Perspective Chapter: Building Damage Estimation and Renovation Proposal System Using Constant Microtremor Measurement for Construction Risk Management

Abstract

In this chapter, the authors propose a system that measures microtremors of the ground and buildings and instantaneously calculates building damage estimation and emergency risk assessment. In the conventional damage estimation and emergency risk assessment due to microtremors, resonance judgments are mainly based on the initial state due to microtremors of the ground, and buildings are often seen. However, there is a possibility that the ground and the building will resonate continuously as the ground and the building are cracked and damaged by disturbances such as earthquakes and winds. In addition, the resonance causes great damage to the building. In the system proposed this time, damage estimation and emergency risk assessment are performed not only in the initial state but also during the duration. In addition, as a renovation proposal, a system is proposed that predicts the damage and emergency risk of the building during renovation with earthquake resistance, seismic isolation, and damping. In this system, the building hazard levels are classified into four categories based on two methods: one method involves calculating the response spectrum and transfer function of the building and ground using continuous microtremor measurements to determine the resonance performance, and the other method utilizes the PML (Probable Maximum Loss) values.

1. Introduction

In Japan, after an earthquake, local government officials and building experts conduct building damage surveys for the purpose of preventing secondary damage and confirming the availability afterward. According to technical guidelines on the damaged building damage classification criteria and restoration, the restoration procedure is divided into three stages: (1) “This building is usable” (2) “Caution is required when entering” and (3) “Entering this building is dangerous” [1], and investigations and restoration measures according to those stages are examples of restoration strategies. It is listed as. Looking at the implementation status of past earthquakes, it takes about two weeks to complete the emergency risk assessment and about three months to complete the disaster classification assessment. The current situation is that recovery activities are further delayed due to confusion after the earthquake, shortage of people, and insufficient number of contractors. We searched for a system construction method for emergency risk determination and verified its validity. Technical Guidelines have been produced for seismic damage classification criteria on affected buildings and restoration after the disaster [1].

Earthquake preparedness guidelines and reality in Japan, the information required for building analysis based on assumed earthquakes is often accessible to the public through the Internet. The information required for building analysis based on assumed earthquakes includes, for example, the publication of borehole survey results for the specific area on websites [2]. These results can be used for earthquake analysis. Additionally, the “J-SHIS Earthquake Hazard Station” (https://www.j-shis.bosai.go.jp/: National Research Institute for Earth Science and Disaster Resilience) [3] provides a probabilistic seismic motion prediction map, which uses a finely classified terrain model in 250m mesh, to present the probability of future earthquake motion based on the prediction of seismic source faults. Data collection based on location can be retrieved from these server spaces. Furthermore, some of this information is also available as web-based APIs, indicating a trend toward systematizing their construction.

The purpose of this study is to reduce human or building damage in the event of an earthquake, clarify the dangers of individual buildings before taking concrete disaster prevention measures, and effectively call for countermeasures against earthquake damage.

Research on emergency risk assessment is often measured with a dedicated accelerometer that is not widely used, but in this research, the accelerometer built into a mobile device such as a smartphone is used, and it is generally used via Internet access [4]. Will build a system that can easily perform predictive diagnosis of emergency risk judgment. In addition, the system to be constructed will be constructed, as shown in Figure 1, by unifying the load on the mobile terminal, the storage of research information, and the judgment processing. We will bring the analysis method by constant tremor and PML to this system, actually distribute it, and consider the tendency of the judgment result about the obtained result.

Figure 1.

Overview of damage prediction judgment system using mobile terminals.

2. Construction of mobile terminal system

2.1 System overview

Currently, Android, iOS, etc., are the mainstream operating systems for smartphones, and the system construction method is different for each. In this research, we focus on Android as an example (Figure 2), but if a similar API is used, it will be possible to develop the same for iOS.

Figure 2.

Appearance of damage prediction judgment system using mobile terminals.

In addition to using Android Studio in Android development, functions such as map function and geocoding use google API, and it is necessary to use google developers console to use these(Table 1).

Function, environment	Explanation
Android	Operation system
Android studio	Platform
Google MAP	Screen using Googlemap
Firebase	Communication between server and terminal

Table 1.

Terminal development environment.

2.2 Construction of measurement system

It is difficult to acquire the values of the acceleration sensor and the azimuth sensor of the mobile terminal at equal intervals because both of them are postprocessed when the values are changed. Therefore, it is necessary to store it in a global variable or static variable in the posted process, and it is also necessary to post a message independently by using a timer function, etc., in parallel processing. In addition, among the processes posted by the sensor, the acceleration in the world coordinate system is acquired by performing the attitude control process of the mobile terminal. For attitude control, after acquiring the rotation matrix with the geomagnetic sensor, the inverse matrix is multiplied by the acceleration matrix to convert it into the world coordinate system. At this time, gravity is removed. The post from the timer mainly runs the export process. The flow chart of the measurement process is shown in Figure 3.

Figure 3.

Appearance of damage prediction judgment system using mobile terminals.

