JP2002213700A - Earthquake damage predicting receiver for pipe line - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an earthquake damage predicting device for a pipe line, capable of predicting a damage occurring place of a pipe line due to earthquake. SOLUTION: Data is entered from a damage predicting regulation file 11, a damage-predicting object pipe line data file 12 and a seismic intensity data input function 13 into a damage-predicting function 14. The damage-predicting function 14 predicts damaged parts of the pipe line, when an earthquake occurs, on the basis of these data. A predicted result is sent to a damage predicting result output function 15 and indicated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a pipeline seismic damage predicting apparatus for predicting a damaged portion of a pipeline such as water or gas due to an earthquake.

[0002]

2. Description of the Related Art When damage to pipelines due to an earthquake occurs, it is difficult to specify the damage location. For example, in the case of a water pipe, first, water is supplied to a pipe in a certain section, pressure is applied to search for a leak point, and when the leak point is found, the leak is repaired.

[0003] After all the leak points in the section are repaired, the section is expanded and the same is repeated. Since this is repeated until all the pipelines are restored, it takes time to restore the pipelines.

[0004]

It is necessary to quickly restore pipelines such as water and gas, which are lifelines. However, in the conventional method, it is troublesome to find a damaged part, and it takes a long time to restore the pipeline. This is the actual situation.

The present invention has been made in view of the above points, and an object of the present invention is to provide a pipeline earthquake damage prediction device capable of promptly detecting a damage point caused by an earthquake and performing a quick recovery operation. With the goal.

[0006]

SUMMARY OF THE INVENTION The present invention provides a damage prediction rule file for predicting damage caused by an earthquake in a pipeline, a damage prediction target pipeline data file in which the type and diameter of the pipeline are recorded, Function to input seismic intensity data of the earthquake
A damage prediction rule file, a damage prediction target pipe data file, a damage prediction function that reads data from the seismic intensity data input function to predict the damage location, and a damage prediction result output function that displays the prediction results from the damage prediction function And a seismic damage prediction device for pipelines.

The present invention further comprises a damage prediction target geological data file describing surface geology and the like, and the damage prediction function predicts a damage location in consideration of data from the damage prediction target geological data file. This is an earthquake damage prediction device for pipelines.

According to the present invention, the damage prediction rule file has a structure representing a tree structure classification rule, and the damage prediction function has a tree structure represented by the damage prediction rule file as an internal structure. An earthquake damage prediction device for a pipeline characterized by predicting damage by tracing a tree structure from the root according to the content of data for each pipeline of a target pipeline data file.

In the present invention, the damage prediction rule file is an if-
The damage prediction function is expressed by the then-type rules, and the damage prediction function operates as a production system using the damage prediction rule file as a rule base and predicts damage to pipelines.

According to the present invention, the damage prediction function has a neural network that inputs seismic intensity data, geological data, and pipeline data and outputs a damage location and a damage form, and the damage prediction rule file includes nodes and nodes of the neural network. The damage prediction function calculates the weight of each side of the neural network as the value of the damage prediction rule file, and uses the data of the damage prediction target area geological data file, damage prediction rule file, and seismic intensity data file to calculate the weight. An earthquake damage prediction device for a pipeline characterized by predicting damage to a road.

The present invention reads a seismic intensity data file of an earthquake that occurred in the past in the damage prediction target area and data from an earthquake damage data file due to the earthquake into the damage prediction rule file, 2. The pipeline earthquake damage prediction device according to claim 1, further comprising a damage analysis function for extracting a relationship between the two.

The present invention reads a seismic intensity data file of an earthquake that occurred in the past in an area different from the damage prediction target area and data from an earthquake damage data file caused by the earthquake into the damage prediction rule file, and 2. A pipeline earthquake damage prediction device according to claim 1, wherein a damage analysis function for extracting a relation of the earthquake damage is connected.

According to the present invention, the damage analysis function uses a KDD method to generate a decision tree for classifying a damage location and a damage form based on seismic intensity, geology, and pipeline data from data in an input data file, and to construct a structure of the decision tree. An earthquake damage prediction device for a pipeline characterized by outputting to a damage prediction rule file.

According to the present invention, the damage analysis function uses the KDD method to generate a decision tree for classifying a damage location and a damage form based on seismic intensity, geology, and pipeline data from data of an input data file, and to generate a leaf node of the decision tree. Each node of the path from the root node to the parent node of the leaf node is set to AN
If- as the condition part of the D condition and the leaf node as the action part
A pipeline seismic damage prediction device characterized in that the rule is converted into a rule of the then rule format and output to a damage prediction rule file.

According to the present invention, a seismic intensity data file, a seismic damage data file, a geological data file for a damaged area, and data from the pipeline data file for the damaged area are read into a damage prediction rule file, and the seismic intensity data and the pipeline data are read. A seismic damage prediction device for pipelines, wherein a damage analysis function for extracting a relationship between data and seismic damage of pipelines is connected.

