JPH03291535A - Seismoscope - Google Patents

Abstract

PURPOSE:To accurately decide earthquake waves which are different, area by area, by superposing two conditions of a vibration frequency and duration one over the other and deciding the cause of the vibration. CONSTITUTION:A sensor part 1 supplies the electric signal of the vibration to a signal processing decision part 3 through a preamplifier 2 and the decision part 3 extracts the vibration frequency and duration of the signal wave. Then a fuzzy reasoning part 4 uses the vibration frequency, amplitude, and duration of the vibration and the number of peaks of the amplitude within a specific time which have acceleration larger than a specific threshold value as fuzzy variables to find the magnitude alphaof the earthquake as the sum of sets of definition variables alpha1, alpha2, and alpha3. Further, a tuning function pat 6 initializes a weight coefficient and a learning function part 5 increases or decreases the weight coefficient of a definition variable alphai which exerts maximum influence of the calculation of the magnitude alpha according to the decision result. Then the decision part 3 decides whether the vibration is caused by the earthquake or others by fuzzy logic of the reasoning part 4 where the two condition of the vibration frequency and duration are superposed one over the other, and outputs a specific signal to an output part 7 when it is decided that the magnitude alpha is larger than the threshold value TH and the earthquake causes the vibration.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a seismic sensor, and particularly to a seismic sensor that is attached to a gas meter or the like to sense earthquakes.

(Prior Art) Japan is one of the most earthquake-prone countries, and has experienced numerous felt and non-sensible earthquakes.

Disasters caused by earthquakes include primary disasters caused by the earthquake itself;
There are secondary disasters that occur along with earthquakes. Secondary disasters are often man-made such as fires, and if the occurrence of an earthquake is accurately detected in real time and appropriate measures are taken based on this, the occurrence of secondary disasters can be significantly prevented.

For this reason, there are many types of seismic sensors that use vibration sensors to detect the occurrence of earthquakes using predetermined algorithms such as measuring the number of amplitude peaks within a certain period of time. Proposed.

(Problem to be solved by the invention) However, during the propagation process from the epicenter to the observation point, seismic waves undergo numerous reflections and refractions while being influenced by many factors such as the ground and soil. , complex,
There are many variations. In particular, the waveform is thick (changes) depending on the ground and soil, and even in the same earthquake, the waveform differs greatly depending on the region.For this reason, conventional algorithms are used to select seismic waves from various vibration waves that are bundled in daily life. Even if they tried, the determination rate would vary greatly depending on the region.

Generally, in areas with hard ground, seismic waves have short periods and small amplitudes, whereas in areas with soft ground, seismic waves have long periods and large amplitudes.

Seismic waves, which vary greatly from region to region, cannot be accurately determined using a single conventional algorithm, such as an algorithm that determines seismic waves by measuring the number of amplitude peaks within a certain period of time.

For this reason, in order to obtain a high detection rate in all regions, it is necessary to prepare a different algorithm for each region, which requires enormous development costs and other expenses. Practically speaking, this is close to impossible.

The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to provide a seismic sensor that can accurately determine seismic waves that vary from region to region.

(Means for Solving the Problem) According to the present invention, the above-mentioned object includes a vibration sensor that converts vibration into an electric signal, a vibration frequency and a duration extracted from the electric signal from the vibration sensor, A signal processing/judgment unit that determines whether the vibration is due to an earthquake or another cause based on the superposition of two conditions, frequency and duration, and outputs a specific signal when it is determined that the vibration is due to an earthquake. This is achieved by a seismic sensor characterized by having the following.

The seismic sensor according to the present invention has a learning function section that learns a judgment condition based on the superposition of two conditions of vibration frequency and duration based on the suitability of the judgment result as to whether the vibration is caused by an earthquake or another cause. In addition, a manual operation section may be provided in the house for supplying a signal to the learning function section regarding the suitability of the determination result as to whether the vibration is caused by an earthquake or another cause.

