JPH022528B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for detecting an accident in a liquid delivery pipeline.

Conventionally, there has been a device of this type as shown in FIG. (Refer to USP3851521) In the same figure,
1 is the liquid feed pipe, 2 is the rupture point of the liquid feed pipe 1, 3 ₁ to 3 ₃
are _a plurality of detectors, for example, first, second, and third three detectors, which are distributed in the axial direction of the liquid feeding tube 1 and detect an abnormal drop in the pressure in the tube;
4 ₃ is a frequency phase shift transmitter that is provided corresponding to each of the detectors 3 ₁ to 3 ₃ and shifts the center frequency of the carrier wave according to the drop in hydraulic pressure; 5, 6, and 7 are first, second, and 3, each of which is composed of measuring devices in the axial direction of the conduit 1, that is, a detector 3 and a frequency phase shift transmitter 4. 8, 9, and 10 are signal transmission paths corresponding to the first, second, and third detection devices 5, 6, and 7, and 11 is each frequency phase shift transmitter 4 ₁
.about.4 ₃ , 12 is a display means, for example a pen recorder, for displaying the output from the receiver 11 as a function of time.

Next, the operation will be explained.

It is a well-known fact as a property of compressible fluids that when a rupture occurs at a certain point 2 in the liquid pipe 1, an instantaneous pressure drop occurs and this propagates in the axial direction at the speed of sound. Now, the flow propagation of the pressure drop that has occurred at point 2 is first transmitted to the second detection device 6, then to the third detection device 6.
detection device 7 and finally reaches the first detection device 5. At this time, each of the detectors 3 ₁ to 3 ₃ detects that the pressure in the liquid pipe 1 has decreased compared to the normal value,
Each of the frequency phase shift transmitters 4 ₁ to 4 ₃ shifts the center frequency of a carrier wave in proportion to the detected falling pressure and sends the carrier wave to the receiver 11 . When the receiver 11 receives this, it sends the output to the pen recorder 12, so the pen recorder 12 outputs the first, second and third three
According to the three detection devices 5, 6, and 7, the moment of pressure drop with the horizontal axis as time is displayed. At this time, the rupture point 2 of the liquid sending pipe 1 is determined from the time difference τ of pressure drop.
The rupture position is determined by d=Dc·τ/2. Here, D is the distance between the second detector 3 ₂ and the third detector 3 ₃ , c is the speed of sound, and d is the distance from the second detector 3 ₂ to point 2.

Since the conventional method for detecting an accident in a liquid transfer pipe is configured as described above, in the case of an accident in which a liquid transfer pipe collapses, a pressure drop occurs on the downstream side and a pressure increase occurs on the upstream side. Only the detectors on the downstream side were activated, which had the disadvantage of erroneously detecting accident conditions. In addition, in the case of a rupture accident of a liquid pipe, the pressure drop propagates both upstream and downstream from the rupture point, but there is a phenomenon in which it is reflected again at the end and propagates back, and these waves converge and create a complex wave. Since pressure fluctuations tend to occur, conventional methods are difficult to deal with and have the disadvantage of being prone to detection errors.

This invention was made in order to eliminate the above-mentioned drawbacks of the conventional system, and since abnormal fluctuations in pressure or flow rate propagate through the liquid at the speed of sound, the pressure or flow rate at both ends of the liquid supply pipe must remain constant. By focusing on a relational expression that holds equally true and using a discriminant derived from this relational expression, it is possible to measure both pressure drop and pressure rise, and detect not only rupture accidents but also crushing accidents, and also detect pressure or flow rate. It is an object of the present invention to provide a method for detecting an accident in a liquid sending pipe that can accurately detect an accident in a liquid sending pipe in response to complex pressure fluctuations that occur due to the reflection of fluctuation waves.

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 is a block diagram of a device used in the method for detecting an accident in a liquid supply pipeline according to the present invention, and the same parts as in FIG.