According to the present invention, a damage prediction rule file includes a seismic intensity data file of an earthquake that occurred in the past, an earthquake damage data file due to the earthquake, a geological data file of a damaged area, a pipe data file of a damaged area, A seismic ground motion analysis function that reads data from the epicenter data file of an earthquake that occurred in the past and reproduces the seismic ground motion, seismic intensity data file, seismic damage data file, damaged area geological data file, and damaged area pipeline data file A seismic intensity data, pipeline data, and a damage analysis function that extracts the relationship between seismic motion and seismic damage to the pipeline by reading data from the seismic motion analysis function are connected.
This is a pipeline earthquake damage prediction device, which is connected to a source information input function for reading real-time source data and a seismic motion analysis function for reproducing a seismic motion from the information to predict a damage location.

According to the present invention, the damage prediction function is connected to a seismic damage number predicting function for estimating the number of pipeline damages in each area in response to an input from the seismic motion analysis function. A seismic damage prediction device for pipelines, which reads in and outputs the estimated number of damages from those having a high probability of occurrence of damage in the corresponding area.

The present invention is characterized in that the damage prediction function receives data from a simulated seismic intensity data input function of reading a simulated seismic intensity data file and inputting simulated seismic intensity data instead of inputting from the seismic intensity data input function. This is an earthquake damage prediction device for pipelines.

According to the present invention, the damage prediction result output function reads the damage prediction result from the damage prediction function and the data from the map data file and the pipeline drawing data file, and superimposes the map, the pipeline drawing and the damaged portion or independently. A seismic damage prediction apparatus for pipelines, characterized in that a graphic creation function for generating a graphic screen is output and the screen created by the graphic creation function is output.

The present invention is characterized in that the graphic creation function accepts an input from the actual damage / recovery status input function and displays the actual damage location, the restored damage location, and the predicted damage location separately. It is an earthquake damage prediction device for roads.

The present invention is characterized in that the graphic creation function inputs only actual damage data to the actual damage recording function and saves the actual damage as an actual damage data file in the same format as the earthquake damage data file. This is an earthquake damage prediction device for pipelines.

[0022]

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

FIGS. 1 to 3 show a first embodiment of a pipeline earthquake damage prediction apparatus according to the present invention.

As shown in FIG. 1, the apparatus for predicting earthquake damage in a pipeline includes a damage prediction rule file 11 for predicting damage.
A damage prediction target pipe data file 12 that records the type and diameter of the pipe, and an input function (input device) 13 for seismic intensity data of the generated earthquake.

Damage prediction rule file 11, damage prediction target pipeline data file 12, seismic intensity data input function 13
Is sent to the damage prediction function (damage prediction device) 14, where the damage prediction function 14 predicts the damage location. The prediction result predicted by the damage prediction function 14 is displayed from a damage prediction result output function (damage prediction result output device) 15.

Further, as shown in FIG. 1, the earthquake damage prediction device includes an other damage analysis function 24 and an epicenter information input function 41.
Earthquake motion analysis function 42, earthquake motion analysis function 44, earthquake damage number prediction function 51, and simulated seismic intensity data input function 6
1, a simulated seismic intensity data generation function 71, a graphic creation function 83, an actual damage / restoration status input function 91, and an actual damage recording function 1, and these functions are realized as processes on a computer.

The earthquake damage prediction device further includes a damage prediction target geological data 16 and an earthquake damage data file 21.
, Seismic intensity data file 22, damage occurrence area geological data file 23, damage occurrence area pipeline data file 3
1, an epicenter data file 43, a simulated seismic intensity data file 62, a map data file 81, a pipeline drawing data file 82, and an actual damage data file 2,
These data files are realized as files on a computer.

Next, a first operation of the present embodiment having the above-described configuration will be described.

(First Operation) First, signals from the damage prediction rule file 11, the damage prediction target pipe data file 12, the seismic intensity data input function 13, and the damage prediction target area geological data 16 are input to the damage prediction function 14. Then, the prediction result from the damage prediction function 14 is displayed from the damage prediction result output function 15.

Further, the seismic intensity data input function 13 is transmitted to the seismic intensity meter 12 of each place through the seismic intensity meter network 124 during this time.
Signals 1, 122 and 123 are input.

Next, the operation of each file will be described in detail. In the damage prediction rule file 11, if-then damage prediction rules are described (recorded) in one line.

<Example of description in damage prediction rule file 11> if pipe type = DCIPand seismic intensity> = 5and and geology = landfill site then joint damage if pipe type = CIPand seismic intensity> = 5and and geological = landfill site then damage (Omitted hereinafter) The pipe type CIP represents a cast iron pipe, and DCIP represents a ductile cast iron pipe. The first rule in the above description example indicates that in a landfill where the seismic intensity is more than 5 or more, there is a high possibility that the joint of the ductile iron pipe is damaged.

