Earthquake prediction device

Abstract

An earthquake prediction device comprises a predicted value calculation unit 16 that calculates, from a prediction formula below, a predicted value (MMIap) indicating a predicted intensity of a ground motion on the Modified Mercalli Intensity scale, using a maximum acceleration value (Aumax), which is a maximum absolute value among absolute values of vertical acceleration component of the ground motion, after a sensor starts detecting the ground motion caused by an earthquake. The prediction formula is: MMIap=a log.sub.10(Aumax)+a.

Claims

1. An earthquake prediction device comprising: a sensor configured to detect a ground motion; a vertical acceleration acquisition unit that sequentially acquires vertical acceleration information indicating a vertical acceleration component of the ground motion from the sensor, when the sensor starts detecting the ground motion; a predicted value calculation unit that calculates a predicted value (MMIap) indicating an intensity of earthquake shaking by an index value on the Modified Mercalli Intensity scale, using a maximum absolute value among absolute values of a vertical acceleration component of the ground motion indicated by the vertical acceleration information sequentially acquired by the vertical acceleration acquisition unit as a maximum acceleration value (Aumax), using a prediction formula below: the prediction formula: MMIap=a log.sub.10(Aumax)+a+a, wherein a and a are regression coefficients calculated in advance by regression analysis, and a is an adjustment factor adjustable between 1 and 1; a turn-style adjustment knob configured to be adjusted by a manual operation by a user; and an adjustment factor setting unit that sets the adjustment factor (a) in the prediction formula, based on the manual operation of the adjustment knob, in which the regression coefficients a and a are calculated in advance.

2. The earthquake prediction device according to claim 1, further comprising a warning unit that compares the predicted value (MMIap) calculated by the predicted value calculation unit and a warning reference value set in advance with each other, and issues a warning when the predicted value (MMIap) is larger than the warning reference value.

3. The earthquake prediction device according to claim 2, further comprising an earthquake occurrence determination unit that determines occurrence of an earthquake by presence/absence of the ground motion, wherein the warning unit issues the warning when the earthquake occurrence determination unit determines that the earthquake is occurring.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a block diagram showing, with blocks, various functions of an earthquake prediction device of a first embodiment.

(2) FIG. 2 is an exponential graph, with the horizontal axis denoting an acceleration (unit: gal) and with the vertical axis denoting an index value on the MMI scale, on which a maximum absolute value among absolute values of vertical acceleration component of a ground motion of each earthquake that occurred in the past is plotted as an abscissa and an index value indicating each intensity of earthquake shaking on the MMI scale is plotted as an ordinate.

(3) FIG. 3 is a table in which earthquakes that occurred in the past are divided on the basis of whether a calculated value (MMIa) and a predicted value (MMIap) indicating each intensity of earthquake shaking are each at level 5.5, and the divided numbers of the earthquakes are indicated.

(4) FIG. 4 is a bar graph in which earthquakes that occurred in the past indicating level 5.5 or larger both in the calculated value (MMIa) and the predicted value (MMIap) are divided on the basis of a difference between a timing when the predicted value (MMIap) reached level 5.5 and a timing when the calculated value (MMIa) reached level 5.5, and the divided numbers of the earthquakes are indicated.

(5) FIG. 5 is a flowchart of an earthquake warning process executed by the earthquake prediction device of the first embodiment.

(6) FIG. 6 is a block diagram showing, with blocks, various functions of an earthquake prediction device of a second embodiment.

(7) FIG. 7 is a graph showing a state in which a warning success rate and a cry-wolf false warning rate change when an adjustment factor (a) is adjusted.

(8) FIG. 8 is a block diagram showing, with blocks, various functions of an earthquake prediction device of another embodiment.

(9) FIG. 9 is a flowchart of an earthquake warning process executed by the earthquake prediction device of the another embodiment.