2.3 Creating a summary file on a mobile terminal and communicating with the server

In the terminal, it is necessary to create comprehensive data including measurement data and property information for measurement data. After transferring the file by FTP communication, the terminal creates a queue for the server to start analysis and sends it. For message queues, information is sent to SQL and processed to replace the queue.

3. Building a system for a server

3.1 System overview

The server constantly evaluates the resonance by tremors and analyzes by PML (Probable Maximum Loss) value, and makes a prediction judgment of the damage degree classification judgment standard of the damaged building. In the evaluation of resonance by constant tremor, ground analysis by Shake is performed, seismic motion in the surface layer ground is created, transfer function and H/V spectrum are obtained, and resonance is evaluated from the ground predominant period and the natural period of the building. In addition, noise removal such as smoothing and low-pass filter is also processed on the server side for the measured value data. On the other hand, in PML, the damage rate of the building is calculated by obtaining the inter-story deformation angle from the bilinear model of the building consisting of the stochastic seismic velocity and the base shear coefficient of the building.

The operation is to receive the measurement result of constant tremor from the terminal, execute the analysis, and then save the data in SQL. In addition, considering the changes in the analysis method in the future, it is a prerequisite that the input data is appropriately stored in the server and that the data stored in SQL is specified or provided in a form that is easy to see for the majority. The hardware or software environment of the server is selected in all aspects such as capacity, speed, stability, etc., to meet these assumptions, but by consolidating analysis or data storage, etc., in one place, data falsification can be done. There is no room stable results can be provided, and there is the merit that the burden of analysis on the terminal can be removed. To execute these tasks, the server development environment utilizes Table 2.

Function, environment	Explanation
Windows server	Operation system
Visual studio	C++,C#, compiler
Google geocording	Latitude and longitude from the address
Firebase	Communication between server and terminal
PHP	Dialogue with the database
JSHIS	Internet site
MYSQL	database
SHAKE	Ground analysis

Table 2.

Server development environment.

3.2 Selection of boring logs and analysis by Shake

In this research, XML data of boring log is obtained from the national land information retrieval site “Kunijiban” [2]. The coordinate values and data in the data are saved in SQL in advance, and the closest ground data is acquired from the data corresponding to the search near the area of 200 m, and the data is analyzed. The data of the boring log obtained by this is automatically generated according to the input specifications for Shake, and the surface seismic motion is generated. In addition, the shear wave velocity is input at the time of input, but in this study, the commonly used formula shown in the road bridge specification was used to estimate Vs (Figure 4).

Figure 4.

Overall analysis flow (1) and judgment flow (2).

3.3 Creation of H/V spectrum and transfer function

The H/V spectrum can be obtained by using the amplitude spectrum for each direction obtained by the Fourier transform in the following equation.

The period in which the H/V spectral ratio shows the maximum value is the natural period of the surface layer ground, and the maximum value is the estimated value of the amplification factor. In this study, the period of the maximum value of the H/V spectrum is treated as the constant ground predominant period of the surface layer. In addition, the transfer function is obtained by the following equation when the dynamic interaction between the ground and the structure is taken into consideration. The period of the maximum value of the transfer function is the building natural period:

$$S_i=\sqrt{\frac{X_i{}^2+Y_i{}^2}{Z_i{}^2}}$$E1

Si : H/V spectral ratio at i　　　Xi : Spectrum in the X direction at i

Yi : Y-direction spectrum at i　　 Zi : Z-direction spectrum at i

$$T{f}_i=Bi/Gi$$E2

Tfi 　: Transfer function at i

Bi :　Fourier spectrum of the building at i

Gi :　Fourier spectrum of the ground at i

3.4 Natural period of buildings and ground predominance period at the time of earthquake

The range in which the natural period of the building transitions can be calculated as the resonance range by specifying the interlayer deformation angle for the assumption of the interlayer deformation at all times and the assumption of the interlayer deformation angle at the time of an earthquake. Table 3 shows this study.

	Steel	Reinforced concrete	Wood
At design time	1/300	1/400	1/200
During an earthquake (LV1)	1/200	1/250	1/120
During an earthquake (LV2)	1/100	1/125	1/60

Table 3.

Setting the interlayer deformation angle between design and earthquake.

It is the specification of the specified interlayer deformation angle. In addition, if the calculation of the ground predominant period Shake at the time of an earthquake is available, the predominant period is always obtained from the specified seismic wave by the same method, but if the acquisition of the boring columnar map fails, the ground is always obtained. The multiplication value α is defined from the value of the predominant cycle and is defined as the ground predominant cycle at the time of an earthquake (Table 4). Each cycle is considered in a band shape, and the future emergency risk judgment is predicted from the occupancy rate of each area (Figure 5):

Ground type	Natural period	α
First-class ground	~0.4	1.0
Second-class ground	0.4~0.6	1.5
Third-class ground	0.6~	2.0

Table 4.

Setting of ground type and multiplication value α.

Figure 5.

