JPH0464018B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a water leakage detection device that detects water leakage from a water pipe based on changes in sound pressure and vibration.

The amount of water leaking from water treatment plants or distribution pipes and water supply pipes on the way from water distribution reservoirs to customers currently accounts for over 10% of the total amount of water distributed, and the development of new water sources requires a huge amount of funding. shall be. For this reason, it is extremely urgent to control the amount of water leakage, and it is necessary to promptly know when water leaks occur, but most water leaks occur underground, so it is difficult to detect them from above ground.

A typical method for detecting underground water leakage that has been used in practice is a method using an acoustic rod. This can be done by placing a sound rod on the ground where the pipe is buried, by directly contacting the buried pipe through a bored hole, or by placing it on an exposed part of the pipe, such as a fire hydrant or water meter. In this method, the vibration sound transmitted from the sound rod is mechanically or electrically amplified by contacting the sound rod, and the inspector listens to the sound through a headphone to determine whether there is a water leak. However, this method requires skilled techniques to distinguish between water leak sounds and other noises, and it takes a huge amount of time and effort for surveyors with this technology to patrol all urban areas. The problem is that progress in discovering these areas is slow.

For this reason, there is a correlation type water leak detection device that has recently been put into trial use with the aim of automating water leak detection. This is an attempt to determine the location of water leakage by cross-correlating signals from vibration sensors attached to two fire hydrants. However, this has the following drawbacks. First, we must have accurate data on the branching of the pipes in the survey section, the pipe material, and the length of the pipes. Second, if there are branch pipes along the way, the branch pipes must be investigated separately. Thirdly, although it does not require skilled technology, it does require patrolling inspections within the city.
There are limits to early detection of water leaks.

Approximately 90% of water leaks currently occur in the water supply pipes leading to consumers, including branch points from the distribution pipes, so it is necessary to permanently install water leakage detection devices on the water supply pipes of each consumer. This enables early detection of water leaks around customers. The sensors used in such a water leak detection device include a pressure detector that detects water pressure fluctuations due to water leak sound propagating in the pipe as minute water pressure fluctuation waves, or a vibration detector that detects mechanical vibrations of pipe walls, etc. In Although,
It is difficult to completely eliminate noise from sources other than water leakage or from vibration sources.

In addition to the noise that usually comes from the water faucet,
The main sources of noise include noise from passing vehicles, noise from machine operation in factories, and noise from work at construction sites. In order to simplify the explanation, sound will be described below, but the same applies to vibration.
Methods for distinguishing these noises from water leakage sounds include (1) discrimination based on signal size, (2) discrimination based on signal frequency, and (3) measurement late at night when the noise source is reduced. There are ways to do this. Therefore, by using these methods together or singly, it is possible to improve the ability to discriminate between water leakage sound and noise.

An object of the present invention is to provide a water leakage detection device that can effectively detect the presence or absence of water leakage while preventing false detections due to noise.

Water leaks occur due to corrosion of water supply and distribution pipes, water valves, etc., loosening of joints destroyed by external force, poor execution, etc. Once a leak occurs, it will continue to occur unless repaired and will recover naturally. That is impossible. On the other hand, the biggest source of noise is the sound of the water faucet.
It is obvious that the sound is produced only when using water. Other noise sources such as vehicle sounds, various mechanical sounds,
It is extremely rare for construction work noise to occur continuously for 24 hours. Therefore, the present invention integrates the duration of the sound for a certain period of time set in advance,
This system distinguishes it from other noises by determining water leakage only when this integral value exceeds a preset criterion, that is, when the ratio of the integral value during the above-mentioned fixed time exceeds the criterion. It is.

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

FIG. 1 shows an outline of the configuration of the present invention, in which water leakage sound and noise detected by a sensor 1 are converted into electrical signals and input to a waveform shaping circuit 2. The waveform shaping circuit 2 amplifies the electrical signal from the sensor 1,
Filtering by frequency and discrimination by wave height value are performed to identify signals that are significant as water leakage signals.
In other words, if a signal is greater than the planned reference level,
For example, waveform shaping is performed to a high level and non-significant signals to a low level. The signal duration integration circuit 3 is
The time when the input signal is at a high level is integrated and a time-integrated signal T0 is output. The water leakage determination circuit 4 is a circuit that outputs a water leakage determination signal D0 when the time integrated by the signal duration integration circuit 3 exceeds a determination reference value.

FIG. 2 is a configuration diagram of an embodiment of the waveform shaping circuit 2. In FIG. The amplifier circuit 20 amplifies the electrical signal from the sensor 1 by 1×10 ⁵ to 1×10 ⁶ times, and extracts the main component of the signal by using a bandpass filter 21 with a low cutoff frequency of 1 kHz and a wide cutoff frequency of 10 kHz, for example. Let it pass. This alternating current electric signal is converted into a positive signal by an absolute value amplification circuit 22, and a comparator circuit 23 converts the negative signal into a positive signal.
is input. The comparison circuit 23 has a judgment reference voltage +
Converts only input signals higher than Ve to constant voltage signals and outputs them.