It is assumed that the target area for damage prediction is divided into mesh areas. For example, the area of 5km square is 2 sides
If it is divided into a 50-m mesh area,
Pieces, and 20 pieces horizontally, which can be divided into 400 areas in total. This area can be represented by horizontal (00-19) and vertical (00-19) coordinates as shown in FIG.

In the damage prediction target pipeline data file 12, at least the type of pipe laid in the area is described. A description example is shown below.

<Example of description in damage prediction target pipeline data file 12> Area = A00-00, pipe type = DCIP area = A00-00, pipe type = CIP area = A00-01, pipe type = DCIP area = A00- 02, pipe type = CIP area = A00-02, pipe type = SP (hereinafter abbreviated) The first A in the area coordinates means about “A city”. This is a symbol for convenience when the damage prediction target area is composed of several blocks such as municipalities.

Geological data file 16 for damage prediction target area
Describes the typical geology of the area. A description example is shown below.

<Geological data file 1 for damage prediction target area
Description example in No. 6> Area = A00-00, Geology = Landfill Area = A00-01, Geology = Landfill Area = A00-02, Geology = Alluvium (hereinafter abbreviated) Next, the seismic intensity data input function 13 As described above, the seismic intensity data from the seismic intensity meters 121, 122, and 123 in various places are input in real time through a seismic intensity meter network 124 using a dedicated line or the like. The seismic intensity data input function 13 generates seismic intensity data of each area and inputs the data to the damage prediction function 14.

An example of data input from the seismic intensity data input function 13 to the damage prediction function 14 is shown below.

<Data example of seismic intensity data input function 13> Area = A00-00, seismic intensity = 6 strong Area = A00-01, seismic intensity = 5 strong Area = A00-02, seismic intensity = 5 weak (hereinafter abbreviated) Damage prediction function 14, the damage prediction rule file 1
1 as a rule base, a damage prediction target pipeline data file 12 and a damage prediction target geological data file 16
And the input from the seismic intensity data input function 13, and operates as a production system to predict the damage location.

Next, the damage prediction function 14 calculates the data from the damage prediction rule file 11, the data from the damage prediction target pipeline data file 12, the data from the damage prediction target geological data file 16, and the vibration data. An operation when data from the input function 13 is received as an input will be described.

The damage prediction function 14 first receives geological data and seismic intensity data for each area from the damage prediction target area geological data file 16 and seismic intensity data input function 13, and combines them with the data from the damage prediction target pipeline data file 12. For example, the area A00-00 has a geological landfill and a seismic intensity of 6 or more. Therefore, when combined with the data of the damage prediction target pipeline data file description example, the result is as follows.

Area = A00-00, pipe type = DCIP, geology = landfill, seismic intensity = high 6 area = A00-00, pipe type = CIP, geology = landfill,
Seismic intensity = strong 6 When this operation is performed for all areas, the following data is obtained.

Area = A00-00, pipe type = DCIP, geology = landfill, seismic intensity = high 6 area = A00-00, pipe type = CIP, geology = landfill,
Seismic intensity = high 6 Area = A00-01, Pipe type = DCIP, Geological = landfill, Seismic intensity = High 5 Area = A00-02, Pipe type = CIP, Geological = alluvial,
Seismic intensity = less than 5 Area = A00-02, pipe type = SP, geology = alluvium, seismic intensity = less than 5 (abbreviated below) If rules from damage prediction rule file 11 are applied to this, area = A00-00, pipe type = DCIP, geology = landfill, seismic intensity = above 6 area = A00-01, pipe type = DCIP, geology = landfill, seismic intensity = above 5 The above two are if pipe type = DCIPand seismic intensity> = 5
Since the condition of strong and geology = damage to the landfill joint is satisfied, it is determined that there is a high possibility that damage has occurred to the joint in the DCIP pipelines in areas A00-00 and A00-01.

Similarly, area = A00-00, pipe type = CI
P, geology = landfill, seismic intensity = slightly over 6, if pipe type = CIPa
Since the condition of nd seismic intensity> = more than 5 and geology = the landfill then damage to the fixtures is satisfied, it is determined that there is a high possibility that the fixtures are damaged in the CIP pipeline in the area A00-00.

The damage prediction result output function 15 outputs prediction results from the damage prediction function 14 as follows.

Area = A00-00, Pipe Type = CIP-> Joint Damage Area = A00-00, Pipe Type = DCIP-> Damage Damage Area = A00-01, Pipe Type = DCIP-> Damage Damage (Second In addition to the operation described above, data from the earthquake damage data file 21, the seismic intensity data file 22, and the geological data file 23 where the damage occurred is input to the damage analysis function 24, and the information from the damage analysis function 24 is input. The damage location may be predicted by the damage prediction function 14 in consideration of the above (second operation).

