JPS6319806B2 - - Google Patents

Description

[Detailed description of the invention]

The purpose of this invention is to detect sudden large-scale leaks at an early stage and to minimize the amount of leakage outside the pipe by measuring sudden abnormal drops in pressure during operation of liquid transportation pipelines. This invention relates to an improved method for detecting leaks in pipelines for transporting liquids. Conventionally, methods for detecting leaks in liquid transport pipelines using abnormal pressure drops are shown in Figure 1 (see Japanese Patent Application Laid-Open No. 138785/1983) and Figure 2 (Japanese Patent Laid-Open No. 138785).
55-17737)), methods using an apparatus shown in a schematic diagram are known. In Figure 1, 1 is the pipeline, 2 is the departure base of pipeline 1, and 3 is the same terminal base, both bases 2,
3 has a tank. 4 is pipeline 1
These pressure transmitters 4 are installed at approximately regular intervals along the way, and these pressure transmitters 4 measure the liquid pressure (operating pressure) in the pipeline 1 at the installation point and convert it into an electrical signal, etc. Convert. 5 is a telemeter slave station for transmitting the liquid pressure measurement signal from the pressure transmitter 4 to the central control room 6. 7
is a line cable connecting the telemeter slave station 5 and the central control room 6. This line cable 7 is unnecessary when transmitting the signal from the telemeter slave station 5 to the central control room 6 by radio. Reference numeral 8 denotes a telemeter master station provided in the central control room 6, and this telemeter master station 8 receives signals from the telemeter slave stations 5. Reference numeral 9 denotes a computer as a calculation device for processing the liquid pressure measurement signal received by the telemeter master station 8. In addition, as an example of the pressure transmitter 4, for example, it is equipped with a predetermined withstand pressure,
Moreover, the measurement range is limited to the normal pressure fluctuation range at the installation point (for example, withstand pressure 50 kg/
^cm2 , measurement range 10-20Kg/ ^cm2 ). At each pressure measurement point, liquid pressure is constantly measured at predetermined sampling time intervals, while
In the central control room 6, the computer 9 calculates the difference between the measured values of temporally adjacent liquid pressure measurement signals, and the calculated difference value (average slope rate) thus obtained is set as a predetermined set value. Leakage is detected if it exceeds the limit. When there is no leakage, the calculated value is almost zero (the liquid pressure is almost constant at each measurement point, indicating that there is no leakage). When a large amount of leakage suddenly occurs, a negative pressure wave is generated along with the leakage, and when the leading edge of this negative pressure wave passes through the pressure measurement point, the pressure transmitter 4 at that pressure measurement point The measured pressure (value) changes significantly. Therefore, the arrival of this negative pressure wave causes a large change in the liquid pressure measurement signal, which is detected in the central control room 6 as a leak. Further, the leak position is detected by calculating, by the computer 9, the difference in arrival time of the negative pressure waves to the pressure measurement points sandwiching the leak point. In FIG. 2, reference numeral 10 calculates the difference between the measurement values of the measurement signals sampled at predetermined intervals from the pressure transmitter 4 and which are adjacent to each other in time, and the calculated difference value obtained in this way corresponds to a predetermined set value. When you cross it,
The difference calculation result (signal) is sent to the telemeter slave station 5.
It is a microprocessor that outputs to and,
The telemeter slave station 5 transmits the difference calculation signal from the microprocessor 10 and the time at which this difference calculation signal is generated to the central control room 6. The central control room 6 issues a command to transmit liquid pressure measurement signals at other measurement points at the same time as the difference calculation signal to the telemeter master station 8 via the microprocessor 10 and the telemeter slave station 5. issue. Therefore, in the central control room 6, in the same manner as explained in the apparatus shown in FIG.
Based on the difference calculation signal and the liquid pressure measurement signal from another pressure transmitter 4 at the same time, calculations for confirming leakage in the pipeline 4 and calculations for detecting the leakage position are performed. In the conventional detection method described above, the data processing such as the sampling interval of pressure at each measurement point along the pipeline and the calculation for calculating the difference value are as follows. △Sampling interval for pressure measurement
When T is the measured value of the nth pressure and P(n) is the measured value of the nth pressure, the difference value DP(n) is calculated by DP(n) = P(n) - P(n-1) ......(1) There is. When this DP(n) exceeds the alarm value, it is determined that a leak has occurred. As described above, in the conventional detection method, the difference calculation is performed at the same period as the sampling interval for pressure measurement. Note that the amount of pressure change and the amount of leakage at the time of sudden leakage are in a proportional relationship, and the alarm value is set in accordance with the amount of leakage that is the detection target. However, in the conventional detection method described above,