6a in FIG. 6 is an example of the output signal waveform of a bandpass filter, and FIG. 6b is an example of the output signal waveform of the absolute value amplifier circuit, in which the negative signal of a is folded back at 0V. FIG. 3c shows the output signal waveform of the comparator circuit which becomes high level when the signal is larger than the determination reference voltage +Ve, and becomes the output signal of the waveform shaping circuit 2.

Figure 3 shows the signal duration integration circuit 3 in Figure 1.
This is an example. The frequency of the pulse oscillator 30 is set to be several times higher than the high cutoff frequency of the filter included in the waveform shaping circuit 2 to ensure resolution. The timing circuit 34 has a built-in timer and outputs an integration start signal at the time set by the control input.
It has the function of generating a series of command signals such as INI, integration end signal INO, judgment execution signal DIO, and counter 33 reset signal RS.

These signals normally occur on a daily basis; for example, the integration start signal INI at midnight is generated at 5 am.
It is set so that the integration end signal INO is output at the same time.
The flip-flop 35 maintains the logic "1" state from the integration start signal INI to the integration end signal INO, performs an AND with the pulse by the AND gate 31, and supplies the pulse to the AND gate 32. The AND gate 32 digitizes the duration of the output signal of the waveform shaping circuit 2 by taking the AND of the output pulse of the AND gate 31 and the output signal of the waveform shaping circuit 2. By counting the output pulse train of the AND gate 32, the counter 33 performs time integration of the waveform shaping circuit output signal from the generation of the integration start signal INI until the generation of the integration end signal INO.

6d is the output pulse train of the AND gate 32, and the number of pulses is proportional to the high level period of the output signal of the waveform shaping circuit 2. e is the integration start signal INI, f is the integration end signal INO, g is the integration period signal indicating the logic "1" state of the flip-flop, and h is the determination execution signal DIO, which is issued after the integration end signal INO, and is the water leakage determination circuit 4. It is for operating. A reset signal RS is issued further after this and is used to initialize the counter 33.

FIG. 4 shows another example of the signal duration integration circuit 3.
The difference with FIG. 3 is that instead of the pulse oscillator 30,
The voltage-frequency conversion circuit 40 is used. The voltage-frequency conversion circuit 40 is a circuit that outputs a pulse with a frequency proportional to the input voltage, but since the magnitude of the output signal of the waveform shaping circuit 2 is a constant value, it remains constant while the input signal continues. Since it outputs frequency pulses, it has the same function as the circuit in FIG. 3, and the signal waveform in FIG. 6 is also the same.

FIG. 5 is an embodiment of the water leakage determination circuit 4, and shows a binary parallel comparison circuit. Now, assume that two binary numbers with _k bits are A (a digital value of _k bits counted by the counter 33) and B (a digital value with the same number of bits set in advance as a judgment reference value). ), and the logic “1” of the LSD of binary number A
A ₁ , MSD's logic “1” is A _K , similarly, LSD's logic “1” of binary number B is B ₁ MSD's logic “1”
Let B _K be, and define p _k , q _k , and r _k as follows. p _k = A _k ∩B− _k (k=1, 2,…, K) q _k = A− _k ∩B _k (k=1,2,...,K) r _k =A _k ∩B _k ∪A- _k ∩B- _k (k=1,2,...,K)
...(1) However, A- _k and B- _k are the logic "0" signals of Ak _{and Bk} , respectively, ∩ is a symbol representing logical product, and ∪ is a symbol representing logical sum.

Part A in FIG. 5 is a circuit configuration of the above equation (1), and is constructed using an AND gate 50, an OR gate 51, and a NOT gate 52 as shown. In other words, p _k (k=1, 2,..., K) is expressed as follows: if the corresponding bit (k) is A>B (A _k is logic "1")
When B _k is logic "0"), it becomes logic "1". In addition, r _k (k=1, 2,..., K) is a logic "1" when the corresponding bit (k) is A=B (both A _k and B _k are logic "1").
or in the case of logic "0") becomes logic "1". Furthermore, q _k (k = 1, 2, ..., K) is defined when the corresponding bit (k) is A < B (A _k is logic "0", B _k
is logic "1") only when the logic "1" is logic "1".

Also, the logical symbols P, Q, and R of the judgment results regarding the magnitude of binary numbers A and B are expressed as P=1 (A>B) Q=1 (A<B) R=1 (A=B) ...(2) If we decide as follows, P=p _K ∪(p _K-1 ∩r _K )∪(p _K-2 ∩r _K ∩r _K-1 )
∪… Q=q _k ∪(q _K-1 ∩r _K )∪(q _K-2 ∩r _K ∩r _K-1 )∪… R=r _K ∩r _K-1 ∩r _K-2 ∩…∩ r ₁ …(3). Part B of FIG. 5 uses an AND gate 50 and an OR gate 51 to construct the logic of equations (1), (2), and (3).