EXPLANATION OF REFERENCE NUMERALS

(10) 1 . . . earthquake prediction device, 3 . . . acceleration sensor device, 5 . . . external warning device, 10 . . . acceleration acquisition unit, 14 . . . acceleration recording unit, 16 . . . predicted value calculation unit, 18 . . . first warning unit, 20 . . . earthquake occurrence determination unit, 20a . . . flag storage region, 22 . . . adjustment factor setting unit, 24 . . . general earthquake determination unit, 26 . . . second warning unit, 30 . . . vertical acceleration sensor, 32 . . . east-west acceleration sensor, 34 . . . north-south acceleration sensor

MODE FOR CARRYING OUT THE INVENTION

(11) Embodiments of the present invention will be explained below with reference to the drawings.

First Embodiment

(12) 1. Earthquake Prediction Device 1

(13) An earthquake prediction device 1 of a first embodiment will be explained with reference to FIG. 1. It is to be noted that the first embodiment will be referred to as the present embodiment in the sections below in which the first embodiment is explained.

(14) The earthquake prediction device 1 of the present embodiment is a computer device including a CPU, a ROM 1a, a RAM, and so on. The CPU and the RAM are not illustrated in FIG. 1.

(15) Connected to the earthquake prediction device 1 are an acceleration sensor device 3 and an external warning device 5.

(16) Among these, the acceleration sensor device 3 comprises three acceleration sensors (a vertical acceleration sensor 30, an east-west acceleration sensor 32, and a north-south acceleration sensor 34) that detect a ground motion as acceleration components in three directions (vertical, east-west, and north-south) orthogonal to each other.

(17) In the present embodiment, observation points are set up in a dispersed manner at areas where occurrence of an earthquake is expected, and the earthquake prediction device 1 and the acceleration sensor device 3 are placed at each of the observation points.

(18) When a seismic wave arrives at the observation points, the respective sensors 30 to 34 each start detecting the corresponding acceleration component of the ground motion at each observation point, and the acceleration sensor device 3 starts outputting analog signals indicating the respective acceleration components.

(19) The external warning device 5 is placed in a place apart from the respective observation points, and is connected to a plurality of the earthquake prediction devices 1 placed at the respective observation points, so as to be able to communicate with them via public lines.

(20) Upon receipt of a warning signal from any of the earthquake prediction devices 1, the external warning device 5 performs a warning action such as output of a warning sound and display of warning information.

(21) Further, if the external warning device 5 works in conjunction with, for example, a train control device, upon receipt of the warning signal, the external warning device 5 performs a warning action to instruct the train control device to stop trains.

(22) As shown in FIG. 1, the earthquake prediction device 1 comprises an acceleration acquisition unit 10, an acceleration recording unit 14, a predicted value calculation unit 16, a first warning unit 18, and an earthquake occurrence determination unit 20.

(23) Functions of these respective units 10 to 20 are fulfilled by the earthquake prediction device 1's execution of an earthquake warning process A stored in the ROM 1a, which will be described later.

(24) The acceleration acquisition unit 10 sequentially inputs the analog signals indicating the acceleration components in the three directions (vertical, east-west, and north-south) outputted when the respective sensors 30 to 34 in the acceleration sensor device 3 detect the ground motion, and samples these analog signals in each sampling period set in advance.

(25) Then, the acceleration acquisition unit 10 sequentially outputs digital signal generated by sampling the analog signal indicating the vertical acceleration component of the ground motion to the acceleration recording unit 14 and the earthquake occurrence determination unit 20.

(26) Further, the acceleration acquisition unit 10 sequentially outputs digital signals generated by sampling the analog signals indicating the acceleration component in the east-west direction and the acceleration component in the north-south direction to the earthquake occurrence determination unit 20.

(27) In the present embodiment, the sampling period is set at 100 Hz, but is not limited to this. (A configuration may be adopted in which the acceleration acquisition unit 10 is arranged in the acceleration sensor device 3 and the digital signals are transmitted from the acceleration sensor device 3 to the earthquake prediction device 1.)

(28) Each time the acceleration recording unit 14 inputs the digital signal from the acceleration acquisition unit 10, the acceleration recording unit 14 executes a process for sequentially storing information indicated by the digital signal.