Here, the seismic damage data file 21, the seismic intensity data file 22, and the damage occurrence geological data file 23 contain data relating to the earthquake damage that occurred in the past. It doesn't have to be data.

The seismic intensity data file 22 describes seismic intensity data of a past earthquake. A description example in the case of using the damage data of the past earthquake of City B is shown below. 5k as well as City A
It is assumed that m squares are separated by a 250 m mesh.

<Description example in seismic intensity data file 22> Area = B00-00, seismic intensity = high 6 Area = B00-01, seismic intensity = high 5 Area = B00-02, seismic intensity = low 5 (hereinafter abbreviated) Seismic damage data file 21 describes the damage caused by the past earthquake. Describe at least the following information.

<Description example in the earthquake damage data file 22> Area = B00-00, Pipe type = DCIP, Diameter = 75m
m, installation year = 1980, damage location = accessory area = B00-00, pipe type = DCIP, diameter = 100
mm, installation year = 1980, damage location = accessory area = B00-00, pipe type = CIP, diameter = 75m
m, installation year = 1977, damage occurrence location = accessory area = B01-00, pipe type = DCIP, diameter = 100
mm, installation year = 1981, damage location = joint (hereinafter abbreviated) The damage area geological data file 23 describes the geological data of the area where damage occurred in the past earthquake. The representative geology is described for each area. A description example is shown below.

<Example of description in geological data file 23 of damage occurrence area> Area = B00-00, geology = terrain area = B00-01, geology = alluvial area = B00-02, geology = landfill (abbreviated below) Damage analysis The function 24 includes an earthquake damage data file 21,
Data from the seismic intensity data file 22 and the geological data file 23 where the damage occurred is read. First, the seismic intensity data file 22 and the geological data file 23 where the damage occurred.
Receive seismic intensity data and geological data for each area from
Match with the data from the earthquake damage data file 21.
When the data from the above-mentioned seismic intensity data file 22, the data from the earthquake damage data file 21, and the data from the damage occurrence area geological data file 23 are received,
Obtain the following data.

Area = B00-00, tube type = DCIP,
Diameter = 75mm, installation year = 1980, damage location =
Accessories, seismic intensity = 6 or more, geology = terrace area = B00-00, pipe type = DCIP, caliber = 100
mm, installation year = 1980, damage location = attachment, seismic intensity = 6 strong, geology = terrace area = B00-00, pipe type = CIP, diameter = 75m
m, installation year = 1977, damage location = pipe, seismic intensity =
6 strong, geology = terrace area = B01-00, pipe type = DCIP, caliber = 100
mm, construction year = 1981, damage location = joint, seismic intensity = 5 or more, geology = alluvial deposit (abbreviated below) The damage analysis function 24 analyzes these data and analyzes the data between the damage occurrence location and other items. Extract relationships as rules. This includes KDD (Knowledge Discovery and Data-m
ining). For example, a method of generating a decision tree for classifying the damage occurrence location and expressing this in the form of a rule is adopted. In the above four examples, the damage occurrence locations are three types of joints, fittings, and pipes. There are several possible decision trees for classifying damage occurrence locations based on information on pipe type, diameter, installation year, seismic intensity, and geology, one of which is shown in FIG.

This decision tree indicates that when the pipe type is DCIP and the seismic intensity is a little over 6, there is a high possibility that damage will occur to the accessories. When this decision tree is converted into the form of the if-then rule, the following is obtained.

If pipe type = DCIPand seismic intensity = 6 strong then damage to if tube type if pipe type = DCIPand seismic intensity = 5 strong then body damage if pipe type = CIPthen joint damage Actually, there are many data, so a more complicated decision tree Are generated, and many rules are generated. The damage analysis function 24 performs the above processing, outputs the rule generated in the form of a decision tree to the damage prediction rule file 11, and the damage prediction file 11 performs damage prediction in the event of a future earthquake according to the rule of the decision tree. Use for.

(Third Operation) In addition to the above-described operation, the damage prediction function 14 may predict a damage location in consideration of data from the damage occurrence area pipeline data file 31 (third operation). ).

In the damage occurrence area pipeline data 31, detailed pipeline data of an area where damage has occurred in the past earthquake is described. A description example is shown below.

<Example of description in damage occurrence area pipeline data file 31> Area = B00-00, pipeline section = M0001, pipe type =
DCIP, diameter = 75mm, installation year = 1980, installation direction = 90, total length = 6m area = B00-00, pipeline section = M0002, pipe type =
DCIP, diameter = 75 mm, installation year = 1980, installation direction = 180, total length = 10 m (omitted) area = B00-00, pipeline section = M0103, pipe type =
DCIP, diameter = 100 mm, installation year = 1980, installation direction = 90, total length = 6 m, area = B00-00, pipeline section = M0104, pipe type =
DCIP, diameter = 100 mm, installation year = 1980, installation direction = 180, total length = 10 m (omitted) area = B00-00, pipeline section = M0201, pipe type =
CIP, diameter = 75mm, installation year = 1977, installation direction = 60, total length = 6m area = B00-00, pipeline section = M0202, pipe type =
CIP, diameter = 75 mm, installation year = 1977, installation direction = 180, total length = 6 m (omitted) area = B01-00, pipeline section = M0301, pipe type =
DCIP, diameter = 100mm, installation year = 1981, installation direction = 150, total length = 6m area = B01-00, pipeline section = M0302, pipe type =
DCIP, diameter = 100 mm, laying year = 1981, laying direction = 60, total length = 6 m (hereinafter abbreviated) At this time, the earthquake damage data file 21 describes the pipeline section where the damage occurred and the damage situation. A description example is shown below.