In the figure, 13 ₁ and 13 ₂ are the first and second pressure gauges 1 installed at both ends of the liquid pipe 1, respectively.
Signals from 30 ₁ and 130 ₂ , 14 ₁ and 14 ₂ are the first and second signals installed at both ends of the liquid pipe 1, respectively.
Signals from flowmeters 140 ₁ , 140 ₂ , 15 ₁ ,
15 ₂ is the first and second transmitting signal.
16 is a data storage device which also serves as a receiver for the above-mentioned signal; 17 is a data processing device; and 18 is a display device for displaying the processed results. This is what we do. In other words, if a rupture occurs in the liquid delivery pipe, the location of the accident point, along with the accident accuracy, can be determined.
Display to user. In the figure, A and B indicate the positions of measurement points, and F indicates the position of the accident point.

Next, the accident detection method of the present invention will be explained.

Generally, a liquid pumped through a pipe is treated as a compressible fluid, so any distance x, any time t
The pressure p(x, t) or flow rate q(x, t) is
As is well known, it is expressed by the following equation.

p(x, t)=F(t+x/c)+f(t-x/c)...(1) q(x, t)=-gD/c{F(t+x/c)-f(t
- x/c)} ...(2) Here, F(t+x/c) is the backward wave component that propagates from downstream to upstream, f(t-x/c) is the forward wave component that propagates from upstream to downstream, D is the cross-sectional area of the liquid flowing through the pipe, g is the gravitational acceleration, and c is the wave propagation speed, which is equal to the speed of sound propagating in the liquid. In the liquid delivery pipe shown in Figure 2, if measurement point A is the origin of distance 0 and the distance between measurement point B is l, then the time τ for the wave to propagate between these two points A and B is τ = l/ It is c.

By substituting the value of x=0 or x=l in equations (1) and (2) and considering τ=l/c, the following equations (3) and (4) can be obtained.

p _A (t-τ)+k・q _A (t-τ)=p _B (t)+k・q _B
(t)……(3) p _A (t)−k・q _A (t)=p _B (t−τ)−k・q _B (t
−τ)……(4) Here, corresponding to the distance 0, add the subscript A to the distance l
It is represented by the subscript B corresponding to . Further, k is a constant and k=c/gD. Equation (3) is the sum of the pressure and flow rate values measured at the left measurement point A in the liquid supply pipe in Figure 2, p _A (t-τ) + k・q _A (t-τ)
However, after time τ, the sum of the pressure and flow values measured at point B on the right side of the liquid sending pipe is p _B (t) + k・q _B (t)
It shows that it is equal to . Similarly, in equation (4), the difference between the pressure and flow rate at point A, p _A (t) - k・q _A (t), is determined by the time τ
This shows that the difference between the pressure and the flow rate previously measured at point B on the right side of the liquid sending pipe is equal to p _B (t-τ) - k·q _B (t-τ).

Now, in the method of the present invention, based on the above equations (3) and (4), a determination is made to determine the state of the liquid supply pipe, that is, whether it is normal or an accident. The present invention provides a method of deriving a formula and using the formula to accurately detect accidents in the pipeline. That is,
From the measurement data collected moment by moment and the data accumulated τ hours ago in the data storage device 16, λ(t)=P _A (t-τ)+k・q _A (t-τ)-p _B (t
)−k・q _B (t)……(5) μ(t)=p _A (t)−k・q _A (t)−p _B (t−τ)+
The following discrimination is performed by calculating the discriminants λ(t) and μ(t) as follows: k·q _B (t−τ) (6).

() max{|λ(t)|, |μ(t)|}≦ε, then normal...(7) () max{|λ(t)|, |μ(t)|}>ε There is an accident, and if λ(t)-μ(t)<δ, then it is a crushing accident...(8) If λ(t)-μ(t)≧δ, it is a bursting accident...(9) ).

Here, ε and δ are constants determined in consideration of the properties of the target pipe and liquid.