(29) The information indicated by the digital signal is vertical acceleration information indicating the vertical acceleration component of the ground motion inputted as the analog signal from the acceleration sensor device 3.

(30) Each time the digital signal is outputted from the acceleration acquisition unit 10, the predicted value calculation unit 16 sequentially calculates, on the basis of a prediction formula, which will be described later, a predicted value (MMIap) indicating the intensity of earthquake shaking on the MMI scale, using a maximum acceleration value (Aumax), which is a maximum absolute value among absolute values of the vertical acceleration component in the vertical acceleration information recorded in the acceleration recording unit 14.

(31) When the earthquake occurrence determination unit 20 determines that an earthquake is occurring, the first warning unit 18 outputs the warning signal to the external warning device 5 if the predicted value (MMIap) calculated by the predicted value calculation unit 16 is determined to be larger than a warning reference value (level 5.5 on the MMI scale) set in advance.

(32) The earthquake occurrence determination unit 20 comprises a flag storage region 20a in which flag information used in the earthquake warning process A (see FIG. 6) to be described later is stored. The flag information indicates whether a ground motion has been detected at the observation point, i.e., whether an earthquake is occurring now.

(33) Each time the earthquake occurrence determination unit 20 inputs the digital signals indicating the acceleration components of the ground motion in the orthogonal three directions from the acceleration acquisition unit 10 in each sampling period, the earthquake occurrence determination unit 20 calculates the absolute value of the acceleration obtained by vector-synthesizing these acceleration components in the three directions.

(34) Then, if the absolute value of this acceleration is larger than an earthquake occurrence reference value set in advance in order to determine whether an earthquake is occurring, the earthquake occurrence determination unit 20 executes a process for setting the flag information stored in the flag storage region 20a at 1.

(35) In contrast, if the absolute value of this acceleration is equal to or smaller than the earthquake occurrence reference value, the earthquake occurrence determination unit 20 executes a process for setting the flag information stored in the flag storage region 20a at 0.

(36) Then, the earthquake occurrence determination unit 20 outputs the flag information stored in the flag storage region 20a to the first warning unit 18.

(37) 2. Regarding Method of Calculating MMIap

(38) Next, an explanation will be given of the prediction formula below used in the present embodiment.

(39) The prediction formula: MMIap=a log.sub.10(Aumax)+a

(40) This prediction formula is used to obtain the predicted value (MMIap) indicating the intensity of earthquake shaking with an index value on the Modified Mercalli Intensity scale.

(41) Aumax is the maximum absolute value among the absolute values of the vertical acceleration component of the ground motion stored in the acceleration recording unit 14.

(42) As described above, when the vertical acceleration sensor 30 starts detecting the ground motion, the sampled vertical acceleration information indicating the vertical acceleration component of the ground motion is sequentially stored in the acceleration recording unit 14.

(43) When calculating the predicted value (MMIap) using the above prediction formula, the predicted value calculation unit 16 acquires the maximum acceleration value (Aumax) from the acceleration recording unit 14.

(44) On the other hand, a and a are coefficient values calculated in advance using waveform data recorded by the K-NET, which is a database of a seismic observation network operated by the National Research Institute for Earth Science and Disaster Prevention.

(45) Regarding 2323 waveform data recorded by the K-NET at 13 earthquakes that occurred in the past, a maximum absolute value (gal) of a vertical acceleration component and an index value on the MMI scale for each recorded waveform are obtained, and they are plotted on a semilogarithmic graph with the horizontal axis denoting the maximum value and with the vertical axis denoting the index value. As a result, a relationship as shown in FIG. 2 is obtained.

(46) a and a are calculated as regression coefficients by regression analysis using the maximum absolute value of the vertical acceleration component in FIG. 2 as an explanatory variable (X) and using the index value on the MMI scale in FIG. 2 as a dependent variable (Y).

(47) When the regression analysis is performed using the data of the ground motions of the earthquakes recorded by the K-NET, the result is Y=3.56 log.sub.10 X0.07, and thus, a in the above prediction formula is set at 3.56, and a is set at 0.07.