<Example of description in earthquake damage data file 21> Area = B00-00, pipeline section = M0001, damage location = parts area = B00-00, pipeline section = M0103, damage location = parts area = B00-00, pipeline section = M0201, damage location = accessory area = B01-00, pipeline section = M0302, damage location = joint (hereinafter abbreviated) The damage analysis function 24 ,
The data from the seismic intensity data file 22, the geological data file 23 where the damage occurred and the pipeline data file 31 where the damage occurred is read.
And the seismic intensity data and geological data for each area from the geological data file 23 where the damage has occurred, and combine the data with the data from the earthquake damage data file 21. This is further combined with the data from the damaged area pipeline data file 31. When receiving the data from the above-mentioned earthquake damage data file 21, the data from the damage occurrence area pipeline data file 31, the data from the seismic intensity data file 22, and the data from the damage occurrence area geological data file 23, To get the data.

Area = B00-00, pipeline section = M0001, pipe type =
DCIP, caliber = 75mm, laying year = 1980, laying direction = 90, total length = 6m, damage occurrence point = attachment, seismic intensity =
6 strong, Geology = Terrace Area = B00-00, Pipe section = M0002, Pipe type =
DCIP, diameter = 75 mm, installation year = 1980, installation direction = 180, total length = 10 m, damage location = none, seismic intensity = 6 strong, geology = terrace (omitted) area = B00-00, pipeline section = M0103, pipe Seed =
DCIP, diameter = 100 mm, installation year = 1980, installation direction = 90, total length = 6 m, damage location = belongings, seismic intensity = 6 strong, geology = terrace area = B00-00, pipeline section = M0104, pipe type =
DCIP, diameter = 100 mm, installation year = 1980, installation direction = 180, total length = 10 m, damage location = none,
Seismic intensity = 6 strong, Geology = terrace (omitted) Area = B00-00, Pipe section = M0201, Pipe type =
CIP, diameter = 75 mm, installation year = 1977, installation direction = 60, total length = 6 m, damage location = attachment, seismic intensity = 6
Strength, geology = terrace Area = B00-00, pipeline section = M0202, pipe type =
CIP, diameter = 75mm, installation year = 1977, installation direction = 180, total length = 6m, damage location = none, seismic intensity =
6 strong, geology = terrace (omitted) area = B01-00, pipeline section = M0301, pipe type =
DCIP, diameter = 100mm, installation year = 1981, installation direction = 150, total length = 6m, damage location = none, seismic intensity = 5, geology = alluvial area = B01-00, pipeline section = M0302, pipe type =
DCIP, diameter = 100 mm, installation year = 1981, installation direction = 60, total length = 6 m, damage occurrence point = joint, seismic intensity = 5 strong, geology = alluvium (hereinafter abbreviated) Damage analysis function 24 The obtained data is analyzed, and the relation between the damage occurrence location and other items is extracted as a rule. Here, the damage analysis function 24 has a classification of “damage location = none”, and based on detailed data on the pipelines, rules for classifying damaged and non-damaged pipelines and classification of damage locations. Get rules to do.

In order to effectively use these rules, detailed pipe data similar to the damage area pipe data file 31 is described in the damage prediction target pipe data file 12. A description example is shown below.

<Description example in damage prediction target pipeline data file 12> Area = A00-00, pipeline section = N0001, pipe type =
DCIP, diameter = 75mm, installation year = 1980, installation direction = 90, total length = 6m area = A00-00, pipeline section = N0002, pipe type =
DCIP, diameter = 75mm, installation year = 1980, installation direction = 180, total length = 10m (omitted) area = A00-00, pipeline section = N0101, pipe type =
CIP, diameter = 75mm, installation year = 1975, installation direction = 90, total length = 6m area = A00-00, pipeline section = M0102, pipe type =
CIP, diameter = 75 mm, laying year = 1975, laying direction = 120, total length = 6 m (hereinafter abbreviated) The method of predicting the damage location is the same as the first action and the second action.

(Fourth Operation) The epicenter information input function 41
Epicenter information is input from the Meteorological Agency server 131 via the Internet 132, and this epicenter information is further input to the seismic motion analysis function.

