JPH07140034A - Leakage detection method for pipe line - Google Patents

Abstract

PURPOSE:To decide the occurrence of leakage by determining the 4 difference between an actually measured value and a calculated value of flow state amount obtained by flow simulation thereby determining the variation rate of line pack volume, multiplying the difference by a scaling factor to obtain a corrected difference, and then comparing the corrected difference with a threshold value or employing a statistic method. CONSTITUTION:A measurement input function 1 fetches output values from a manometer, flowmeter, etc., and delivers them to a fluid simulation function 2 which effects unsteady flow simulation based on these output values. A difference calculating function 3 fetches pressure and flow rate from the function 1 and determines the difference from corresponding values calculated by the function 2. A line pack volume variation rate calculating function 4 calculates a variation rate from the calculation results of the function 2. A difference scaling function 5 multiplies the difference at the function 3 by a scaling factor for decreasing the large absolute value whereas increasing the small absolute value of the variation rate. The corrected difference is fed to a leakage decision function 6 where it is compared with a threshold value or the dispersion of fluctuation is subjected to 3SIGMA management method thus deciding abnormality.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a pipeline for transporting fluids such as oil and natural gas, and more particularly to a method for detecting leakage of fluid from the pipeline during transportation.

[0002]

2. Description of the Related Art Generally, means for detecting pipeline leakage can be classified into a direct method using a gas detector or the like and an indirect method using a flow simulation calculation value or the like. For the latter method relating to, for example, "PIPELINE LEAK D
ETION 1988, ASME PD-Vo
l. 19, pp. 55-60 ", in which a flow simulation is performed under the boundary condition of inlet pressure-outlet flow rate in a pipeline in which a pressure gauge and a flow meter are installed at both ends and a plurality of pressure gauges are installed in the middle. A method of detecting leakage by comparing the calculated value with the measured value for the inlet flow rate and the pressure other than the inlet is adopted.
In Japanese Patent Publication No. 4-51719, in a liquid pipeline, a flow simulation in the case where there is no leakage is performed by a characteristic curve method, and the calculated pressure and the measured value are compared, and the difference is a certain value or more. There is disclosed a method of determining that a leakage has occurred when it becomes.

In the leakage detection method of the simulation method, generally, the flow simulation is executed with the measured value of the pressure or the flow rate (which may be the flow velocity) at the boundary end of the pipeline as the boundary condition, and other measured values not used as the boundary condition. And the corresponding calculated value. When a leak occurs in the pipeline, the flow rate between the inlet and the leak point becomes larger than the flow rate between the leak point and the exit, so the pressure distribution along the pipeline bends at the leak point. It becomes the shape. Since the occurrence of leakage is not considered in the flow simulation, there is a contradiction between the calculation result and the measured value (actual phenomenon) regarding the pressure and flow rate that represent the flow state. The magnitude and distribution shape of the deviation of this measured value differ depending on how the boundary conditions of the flow simulation are adopted. For example, even in a simple pipeline having one inflow end and one outflow end, four boundary conditions for calculation can be considered: pressure-pressure, pressure-flow rate, flow rate-pressure, flow rate-flow rate. However, in any case, when a phenomenon such as leakage that is not included in the calculation model occurs, the deviation amount is increased.

[0004]

In the pipeline flow simulation, it is not possible to obtain a calculation result that exactly matches the measured value (actual phenomenon) even if any calculation model and solution method of the state of the art are used. In particular, during unsteady operation in which the flow rate changes with time, a non-negligible error occurs between the calculated value and the measured value. In the method of temporally monitoring the deviation between the two and determining that leakage has occurred when the value becomes abnormally large using, for example, the 3σ management method, the deviation due to the start of unsteady flow fluctuation The increase in volume is often confused with that when a leak occurs, and a false alarm is often output. In particular, if the leak detection accuracy is improved and even minute leaks are to be detected, it is necessary to reduce the leak determination threshold value of the deviation amount to a smaller value. However, there was a problem that it inevitably grew.

The problem of such false alarms is greater than the case where the fluid inside the pipeline is a liquid that can be considered incompressible.
In the case of highly compressible gas, the influence is larger. This is because, in a compressible fluid, when the flow rate at one boundary end of the line changes, there is a time delay before the other boundary ends are affected by the change in flow rate. Is very complicated.