(48) It is to be noted that, in order to indicate each intensity of earthquake shaking with the index value on the MMI scale, a calculation formula proposed in Non-patent Document 1 by Wald et al. is used for calculation of the index value (hereinafter referred to as a calculated value (MMIa)).

(49) The calculation formula: MMIa= log.sub.10(Amax)+

(50) Here, Amax is an absolute value of a maximum velocity of a ground motion.

(51) Further, is 3.66, and is 1.66.

(52) Time history changes of the predicted value (MMIap) and the calculated value (MMIa) with respect to time have been simulated on the basis of the seismic waveform data recorded by the K-NET, using the above prediction formula and the calculation formula, and the simulation results have been compared with one another. The comparison results will be explained next.

(53) As shown in FIG. 3, among the 2323 cases of the seismic waveform data that have been considered here, the number of cases in which both the predicted value (MMIap) and the calculated value (MMIa) indicate level 5.5 or larger by the index value on the MMI scale is 450.

(54) Among these, the number of cases in which the predicted value (MMIap) reached level 5.5 by the index value on the MMI scale earlier than the calculated value (MMIa) in the above simulation is 384, and in contrast, the number of cases in which the calculated value (MMIa) reached level 5.5 earlier is 66.

(55) Upon further consideration of the seismic waveform data of the above 384 cases, as shown in FIG. 4, it has been found that the number of the seismic waveform data in which the predicted value (MMIap) reached level 5.5 by the index value on the MMI scale earlier than the calculated value (MMIa) by a range not less than 0 second and less than 2 seconds is 126, and the like.

(56) It has also been found that, on average, the predicted value (MMIap) reached level 5.5 on the Modified Mercalli Intensity scale about 5.4 seconds earlier than the calculated value (MMIa).

(57) That is, the earthquake prediction device 1 of the present embodiment can predict early, at the time of initial tremor of the ground motion, whether a destructive shaking that requires warning will arrive, using the MMI scale as a ground motion index.

(58) 3. Earthquake Warning Process

(59) Next, an explanation will be given of the earthquake warning process A executed by the earthquake prediction device 1 of the present embodiment with reference to FIG. 5.

(60) The earthquake warning process A of the present embodiment is started when a not-shown power switch of the earthquake prediction device 1 is turned on, and is subsequently executed repeatedly until the power switch is turned off in each sampling period.

(61) In the earthquake warning process A, an acceleration acquisition process of S10 is first executed.

(62) In this S10, the acceleration acquisition unit 10 executes a process for inputting the analog signals indicating the acceleration components of the ground motion in the three directions (east-west, north-south, vertical) inputted from the acceleration sensor device 3 and for sampling the inputted analog signals.

(63) Then, in this S10, a process is executed in which the digital signal indicating the sampled vertical acceleration component of the ground motion is outputted to the acceleration recording unit 14 and the earthquake occurrence determination unit 20, and the digital signals indicating the acceleration component in the east-west direction and the acceleration component in the north-south direction are outputted to the earthquake occurrence determination unit 20. The acceleration recording unit 14 records the vertical acceleration information regarding the vertical acceleration component of the ground motion indicated by the digital signal.

(64) Next, an MMIap calculation process of S12 is executed.

(65) In this S12, the predicted value calculation unit 16 executes a process for calculating the predicted value (MMIap) using the maximum acceleration value (Aumax), which is the maximum absolute value among the absolute values of the vertical acceleration component in the vertical acceleration information recorded by the acceleration recording unit 14.

(66) Next, in S14, the earthquake occurrence determination unit 20 executes a process for calculating an acceleration of the ground motion at the observation point from the acceleration components of the ground motion in the three directions converted into the digital signals by the acceleration acquisition unit 10.

(67) Next, a process of S16 is executed.

(68) In this S16, a process is executed in which whether an earthquake is occurring is determined.