If λ(t) or μ(t) in () is smaller than ε, that is, it is zero in equations (5) and (6) above, but a certain tolerance is set in consideration of measurement errors, frictional pressure loss, etc. If it is within the range ε, it is determined to be normal, and otherwise it is determined that there is an accident as described in () above. Next, if it is determined to be an accident, it is classified into both collapse and rupture, with the former being a case in which there is no leakage, and the latter being a case in which there is a leakage.

Figure 3 shows the wave propagation of pressure or flow rate at the time of an accident in the liquid supply pipeline. Assume that there are measurement points at points A and B on a conduit, and an accident occurs at point F, which is a distance d from measurement point A. In addition, the horizontal axis represents distance and the vertical axis represents time. Solid line 19 represents a forward wave that travels in the direction of flow of the liquid, and 20 represents a backward wave that travels in the opposite direction to the flow direction. propagate. Since wave propagation relational expressions (1) and (2) also hold between points AF and between points FB, respectively, the following equations (10) and (11) are obtained. Focusing now on time tn, it takes τ ₂ times to pass through point F and reach point B, while the backward wave passing through point F at time tn reaches point A in τ ₁ time. do.

λ(tn+τ ₂ )=Δh _F (tn)+kΔq _F (tn) ……(10) μ(tn+τ ₁ )=Δh _F (tn)−kΔq _F (tn) ……(11) Here, Δh _F (tn ) is the pressure difference between the point immediately before and after the accident point F, and Δq _F (tn) is the pressure difference between the point immediately before and after the accident point F.
This is the difference in flow rate between the point immediately before and the point immediately after, that is, the leakage flow rate due to rupture. When the accident is a collapse and there is no leakage, Δq _F (tn) is zero. Furthermore, in the case of a bursting accident, the above Δh _F is smaller than that in a crushing accident. In the normal case without any accidents, Δh _F (tn)=
0, Δq _F (tn) = 0, so the above (10) and (11)
Each expression is zero. Now, in the case of a bursting accident, assuming Δh _F (tn) = 0, λ (tn + τ ₂ ) = -μ (tn
+
τ ₁ ), so if we replace it as t=t _o +τ ₁ , we get λ(t+τ ₂ −τ ₁ )=−μ(t). Therefore, it can be seen that λ and −μ have equal values with time advance ▽=τ ₂ −τ ₁ . This time difference ▽=τ ₂ −τ ₁
Therefore, the distance d of the accident point F is d=(l-c・▽)/
It is found by 2. On the other hand, in the case of a crushing accident,
Assuming Δq _F (tn)=0, λ(tn+τ ₂ )=μ(tn+τ ₁ )
Therefore, if we replace it as t=tn+τ ₁ , we get λ
(t+τ ₂ −τ ₁ )=μ(t). Therefore, λ and μ
It can be seen that they have equal values with a time difference ▽=τ ₂ −τ ₁ . Distance d from this time difference ▽ = τ ₂ − τ ₁ to the accident point F
is determined by d=(l-c・▽)/2. In the example, the time difference between λ and μ is calculated based on their absolute value |λ so that it can be applied to both bursting and crushing accidents.
This is done by comparing time series data of | and |μ|.

Summarizing the above description, the operation when implementing the present invention will be explained with reference to FIG. 4.