There are problems as shown below. The measured values of pressure changes caused by sudden leakage are shown schematically in the upper part of Figure 3 (in the figure, △P
is the amount of pressure change, and Δt is the pressure change time).
When sampling such a step-like pressure change and calculating the difference, even if the calculation interval is the same, the relative sampling timing with respect to the pressure change is as shown in Figure 3 (in the figure, △
T is the sampling interval, and the triangle mark indicates the sampling time point). In this way, the pressure difference calculation value differs due to the difference in sampling timing. That is, the difference calculation values according to the above formula (1) are shown in Figure 4 (corresponding to timing 1), Figure 5 (corresponding to timing 2), and Figure 6 (corresponding to timing 3).
The results are different as shown in the figure. As shown in the figure, in the worst case, only 1/2 of the actual pressure change can be obtained (DP ₁ (2) in Figure 4 corresponds to △P, and in Figure 5, DP ₂ (2) ), DP ₂ (3) corresponds to 1/2△P). In this way, in the conventional method, even if the pressure change is the same, the difference calculation value varies between a value equal to the amount of pressure change and a value equal to 1/2 of it depending on the sampling timing, and the difference calculation value is always the same. You can't get the value. In an actual pipeline, it is impossible to predict when a sudden leak will occur, and the timing of sampling the measured value of the accompanying pressure change may be in any of the cases shown in FIG. Therefore, even if the actual pressure change amount becomes larger than the set alarm value, the calculated difference value will be smaller than the alarm value, and there is a possibility that leakage cannot be detected. When we examine how much pressure change ΔP between time Δt can be detected by the pressure difference calculation value using equation (1) when the sampling interval is ΔT, we find the following for each range of time Δt. become that way. (i) When 0<Δt<ΔT Difference calculation value DP(n)=P(n)−P using equation (1)
The best way to detect the pressure change ΔP using (n-1) is when the sampling timing is
This is a case where the timing has sampling points before and after the pressure change, such as timing 1 in FIG. 3. In this case, the difference calculation value at the time after calculating the difference between the pressures before and after the pressure change becomes the maximum, and the value is ΔP. When the sampling timing is such that the sampling time point is in the center of the pressure change, as in the timing shown in FIG. 3, the detection of the pressure change amount ΔP using the difference calculation value is the worst. In this case, a difference calculation is performed at a central point in the pressure change or at a point after the pressure change, which is the difference between the pressure at the center point in the pressure change and the pressure at a point before or after the pressure change. The maximum value is ΔP/2. If the sampling timing is such that the sampling point within the pressure change is shifted from the center of the pressure change, such as timing 3 in Fig. 3, the pressure change amount ΔP by the difference calculation value is The degree of detection is intermediate between the above two cases. In this case, the difference calculation value at that point in the pressure change, which is the difference between the pressure at that point in the pressure change and the pressure at the point before the pressure change, is the maximum, and the value is ΔP / 2 greater than ΔP. Therefore, when 0<Δt<ΔT, the maximum value of the pressure change amount that can be detected by the difference calculation value according to equation (1) with respect to the pressure change amount ΔP is ΔP,
The minimum value is ΔP/2. (ii) When ΔT<Δt<2ΔT Difference calculation value DP(n)=P(n)−P using equation (1)
The best way to detect the pressure change ΔP using (n-1) is when the sampling timing is
This is a case where there are two sampling points within the pressure change, such as timing 1 in FIG. 9. In this case, the difference calculation value at the later of the two points in the pressure change, which is the difference between the pressures at two points in the pressure change, is the maximum, and that value is the rate of change in pressure. It is ΔT×ΔP/Δt, which is ΔP/Δt multiplied by the sampling interval ΔT. When the sampling timing is such that the sampling time point is in the center of the pressure change, such as timing 2 in FIG. 9, the detection of the pressure change amount ΔP using the difference calculation value is the worst. In this case, the difference calculation value at the central point within the pressure change or at the point after the pressure change is maximum, and the value is ΔP/2. If the sampling timing is such that there is only one sampling point within the pressure change and it is off the center, the degree of detection of the pressure change amount ΔP by the difference calculation value will be between the above two cases. Become. Therefore, when ΔT<Δt<2Δt, the maximum value of the pressure change amount that can be detected by the difference calculation value according to equation (1) with respect to the pressure change amount ΔP is ΔT
×ΔP/Δt, the minimum value is ΔP/2. (iii) When 2ΔT<Δt In this case, as shown in Figure 10, there are at least two sampling points within the pressure change regardless of the sampling timing, so (1 ) There is no distinction between a maximum value and a minimum value for the amount of pressure change that can be detected by the difference calculation value using the formula. Δt. Table 1 shows the maximum and minimum values of the amount of pressure change that can be detected by the difference calculation value according to equation (1) for the above pressure change rate ΔP.