(Function) In general, seismic waves continue to vibrate at a frequency in a predetermined band for a predetermined period of time, so as mentioned above, the combination of the two conditions of frequency and duration determines whether the vibration is due to an earthquake or due to other causes. An earthquake is detected by determining whether it is a thing or not. Then, based on the suitability of the judgment result of whether the vibration is caused by an earthquake or another cause, a judgment condition based on the superposition of two conditions of vibration frequency and duration is learned, and this is a judgment condition that determines whether the vibration is caused by earthquake waves in the installation area. It becomes compatible with the judgment. If a manual operation unit is provided in the house that provides a signal regarding the suitability of the judgment result as to whether the vibration is due to an earthquake or another cause to the learning function unit, this operation is performed inside the house.

(Example) The present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 shows an embodiment of a vibration sensor according to the present invention.

In Fig. 1, the reference numeral 1 indicates a seismic sensor, and the seismic sensor 1 is composed of a combination of a cysmo system and a transducer, and is a sensor that receives vibration and converts it into an electrical signal. section 1, and S is given an electric signal from the sensor section 1.
A preamplifier 2 that improves the /N ratio and performs amplification, a signal processing determination unit 3 that receives the electrical signal amplified by the preamplifier 2, a fuzzy inference unit 4, a learning function unit 5, and a tuning function. 6 and an output section 7 to which the electrical signal of the determination result from the signal processing determination section 3 is applied.

The output unit 7 outputs an activation signal to the buzzer 8, and when the seismic sensor 1 is installed in a gas meter of a gas consumer,
The output unit 7 outputs a control signal to the cutoff well 9 that cuts off the gas supply, and operates the cutoff valve 9 to cut off the gas supply in the event of an earthquake.

The signal processing/determination section 3 is given the electric signal from the preamplifier 2, that is, the signal wave to be determined, extracts the frequency and duration of this signal wave from this, and divides the frequency and duration into two. It determines whether the vibration is due to an earthquake or another cause by superimposing two conditions, and when it is determined that the vibration is due to an earthquake, it outputs a specific signal indicating that it is an earthquake to the output section 7. It is configured as follows.

The above-mentioned determination in the signal processing determination section 3 is performed using a fuzzy inference section 4 based on fuzzy theory. In this case, the fuzzy variables are the frequency and
The amplitude, the duration of the vibration, and the number of peaks of the amplitude within a predetermined time period having an acceleration equal to or greater than a predetermined threshold are used, and fuzzy inference is performed based on the fuzzy rule defined as follows.

α1=W13μm (χ1) V μ4 (χ4)・W4 ... (1) α2=
W2 ・μ2 (χ2) Δ μ3 (χ3)・w3 ... (2) α3
= w1 ・μm (χ1) Δ μ3 (χ3)・w3 ... (3) α
= α1 Vα2 Vα3 ... (4)
μ, (χ1) is the earthquake-likeness regarding the frequency, μ2 (
χ2) indicates the likeness of an earthquake regarding the amplitude, μ3 (χ3) indicates the likeness of an earthquake regarding the duration of vibration, μ4 (χ4) indicates the likeness of an earthquake regarding the number of amplitude peaks of acceleration above a predetermined threshold, and Wl, w2, W3, W4 is a coefficient that determines the weight of the fuzzy variables χ1, χ2, χ3, and χ4, respectively.

α1 is a defined variable that is the union of the earthquake-likeness in terms of frequency and the earthquake-likeness in terms of the number of amplitude peaks, and is used exclusively as a countermeasure against running vibrations of automobiles, trucks, etc.

α2 is a defined variable that is the product of the earthquake-likeness related to the amplitude and the earthquake-likeness related to the duration of vibration, and this is a measure against vibration caused exclusively by being hit by a game ball or the like.

α3 is a defined variable that is the product of the earthquake-likeness in terms of frequency and the earthquake-likeness in terms of duration of vibration, and is a measure against vibrations caused mainly by people leaning on the ground.

