JPS608468B2 - Earthquake early detection warning system - Google Patents

Description

[Detailed Description of the Invention] The present invention relates to an earthquake early detection and warning system that detects earthquakes as quickly as possible, estimates the damage area, and issues warnings only for earthquakes that are truly harmful. .

Trains traveling at high speed on railway tracks are in an extremely dangerous situation if they encounter a large earthquake.

For this reason, equipment for detecting earthquakes is installed along each railway line, and they constantly monitor the occurrence of earthquakes. Conventional earthquake detection and warning systems issue a warning that a dangerous earthquake has occurred when the amplitude value of an earthquake detected by a detection device exceeds a predetermined limit value (for example, horizontal ground motion 4 foundation a!). The plan was to temporarily stop trains by ordering them to apply sudden brakes, and then resume operation after confirming the safety of the earthquake. Here, ``In railways such as the Japan National Railways Shinkansen, which conducts highly centralized train management, the amount of information exchanged between the train command and the trains is enormous, and the amount of information is limited. In order to prevent mistakes, emergency train stop commands are realized by stopping the turbulence in the relevant section.Once the turbulence has been stopped, this substation equipment is designed to carry out ``safety confirmation work in accordance with the regulations''. It is difficult to restart the haze after a short period of time.Therefore, once ground motion that exceeds the critical level is detected and the substation is instructed to stop exposure, the earthquake can be safely detected in a relatively short period of time. Even if it is confirmed that the dew has occurred, it will take a considerable amount of time for the dew to resume.Therefore, when setting this limit level, it is important to consider "It is necessary to minimize the interruption of train travel due to non-hazardous earthquakes by setting a high value. However, in order to ensure the safety of train travel, setting a limit level It is also necessary to allow for a certain degree of safety factor.For this reason, ``In such an earthquake detection and warning system, it is inevitable to some extent that trains will be stopped even by an earthquake that is not dangerous.'' The present invention detects the occurrence of an earthquake as quickly as possible, and when an earthquake is detected, calculates the epicenter and estimates the magnitude based on the period and amplitude of the initial motion, and uses the results to estimate the damage. By estimating the area and determining whether or not the railway line being laid is included in the damaged area, whether or not it is causing damage, and if it is determined that it is causing damage, we will immediately stop the dew in the relevant section, etc. If the data at this point does not determine that there is a cause of harm, new input data is taken in and a determination is made that the earthquake is a cause of harm. The above-mentioned procedure is repeated to determine whether or not the earthquake is a cause of harm until the earthquake is complete. By doing so, we will be able to secure sufficient time, and this will enable us to issue an early warning only if the detected earthquake is one that is truly causing damage, and for all earthquakes that are truly causing harm. The system provides a detection and warning system, and when applied to the earthquake detection and warning system along railway lines as described above, ``The fact that dew stops at the substation and trains stop traveling means that the detected earthquake is truly dangerous. Since it is limited only to cases where this is included, disturbances in the entrainment timetable can be minimized and stopped, making it extremely effective.

A detailed explanation will be given below based on examples. FIG. 1 is a diagram showing an example of the arrangement of detection points in the present invention.

In the figure, 1 is the railway track, 2 is the substation for monitoring the railway station, 3 is the observation point placed at about 100 intervals along the railway line, and 41 is the substation for monitoring the railway line. These are detection points distributed at several locations around 3 to 5 locations around this observation point 3. Each substation 2 supplies electric power for train operation to the track turtles in the assigned section, and adjacent substations 2 are connected by remote control lines 5 to enable mutual information exchange. It is being done. In addition, as shown in Figure 2, each observation point 3 is equipped with an observation point device 30 connected to the nearest substation 2 through a communication line 6, and information such as the above-mentioned Kiryu stop command is transmitted through this communication line. 6 (also the control line 6 if necessary) from the observation point device 30 to the corresponding substation 2.
sent to. A detection point device 40 is installed at each detection point 4', and each of them is connected to the observation point device 3 via a transmission line 7.
0, the seismic data at the detection point is sent to the observation point device 3Q.
I am sending it to Here, in order to detect the occurrence of an earthquake more quickly, it is advantageous to install the sensor in an earthquake-prone area, including the ocean floor, or in a location closer to the area, taking into consideration the propagation speed of seismic waves.