Pressure p _A (t) and flow rate q _A (t) at measurement point A (Fig. 3)
and the pressures p _B (t), q _B (t) at measurement point B in Figure 2.
is sent to the data storage device 16 at a period of Δτ, and in this data storage device 16, the measurement data time series p _A (t), p _A (t−Δτ), ..., q _A t, q _A (t−
Δτ),..., p _B (t), p _B (t-Δτ),..., q _B (t)
,q _B
(t-Δτ),... are stored sequentially (processing step
401). Next, for each period Δτ, the discriminant function λ(t),
μ(t) and convert this data time series into λ(t),
λ(t-Δτ), λ(t-2Δτ), ..., λ(t-iΔτ
),
..., λ(t-nΔτ) and μ(t), μ(t-Δτ)
,
μ(t-Δτ), μ(t-2Δτ), ..., μ(t-iΔτ
),
..., μ(t-nΔτ) (processing step 402).
That is, at time t, λ and μ have accumulated data for the past nΔτ as a time series, along with the values λ(t) and μ(t) at the current time. Now, the contents of the process are as follows: First, if λ(t) or μ(t) is small (equation (7)), it is determined that the conduit is normal (judgment step 403). Next, if equation (7) is not satisfied, it is an accident, and the type of accident, that is, whether it is a crushing accident or a bursting accident, is determined according to equations (8) and (9) (determination step 404). At this time, from the pattern of time series data of the two discriminant functions λ and μ, the time difference between the two ▽=τ2
−τ1, and the distance d of the accident point F is calculated by {l−c(τ2−
τ1)}/2 (processing steps 405, 406).
In the case of a bursting accident, the leakage flow rate Δq _F is calculated (processing step 407), and in the case of a crushing accident, the pressure drop amount Δh _F is calculated (processing step 408). Figure 5 shows the discriminant, the absolute values of λ and μ, |λ|, |μ|
An example of time series data is shown. Current time t
The determination is made based on data from t−nΔτ hours before. In the early stages, the values of |λ| and |μ| are small;
The pipeline is determined to be normal, but if an accident occurs, |
Since both λ| and |μ| change, it is determined that there is an accident. Whether the accident is a rupture or a crushing is determined by λ
It is determined by (t)-μ(t)=δ. next,
By comparing the time series patterns of |λ| and |μ|, the time shift ▽ between the two is determined. In the example shown in Figure 5, ▽
=3Δτ, and the distance to the accident point is d=(l-3・
It is obtained as Δτ・c)/2. When a crushing accident is determined, the pressure drop is calculated as Δh _F = 1/2 {λ(t) + μ(t)}, and when a bursting accident is determined, Δq _F = 1/2k {λ(t) −μ(t)}, the leakage flow rate is determined.

As described above, according to the present invention, since the time series data of two discriminants are used, it is possible to distinguish between a crushing accident and a bursting accident, and the wave motion caused by the accident is reflected and the measured pressure and flow rate are Even if there are fluctuations, it is possible to detect accidents in the liquid supply pipeline with high accuracy.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a device used in a conventional method for detecting an accident in a liquid supply pipeline, FIG. 2 is a block diagram of a device used in a method for detecting an accident in a liquid supply pipeline according to an embodiment of the present invention, and FIG. 4 is an explanatory diagram of wave propagation for determining the discriminant used in the method for detecting accidents in liquid pipe lines according to an embodiment of the present invention. FIG. FIG. 5 is an explanatory diagram of time-series data of the values of two discriminants used in the method of detecting an accident in a liquid supply pipeline according to an embodiment of the present invention. 1...Liquid feeding pipe line, 2...Rupture point, 13 ₁ , 13 ₂
... Signal from the pressure gauge, 14 ₁ , 14 ₂ ... Signal from the flow meter, 15 ₁ , 15 ₂ ... Signal transmitting device,
16... Data storage device, 17... Signal reception and data processing device, 18... Data processing result display device, 19... Locus of forward wave component propagating in liquid, 20... Propagating in liquid Trajectory of backward waves, 130 ₁ , 130 ₂ ... Pressure gauge, 140 ₁ , 1
40 ₂ ...Flowmeter. Note that the same reference numerals in the figures indicate the same or corresponding parts.

Claims (1)

[Claims] 1 Install a pressure gauge and a flow meter at both ends of the liquid supply pipe,
The pressure value and flow rate value measured in a certain period are stored in a data storage device, and among the stored measurement data, the measured value at the current time and the liquid movement between the measurement points at two positions in the liquid transfer pipe are calculated. The values of two discriminants are calculated from the measured values at the past time, which is the time of propagation, and the values of the discriminants are used to determine whether or not the liquid delivery pipeline is normal. Liquid delivery characterized by locating the location where the accident occurred based on the time lag of time series data of two discriminant formulas and determining whether it is a pipeline crushing accident or a pipe tearing accident. Accident detection method for pipelines.