RU2571878C1 - Method of leak detection on pipelines with pump supply of transported media - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: method includes measurement of intrachannel pressure at serially arranged pipeline sections and correlation treatment of produced data for detection and localisation of a leak. Besides, prior to correlation treatment of data from at least one measured value of the pressure they directly subtract the component of the measurement signal specified by action of the pump equipment.

EFFECT: improved selectivity of control to leaks on pipelines with pump supply of transported medium.

4 cl, 7 dwg

Description

The invention relates to means for monitoring the condition of pipelines, namely, to detect and localize leaks in pipelines with pumping of the transported medium. This technical solution can be used to monitor the tightness of technological sections of main oil pipelines and oil pipelines, as well as gas pipelines.

During the operation of pipelines, great importance is attached to monitoring their integrity. At the same time, they strive to ensure high sensitivity to leaks, short detection time, high localization accuracy and the absence of false alarms. One of the factors that reduces the reliability of monitoring the status of pipelines with a pumped feed of the transported medium is due to the operation of pumping units that create a noise background.

From the patent document RU 2462656 C2, a combined hydroacoustic system for detecting leaks in an oil product pipeline is known, which implements a method for detecting leaks in pipelines with a pumped feed of a transported medium, including the measurement of intra-channel static and dynamic pressure in successive sections of the pipeline, correlation processing of measurement data from mutually adjacent to each other to other sections of the pipeline to detect a leak and determine the location of the leak in the control area. In order to reduce the interference hydroacoustic component in the oil product pipeline, the in-channel dynamic pressure sensors are located at a distance of at least 100-300 m from the pump station. However, in practice, the specified distance is not enough to effectively reduce noise from a running pump unit, which can lead, in particular, to the occurrence of false positives. In addition, the known leak detection system contains a signal filtering unit, however filtering, that is, indirect subtraction of interference, is ineffective for cleaning the measuring signals from the noise background of pumping equipment due to the superposition of leakage spectra and operating pumps that is usually observed in practice, as well as implementation difficulties adaptive filtration, necessary to obtain acceptable results in real operating conditions of pipelines.

The objective is to increase the reliability of control.

The technical result provided by the present invention is to increase the selectivity of control to leaks in pipelines with pumping of the transported medium.

This technical result is achieved due to the fact that a method for detecting leaks in a pipeline with a pumped feed of the transported medium, which includes measuring the channel pressure in successive sections of the pipeline and correlating processing of the data to detect and localize the leak, is characterized by direct subtraction of the component due to the action of the pump equipment, from at least one measured pressure value to correlation processing is given s.

, где: $${\tilde{D}}_i(t)$$

- результат очистки сигнала ближайшего к насосному оборудованию датчика динамического давления от помехового фона насосного оборудования; D_i(t) - измеренное значение давления с ближайшего к насосному оборудованию датчика динамического давления; ΔS_i,i+1(t) - составляющая, обусловленная действием насосного оборудования, вычисляемая по выражению $$\Delta S_{i,i+1}(t)=S_i(t)-S_{i+1}(t)$$

, где: S_i(t) - измеренное значение давления с ближайшего к насосному оборудованию датчика статического давления; S_i+1(t) - измеренное значение давления со следующего датчика статического давления.In the particular case, direct subtraction of the component due to the action of the pumping equipment is performed according to the expression $${\tilde{D}}_i(t)=D_i(t)-\Delta S_{i,i+\mathrm{one}}(t)$$

where: $${\tilde{D}}_i(t)$$

- the result of cleaning the signal closest to the pumping equipment of the dynamic pressure sensor from the interference background of the pumping equipment; D _i (t) is the measured pressure value from the dynamic pressure sensor closest to the pumping equipment; ΔS _{i, i + 1} (t) is the component due to the action of pumping equipment, calculated by the expression $$\Delta S_{i,i+\mathrm{one}}(t)=S_i(t)-S_{i+\mathrm{one}}(t)$$

where: S _i (t) is the measured pressure value from the static pressure sensor closest to the pumping equipment; S _{i + 1} (t) is the measured pressure value from the next static pressure sensor.

Also in the particular case, direct subtraction of the component due to the action of the pumping equipment is made from the measured values of at least two sensors in one section of the pipeline.

In another particular case, the decision to detect a leak is made by jointly registering the correlation peak of the leak and the pressure drop that is not typical for the normal operation of the pipeline.

