JP6189226B2 - Earthquake information distribution system and noise judgment method - Google Patents

Description

  In the building where the portable information terminal is installed, in the event of vibration, it avoids malfunction due to noise, informs the seismic intensity in case of an earthquake and the degree of damage to the building, and can the use of the building be continued? The present invention relates to an earthquake information distribution system and a noise determination method using a portable information terminal that can determine whether or not.

  Conventionally, for example, as shown in FIG. 10 (A), in the earthquake information system, since all devices are dedicated devices for earthquake observation, the means of cooperation with general-purpose IT devices (such as personal computers) is limited, This is a system in which a flexible system configuration cannot be taken and the cost of equipment increases. Also, as shown in FIG. 10B, one form using a form information terminal or an earthquake damage determination system is known (see Patent Documents 1 and 2).

JP 2012-168008 A JP 2008-39446 A

  However, in the above example, there is no discrimination algorithm for small earthquakes and noise. There is no cooperation with IT equipment in the building. Furthermore, it is difficult for parties outside the building to know the earthquake observation results. Moreover, the acceleration sensor used for a portable terminal is noisy, and in order to avoid a malfunction due to noise, it is necessary to limit the acceleration sensor to only a relatively large earthquake having a seismic intensity of 3 to 4 or more. In the conventional damage level estimation system, the earthquake response analysis of the building in advance is indispensable, the estimation algorithm is complicated, high accuracy is required for the sensor, and automatic estimation is difficult due to the characteristics of the estimation method, There is a problem that an expert's judgment is required and it takes much time. The earthquake information distribution system and noise determination method according to the present invention have been proposed to solve such problems.

The gist for solving the above-mentioned problems of the earthquake information distribution system according to the present invention is to achieve vibration detection means, seismic intensity analysis program for analyzing the acceleration data by vibration and calculating seismic intensity, information distribution means, Comprising at least a portable information terminal with an earthquake sensor, a communication network connected to the portable information terminal, and an information receiving / display device connected to the communication network,
The portable information terminal calculates a correlation coefficient value between the original waveform and a waveform obtained by shifting the time of the original waveform by one frame at an appropriate time from vibration acceleration data. And a noise determination means for determining whether the obtained vibration is noise or earthquake.

The acceleration data is acceleration data in two horizontal directions and one vertical direction, and the value of the correlation coefficient in these three directions is obtained by the noise determination means.
Also, in the acceleration data in three directions, two horizontal directions and one vertical direction, if the threshold value is 0.2 and the correlation coefficient value in any one direction is 0.2 or more, the noise determination means It is to determine that the vibration related to the acceleration data is an earthquake.

Furthermore, after the vibration is detected by the vibration detecting means, the seismic intensity calculation is performed by the seismic intensity analysis program while the acceleration data of the vibration is being recorded. After the measurement of the vibration is completed, it is determined whether the vibration is noise or earthquake. Is to do it by means.
In addition, the earthquake damage in which the data for specifying the structure of the target building and the damage information corresponding to the seismic intensity are stored in the storage unit of the portable information terminal only when the noise determination means determines that the vibration is an earthquake. From the database, the damage prediction information of the target building corresponding to the seismic intensity of the earthquake is extracted by the damage analysis means, and the damage prediction information of the target building is distributed together with the seismic intensity information via the communication network by the information distribution means. Is included.

  The gist for solving the above-described problems and achieving the object of the noise determination method according to the present invention is based on the acceleration data of vibration, and the time of the original waveform and the original waveform in three directions of two horizontal directions and one vertical direction. The correlation coefficient value with the waveform shifted by one frame at an appropriate time is obtained, the threshold value is 0.2, and the correlation coefficient value in any one of the three directions is 0.2 or more. If there is, the noise determining means determines that the vibration related to the acceleration data is an earthquake.

  After the vibration is detected by the acceleration sensor, the seismic intensity is calculated by the seismic intensity analysis program during recording of the acceleration data of the vibration, and after the measurement of the vibration is completed, the noise determination means determines whether the vibration is noise or earthquake. Is included.

According to the earthquake information distribution system and the noise determination method of the present invention, a value of a correlation coefficient (autocorrelation coefficient, hereinafter the same) is obtained from acceleration data, and the autocorrelation coefficient value is compared with a threshold value. For example, there is an excellent effect that malfunction due to noise can be avoided while ensuring the sensitivity to detect an earthquake having a seismic intensity of 2 or more.
Further, from the acceleration data of vibration, the value of the autocorrelation coefficient between the original waveform and the waveform obtained by shifting the time of the original waveform by an appropriate time by one frame in three directions of two horizontal directions and one vertical direction is obtained. If the value of the autocorrelation coefficient in any one of the three directions is 0.2 or more, by determining that the vibration related to the acceleration data is an earthquake, the earthquake is increased in only one direction. In consideration of the fact that there is a case of shaking, the determination method is suitable for noise determination.