This earthquake-prone area has been clarified to some extent by the results of research on seismic activity to date.In other words, taking the Tohoku region as an example, there are areas such as off the coast of Tokachi, off the coast of Sanriku, off the coast of Miyagi Prefecture, off the coast of Fukushima Prefecture, and the Sea of Kashima. Many of the earthquakes have their epicenters on the ocean floor, concentrated on the Pacific Ocean side, and are extremely rare on the Japan Sea side.Therefore, earthquake countermeasures for railways in this region include, for example, Hachinohe, Miyako, Ofunato, Mt. Kinka, It is effective to place observation points 3 at appropriate intervals along the Pacific coast, such as Soma, Iwaki, Choshi and Miura Peninsula.There are about 5 detection points 4 around observation points 3 selected in this way. This means that when calculating the source element from observation data, the unknowns are the coordinates X and Y of the source, the depth Z of the source, and the propagation velocity
, oscillation time, etc., and in order to calculate these, it is necessary to create a five-element simultaneous equation using data from five locations.In this layout example, the possibility of missing side, etc. Taking this into consideration, we set five detection points 4 and calculated the epicenter element based on observation data from six locations, including that from observation point 3.Therefore, for economic reasons, etc. If you want to reduce the number of points 4, you can pre-assume one or both of the values of the propagation velocity v and the depth of the epicenter Z, and if you have observation data from 4 or 3 locations, the epicenter element can be Calculations become possible, and ``if you set up at least two detection points, you can achieve the purpose.''Here,
Each of these detection points 4 is installed at 3 to 5 shoes around the observation point 3. Since the propagation speed of incoming seismic waves, especially longitudinal waves (hereinafter referred to as P waves), is usually around 8 f/sec near the ground surface, the initial part of the seismic waves is transmitted through observation point 3 and all detection points 4, which are the lotus. It passes within 1-2 seconds. Therefore, if initial movements are detected at multiple locations as a condition, false detections due to malfunctions, noise, etc. can be reduced. Here, the transmission line 7 connecting each detection point device 40 and the observation point device 30 or the communication line 6 connecting the observation point device 30 and the substation 2 includes an optical fiber or the like. When used, it is extremely effective as a noise countermeasure because it is not affected by induced noise such as lightning.It is also effective to use wireless as the transmission line 7 and communication line 6.In other words, if it is wireless If the transmission line or communication line itself is not cut off due to earthquake damage, and if there is no damage to the transmitting or receiving equipment, communication will not be completely cut off.Figure 2 shows the layout and settings as shown in Figure 2. It is a block diagram showing an example of an observation point device 30 and a detection point device 40 installed at an observation point 3 and a detection point 4.

In the figure, numeral 31 is a sensor for detecting ground motion, and is of a type that detects, for example, three components of ground motion in the vertical direction, east-west direction, and north-south direction. Further, 32 is an amplifier that amplifies the detection signal from the sensor 31, 33 is a buffer amplifier, and 34 is a receiving device that receives signals sent from each detection point device 40 via the transmission line 7. 35 analyzes ground motion observation data sent moment by moment from each detection point device 40, etc., identifies the occurrence of an earthquake and determines its culpability, and issues a warning signal if it is determined to be culpable. The control processing device 36 that generates the alarm signal is a transmitter that sends this alarm signal to the nearest substation 2 via the communication line 6, and the reference numeral 37 is a clock device that generates accurate time information. The observation point device 30' is composed of these devices. Further, in the figure, 41 is a sensor for detecting ground motion, for example, a type that detects only one vertical component of ground motion, and 42 is a sensor for sending the detection output signal to the transmission line 7. It is a transmitting device. The detection point device 40G is mainly composed of these two devices. In this embodiment, the observation point device 3 river sensor 31
However, it is not necessarily necessary to include the sensor 38, and the number of detection points is increased by one, and the observation point device 3 is installed here.
By installing a sensor equivalent to the sensor 31 of 0, it is possible to construct a system with exactly the same function as the one described above.