[Table] Figure 7 is a diagram of the detection ability of the difference calculation value using equation (1), calculated based on Table 1, when the alarm value is P _A for a pressure change of pressure change amount ΔP and change time Δt. It is. In order for pressure changes to be detected,
The maximum value of the amount of pressure change that can be detected by the difference calculation value shown in Table 1 must be greater than or equal to the alarm value P _A. That is, expressed as the following equation. In the case of (i) ΔP≧P _A ………(2) In the case of (ii) ΔT×ΔP／Δt≧P _A ∴ΔP≧P _A ×Δt／ΔT ………(3) In the case of (iii) ΔT× ΔP/Δt≧P _A ∴ΔP≧P _A ×Δt/ΔT ………(4) If the range of equations (2) to (4) above is illustrated, the straight line g (ΔP=P _A ) and Straight line h (ΔP=P _A ×Δt/
ΔT). In order to reliably detect a pressure change, the minimum value of the amount of pressure change that can be detected by the difference calculation values shown in Table 1 must be greater than or equal to the alarm value P _A. That is, expressed as the following equation. In the case of (i) ΔP/2≧P _A ∴ΔP≧2P _A ………(5) In the case of (ii) ΔP/2≧P _A ∴ΔP≧2P _A ………(6) In the case of (iii) ΔT ×ΔP/Δt≧P _A ∴ΔP≧P _A ×Δt/ΔT ………(7) If the range of equations (5) to (7) above is illustrated, the above value line g and straight line h in Figure 7 This is the region A above the straight line l (ΔP=2P _A ) among the regions above.
Therefore, the area where a pressure change may be detected but is not certain is the straight line g, h, which excludes area A from the area where a pressure change may be detected.
and region B surrounded by l. In this way, conventional detection methods are extremely unstable, and no matter how you set the sampling interval, for example, even if you set an alarm value for the purpose of detecting a leak of 10% or more of the oil flow, it cannot be reliably detected. It can only be detected when the amount of oil supplied is 20% or more. Therefore, this invention was made to solve the above-mentioned problems, and it measures liquid pressure at predetermined sampling time intervals at measurement points provided at predetermined intervals along a liquid transportation pipeline, and thus measures the liquid pressure at predetermined sampling time intervals. A pipeline for liquid transportation, in which a difference is calculated between the measured values of the obtained measurement signals that are temporally adjacent to each other, and leakage detection, or leakage detection and leakage position detection of the pipeline is performed based on the difference calculation value. The leakage detection method for a liquid transport pipeline is characterized in that the difference calculation is performed on measured values of the measurement signals at a time interval that is twice or more the sampling time interval. . Next, the present invention will be explained using the apparatus shown in FIG. In the present invention, the operation of the apparatus shown in FIG. 1 differs only in difference calculation processing. That is,
If the sampling interval is △T and the measured value of the nth pressure is P(n), then in the present invention the difference value is DQ(n) = P(n) - P(n-m)
......(8) (m>1, integer constant) Calculate based on. When this DP(n) exceeds the alarm value, it is determined that a leak has occurred. The meaning of the calculation according to equation (8) is as follows. Looking at Figures 5 and 6, the individual difference values DP ₂
(2), DP ₂ (3), DP ₃ (2), DP ₃ (3) are pressure changes △P
This is a small value compared to , but this is because the pressure change is divided into two difference values. Therefore, instead of the difference value according to equation (1), (1)
Using m consecutive difference values of the formula (m>1 integer constant), ADP (n) = DP (n) + DP (n-1) +... + DP (n-m + 1) ...... (9) ADP (n), this ADP(n) is expected to be approximately the same value for the same pressure change regardless of the timing of sampling, and leakage detection can be performed by comparing this value with the alarm value. It becomes reliable and stable. Using the relationship of equation (1) to equation (9), ADP(n)=DP(n)+DP(n-1)+…+DP(n
-m+1)=(P(n)-P(n-1))+(P(n-
1)
-P(n-2))+...+(P(n-m+1)-P(n-
m))=P(n)-P(n-m)≡DQ(n), which is the same relationship as shown in equation (8).
In other words, the difference value based on equation (8) is the sum of m consecutive difference values based on equation (1), and as mentioned above, this is almost the same value for the same pressure change regardless of the sampling timing. becomes. When we examine how much pressure change ΔP between time Δt can be detected by the pressure difference calculation value using equation (8) when the sampling interval is ΔT, we find the following for each range of time Δt. become that way. (i) When Δt<(m-1)ΔT In this case, as shown in Fig. 11, there are only m-1 sampling points at most within the pressure change, regardless of the sampling timing. Difference calculation DQ(n)= by formula (8)
There is no distinction between a maximum value and a minimum value in the amount of pressure change that can be detected by P(n)-P(n-m), and the value is ΔP. (ii) (m-1) When ΔT<Δt<mΔT In this case, depending on the sampling timing, there are a maximum of m timing points and a minimum of m-1 timing points within the pressure change. Difference calculation value DQ (n) = P (n) - P according to formula (8)
The best way to detect the pressure change ΔP using (n-m) is when the sampling timing is
This is a case where there are m-1 sampling points within the pressure change, such as timing 1 in FIG. 12. In this case,
The difference calculation value obtained by calculating the difference between the pressures before and after the pressure change becomes the maximum, and the value is ΔP. In the case of timing such as timing 2 in Fig. 12, where there are m sampling points within the pressure change, and the midpoint position of the m points is located at the center of the pressure change. The detection of the pressure change amount ΔP using the difference calculation value is the worst.
In this case, the difference calculation value obtained by calculating the difference between the pressure at the time before (or after) the pressure change and the pressure at the last (or first) time within the pressure change becomes the maximum, and that value is [ Δt−{Δt−(m−1)
ΔT}/2]×ΔP/Δt={(m-1)ΔT×ΔP/
Δt+ΔP}/2. (iii) mΔT<Δt<(m+1)ΔT In this case, depending on the timing of sampling, the number of sampling points within the pressure change is m+1 at maximum and m at minimum. Difference calculation value DQ (n) = P (n) - P according to formula (8)
The best way to detect the pressure change ΔP using (n-m) is when the sampling timing is
This is a case where there are m+1 sampling points within the pressure change, such as timing 1 in FIG. 13. In this case,
The difference calculation value obtained by calculating the difference between the pressure at the first point in the pressure change and the pressure at the last point in time is the maximum, and the value is mΔT×ΔP/Δt. In the case of timing such as timing 2 in Fig. 13, where there are m sampling points within the pressure change, and the midpoint position of the m points is located at the center of the pressure change. The detection of the pressure change amount ΔP using the difference calculation value is the worst.
In this case, the difference calculation value obtained by calculating the difference between the pressure at the time before (or after) the pressure change and the pressure at the last (or first) time within the pressure change becomes the maximum, and the value is { (m-1)ΔT×ΔP/Δt+
ΔP}/2. (iii) When (m+1)ΔT<Δt In this case, as shown in Fig. 14, there are at least m+1 sampling points within the pressure change regardless of the sampling timing, so the difference according to equation (8) Calculated value DQ(n)=P
There is no distinction between a maximum value and a minimum value in the amount of pressure change that can be detected by (n)-P(n-m), and the value is determined by the pressure change rate ΔP/Δt multiplied by m times the sampling interval ΔT. mΔT×ΔP/Δt multiplied by Table 2 shows the maximum and minimum values of the pressure change amount that can be detected by the difference calculation value using equation (8) for the above pressure change amount ΔP. As shown in Table 2, the difference value shows variation only when the pressure change time △t is (m-1)△T<△t<(m+1)△T, but in other cases, the difference value is Always constant.