  Since it uses a mobile information terminal to discriminate noise and estimate the degree of damage and distribute it to the outside, it is highly mobile, has a high distribution function, and can quickly determine the damage situation of the building and make a decision to continue using it. It has an excellent effect that it can be done.

  In addition, since each portable information terminal is provided with an acceleration sensor and is independent, no special setting change is required for changing / adding / removing the sensor installation point, so that it has an excellent effect that maintenance is easy. Is.

It is a system configuration figure showing the minimum example of composition of earthquake information distribution system 1 concerning the present invention. It is a remote management type system block diagram of the earthquake information distribution system 1 of the same invention. It is a data external management type system configuration diagram of the earthquake information distribution system 1 of the present invention. It is a general-purpose system configuration diagram of the earthquake information distribution system 1 of the present invention. It is a correlation diagram which shows the autocorrelation of the sensor output in a noise determination means. It is explanatory drawing which shows that the probability distribution of the auto (cross) correlation coefficient R is a normal distribution. It is a flowchart figure when the vibration of the earthquake information delivery system 1 of the same invention is detected. It is explanatory drawing (A), (B), (C) which shows the example which calculated | required the value of the autocorrelation coefficient R which concerns on three horizontal and vertical directions by the noise determination means about three times of vibration, respectively. FIG. 5 is a detailed view (A) showing the contents stored in the storage unit as an earthquake damage database in the portable information terminal, and an explanatory view (B) showing a method for estimating the degree of damage corresponding to the shaking of the earthquake. It is a schematic block diagram (A) of the example which shows the distribution system of the earthquake information which concerns on a prior art example, (B).

  As shown in FIGS. 1 and 5, the earthquake information distribution system 1 according to the present invention is provided with noise determination means in the portable information terminal 2, and in the process of obtaining building damage prediction information, malfunctions due to the effects of noise are detected. It is something to be excluded.

  As shown in FIG. 1, the portable information terminal 2 includes at least vibration detection means using an acceleration sensor, a seismic intensity analysis program for analyzing the acceleration data by vibration and calculating seismic intensity, and information distribution means. The portable information terminal 2 includes, for example, a PDA, a smartphone, an iPhone, an ipad, a tablet computer terminal, a small notebook PC having a call function, and the like.

  A hub 3 connected to the portable information terminal 2, a personal computer (hereinafter referred to as a PC) 4 connected to the hub 3, a main Internet 5 as a communication network connected to the hub 4, It comprises at least an information receiving / displaying device (portable information terminal, PC, notebook PC, etc.) for the external user 6 connected to the communication network 5.

,
2 to 4, as another embodiment, FIG. 2 shows a remote management system configuration diagram in which a manager 7 and a terminal control server 8 that the manager 7 accesses via the Internet 5 are provided in FIG. FIG. 3 is a data external management type system configuration diagram in which an external user 6 and a PC 4 can access the data sharing / simple diagnosis server 9 via the Internet 5, and FIG. 4 is a general-purpose system configuration in which these combinations are collected. The figure is shown.

  The portable information terminal 2 calculates the value of the correlation coefficient between the original waveform and the waveform obtained by shifting the time of the original waveform by an appropriate time by one frame from the vibration acceleration data. And a noise determination means for determining whether the obtained vibration is a noise or an earthquake.

  In this application, the correlation coefficient R between the original waveform and the waveform obtained by shifting the time of the original waveform by an appropriate one frame from the acceleration data of the vibration is referred to as an autocorrelation coefficient R in this application. This is because the correlation coefficient is calculated based on the same acceleration data. As a concept of this autocorrelation coefficient R, as shown in FIG. 5, for example, in the case of an output example of a seismic wave, the correlation with a waveform shifted by a time interval dt (for one frame: 1/50 to 1/200 second) As shown in the autocorrelation diagram, since there is almost no change in the data, a high correlation is shown. On the other hand, if the noise is random, the correlation is close to 0 (zero).

As the value of the autocorrelation coefficient R, two vectors (data array) composed of n samples are placed as follows.
An = (a1, a2,..., An), Bn = (b1, b2,..., Bn)
The cross correlation coefficient of these two vectors is
(Formula 1)
Defined by

When the cross-correlation coefficient is the autocorrelation coefficient R, it is considered that the number of observation data vectors is innumerable, and about 1/50 to 1/200 second as an appropriate time for each data. Thus, the distribution of the set R (R ¹ _n ,... R ⁱ _n ,...) Cross-correlated with the data shifted by one frame is a normal distribution according to the mean μ _r and the variance σ ² _r by the central limit theorem.