In this case, since the position of the observation point 3 can be determined without considering whether it is suitable for detecting earthquake data, priority should be given to the arrangement of the observation point device 30, the dispatch of observers, etc. This makes it possible to select a location with convenient transportation. Further, FIG. 3 is a block diagram showing an example of a communication control section in the substation 2. As shown in FIG. In the figure, 21 is a remote control receiving device for receiving signals sent from the adjacent substation 2 via the remote control line 5, and 22 is a remote control receiving device for transmitting or transferring signals to the adjacent substation 2. This is a remote control receiving device for sending signals to the control circuit 5. Further, 23 is a receiving device that receives a feeding stop command, etc. sent from the observation point device 30 via the communication line 6, and 24 is a receiving device that controls the operation of the substation equipment in response to the feeding stop command, etc. A control device 25 is responsive to this control signal. Furthermore, 26 is a transmission/reception control device that controls transmission and reception in these devices, and 28 is a signal bus line that interconnects all of these devices. The communication control section shown in FIG. 3 is for a substation that is directly connected to the observation point device 30 by the communication line 6, and the communication control section in the substation that is not directly connected is the receiving device 23.
Not equipped with In this case, the flea stop command and the like from the observation point device 30 are sent from the adjacent substation via the remote can 8 circuit 5 in the same way as normal remote control information. Below, the first
The operation of the embodiment shown in the block diagrams of FIGS. 3 to 3 will be described.

Each sensor 31 and 41 belonging to the observation point device 30 or the detection point device 401 constantly detects ground motion at the point where each sensor is installed, converts this into an electrical signal, and sends it to the control processing device 35. The control processing device 35 receives information from the sensor 41 at each detection point 4 via the transmitting device 42, transmission line 7, and receiving device 34, and receives information from the sensor 31 at the observation point 3 through the amplifier 32 and buffer amplifier. 3
Receive via 3. The control processing device 35 monitors the ground motion information that is continuously sent, and considers that an earthquake has occurred when the observed value exceeds a predetermined reference value. Observations for detecting the occurrence of an earthquake include various values such as the energy output of seismic waves, the root mean square value of the amplitude of seismic waves, and the cross-correlation of seismic waves from the sensor 31 at observation point 3 and the sensor 41 at each detection point 4. is possible. As mentioned earlier, in this example, we assume a detection and warning system that sends out a warning only when a truly destructive earthquake is detected. The occurrence of an earthquake is detected based on a combination of judgment results. First, we will discuss detection based on the energy output of seismic waves.
The control processing device 35 collects information on each component of the ground motion detected by the sensor 31 and constantly sent, at a predetermined time interval (for example, at a sampling rate of 100 to 300 Hz), and collects information on several past samplings. First, the drift level is calculated based on the average value.

This drift level is constantly updated based on sampling information obtained from time to time. Next, the sensor 31
The vertical direction detected by and sent from time to time,
Values x, , x2 obtained by removing the drift level from the sampling values of ground motion information of three components in the east-west direction and north-south direction
, x3, the energy output y, is calculated. This calculation is based on the corrected sampling value x of each component mentioned above,
, ume, can be found in the form of the sum of squares. Control processing device 3
5 compares the calculation result of this energy output with a predetermined reference value Y, each time, and calculates the energy output y.
．． The arrival of seismic waves is detected by observing whether or not exceeds the reference value Y. Next, detection of the arrival of seismic waves using the root mean square value of ground motion amplitude will be described.

First, the control processing device 35 samples one of the ground information detected by the sensor 31 or 41 and constantly sent, for example, vertical movement information of the sensor 31, in the same manner as described above, and sequentially transfers it to the memory. This memory has n storage areas, and each time new information is input, the oldest information overflows and disappears. The value is stored after removing, for example, the drift level. Next, each time the control processing device 35 writes a new sampling value to the memory, the control processing device 35 writes information x(L) of the entire storage area.
, x (t2), ... y
Calculate 2. The calculation result y2 of the root mean square value is compared with a predetermined reference value Y2 each time, and the root mean square value y2 is determined.
It is monitored whether or not exceeds the reference value Y2, and the arrival of seismic waves is detected. To save calculation time and storage area, the root mean square value y2 can also be calculated by exponential averaging. Furthermore, seismic wave arrival detection using the cross-correlation of seismic waves from the sensor 31 at the observation point 3 and the sensor 41 at each detection point 4 will be described.