[Table] Based on Table 2, the detection ability of the difference calculation value using equation (8) in the present invention is calculated based on Table 2, when the alarm value is P _A for a pressure change of pressure change amount ΔP and change time Δt. If we calculate it in the same way as in the formula, we get the following. In order for pressure changes to be detected,
The maximum value of the amount of pressure change that can be detected by the difference calculation value shown in Table 2 must be greater than or equal to the alarm value P _A. That is, expressed as the following equation. In the case of (i) ΔP≧P _A ………(10) In the case of (ii) ΔP≧P _A ………(11) In the case of (iii) mΔT×ΔP／Δt≧P _A ∴P≧P _A /m ×Δt／ΔT……(12) In the case of (iv) mΔT×ΔP／Δt≧P _A ∴ΔP≧P _A /m×Δt／ΔT……(13) The above equations (10) to (13) If the range is shown as m=6, the straight line G (ΔP=P _A ) and the straight line H in FIG.
The area is above (ΔP= _PA /m×Δt/ΔT, where m=6). In order to reliably detect a pressure change, the maximum value of the amount of pressure change shown in Table 2 must be above the alarm value P _A. That is, expressed as the following equation. In case of (i), ΔP≧P _A ……(14) In case of (ii), {(m−1)ΔT×ΔP/Δt+ΔP}/2
≧P _A ∴ΔP≧2P _A /{(m-1)ΔT/Δt+1)}
......(15) In case of (iii) {(m-1)ΔT×ΔP/Δt+ΔP}/2
≧P _A ∴ΔP _A ≧2P _A / {(m-1)ΔT/Δt+1}
………(16) In case of (iv), mΔT×ΔP/Δt≧P _A ∴ΔP≧P _A /m×Δt/ΔT ………(17) The range of formulas (14) to (17) above is m= 6, the straight line L (ΔP=2P _A /{(m−
1) ΔT/Δt+1}, where m=6). Therefore, the area where pressure changes may be detected but is not certain is surrounded by straight lines G, H, and L, excluding area C from the area where pressure changes may be detected. area D. Note that when m=6, the time difference (time interval) T of the difference calculation becomes 6ΔT. Comparing FIG. 7 and FIG. 8, which show the detection performance of the prior art, it can be seen that the present invention can significantly narrow the range of pressure changes that may not be detected, which is indicated by diagonal lines. It is clear that it will be reliable and stable. The pressure change due to actual leakage is the change time △t
However, in the present invention, by increasing the time difference T of the difference calculation, it becomes possible to detect gradual pressure changes accordingly. For example, if the alarm value is P _A and m in equation (8) is 12, that is, the time difference of the difference calculation is 2T (=12△
T), it is possible to reliably detect the pressure change shown in the range of region C and region D in FIG. The data processing method described above can also be applied to the data processing performed by the microprocessor 10 shown in FIG. 2, and the above-described effects can also be obtained in this case. Furthermore, the leak detection according to the present invention is stable without detection errors, and the leak detection and position detection effects can be reliably achieved. As explained above, in the present invention, leak detection and leak position detection of a liquid transport pipeline can be performed extremely reliably.

