Big sensor corresponding to peak value carries out record；If what the adjacent sensors of the left and right both direction of current sensor recorded has The peak value of the effective anomaly data that the peak value respectively less than current sensor of effect abnormal data records, then record current sensor The peak value of effective anomaly data is as final peak-peak, and this peak-peak and current sensor are carried out record；Will be The half value of whole peak-peak is as circumference abnormal data threshold value；

Step 6.2.2: determine circumference abnormal data；

Method is: the effective anomaly data that the sensor that the left and right both direction of sensor belonging to described peak-peak is adjacent records Peak value more than circumference abnormal data threshold value be considered as circumference abnormal data and the sensor of correspondence is considered as covering in pipeline section defect In the range of lid, obtain circumference abnormal data and the number of sensors of pipeline section defect covering；

And the axial location at the peak value of abnormal data that records of the sensor covering pipeline section defect carries out record, based on sensing Device position difference and detection error, the axial location residing for each peak value has difference, utilizes method of least square to determine relative error Minimum position, as the axial location of circumference abnormal data peak value；

Step 6.2.3: utilize cubic spline interpolation, carries out cubic spline interpolation to circumference abnormal data respectively, obtains circumference The cubic spline interpolation matched curve of abnormal data, is called for short circumference curve；

Step 6.2.4: according to the result of step 6.1.3, it is effective that a plurality of sensor that run of designing defect covers records in real time The area of the cubic spline interpolation matched curve of abnormal data and；

Step 6.2.5: taking the method identical with step 6.1.1 to step 6.1.6, repeated execution of steps 6.1.1 is to step 6.1.6, obtain the feature of circumference curve, including: peak-to-valley value, paddy valley, area, energy, flex point spacing, peak-to-valley value and paddy valley Ratio, area and the ratio of peak-to-valley value, the ratio of area and paddy valley；

Step 7: the eigenvalue of the effective anomaly data of output step 6 gained.

The method of detection generating date in pipe leakage the most according to claim 1, it is characterised in that: described step The interior detection history magnetic flux leakage data according to normal pipeline section described in 2 determines that the method for abnormal data threshold value is as follows:

From the interior detection history magnetic flux leakage data of the normal pipeline section of multistage, determine the maximum of each normal pipeline section magnetic flux leakage data respectively, Try to achieve the meansigma methods of multistage normal pipeline section magnetic flux leakage data maximumMeanwhile, wherein one section of normal pipeline section magnetic flux leakage data is tried to achieve Meansigma methods；Detect one section of defect pipeline section, and try to achieve the meansigma methods of this defect pipeline section magnetic flux leakage data；Ask for the leakage of described defect pipeline section The ratio of the meansigma methods of the meansigma methods of magnetic data and described one section of normal pipeline section magnetic flux leakage data；That is,

$\stackrel{\&OverBar;}{X_{\max }}=\frac{1}{n}\underset{i=1}{\overset{n}{\&Sigma;}}{X}_{\max },\stackrel{\&OverBar;}{X}=\frac{1}{N_1}\underset{i=1}{\overset{N_1}{\&Sigma;}}{X}_i,\stackrel{\&OverBar;}{X_1}=\frac{1}{N_2}\underset{i=1}{\overset{N_2}{\&Sigma;}}{X}_{1i},C=\frac{\stackrel{\&OverBar;}{X_1}}{\stackrel{\&OverBar;}{X}}$

In formula: n is normal pipe hop count amount；X_maxMaximum for every section of normal pipeline section magnetic flux leakage data；For the normal pipeline section of multistage The meansigma methods of magnetic flux leakage data maximum；It it is one section of normal pipeline section magnetic flux leakage data meansigma methods；X_iIt it is one section of normal pipeline section leakage field number Strong point；N₁It is one section of normal pipeline section magnetic flux leakage data point quantity；It it is one section of existing defects pipeline section magnetic flux leakage data meansigma methods；X_1iIt is one Section existing defects pipeline section magnetic flux leakage data point；N₂It is one section of existing defects pipeline section magnetic flux leakage data point quantity；C is one section of defect pipeline section leakage The ratio of the meansigma methods of magnetic data and the meansigma methods of one section of normal pipeline section magnetic flux leakage data；

Then abnormal data threshold value is

The method of detection generating date in pipe leakage the most according to claim 1, it is characterised in that: described step The method of the abnormal data isolated in real-time pipeline detection magnetic flux leakage data described in 3 is as follows:

The method separating described abnormal data is: be considered as exception more than the real-time pipeline detection magnetic flux leakage data of abnormal data threshold value Data.