The invention is illustrated by the following illustrations.

FIG. 1: block diagram of a leak detection system.

FIG. 2: a generalized functional diagram of a device for processing data and giving an alarm.

FIG. 3: algorithm for monitoring the technological section of the pipeline.

FIG. 4, 5: oscillograms of dynamic pressure from the first and second sensors in a controlled section of the pipeline.

FIG. 6: a graphical view of the data after correlation analysis (prior art).

Fig.7: data view after correlation analysis with preliminary cleaning of a component of the working pumping equipment (the present invention).

The implementation of the invention is shown in the following example.

The design of the pipeline is characterized in that a pump 2 is connected to the pipe 1, which creates excess pressure and ensures the flow of the transported medium through the pipeline. Through the wall of the pipe 1, a number of sensors 3-6 are introduced to measure the in-channel pressure. Sensors 3-6 are two-channel combined pressure receivers with a pair of transducers for static and dynamic pressure, respectively. Sensitivity of sensors 3-6 is approximately 1 Pa for static pressure and 0.01 Pa for dynamic pressure. Sensors 3-6 are installed in series so that they divide the pipeline into equal sections A, B, C of control, lying between the sensors 3-6 (Fig. 1), etc. All sensors 3-6 are connected through appropriate controllers to a device 7 for processing data and giving an alarm, thus forming an automated information system to detect possible leaks due to a defect 8 in the pipeline.

The main functional units of the device 7 for data processing and alarm signaling are (Fig. 2): interface part 9, first subtractor 10, correction delay line 11, second subtractor 12, correlation analyzer 13, logic analyzer 14 and monitor 15. Interface part 9 serves for information communication of sensor controllers 3-6 with the main data processing circuits. The signal inputs / outputs of the functional nodes of the device 7 are interconnected as follows. The first input (input 1) of the first subtractor 10 is directly connected to the output of the interface part 9, and the second input (input 2) is connected to the corresponding output of the interface part 9 through the correction delay line 11 to compensate for the time difference between the dynamic pressure signals from neighboring sensors. For example, sensor 4 will detect leakage noise earlier than sensor 3, which will require synchronization of their signals; therefore, the time delay value is determined by constant cross-correlation analysis of a continuous data stream at each monitoring site. The inputs of the second subtractor 12 are connected to the output of the interface part 9 and to the output of the first subtracter 10. The inputs of the correlation analyzer 13 are connected to the output of the interface part 9 and the output of the second subtractor 12, and the inputs of the logic analyzer 14 are connected to the output of the correlation analyzer 13 and the output of the first subtracter 10. The output of the logic analyzer 14 is connected to the input of the monitor 15.

The pumping of the transported medium is accompanied by monitoring the tightness of the pipeline using a leak detection system. In the process of monitoring, sensors 3-6 measure the intra-channel pressure in the pipe 1 (Fig. 3) with a frequency of at least 100 Hz (Figs. 4 and 5). The controllers convert the analog measuring signals into digital form and continuously transmit the stream of received data to the device 7 for cross-correlation (Fig. 6 or 7) and other data processing and the audiovisual alarm in case of leakage detected by the correlation peak 16, including an indication of the calculated coordinates leak points.

A component is constantly present in the measuring signal due to the action of pump 2, which creates a relatively strong pulsating noise with an amplitude of 1 ÷ 100 Pa at a leak noise level of 0.01 ÷ 1 Pa, which makes it difficult to find the correlation peak 16 of the leak due to the appearance of an interfering peak 17 of the pump ( Fig. 6).

In order to isolate the useful signal before the correlation processing of the digitized data begins, in the device 7, a component, caused by the action of the pump 2, is directly subtracted from the at least one measured value of the in-channel pressure in the pipe 1. Since in the section A of the pipeline, the pump 2 has the greatest effect on the sensor 3, it is from the signal of this sensor that the component due to the action of the pump 2 is subtracted, due to which, in this section of the pipeline, the noise is accurately eliminated time and, as a consequence, increasing the selectivity of control to leaks. A greater increase in the selectivity of control for leaks can be achieved by subtracting the component due to the action of the pump 2, both from the signal of the sensor 3 and from the signal of the sensor 4.