(69) In this S16, the first warning unit 18 executes a process for, specifically, determining whether the flag stored in the flag storage region 20a is 1 indicating that an earthquake is occurring or 0 indicating a normal state in which no earthquake is occurring.

(70) In this S16, if the flag is determined to be 0, i.e., normal state, (S16: YES), a process of S18 is executed next. If the flag is determined to be 1, i.e., earthquake is occurring (S16: NO), a process of S22 is executed next.

(71) In S18, a process for determining whether the absolute value of the acceleration of the ground motion at the observation point is larger than the above-described earthquake occurrence reference value is executed.

(72) This S18 is executed by the earthquake occurrence determination unit 20.

(73) In this S18, if the absolute value of the acceleration of the ground motion is larger than the earthquake occurrence reference value, i.e., if an earthquake is occurring (S18: YES), a process for changing the flag stored in the flag storage region 20a from 0 to 1 is executed (S20). Then, the present earthquake warning process A ends, and the processes of and after S10 are executed again.

(74) In contrast, if the absolute value of the acceleration of the ground motion is smaller than the earthquake occurrence reference value, i.e., if no earthquake is occurring (S18: NO), the present earthquake warning process A ends immediately, and the processes of and after S10 are executed again.

(75) Next, an explanation will be given of a process of S22 executed when it is determined in S16 that the flag is 1, i.e., earthquake is occurring (S16: NO).

(76) In this S22, the first warning unit 18 executes a process for determining whether the predicted value (MMIap) calculated in S12 is equal to or larger than the warning reference value, which is a standard for warning, i.e., whether the predicted value (MMIap) is at level 5.5 or larger on the MMI scale.

(77) If it is determined in this S22 that the predicted value (MMIap) is larger than the warning reference value, it is found that, as described above, it is 5.4 seconds on average before occurrence of the shaking of level 5.5 or larger on the MMI scale at the observation point where the earthquake prediction device 1 is placed.

(78) Thus, if it is determined in S22 that the predicted value (MMIap) is at level 5.5 or larger on the MMI scale (S22: YES), a process of S24 is executed next, where a process for transmitting the warning signal from the first warning unit 18 to the external warning device 5 is executed. Then, after this S24, a process of S27 is executed.

(79) In contrast, if it is determined in S22 that the predicted value (MMIap) is at level smaller than 5.5 on the MMI scale (S22: NO), a process of S27 is executed next.

(80) In S27, contrary to S18, a process for determining whether the value of the acceleration of the ground motion at the observation point is smaller than the earthquake occurrence reference value set in advance is executed.

(81) This S27 is executed by the earthquake occurrence determination unit 20. In this S27, similarly to S18, a process for determining whether the absolute value of the acceleration of the ground motion at the observation point is equal to or smaller than the above-described earthquake occurrence reference value is executed.

(82) In this S27, if the absolute value of the acceleration of the ground motion is equal to or smaller than the reference value (S27: YES), a process for changing the flag stored in the flag storage region 20a from 1 to 0 is executed (S28). Then, the present earthquake warning process A ends, and the processes of and after S10 are executed again.

(83) In contrast, if the value of the acceleration of the ground motion is larger than the reference value (S27: NO), the present earthquake warning process A ends immediately, and the processes of and after S10 are executed again.

(84) 4. Characteristic Effects of Earthquake Prediction Device of Present Embodiment

(85) As described above, when the predicted value (MMIap) of the ground motion of the earthquake that occurred in the past and the calculated value (MMIa) are compared with each other, it has been found that the predicted value (MMIap) reached the warning reference value 5.4 seconds earlier than the calculated value (MMIa) at the time of initial tremor of the ground motion.

(86) Consequently, with the earthquake prediction device 1 of the present embodiment, it is possible to predict occurrence of an earthquake that requires warning early at the time of initial tremor of the ground motion using the MMI scale as the ground motion index.

(87) Such an early prediction of occurrence of an earthquake that requires warning enables a braking to be performed 5.4 seconds prior to a severe shaking in a case of a rapid-transit railway with a travel speed of 270 km/h, for example. Thus, with the earthquake prediction device 1 of the present embodiment, the train can be stopped 375 m shorter than in the case where the braking is performed when the severe shaking occurs.