Here, the average μ _r is
(Formula 2)
The variance σ ² _r is
(Formula 3)
The probability distribution of R _n is a normal distribution shown in FIG.

From the autocorrelation coefficient R, the noise and the earthquake are judged. For this purpose, it is necessary to determine a threshold value of the autocorrelation coefficient R that is determined to be noise. Therefore, the following is performed.
When an allowable lower limit of S / N is set as SN ₀ , an expected value of an autocorrelation coefficient R _n that is a cross-correlation coefficient with respect to the lower limit is μ _r0 = SN ₀ / (1 + SN ₀ ).

At this time, a reference in which the autocorrelation coefficient R, which is a cross-correlation coefficient, exceeds a certain value R is expressed by the following expression.
(Formula 4)
Here, N (μ, σ ² ): normal distribution with mean μ and variance σ ² .

Therefore, when the defect rate (probability that data having an S / N value lower than the lower limit value is not excluded due to stochastic variation) is to be kept below Pcr, Rcr serving as a determination criterion is expressed by the following equation: expressed.
(Formula 5)
Note that these notations are respectively min {A};
{X | conditional expression B (x)}: represents a set of all x satisfying conditional expression B (x).

Therefore, when Rcr is obtained, condition 1: First, “S / N ratio is 1 or more in a time zone of 1/10 or more of all time zones in which data is recorded” with respect to actual vibration data. The vectors that do not satisfy the condition are excluded. This is because even if the signal is too weak or the duration is too short, it is not an earthquake even if there is autocorrelation. there,
SN ₀ = 1 × 1/10 + 0 × 9/10 = 0.1,
μ _r0 = SN ₀ / (1 + SN ₀ ) = 0.1 / (1 + 0.1) = 0.091

Furthermore, condition 2: Since the number of data is at least 3,000 (60 seconds at 50 Hz),
Variance σ ² = (0.091) (1-0.091 ² ) / 3000 (0.1)
= 3.0 × 10 ⁻⁴ ,

Condition 3: Moreover, Pcr is Six Sigma used in quality control etc. (Definition: σ meaning standard deviation in statistics. Product failure following a normal distribution with a certain quality characteristic value (mean value μ, standard deviation σ) In the state of occurrence, the defect rate Pcr = 3.4 × 10 ⁻⁶ in “reducing the occurrence rate of defective products to 3.4 times even if the work is performed 1 million times”, which is expressed by the above formula. Substituting into 5,
(Formula 6)
It becomes. Therefore, by setting the threshold value = 0.2, if the autocorrelation coefficient R is 0.2 or more, noise is excluded, and it can be determined that the vibration is an earthquake.

  Therefore, the acceleration data of the vibration is acceleration data in two horizontal directions and one vertical direction, and the value of the autocorrelation coefficient R is obtained by the noise determination means for each of these three directions.

  Then, in the acceleration data in the three directions of two horizontal directions and one vertical direction, if the threshold value is 0.2 and the value of the autocorrelation coefficient R in any one direction is 0.2 or more, noise determination is performed. The means determines that the vibration related to the acceleration data is an earthquake. This is the basis for setting the threshold value to 0.2 in the present specification and claims.

  In the earthquake information distribution system 1 according to the present invention, a processing procedure when vibration occurs is shown. As shown in FIG. 7, in the portable information terminal 2, vibration detection means, seismic intensity analysis program, noise determination means, information distribution Initial setting necessary for starting up and starting up the means is performed (step 1: hereinafter, abbreviated as ST1).

  Then, after waiting in ST1, vibration is detected by the vibration detecting means (ST2) in the portable information terminal 2, and acceleration data of the vibration is recorded by the CPU (ST3). The seismic intensity calculation is performed with the seismic intensity analysis program installed in (ST4).

  The measurement of the vibration is completed (ST5). Thereafter, it is determined whether the vibration is noise or earthquake by a determination program as noise determination means (ST6). If autocorrelation coefficient R is less than 0.2, the vibration is determined to be noise, and the process returns to the standby state (ST1). On the other hand, the correlation coefficient value between the original waveform and the waveform obtained by shifting the time of the original waveform by an appropriate time by one frame is obtained from the vibration acceleration data in three directions of two horizontal directions and one vertical direction. Is 0.2, and if the value of the correlation coefficient in any one of the three directions is 0.2 or more, the noise determination means determines that the vibration related to the acceleration data is an earthquake.

  As a specific example of the determination of vibration, as shown in FIG. 8A, the upper two of the data are the horizontal direction, the lower is the vertical direction, and the values of the autocorrelation coefficient R are both 0. It is not 2 or more. Therefore, this vibration is noise. The data shown in FIG. 8B exceeds the threshold value 0.2 because the values of the upper two horizontal autocorrelation coefficients R are 0.689 and 0.492, and this vibration is an earthquake. In the data shown in FIG. 8C, only the upper one has an autocorrelation coefficient R of 0.223 and exceeds the threshold value 0.2, and the other is less than 0.2. Since the value of the autocorrelation coefficient R in any one direction should be equal to or greater than the threshold value, this vibration is determined to be an earthquake.