First, the control processing device 35 samples the vertical motion component of the ground shift information from the sensor 31 and the ground information from each sensor 41 as in the case described above, and sequentially stores the memorandum in the overflow format as described above. I'm going to go. This sampled value is stored after removing the drift level, for example. The cross-correlation is between sensor 31 at observation point 3 and detection point 4.
The information of the sensor 41 can be obtained in the form of the sum of products of each piece of stored information in the stored memory. Therefore, there are five cross-correlations ap, , ap between observation point 3 and each detection point 4.
2, . . . ap5 is obtained. The control processing device 35 constantly compares each cross-correlation ap obtained in this way with a predetermined reference value ap;
When i exceeds the reference value api, it is detected that the seismic wave has passed. The control processing device 35 determines the arrival of seismic waves based on a combination of the determination results based on the energy output, root mean square value, cross-correlation, etc. described above.

Therefore, the arrival of seismic waves will not be erroneously detected due to the introduction of large noise or malfunction of instruments.

Furthermore, the observed values used to determine the arrival of seismic waves need not be limited to these three types, and other observed values such as the amplitude value of seismic waves may be used.

The control processing device 35 that detected the arrival of seismic waves in this way
The ground motion information from each sensor 31 and 41 is sampled in the same manner as described above, and the maximum value of each in the memory is constantly updated based on this, and the magnitude is estimated based on this. Here, the calculation of the magnitude is based on the maximum amplitude Am of the seismic wave, the period T at that time, and the epicenter distance R.
Various methods have been proposed for calculation using the formula M=Qlog(Am/T)+81ogR+y (constant Q, 8, y
are slightly different).

In addition, using the earthquake duration (F-P) and epicenter distance △,
=. A calculation method using the formula log(F-P)+81ogΔ10y has also been proposed.

However, calculating magnitude estimation using such a method is not suitable for early warning. That is. When calculating using these methods, calculations cannot be performed until the arrival of the maximum motion or the end of the earthquake, which can take several minutes to several tens of minutes in the case of a large earthquake, and it is difficult to issue a warning after the earthquake has passed. Situations can also occur. Furthermore, ``When calculating the epicenter distance R using only the arrival time of the P wave, the epicenter direction is calculated accurately, but the epicenter distance R is often calculated close to the observation point. This is especially noticeable when using values. However,
Increasing the size of the observation network is disadvantageous for early detection and warning of earthquakes because it takes a long time for seismic waves to pass through all observation points. Therefore, in the present invention, the estimation calculation of the magnitude and the calculation of the direction and distance of the seismic source are performed using the period np and amplitude v of the initial motion portion of the seismic wave.

When the control processing device 35 detects the arrival of a seismic wave, it first determines whether it is a longitudinal wave (concentration wave, hereinafter referred to as P wave) or a transverse wave (torsion wave, hereinafter referred to as S wave).

This determination is made using the horizontal amplitude. That is, S waves and P
As shown in FIG. 4, waves are extremely easy to distinguish because the amplitude ratios of vertical motion and horizontal motion are significantly different. If it is determined that it is a P wave in this discrimination process, the control processing device 3
Step 5 calculates the direction of the epicenter based on the arrival time of the P wave. Next, the empirical formula Mp = alo obtained from the fact that the magnitude Mp is approximately proportional to the logarithm of the initial period Tp of the seismic wave
gTp×b and the formula for estimating the magnitude from the amplitude of the initial motion derived based on Gudenberg's (Gu, where n is rg) formula for determining the body wave magnitude Mp=c (logv+lo station)×d, where v is the vertical initial motion The magnitude Mp and the epicenter distance R are estimated and calculated based on the amplitude (mkine).

The hypocenter distance R calculated in this manner is not affected by the size of the observation network and is relatively accurate. Here, the above-mentioned experimental formula was obtained as follows.