The present invention has been made in order to solve the above problems, and does not give an erroneous warning of leakage due to an unsteady fluctuation of the flow even if the flow is unsteady.
An object of the present invention is to provide a pipeline leak detection method capable of reliably detecting leak occurrence.

[0007]

A pipeline leak detection method according to the present invention executes a flow simulation of a pipeline for transporting a fluid in real time, and obtains a flow state quantity (pressure, flow rate, etc.) of the fluid obtained by the simulation. ) And the measured values by the sensors installed in the pipeline are compared to obtain the deviation amount between the two, and the difference between the sum of the inflow amount into the pipeline and the sum of the outflow amount from the pipeline. The step of obtaining the absolute value of the line pack amount change speed is defined as follows, and it is small when the absolute value of the line pack amount change speed is large with respect to the deviation amount, and becomes large when the absolute value of the line pack amount change speed is small. A step of obtaining a modified deviation amount by multiplying a scaling coefficient having a different function form, and if the modified deviation amount exceeds a preset threshold value, or Statistical probability of occurrence of a positive deviation amount and a step of determining a leakage occurs when it becomes less than a predetermined significance level.

[0008]

In the present invention, it is determined that the greater the rate of change in the amount of line packs, the greater the unsteadiness of the flow. Therefore, the deviation amount is corrected in the direction of decreasing. On the contrary, since it is determined that the unsteadiness of the flow is smaller as the change rate of the line pack amount is smaller, the deviation amount is corrected to be increased.
In this way, the pipeline leak is judged based on the corrected deviation amount, that is, the corrected deviation amount. On the other hand, it is possible to reliably prevent the occurrence of the leakage without overlooking the increase in the deviation amount when the leakage actually occurs.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a functional block diagram showing a processing function of, for example, a computer for implementing a pipeline leak detection method according to an embodiment of the present invention. In the figure, a measurement value acquisition function 1 is a function for fetching the output values of sensors such as pressure gauges and flow meters provided in various places of the pipeline in real time from a telemetering device at regular intervals. Of these, some predetermined measurement values and their acquisition times are passed to the flow simulation function 2. In the flow simulation function 2, the unsteady flow simulation of the entire pipeline is performed with these as boundary conditions for the time increment from the end of the previous calculation to the measurement time of this time. After the simulation calculation is completed, the deviation calculation function 3 measures values (pressure, flow rate) other than those used as boundary conditions in the flow simulation function 2.
Is received from the measured value acquisition function 1, and the deviation from the calculated value of the corresponding flow simulation function 2 is calculated.

On the other hand, the line pack amount change speed calculation function 4
At, the calculated line pack amount change rate ΔLPcalc is calculated from the calculation result of the flow simulation function 2. As shown in FIG. 2, when the calculation range of the flow simulation has N boundary ends, this value is represented by the calculated flow rate at the i-th boundary end as Qi, calc,

[0011]

[Equation 1]

[0012] At this time, the sign of the flow rate is positive in the direction of flow into the pipeline. Further, this value may not be an instantaneous value at a certain time point but may be an integrated value within a certain time period before the current time point.

Next, in the deviation scaling function 5,
The deviation between the measured value and the calculated value obtained by the deviation calculation function 3 is multiplied by an appropriate scaling coefficient. here,
The calculation formula of the scaling coefficient is given in advance such that the larger the absolute value of ΔLPcalc, the smaller it becomes, and the smaller the absolute value of ΔLPcalc, the larger the functional form. For example, the following formula can be considered as a simple one. Cs = MIN (1, | ΔLPmeas−ΔLPcalc | / | ΔLPcalc |) (2) Cs = | ΔLPmeas−ΔLPcalc | / (| ΔLPmeas−ΔLPcalc | + | ΔLPcalc |) (3) where ΔLPmeas is the expression (1). It is the line pack amount change speed in the measurement obtained in the same manner as. The scaling coefficient Cs thus obtained is multiplied by the deviation amount obtained by the deviation calculation function 3, and the scaled deviation amount (correction deviation amount) is sent to the leakage determination function 6. The leakage determination function 6 observes the magnitude of the scaled deviation amount (correction deviation amount) over time, and when the value becomes abnormally large, determines that leakage has occurred and outputs a leakage alarm. At this time, a fixed value may be given in advance as the leakage determination threshold value, or the variance is obtained from the variation state of the corrected deviation amount before the current time, and the statistical method such as the 3σ management method is used. The presence or absence of abnormality may be tested.