  And only when it is determined that the vibration is an earthquake by the noise determination means, the data for specifying the structure of the target building and the damage information corresponding to the seismic intensity are stored in the storage unit of the portable information terminal 2 From the damage database (ST7), damage prediction information of the target building corresponding to the seismic intensity of the earthquake is extracted by a program as damage analysis means (ST8).

  In the earthquake damage database, for example, as shown in FIG. 9A, a building where the portable information terminal 2 is installed is a target building for measuring the degree of damage caused by an earthquake. Regarding the target building, as the building conditions, data such as the structure, the number of floors, and the age of the building are stored in the storage unit. There are four types of structures: reinforced steel concrete (SRC), reinforced concrete (RC), wooden, and steel (S).

  Further, as shown in FIG. 9B, the damage caused by the seismic intensity of the past earthquake (seismic intensity of about 2 to 6) of the target building is stored as data in the storage unit. The damage analysis means extracts from the database the degree of damage that matches the condition of the target building corresponding to the earthquake.

  Next, as shown in FIG. 7, in ST9, the information distribution means of the portable information terminal 2 sends the damage prediction information of the target building to the information distribution destination through the communication network 5 together with the seismic intensity information. Distribution is made to a terminal in the LAN in the building from an external external server to a certain registrant or by mail distribution (ST10). Thereby, information on the degree of earthquake damage is displayed on each personal computer or portable information terminal.

  As described above, in the event of an earthquake, the damage level of the target building can be immediately determined, so that the continuation of use of the building can be determined.

  According to the earthquake information distribution system and the noise determination method according to the present invention, it is possible to distribute the disaster information using the portable information terminal, particularly in the case where the vibration is small, to eliminate the influence of noise and to cause the disaster information without malfunction. it can.

1 Earthquake information distribution system,
2 portable information terminals,
3 Hub,
4 Personal computer (PC),
5 Communication network (Internet),
6 external users,
7 Administrator,
8 terminal control server,
9 Data sharing / simple diagnosis server.

Claims (7)

A mobile information terminal with an earthquake sensor having at least vibration detection means, a seismic intensity analysis program for analyzing acceleration data by vibration and calculating seismic intensity, and an information distribution means;
A communication network connected to the portable information terminal;
An information reception / display device connected to the communication network;
Comprising at least
The portable information terminal calculates a correlation coefficient value between the original waveform and a waveform obtained by shifting the time of the original waveform by one frame at an appropriate time from vibration acceleration data. And having noise determination means for determining whether the obtained vibration is noise or earthquake.
Earthquake information distribution system characterized by The acceleration data is acceleration data in two horizontal directions and one vertical direction, and the value of the correlation coefficient in these three directions is obtained by the noise determination means.
The earthquake information distribution system according to claim 1. In acceleration data in three directions of two horizontal directions and one vertical direction, if the threshold value is 0.2 and the value of the correlation coefficient in any one direction is 0.2 or more, the noise determination means performs the acceleration. Determining that the vibration associated with the data is an earthquake;
The earthquake information distribution system according to claim 2, wherein: After the vibration is detected by the vibration detection means, the seismic intensity calculation program calculates the seismic intensity while recording the acceleration data of the vibration, and after the measurement of the vibration is completed, the noise determination means determines whether the vibration is noise or earthquake. about,
The earthquake information distribution system according to claim 2 or 3, wherein From the earthquake damage database in which the data identifying the structure of the target building and the damage information corresponding to the seismic intensity are stored in the storage unit of the portable information terminal only when the vibration is determined to be an earthquake by the noise determination means, Extracting the damage prediction information of the target building corresponding to the seismic intensity of the earthquake with the damage analysis means, and distributing the damage prediction information of the target building with the seismic intensity information via the communication network with the information distribution means;
The earthquake information distribution system according to any one of claims 1 to 4, wherein: From the vibration acceleration data, a correlation coefficient value between the original waveform and a waveform obtained by shifting the time of the original waveform by an appropriate time by one frame in three directions of two horizontal directions and one vertical direction is obtained, and the threshold value is set to 0. .2, if the value of the correlation coefficient in any one of the three directions is 0.2 or more, the noise determining means determines that the vibration related to the acceleration data is an earthquake,
A method for determining noise. After the vibration is detected by the acceleration sensor, the seismic intensity calculation is performed by the seismic intensity analysis program during the recording of the vibration acceleration data. After the measurement of the vibration is completed, the noise determination means determines whether the vibration is noise or earthquake. What to do,
The noise determination method according to claim 6.