Here, the epicenter information is, specifically, data such as the direction and acceleration of the shaking of the seismic ground motion. The seismic ground motion function 42 uses the information of the epicenter such as the epicenter, depth of the epicenter, and magnitude. Calculate data such as the direction and acceleration of the sway. Similarly, the epicenter information data file 43 records the epicenter information such as the latitude and longitude of the epicenter of the earthquake that occurred in the past, the depth, the magnitude, and the type of the epicenter in the epicenter, and the direction of the shaking of the seismic motion in the seismic motion analysis function 44. Calculate data such as acceleration and acceleration.

<Example of description in epicenter information data file 43> Latitude = N34.1, longitude = E139.3, depth of epicenter =
10 km, magnitude = M4.8, seismic type = ocean type The seismic motion analysis function 44 reads the data from the epicenter data file 43, calculates the direction and acceleration of shaking for each area of the damage occurrence area, and sends it to the damage analysis function 24. input. An example of calculation data of the seismic motion analysis function 44 is shown below.

<Example of Calculation Data of Earthquake Motion Analysis Function 44> Area = B00-00, direction = 60, acceleration = 200g
al area = B00-01, direction = 65, acceleration = 180g
al area = B00-02, direction = 70, acceleration = 180g
al (hereinafter abbreviated) The damage analysis function 24 performs the same processing as that of the second action and the third action, including the analysis result of the seismic motion.
As a result, a rule relating to pipeline damage including information on the seismic motion is generated and stored in the damage prediction rule file 11.

The epicenter information input function 41 is connected to the Meteorological Agency server 131 via the Internet 132 as described above, acquires epicenter information from the Meteorological Agency server 131, and inputs it to the seismic motion analysis function. The earthquake motion analysis function 42 calculates the direction and magnitude of shaking for each area based on the input data, and inputs the calculated direction and magnitude to the damage prediction function 14.

The damage prediction function 14 includes a damage prediction rule file 11 containing rules on earthquake motion and damage, and a second data including a data input on earthquake motion from the earthquake motion analysis function 42.
By performing the same processing as that of the third operation and the third operation, the damage location of the pipeline is predicted.

(Fifth Action) A relational expression for each pipe type of the seismic motion and the number of damage to the pipeline has been proposed based on the results of pipeline damage due to past earthquakes. The earthquake damage number prediction function 51 uses these relational expressions to calculate the number of pipeline damage cases for each pipe type in the area. The damage prediction function 14 outputs as many damages as damage prediction results from those having the highest damage occurrence probability. The damage occurrence probability is the reliability of the rule. The reliability of a rule is defined as the accuracy rate of whether the damage was correctly classified when the rule was applied to the damage results of past earthquakes.

(Sixth Operation) The seismic intensity data input function 13 of the first operation, the seismometer network 124, and the seismometer 1
Instead of 21, 122, and 123, the data from the simulated seismic intensity data file 62 is input to the simulated seismic intensity data input function 62. In this case, instead of using the seismometer network 124, the damage to the assumed earthquake is examined.

The simulated seismic intensity data file 62 describes the seismic intensity data of the assumed earthquake. A description example is shown below.

<Example of description in simulated seismic intensity data file 62> Area = A00-00, seismic intensity = slight 7 Area = A00-01, seismic intensity = slightly 7 Area = A00-02, seismic intensity = strong 6 (hereinafter abbreviated) Simulated seismic intensity data The input function 61 reads data from the simulated seismic intensity data file 62 and inputs the data to the damage prediction function 14. A process similar to the first operation is performed. This makes it possible to predict pipeline damage to the assumed earthquake, and to take measures in advance.

(Seventh operation) Simulated seismic intensity data generation function 7
1. Epicenter latitude and longitude, epicenter depth, magnitude,
Enter the type of earthquake. Thereby, simulated seismic intensity data corresponding to the simulated seismic intensity data file 62 in the sixth operation can be generated. Since a method of calculating the seismic intensity of each place from the epicenter information has been proposed, it is possible to easily estimate the damage of the assumed earthquake and take measures.

(Eighth Function) Graphic Creation Function 83
Are data from a map data file 81 in a bitmap format, data from a pipeline drawing data file 82 expressed as a CAD data format, and a damage prediction function 1.
4 is received as an input. Pipeline attributes (pipe type, diameter, etc.) are also attached to the pipe drawing data 82. The graphic creation function 83 superimposes and displays the map and the pipeline drawing, and displays the damage prediction result on the screen. When the damage situation for each pipe type of each mesh is known, the damage situation is superimposed and displayed for each mesh. When detailed pipeline data is known as in the third operation, the damage status is displayed by changing the color of the pipeline, for example.

As a result, the damage prediction situation can be visualized, and the damage situation can be easily grasped.

(Ninth Operation) The actual damage / recovery status input function 91 is a function for inputting the actual damage /
The input of the damaged portion, which is not the predicted damage occurrence portion, or the damaged portion is received and displayed on the screen via the graphic creating function 83.