FIG. 3 shows a situation in which air is flowed through a 4 in.times.1400 m experimental pipe to cause a leak in the center of the pipe while the flow rate at the outlet is being rapidly reduced. A pressure gauge and a flowmeter were installed at the inlet and outlet of the pipe, respectively, and the outputs were taken into the computer at regular intervals. The computer has all the functions of the above-mentioned leak detection method programmed therein. The flow simulation was performed at the inlet pressure-outlet pressure boundary. The state quantity used for leakage determination was the difference between the inlet flow rate and the outlet flow rate, and the deviation amount between the measured value and the calculated value was calculated. The curve 9 in the figure represents the change with time of this deviation amount,
It increases immediately after the start of the flow rate fluctuation, and if this is used as it is for leak determination, a false alarm will be issued. On the other hand, the calculated line pack amount change speed (curve 7), the measured line pack amount change speed (curve 8), and the deviation amount (correction deviation amount) scaled by the equation (3) are the curve 10
If a leak is determined using this value, there will be no false alarm due to flow fluctuations, and 20
It could be detected after a second. At this time, the leakage determination criterion is that the significance level is 0.5% by t-test for the variance of the scaled deviation amount (correction deviation amount: curve 10).
And

When a leak occurs, the deviation between the measured value and the calculated value increases not only in the flow rate but also in the pressure. Therefore, it is not necessary to limit the flow rate shown here to the state used for the leak determination. You may pay attention. As a solution method of the flow rate simulation, there are many known explicit solution systems and implicit solution systems, but the present invention does not limit the applied flow simulation solution, and any solution method can be used. Good. Further, any of the above-mentioned boundary conditions may be used for the flow simulation, but it is preferable to use the pressure boundary as much as possible. This is because, as the number of flow rate boundaries among all pipeline boundaries increases, the calculated line pack rate of change apparently changes the line pack rate (the difference between the total inflow rate measurement value and the total outflow rate measurement value). ), And when leakage occurs, the amount of deviation is scaled in the direction of being reduced.

[0016]

As described above, according to the present invention, the flow simulation is executed, and the deviation amount between the calculated value and the measured value of the flow state amount of the liquid is reduced as the line pack amount change speed increases. Direction, and conversely, the smaller the line pack amount change speed, the larger the deviation amount is corrected, and the pipeline leakage is determined based on the corrected deviation amount thus corrected. While it is possible to prevent a false alarm of leakage by compensating for the decrease in the accuracy of flow simulation that occurs when there is a large amount of steady fluctuation, it is possible to make sure that leakage does not occur without overlooking the increase in deviation when an actual leakage occurs. Occurrence can be detected. Therefore, as compared with the conventional method of the same type, it is possible to reduce false alarms due to variable operation of the pipeline and to perform highly reliable leak detection without adding a new device.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing processing functions of a computer for implementing a pipeline leak detection method according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of the shape of a pipeline for performing a flow simulation.

FIG. 3 is a diagram showing a situation when a leak occurs in an experimental pipe.

[Explanation of symbols]

1 Measurement data acquisition function 2 Flow simulation function 3 Deviation calculation function 4 Line pack amount change speed calculation function 5 Deviation scaling function 6 Leakage judgment function 7 Calculation line pack amount change speed change curve 8 Line pack amount change in measurement Speed change curve 9 Deviation amount change curve 10 Scaled deviation amount (corrected deviation amount) change curve

Front Page Continuation (72) Inventor Yoshiaki Asano 1-2-1, Marunouchi, Chiyoda-ku, Tokyo Nihon Steel Pipe Co., Ltd.

Claims (1)

[Claims] 1. A flow simulation of a pipeline for transporting a fluid is executed in real time, and the calculated flow state quantity of the fluid obtained thereby is compared with those measured by a sensor installed in the pipeline. Obtaining the deviation amount of both, and obtaining the absolute value of the line pack amount change speed defined as the difference between the sum of the inflow flow rate to the pipeline and the sum of the outflow flow rate from the pipeline; And a step of determining a correction deviation amount by multiplying by a scaling coefficient having a functional form that is small when the absolute value of the line pack amount change speed is large and is large when the absolute value of the line pack amount change speed is small, When the correction deviation amount exceeds the preset threshold value,
Or a leak detection method for a pipeline, which comprises a step of determining that a leak has occurred when the statistical appearance probability of the correction deviation amount is equal to or lower than a predetermined significance level.