As shown in equation (4), α is the seismic intensity determined by the union of α1, α2, and α3, and if this seismic intensity α is greater than or equal to a predetermined threshold TH, it is considered an earthquake. If not, it is determined that it is not an earthquake but a disturbance.

Figure 2 shows an example of the membership function μm of the fuzzy variable χ1 related to the frequency, Figure 3 shows an example of the membership function μ2 of the fuzzy variable χ2 related to the amplitude, and Figure 4 shows the membership function μm of the fuzzy variable χ3 related to the duration of vibration. FIG. 5 shows an example of the friendship function .mu.3, and FIG. 5 shows an example of the membership function .mu.4 of the fuzzy variable .chi.4 regarding the number of peaks of the amplitude of acceleration above a predetermined threshold.

The tuning function unit 6 has an adjustment unit that performs initial settings of the weighting coefficient W, etc., and a switch button such as a reset button that generates a signal as to whether or not the determination has been made correctly based on the user's judgment. Based on this, the learning function unit 5 determines αi that had the greatest influence on the calculation of the earthquake intensity α, and if the determination result is determined to be correct, it increases the weighting coefficient wi related to that αi. On the other hand, when it is determined that the judgment result is wrong, the weighting coefficient wi related to αi
Learning control is now in place to reduce this.

FIG. 6 shows a flowchart for earthquake determination and learning based on a determination algorithm using a fuzzy theory such as "double series".

First, in step 10, the earthquake intensity α is calculated according to the above equation (4), and then in step 20, the defining variables α1, α2, and α3 used this time are memorized. be exposed.

Next, in step 30, an earthquake determination is made based on whether the earthquake intensity α is greater than a predetermined threshold TH. If α > TH, it means that an earthquake has been detected, and in this case, the process proceeds to step 40 to perform earthquake determination output processing, whereas when α > TH, it means that there is no earthquake, and the process returns to step 10. .

In step 40, for example, the buzzer 8 is operated and the cutoff valve 9 is cut off.

In step 50, it is determined whether or not the current earthquake determination result is correct based on the user's operation of a switch button provided in the tuning function section 6. If it is correct, the process proceeds to step 60; otherwise, the process proceeds to step 60. If an earthquake is determined even though it is not an earthquake, the process proceeds to step 80.

In step 60, αi that had the greatest influence on the calculation of earthquake intensity α is calculated from α1 and α stored in step 20.
2, α3 is determined, and then, in step 70, the weighting coefficient wi associated with αi is increased. For example, αi has the greatest influence on the calculation of earthquake intensity α.
is α1, the weighting coefficients W1 and W4 are increased.

In step 80, as in step 60, α1 that has the greatest influence on the calculation of α is determined, and in step 90, the weighting coefficient wi associated with αi is reduced. will be held.

As shown in the double series, by learning the weighting coefficient wi based on the suitability of the judgment result, the probability of earthquake judgment by fuzzy theory becomes even more positive in the process of use, and in the process of use, the probability of earthquake judgment Each of the vessels 1 will be adapted to each region.

FIG. 7 shows one embodiment in which the seismic sensor according to the present invention is incorporated into a gas meter. In FIG. 7, reference numeral 11 indicates a gas meter as a whole, and the gas meter 11 includes a sensor section 1 of the seismic sensor 10 in addition to a gas meter display section 12.
, a controller 13 configured with a signal processing determination unit 3, etc.
A fuzzy inference microprocessor 14, a battery 15 for powering the seismic sensor 10, etc., a shutoff valve 9, an indicator lamp 16 that displays the reason for shutoff and a warning, and an operation switch 17
have. The operation switch 17 is for double learning; for example, if the number of times of suppression is one, it is assumed that the judgment result is correct, and if it is pressed twice in succession, the judgment result is incorrect.・It would be good if the ψ production promise was fixed.