First, the predicted damage occurrence location is displayed in color 1. Next, input of the damage spots revealed during the inspection and restoration work is accepted. If it matches the damage prediction location, click the corresponding damage prediction location on the mouse screen,
When the menu is displayed and “Damage confirmation” is selected, the corresponding predicted damage location changes to color 2. If damage has occurred in a location other than the damage prediction location, click on the location, display a menu, and select “New Damage” to register new damage and display it in color 3. As a result of checking the damage prediction location, if it is found that there is no damage, the color is changed to color 4 by selecting the corresponding damage prediction location, displaying a menu and selecting “no damage”. The location where damage has been confirmed is indicated by color 2 or color 3. When the restoration of these damaged portions is completed, the affected portions are clicked, a menu is displayed, and “recovery completed” is selected, so that the color changes to color 5 and color 6, respectively.

In the manner described above, the earthquake damage prediction device can perform not only damage prediction, but also damage status, recovery status,
Accuracy of damage prediction can be easily grasped.

(Tenth Action) The actual damage recording function 1 outputs the actual damage data reflecting the information input from the actual damage / recovery status input function 91 to the actual damage data file 2. The description format of the actual damage data file 2 is the same as that of the earthquake damage data file 21.

In this manner, a new rule can be obtained by newly analyzing the data from the actual damage data file 2 as the data of the earthquake damage data file 21 by the damage analysis function 24.

As described above, according to the present embodiment, it is possible to predict damage to a pipeline caused by an earthquake by using a computer.
Further, a rule for damage prediction can be generated by a computer, and a rule for more detailed damage prediction can be generated by a computer.

Further, the damage prediction in consideration of the information on the seismic motion can be performed by a computer, and the upper limit of the number of damage predictions can be set.

Further, it is possible to take measures in advance by performing damage prediction for the assumed earthquake, and to easily perform damage prediction for the assumed earthquake by providing the epicenter information, and to visually display the damage prediction status. This makes it possible to quickly and accurately grasp the damage situation.

By visually displaying the actual damage situation and the damage prediction situation, the situation at the time of the earthquake can be quickly and accurately grasped, and the earthquake damage can be utilized for the prediction of the future earthquake damage.

[0084]

As described above, according to the present invention, it is possible to reliably predict the location where the damage to the pipeline due to the earthquake has occurred, so that the restoration work can be performed preferentially from the predicted location. As a result, it is expected that the pipelines such as water supply and gas, which are lifelines, will be quickly restored in the event of an earthquake.

[Brief description of the drawings]

FIG. 1 is an apparatus for predicting earthquake damage of a pipeline according to the present invention.

FIG. 2 is a diagram showing a state in which a damage prediction target area is divided into mesh areas.

FIG. 3 is a diagram showing a decision tree for classifying damage occurrence locations.

[Explanation of symbols]

 1 Actual damage recording function 2 Actual damage data file 11 Damage prediction rule file 12 Damage prediction target pipeline data file 13 Seismic intensity data input function 14 Damage prediction function 15 Damage prediction result output function 16 Damage prediction target geological data 21 Earthquake damage data file 22 Seismic intensity data file 23 Damage area geological data file 24 Damage analysis function 31 Damage area pipe data file 41 Source information input function 42 Earthquake motion analysis function 43 Source data file 44 Earthquake motion analysis function 51 Earthquake damage prediction function 61 Simulated seismic intensity data Input function 62 Simulated seismic intensity data file 71 Simulated seismic intensity data generation function 81 Drawing data file 82 Pipeline drawing data file 83 Graphic creation function 91 Actual damage / restoration status input function

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Yukihiko Tomizawa 1-1-1, Shibaura, Minato-ku, Tokyo Inside Toshiba Corporation Head Office (72) Inventor Takatoshi Kato 1-1-1, Shibaura, Minato-ku, Tokyo No. 1 F-term in Toshiba Corporation head office (reference) GG67

Claims (16)