[Brief explanation of the drawing]

Figures 1 and 2 are schematic explanatory diagrams of a leak detection device for liquid transportation pipelines, and Figure 3 is 0<Δt.
A diagram schematically showing the relationship between the pressure change due to liquid leakage and its measurement timing when <ΔT,
Figures 4, 5, and 6 are diagrams showing the relationship between the sampling timing for pressure measurement and the pressure difference value, Figure 7 is a diagram showing leak detection ability by the conventional leak detection method, and Figure 8 9 is a diagram showing the leakage detection ability according to the present invention, FIG. 9 is a diagram schematically showing the relationship between the pressure change due to liquid leakage and the timing of its measurement when ΔT<Δt<2ΔT, and FIG.
Figure 11 is a diagram similar to Figure 9 when 2ΔT/Δt, Figure 11 is a diagram similar to Figure 9 when Δt<(m-1)ΔT,
Figure 12 is a diagram similar to Figure 9 when (m-1)ΔT<Δt<mΔT, and Figure 13 is a diagram where mΔT<Δt<(m
FIG. 14 is a diagram similar to FIG. 9 when +1)ΔT, and FIG. 14 is a diagram similar to FIG. 9 when (m+1)ΔT<Δt. 1... Pipeline, 2... Departing base, 3... Destination base, 4... Pressure transmitter, 5... Telemeter slave station, 6... Central control room, 7... Line cable,
8...Telemeter master station, 9...Computer, 1
0...Microprocessor.

Claims (1)

[Claims] 1. Liquid pressure is measured at predetermined sampling time intervals at measurement points provided at predetermined intervals along a liquid transport pipeline, and the measurements thus obtained are temporally adjacent to each other. In a method for detecting a leak in a pipeline for liquid transportation, the method comprises calculating a difference between measured values of signals, and detecting a leak, or detecting a leak and detecting a leak position in the pipeline based on the calculated difference value, wherein the difference calculation is performed by: A method for detecting leakage in a pipeline for liquid transportation, characterized in that the method is performed on measured values of the measurement signals at a time interval that is twice or more the sampling time interval.