That is, the inventors of this case conducted automatic observation of the foothills on the Pacific coast of the Tohoku region and collected a lot of data. In the process of analyzing the data, the magnitude Mp and the period Tp of the initial seismic wave motion were determined.
By sequentially plotting the observed data on the coordinates with x and y axes, we found that the two showed a strong correlation as shown schematically in Figure 5. The above-mentioned empirical formula was derived based on this correlation graph. When the estimated values of the magnitude Mp and the epicenter distance R are calculated in this way, the control processing device 35 uses them to estimate the area where serious damage is expected, and the damage caused to the railway line 1 by this estimation. Determine about.

In general, the relationship between the extent of the area affected by an earthquake and its magnitude is quite complex. In other words, it is highly unlikely that the disaster-affected areas are distributed in the same circular pattern around the epicenter, and the shape is usually extremely complex. However, there is a strong correlation between the extent of damage and its magnitude, and the distance from Fujio to the farthest point where damage has spread or the radius r of the affected area.
If we take its logarithm, we get: log=A・M−B However, the relationship can be approximated by a simple linear expression such as M: magnitude, A, B: constant.

Next, it is determined whether the railway line 1 passes through an area where serious damage is expected to occur as estimated in this way. If the railway line 1 passes through the estimated area, there is a possibility of damage depending on the area*, so it is determined that there is a nuisance, and a stop command is sent. If it is determined that there is no harm caused by the above determination, it is determined whether the earthquake has ended at that point.

When the end of the earthquake is detected in the end determination process, the process returns to the earthquake monitoring process using energy output, root mean square value, cross correlation, and the like. Furthermore, if the earthquake has not yet ended at the determination processing point, the determination processing of whether it is a P wave or an S wave is restarted. P
When it is determined that it is an S wave by determining whether it is a wave or an S wave, the control processing device 35 calculates the time from the detection of the P wave to the detection of the S wave,
That is, the P-wave continuation time p-s is calculated.

Next, the Keno distance △ from this P wave continuation time p-s is
Furthermore, the direction of the epicenter, etc. will be calculated. Furthermore, the magnitude Ms is estimated based on these calculation results, observation data, etc. Various methods have been proposed for estimating the magnitude Ms, such as using the epicenter distance △ and the maximum amplitude Am. Therefore, you can choose an appropriate one from among them.After obtaining the magnitude Ms and the epicenter distance R, as in the case of determining based on the initial movement of P waves, the degree of harm can be determined based on these estimated calculated values R and Ms. In the calculation using the S wave explained above, the duration time Tp-s of the P wave is determined.
is obtained as an observed value, it is possible to calculate the epicenter distance R or the epicenter distance with high precision.Therefore, it is possible to calculate the estimated magnitude Ms, and also to judge whether there is any harm caused, with high precision. When determining whether the earthquake is a P wave or an S wave, if a P wave is not detected but a direct S wave is detected, the earthquake is either a direct earthquake or an earthquake with an extremely distant epicenter. It is good to think that it is.

Therefore, in the case of a direct type, the presence or absence of a harmful effect is determined from the calculation of the magnitude Ms, and if it is extremely far away, a determination result that there is no harmful effect is sent in the judgment of the harmful effect. If it is determined that there is no culpability based on the above judgment,
The end of the earthquake is identified, and if it has ended, the process returns to the earthquake monitoring process described above. In addition, if the identification result shows that the earthquake has not ended, the maximum amplitude value LM of the seismic wave up to that point
and a predetermined constant value Lc (for example, a value of about 40 saddles 1), and if the maximum amplitude value LM exceeds the constant value LC, a process is performed to send a stop command in consideration of the occurrence of damage. If it does not exceed the predetermined value Lc, this process continues until the end of the earthquake.
The control processing device 35 sends an exposure stop command to the corresponding substation 2 as described above.