In the above configuration, the logic symbol P is defined when the most significant bit A _K of the two binary numbers A and B is logic "1", B _K is logic "0" and p _K is logic "1", or a certain bit For example, in _K-1 bits, if the upper bits are equal (in this case, A _K = B _K ) and the corresponding bits are A > B (in this case, A _K-1 > B _K-1 )
When the corresponding bit (p _K-1 in this case) becomes logic "1", P=1, indicating that the binary number A is greater than B. logical symbol R
When the bits of the two binary numbers A and B are all equal (1=1 or 0=0), R=1, indicating that the binary number A=B.

The logical symbol Q is for two binary numbers A and B,
In the opposite relationship to the case of P above, if A<B, then Q
=1. Therefore, if the binary number A is made to correspond to the counter value and the binary number B is made to correspond to the determination reference value as described above, the output P becomes the determination output. Of course, the part B' related to the signal output of Q is unnecessary in realizing the water leakage determination circuit.

Next, regarding the criteria for determining the water leakage state, if the time integration rate used for water leakage determination is F, its value is determined by the following equation.

Here, T corresponds to the time integration rate calculation interval, and ti corresponds to the interval exceeding the determination reference voltage Ve, that is, the width of each rectangular wave in FIG. 6C. In addition, the suffix i is the number of rectangular waves exceeding the reference voltage Ve (i=1, 2,
..., n).

If this water leakage waveform is approximated as a sine wave with the signal magnitude as the peak value Vp and the judgment reference voltage Ve as shown in Figure 8, and π and θ as shown in the figure, the above equation (4) is as follows. It is expressed as

F=π-2θ/π×100%...(5) Generally, a sine wave is expressed as y=a sin x, and if the peak value a=Vp, then Ve=Vpsinθ ∴θ=sin ^-1 (Ve/Vp) , substituting this into equation (5) yields the following.

F = [1 = 2/π sin ^-1 (Ve/Vp)] ...(6) Figure 7 shows F (time integration rate) in equation (6) above on the vertical axis, and the magnitude of the water leakage signal (Vp). The horizontal axis represents the determination reference voltage Ve as a parameter. In addition, in the figure, the magnitude of a sine wave signal with an effective value of 1V is
It is set to 0dB.

In FIG. 7, for example, if Ve=0.5V and the determination criterion is 40%, it can be determined that there is a water leak if a signal of −7 dB or more is continuously input.
Also, when a +20 dB vibration sound is input, if there is a water leak, the time integration rate will be 98% from Figure 7, but since this is noise, it will occur at a duty rate of 20% during the measurement period, for example. And the time integration rate is 98×
Since it is only 0.2=19.6%, it is considered as noise without being incorrectly determined as a water leak.

The actual determination reference voltage Ve and the water leakage determination value (%) relative to the time integration rate are set for each installation location based on the results of field tests, etc., with reference to the graph in FIG. 7. As mentioned above, the water leak detection device is installed in the water supply pipe to the consumer (general household), so in order to avoid misjudgment due to the sound of the water faucet, which is the biggest source of noise, it is assumed Each value is determined based on the longest continuous use of water.

Note that the sensor in this device is not limited to a pressure-to-electrical converter, but may also be a vibration (displacement, velocity, acceleration)-to-electrical exchanger, etc., and can be applied in exactly the same way.

As described above, according to the present invention, even if the frequency component of the noise is similar to the water leakage sound signal and its magnitude is larger than the water leakage sound signal, for a preset certain period of time,
By integrating the duration of the signal, a water leak can be detected by easily distinguishing between noise and a water leak signal based on the ratio of the integral value to the fixed time.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram showing an embodiment of the water leakage detection circuit according to the present invention, Fig. 2 is a detailed diagram of the waveform shaping circuit shown in Fig. 1, and Fig. 3 is the signal duration integral shown in Fig. 1. A block diagram of the circuit; FIG. 4 is a block diagram showing another example of the signal duration integration circuit; FIG. 5 is a block diagram of the water leakage determination circuit shown in FIG. 1; and FIG. 6 shows the operation of the present invention. FIG. 7 is a diagram showing the relationship between signal magnitude and time integration rate, and FIG. 8 is a waveform diagram when the water leakage state is approximated as a sine wave. 1...Sensor, 2...Waveform shaping circuit, 3...
Signal duration integration circuit, 4...Water leakage determination circuit,
INI...Integration start signal, INO...Integration end signal,
DIO...Judgment execution signal, RS...Reset signal,
T ₀ ...Time integral output, D ₀ ...Judgment output.

Claims (1)

[Claims] 1 A sensor that detects sound pressure fluctuations and vibrations caused by water leakage from water pipes and attached equipment is input, and the output of this sensor is input, and components exceeding a predetermined level are converted into a predetermined peak value signal. a signal duration integration circuit that integrates the output of this waveform shaping circuit for a predetermined time set longer than the longest continuous usage time of water expected in a general household, and this integrated value. A water leakage detection device comprising: a water leakage determination circuit that determines that there is a water leakage if a ratio of the ratio to the predetermined time is greater than a set determination reference value.