[Claims] Claims: 1. A damage prediction rule file for predicting damage caused by an earthquake in a pipeline, a damage prediction target pipeline data file that records the type and diameter of the pipeline, and an input function of seismic intensity data of the generated earthquake Damage prediction rule file, Damage prediction target pipeline data file, Damage prediction function that reads data from seismic intensity data input function to predict damage location, and Damage prediction result output that displays the prediction result from Damage prediction function A pipeline seismic damage prediction device, comprising: 2. A damage prediction target geological data file describing surface geology and the like is further provided, and the damage prediction function predicts a damage location in consideration of data from the damage prediction target geological data file. The pipeline earthquake damage prediction device according to claim 1. 3. The damage prediction rule file has a structure expressing a tree structure classification rule, and the damage prediction function has a tree structure represented by the damage prediction rule file as an internal structure. 3. The pipeline earthquake damage prediction device according to claim 1, wherein damage is predicted by tracing a tree structure from the root according to the content of data for each pipeline of the pipeline data file. 4. The damage prediction rule file is expressed by if-then rules, and the damage prediction function operates as a production system using the damage prediction rule file as a rule base to predict damage to a pipeline. The earthquake damage prediction device for pipelines according to claim 1 or 2. 5. The damage prediction function includes seismic intensity data, geological data,
It has a neural network that inputs pipeline data and outputs the damage location and damage form. The damage prediction rule file indicates the weights of the nodes connecting the nodes of the neural network. 3. The pipe according to claim 2, wherein the weight of the damage prediction rule file is used as the value of the damage prediction rule file to predict damage to the pipeline from the data of the damage prediction target area geological data file, the damage prediction rule file, and the seismic intensity data file. Road earthquake damage prediction device. 6. A seismic intensity data file of an earthquake that occurred in the past in the damage prediction target area and data from the seismic damage data file caused by the earthquake are read into the damage prediction rule file, and the relationship between the seismic intensity data and the seismic damage of the pipeline is read. 2. The pipeline earthquake damage prediction device according to claim 1, wherein a damage analysis function for extracting the damage is connected. 7. A seismic intensity data file of an earthquake that occurred in the past in an area different from the damage prediction target area and data from an earthquake damage data file due to the earthquake are read into the damage prediction rule file, and the seismic intensity data and the earthquake of the pipeline are read. 2. A seismic damage prediction device for pipelines according to claim 1, wherein a damage analysis function for extracting a damage relationship is connected. 8. The damage analysis function uses a KDD method to generate a decision tree for classifying a damage location and a damage mode based on seismic intensity, geology, and pipeline data from data in an input data file,
8. The apparatus according to claim 6, wherein a structure of the decision tree is output to a damage prediction rule file. 9. The damage analysis function uses a KDD method to generate a decision tree for classifying a damage location and a damage mode based on seismic intensity, geology, and pipeline data from data in an input data file,
For each leaf node of the decision tree, the node of the path from the root node to the parent node of the leaf node is set as a condition part of an AND condition, and the rule is converted into an if-then rule format in which the leaf node is an action part, The pipeline earthquake damage prediction apparatus according to claim 6, wherein the apparatus outputs the damage prediction rule file. 10. A seismic intensity data file, an earthquake damage data file, a damage occurrence area geological data file, and data from the damage occurrence area pipeline data file are read into the damage prediction rule file, and the seismic intensity data and the pipeline data are read. 8. A seismic damage prediction apparatus for pipelines according to claim 6, wherein a damage analysis function for extracting a relation of seismic damage to pipelines is connected. 11. The damage prediction rule file may include a seismic intensity data file of an earthquake that occurred in the past, an earthquake damage data file due to the earthquake, a geological data file of a damaged area, a data file of a pipeline in the damaged area, A seismic motion analysis function that reads data from the epicenter data file of the generated earthquake and reproduces the seismic ground motion, seismic intensity data file, seismic damage data file, damaged area geological data file, damaged area pipeline data file, and ground motion analysis A seismic intensity data, pipeline data, and a damage analysis function that extracts the relationship between seismic motion and seismic damage to the pipeline are connected.The damage prediction function is equipped with an epicenter information input function that reads real-time epicenter data. Earthquake motion analysis function that predicts damage locations by reproducing earthquake motion from information The seismic damage prediction device for pipelines according to claim 6, wherein: 12. An earthquake damage number prediction function for receiving the input from the seismic motion analysis function and estimating the number of damage to the pipeline in each area is connected to the damage prediction function, and the data is read by the earthquake damage number prediction function. 12. The earthquake damage prediction device for pipelines according to claim 11, wherein the estimated number of damages is output from those having the highest damage occurrence probability in the area. 13. The damage prediction function receives data from a simulated seismic intensity data input function for reading a simulated seismic intensity data file and inputting simulated seismic intensity data instead of inputting from the seismic intensity data input function. Claims 1 to 1
2. An apparatus for predicting earthquake damage to a pipeline according to any one of 2. 14. A damage prediction result output function, wherein a damage prediction result, data from a map data file and a pipeline drawing data file are read from the damage prediction function, and a map, a pipeline drawing and a damaged portion are superimposed or a single graphic. The seismic damage prediction apparatus for pipelines according to any one of claims 1 to 13, wherein a graphic generation function for generating a screen is connected, and the screen generated by the graphic generation function is output. 15. The graphic creation function receives an input from the actual damage / recovery status input function,
15. The earthquake damage prediction device for a pipeline according to claim 14, wherein the restored damage point and the predicted damage point are displayed separately. 16. The graphic creation function according to claim 1, wherein only the actual damage data is input to the actual damage recording function, and the actual damage is stored as an actual damage data file in the same format as the earthquake damage data file. 15. An earthquake damage prediction device for pipelines according to 15.