FIG. 8 shows a case where the gas meter 11 with a seismic sensor as shown in FIG. 7 is installed outdoors.
FIG. 9 shows an embodiment in which a gas meter 11 with a seismic sensor is installed indoors. In FIGS. 8 and 9, reference numeral 18 indicates a gas pipe, and 19 indicates a wall of a house.

In the embodiments shown in FIGS. 7 to 9, the display lamp 16 and the operation switch 17 are built into the scum meter 11 in all cases, but the sass meter 11 is placed outdoors. In this case, as shown in FIG. 10 or 11, the display lamp 16, operation switch 17, and even the buzzer 8 are incorporated into the unit box 20 separately from the meter 11, and only this unit box 20 is installed indoors. It may be arranged in such a way that the In this case,
The contents of the display lamp 16 can be conveniently viewed indoors, and the operation switch 17 for learning can also be easily operated indoors.

In addition, in the series of embodiments, the present invention is not limited to this, and the present invention is not limited to whether earthquake determination is performed using fuzzy theory using amplitude and number of peaks in addition to frequency and duration. In the present invention, earthquake determination may be performed by superimposing two conditions, namely, minimum limit, vibration frequency, and connection time, and is not limited to determination based on fuzzy theory.

(Effects of the Invention) As is clear from the above explanation, in the seismic sensor according to the present invention, it is possible to detect vibrations caused by earthquakes or other vibrations due to the superposition of at least two conditions of vibration frequency and duration. Since the 'I'11 determination is performed and an earthquake is detected, this earthquake detection can be performed reliably.

Based on the suitability of the judgment result as to whether the vibration is due to an earthquake or another cause, the judgment condition based on the superposition of the two conditions of vibration frequency and duration is learned, so that it can be used in the area where it is installed. The conditions for earthquake determination will become more accurate, and earthquake determination will be performed more reliably and with a higher determination rate.

If a manual operation unit such as an operation switch is provided indoors, which provides a signal to the learning function unit regarding the validity of the judgment result as to whether the earthquake was caused by vibration or other causes,
This operation can be easily performed indoors, eliminating the need to go outside for learning.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing one embodiment of the seismic sensor according to the present invention, and Figs. 2 to 5 show the membership function characteristics used in the fuzzy theory applied to the seismic sensor according to the present invention. FIG. 6 is a flowchart showing an example of the operation procedure of the seismic sensor according to the present invention, FIG. 7 is a perspective view showing one embodiment in which the seismic sensor according to the invention is incorporated into a gas meter, and FIG. FIGS. 11 to 11 are schematic configuration diagrams each showing an embodiment of the combination of a seismic sensor and a gas meter according to the present invention. 1... Sensor section 2... Preamplifier 3... Signal processing determination section 4... Fuzzy inference section 5... Learning function section 6... Tuning function section 7... Output section 8. ... Buzzer 9 ... Shutoff valve 10 ... Earthquake sensor 11 ... Gas meter 12 ... Gas meter display section 13 ... Controller 15 ... Lithium battery 16 ... Display lamp 17 ... Operation Switch 18...Gas pipe 20...Box unit

Claims (1)

[Claims] 1. A vibration sensor that converts vibration into an electrical signal, extracts the frequency and duration from the electrical signal from the vibration sensor, and generates vibration by superimposing the two conditions of frequency and duration. Determine whether it is due to an earthquake or other causes,
A seismic sensor comprising: a signal processing determination section that outputs a specific signal when determining that vibration is caused by an earthquake. 2. It has a learning function section that learns the judgment condition based on the superposition of two conditions of vibration frequency and duration based on the suitability of the judgment result of whether the vibration is caused by an earthquake or another cause. The seismic sensor according to claim 1, characterized in that: 3. A manual operation section according to claim 2, characterized in that a manual operation section is provided in the house for giving a signal to the learning function section regarding the suitability of the judgment result as to whether the vibration is caused by an earthquake or another cause. Seismic sensor.