That is, information specifying the substation to which this silkworm stoppage command applies is added and sent to the communication line 6 by the transmitting device 36. At the substation where the receiving device 23 receives this information, the reception control device 26 determines whether the information is addressed to the own substation or the Ike substation. It is transferred to the destination substation via the control transmitter 22 and the remote control circuit 5. At the substation that receives the flea stop command, the control device 24 controls the relay 25 to stop the operation of the flea equipment. As a result, the power supply to the relevant line section will be stopped and the train will stop. After that, the maintenance personnel will confirm the actual damage.
Recovery processes such as repairing the damaged parts will be carried out. The flowchart in FIG. 6 is a diagrammatic representation of this series of processing procedures. Although the embodiments applied to earthquake detection along railway lines have been described above, the present invention should not be limited to this only, and can be applied to nuclear plants, chemical plants, etc.
Although it takes a lot of time and economic burden to resume operation once the operation has been stopped, it goes without saying that it is extremely effective as a disaster prevention measure.

In this way, the present invention uses the period and amplitude of the initial motion of seismic waves to calculate the magnitude and epicenter, which are the basis for determining the degree of damage caused by the earthquake. Therefore, the true degree of damage caused by the earthquake can be determined at an extremely early stage after the earthquake is detected. It is now possible to issue a warning only for possible earthquakes, which eliminates the needless suspension of train, plant, etc. operations, and when applied to the Shinkansen earthquake warning system, it can prevent unnecessary disruption to timetables. When used in an earthquake warning system for a nuclear or chemical plant, the economic effect will be extremely large.

[Brief explanation of drawings]

Fig. 1 is a diagram showing an example of the equipment arrangement of an earthquake detection and warning system according to the present invention, Fig. 2 is a block diagram showing an embodiment of the present invention; Fig. 3 is a block diagram showing an example of a substation;
Figure 5 is a diagram for explaining the discrimination between P waves and S waves.
FIG. 6, which is a diagram showing the correlation between p and Tp, is a flowchart for explaining the operation of the embodiment. 1...Railway track, 2...Substation 3...Observation point,
4...Detection point "5...Remote control line" 6...
...Communication line 7...Transmission line, 21...Remote control receiver, 22...Remote control transmitter, 23...Receiver, 24...Control device, 25...Relay , 26...
...Transmission/reception control device, 30...Observation point device, 31.
...Sensor, 32...Amplifier, 3...Buffer amplifier, 34...Receiving device, 35...Control processing device 3
6... Transmitting device, 37... Clock device 40...
- Detection point device, 41...sensor, 42...transmission device. Connecting Diagram Figure 2 Figure 3 Figure 4 Figure 5 Figure 6

Claims (1)

[Scope of Claims] 1. A system for detecting the occurrence of an earthquake, determining its harmfulness, and issuing a warning only for earthquakes that are truly considered to be harmful, and for all earthquakes that are truly thought to be harmful. In an earthquake early detection warning system, multiple detection points are selected and a detection point device equipped with a sensor and a transmitter is placed at each of them, and each detection point device is connected to a receiver, a control processing device, and a clock device. The ground motion information detected by each sensor is connected to the equipped observation point equipment via a transmission line, and the ground motion information detected by each sensor is concentrated on the observation point equipment. The arrival of seismic waves is constantly monitored using observed values of oscillation and cross-correlation, and when the arrival of seismic waves is detected, the first step is to determine the period and amplitude of the initial motion of the seismic wave. Based on this, the estimated magnitude and the epicenter are calculated. Based on this, the damage area is estimated and a warning is issued to the relevant area for dangerous areas. For other areas, the epicenter will be continuously monitored until the earthquake ends. , an earthquake early detection and warning system that repeatedly estimates the magnitude and damage area and issues a warning as necessary. 2. The earthquake early detection and warning system according to claim 1, wherein the observation point device has a sensor. 3. Claim 2, characterized in that the process of sending out the alarm is performed by the logical sum of the determination of the degree of harm based on the estimation of the damage area and the observation of the maximum amplitude value larger than a predetermined reference value. Earthquake early detection warning system described. 4 Using the energy output, root mean square value, and cross-correlation of seismic waves as observed values for detecting the arrival of seismic waves, these three
4. The earthquake early detection and warning system according to claim 3, wherein the earthquake early detection and warning system determines that an earthquake wave has arrived when a detection result based on a combination of observed values indicates the arrival of an earthquake wave. 5. The earthquake early detection and warning system according to claim 4, wherein each sensor is arranged in an earthquake-prone area or a region closer to the area within a range of 3 to 5 km.