(88) Besides, when an earthquake that requires warning is predicted by the earthquake prediction device 1 of the present embodiment, actions may be taken, such as stopping an elevator at the nearest floor, stopping a machine vulnerable to shaking, delivering a message to people via a speaker or TV.

(89) Moreover, in the earthquake prediction device 1 of the present embodiment, occurrence of an earthquake that requires warning is predicted early using the MMI scale, and thus, prediction of an earthquake that is easy to understand globally is possible.

(90) Furthermore, in the earthquake prediction device 1 of the present embodiment, a warning is issued only when the predicted value (MMIap) is larger than the earthquake occurrence reference value set in advance (S22.fwdarw.S24), and thus, a useless warning issued when an earthquake that does not require a warning is occurring can be inhibited.

Second Embodiment

(91) Next, a second embodiment of the present invention will be explained.

(92) In the present embodiment, only differences from the first embodiment will be explained. It is to be noted that the second embodiment is referred to as the present embodiment in the sections below in which the second embodiment is explained.

(93) 1. Earthquake Prediction Device 1

(94) As shown in FIG. 6, the earthquake prediction device 1 of the present embodiment is different from the earthquake prediction device 1 of the first embodiment in that an adjustment factor setting unit 22 is provided.

(95) In addition, the present embodiment is different from the first embodiment in that an adjustment value a is added to the prediction formula for calculation of the predicted value (MMIap) to be used by the predicted value calculation unit 16.

(96) The prediction formula: MMIap=a log.sub.10(Aumax)+a+a

(97) In the present embodiment, a can be adjusted between 1 and 1. As the adjustment factor setting unit 22, a turn-style adjustment knob is used, for example, with which a value of a can be adjusted by a manual operation, i.e., by varying a turning amount, etc., of the knob.

(98) The predicted value calculation unit 16 calculates the predicted value (MMIap) using a value set as the adjustment value a set by the adjustment factor setting unit 22 and using the prediction formula including such a.

(99) It is to be noted that the predicted value (MMIap) is calculated using the above prediction formula including the above-described a in S22 too, in the earthquake warning process A executed by the earthquake prediction device 1 of the present embodiment.

(100) 2. Regarding Adjustment Value a

(101) Next, a warning success rate and a cry-wolf false warning rate will be explained with reference to FIG. 7.

(102) The warning success rate and the cry-wolf false warning rate are calculated using the data of the ground motions of the earthquakes recorded by the K-NET.

(103) The warning success rate is a rate of the earthquakes whose predicted value (MMIap) is 5.5 or larger among all the earthquakes whose calculated value (MMIa) is 5.5 or larger.

(104) The cry-wolf false warning rate is a rate of the earthquakes whose calculated value (MMIa) is smaller than 5.5 among all the earthquakes whose predicted value (MMIap) is 5.5 or larger.

(105) As shown in FIG. 7, the warning success rate is higher as a is closer to 1, and is approximately 100% when a is set at 1. In contrast, the warning success rate is lower as a is closer to 1, and is approximately 40% when a is set at 1.

(106) On the other hand, the cry-wolf false warning rate is lower as a is closer to 1, and is approximately 0% when a is set at 1. In contrast, the cry-wolf false warning rate is higher as a is close to 1, and is approximately 40% when a is set at 1.

(107) 3. Characteristic Effects of Earthquake Prediction Device of Present Embodiment

(108) The earthquake prediction device 1 of the present embodiment produces effects below in addition to the effects produced by the earthquake prediction device 1 of the first embodiment.

(109) When occurrence of an earthquake is predicted early and a warning thereabout is issued using the earthquake prediction device 1 of the present embodiment, following two demands from users are expected, for example.

(110) One demand expected is that a warning be issued in every case when occurrence of an earthquake that requires caution is predicted even if there are some cases in which the prediction turns out to be incorrect, regardless of whether the earthquake that requires caution is actually occurring. In this case, increase in the warning success rate is desired.