The direct subtraction of the component due to the action of the pump 2 is performed for the section A of the pipeline at each time point t according to expressions (1) and (2).

where: ΔS _3,4 (t) is the component due to the action of the pump 2;

S ₃ (t) - sensor 3 measured value of the static pressure;

S ₄ (t) is the static pressure value measured by sensor 4;

- результат очистки сигнала динамического давления с датчика 3 от помехового фона насоса 2; $${\tilde{D}}_3(t)$$

- the result of cleaning the dynamic pressure signal from the sensor 3 from the interference background of the pump 2;

D ₃ (t) - measured by the sensor 3 value of dynamic pressure.

The computing process in the device 7 includes the conversion of data when passing through the functional nodes 9-14.

From the interface part 9, data characterizing the value of S ₃ (t) is fed to the correction delay line 11, characterizing the value of S ₄ (t), - to the first input of the first subtractor 10, characterizing the value of D ₃ (t), - to the first input of the second a subtractor 12, characterizing the value of D ₄ (t), which is the dynamic pressure value measured by the sensor 4, to the first input of the correlation analyzer 13.

, поступающие на второй вход корреляционного анализатора 13. Результат взаимно корреляционного анализа с выхода узла 13 поступает на первый вход логического анализатора 14, точно выявляющего факт утечки по единственному корреляционному пику 16. Выходные данные логического анализатора 14 служат управляющим сигналом для монитора 15.In the first computational step, data is received from the output of the correction delay line 11, which eliminates the influence of the pumping speed, temperature, viscosity, density, and other properties of the transported medium on the result of processing the data. Then, the data arrives at the second input of the first subtractor 10. At the second clock, data is received from the output of the first subtractor 10, which is a difference signal characterizing the pulsation of the pumped medium from pump 2, that is, the value ΔS _3.4 (t) supplied to the second input of the second of the subtractor 12 and to the second input of the logic analyzer 14. At the third clock cycle, data is obtained from the output of the second subtractor 12, which is the value of the cleared, that is, the useful signal $${\tilde{D}}_3(t)$$

received at the second input of the correlation analyzer 13. The result of the cross-correlation analysis from the output of the node 13 is fed to the first input of the logical analyzer 14, which accurately reveals the fact of leakage at a single correlation peak 16. The output of the logical analyzer 14 serves as a control signal for the monitor 15.

Subsequent sections of the pipeline B, C, etc. they control in the same way, and the component to be subtracted from the operation of the pump 2 is determined individually for each section. Analysis of each section of the pipeline is carried out on the basis of data obtained from the sensor systems of this control section.

To further increase the selectivity of the control to leaks, the decision to detect the leak is made when the correlation peak of the leak and the pressure drop uncharacteristic for the normal operation of the pipeline are registered at the control section typical of the leak (Fig. 3). At the same time, a false response of the system due to impacts and extraneous noise from operation, in particular, transport or repair equipment, as well as other equipment in the immediate vicinity of the pipeline, which was impossible to implement by filtering the signals, is eliminated. If there is a correlation peak characteristic of a leak, and a static pressure drop in the pipeline does not occur in this section, then the logic analyzer 14 does not make a decision on the occurrence of a leak.

Claims (4)

1. A method for detecting a leak in a pipeline with a pumped feed of the transported medium, which includes measuring the intra-channel pressure in successive sections of the pipeline and correlating the received data to identify and localize the leak, characterized in that prior to the correlation data processing, a direct subtraction of the measuring signal component is performed, due to the action of pumping equipment from at least one measured pressure value. 2. The leak detection method according to claim 1, wherein the direct subtraction of the component due to the action of the pumping equipment is performed according to the expression

where:

- the result of cleaning the signal closest to the pumping equipment of the dynamic pressure sensor from the interference background of the pumping equipment; D _i (t) - the measured pressure value with the closest to the pump equipment dynamic pressure sensor; ΔS _{i, i + 1} (t) is the component due to the action of pumping equipment, calculated by the expression

where: S _i (t) is the measured pressure value from the static pressure sensor closest to the pumping equipment; S _{i + 1} (t) is the measured pressure value from the next static pressure sensor. 3. A method for detecting a leak according to claim 1, wherein the direct subtraction of the component due to the action of the pumping equipment is made from the measured values of at least two sensors in one section of the pipeline. 4. A method for detecting a leak according to claim 1, wherein the decision to detect a leak is made when the correlation peak of the leak and the pressure drop that is not typical for the normal operation of the pipeline are recorded together.

Images