(111) The other demand expected is that a warning not be issued when an earthquake that requires caution is not occurring even if there are some cases in which no warning is issued when an earthquake that requires caution is occurring. In this case, decrease in the cry-wolf false warning rate is desired.

(112) Thus, in the earthquake prediction device 1 of the present embodiment, the predicted value (MMIap) to be calculated is adjusted by adding a in the prediction formula, and the above two demands can thereby be met.

(113) For example, when the warning reference value is set at level 5.5 on the MMI scale and a is set at 1, the cry-wolf false warning rate is closer to 0%, and in contrast, when a is set at 1, the warning success rate is closer to 100%, as shown in FIG. 7.

(114) That is, when a is set at 1, if occurrence of an earthquake that requires caution is predicted, a warning is issued without exception regardless of whether the earthquake that requires caution is actually occurring.

(115) On the other hand, when a is set at 1, there are some cases in which no warning is issued when an earthquake that requires caution is occurring, whereas no case arises in which a warning is issued when an earthquake that requires caution is not occurring.

(116) Consequently, the earthquake prediction device 1 of the present embodiment enables a prediction that meets the users' demands.

Correspondence Relationships

(117) The process executed by the acceleration acquisition unit 10 in the process of S10 in the above embodiment corresponds to one example of a vertical acceleration acquisition unit set forth in the claims.

(118) The process executed by the predicted value calculation unit 16 in the process of S14 in the above embodiment corresponds to one example of a predicted value calculation unit set forth in the claims.

(119) The processes in which the first warning unit 18 transmits the warning signal to the external warning device 5 in the processes of S22 to S24 in the above embodiment correspond to one example of a process in which a warning unit issues a warning, which is set forth in the claims.

Other Embodiments

(120) In the above embodiments, the acceleration sensor device 3 has been explained as being a device separate from the earthquake prediction device 1. However, the acceleration sensor device 3 may be incorporated into the earthquake prediction device 1.

(121) In the above embodiments, the external warning device 5 has been explained as a device that can communicate with the earthquake prediction device 1 via public lines. However, the external warning device 5 may be a warning device that is provided to the earthquake prediction device 1 and emits a warning sound.

(122) Further, as shown in FIG. 8, the earthquake prediction device 1 may be designed to comprise a general earthquake determination unit 24 that determines an earthquake by a conventional determination method and a second warning unit 26 that issues a warning based on such a determination.

(123) In this case, the second warning unit 26 executes a process for issuing a warning to the external warning device 5 when the general earthquake determination unit 24 determines that an earthquake is occurring.

(124) Thus, in the earthquake prediction device 1 of the present embodiment, the external warning device 5 issues a warning when occurrence of an earthquake is determined by either the first warning unit 18 or the second warning unit 26.

(125) In this case, the adjustment factor setting unit 22 may be or may not be provided.

(126) When the general earthquake determination unit 24 and the second warning unit 26 are provided, processes of S25 and S26 may be executed between the processes of S24 and S27, as shown in FIG. 9.

(127) In this case, in S25, it is determined whether an earthquake is occurring by a conventional method. If it is determined that an earthquake is occurring (S25: YES), a process for issuing a second warning, which is different from the early warning of the above embodiment, is executed in S26.

(128) The functions 10 to 260 of the respective units constituting the earthquake prediction device 1 of the present embodiment can be fulfilled by a computer to which the acceleration sensor device 3 and the external warning device 5 are connected by using a program stored in the ROM 1a. This program may be used by being loaded to the computer from the ROM 1a or a backup RAM, or may be used by being loaded to the computer via a network.

(129) Alternatively, this program may be used by being recorded on a recording medium of any forms readable by the computer. Such a recording medium includes, for example, a portable semiconductor memory (e.g., a USB memory, a memory card (registered trademark), etc.).

(130) The present invention is only required to be consistent with the gist of the invention set forth in the claims, and is not limited to the above embodiments.