H/V spectral ratio (1) and band determination (2).

$$T_B=\sqrt{\frac{{\Theta }_A}{{\Theta }_B}}\times T_A$$E3

Θ_{A :} Interlayer deformation angle A　　　 　Θ_{B :} Interlayer deformation angle B　

T_{A :} Period at inter-story deformation angle A T_{B :} Period at inter-story deformation angle B

3.5 Analysis by PML

In the analysis by PML, the probabilistic seismic velocity and s related to the ground [5]. The propagation speed, amplification factor, etc., are obtained from JSHIS, and the building is bilinear.

Apply to the model and calculate its damage rate. Since the PML value at the time of an earthquake obtained here is an index of a general building, considering the validity of the measurement target, the abbreviation formula of the building natural period that is generally used and the building natural period obtained by constant tremor are used. By multiplying the ratio of, the PML value of the seismic motion considering the building to be measured was calculated (Figure 6). Additionally, the defined criteria based on the obtained PML values are presented in Table 5.

Judgment	PML value (%)
green	~15
yellow	15~40
red	40~70
gray	70~

Table 5.

Judgment as PML value.

Figure 6.

Bilinear model at the time of PML calculation.

4. Discussion

4.1 Aggregate results

Distribute the application on Google Play. As a result, there are collect the results of 52 measurements. As a tendency of the measurement target. There were 26 cases of reinforced concrete, 18 cases of steel frame, and 7 cases of wooden construction. Most buildings are 1 to 15 stories high and are located in metropolitan areas.

4.2 Features by category

In terms of structural types, among the 18 cases of steel frame structures, 11% were classified as red, 50% as yellow, and 39% as green. Among the 26 cases focused on reinforced concrete structures, 4% were classified as red, 58% as yellow, and 38% as green. As for the wooden structures, out of the 7 cases, 29% were classified as red, 14% as yellow, 29% as green, and 29% as collapse.

Furthermore, focusing on the aggregation based on the ratio between the probabilistic seismic velocity from JSHIS and the S-wave velocity (Vs), when dividing it by the average, the results showed that for cases below the average, 68% were classified as green and 32% as yellow. On the other hand, for cases above the average, 70% were classified as yellow, 22% as red, and 8% as collapse, confirming the dependence on ground characteristics (Table 6).

Result	Location	V/Vs
Yellow	Kameari, Katsushika-ku, Tokyo	0.83
Yellow	Shinjuku, Katsushika-ku, Tokyo	1.00
Gray	Nishi-Nippori, Arakawa-ku, Tokyo	0.99
Yellow	Hibino, Mihama-ku, Chiba City, Chiba Prefecture	0.99
Red	Nogizaki, Moriya, Ibaraki	0.95
Red	Kagurazaka, Shinjuku-ku, Tokyo	0.67
Yellow	Oi, Kashiwa, Chiba	0.83
Yellow	Ueno, Taito-ku, Tokyo	1.01
Yellow	Ueno, Taito-ku, Tokyo	1.00
Red	Nishigotanda, Shinagawa-ku, Tokyo	1.00
Yellow	Kumotoji Tanida, Kuki City, Saitama	0.76
Gray	Saitama Prefecture Kita-Katsushika-gun Sugito Town Koya Tainan	0.82
Red	Kamitakano, Saitama	0.81
Red	Saitama Prefecture Minami Saitama District Miyashiro Town Gakuendai	0.81
Yellow	Saitama Prefecture Kita-Katsushika-gun Sugito Town Koya Taito	0.81
Yellow	Kameari, Katsushika-ku, Tokyo	0.83
Yellow	Tabata Shinmachi, Kita-ku, Tokyo	1.02
Yellow	Nishi-Nippori, Arakawa-ku, Tokyo	0.98
Yellow	Kanamachi, Katsushika-ku, Tokyo	1.02
Yellow	Kanamachi, Katsushika-ku, Tokyo	1.02
Yellow	Kanamachi, Katsushika-ku, Tokyo	0.73
Yellow	Toganecho, Katsushika-ku, Tokyo	1.01
Yellow	Senjujucho, Adachi-ku, Tokyo	1.00
Result	location	V/Vs
Yellow	Tomita, Oamishirasato, Chiba	0.43
Yellow	Kasugacho, Nerima-ku, Tokyo	0.39
Green	Aoba-ku, Sendai, Miyagi	0.12
Green	Hojo Higashi, Chuo-ku, Sapporo, Hokkaido	0.10
Green	Senju, Adachi-ku, Tokyo	0.33
Green	Yagiyamamotocho, Taihaku Ward, Sendai City, Miyagi Prefecture	0.09
Green	Kamihatagi, Hokota, Ibaraki	0.38
Yellow	Kashiwa, Kashiwa, Chiba	0.50
Green	Kashiwa, Kashiwa, Chiba	0.41
Green	Kashiwa, Kashiwa, Chiba	0.41
Green	Kashiwa, Kashiwa, Chiba	0.50
Green	Nishi-Nippori, Arakawa-ku, Tokyo	0.33
Green	Yaesu, Chuo-ku, Tokyo	0.34
Green	Japan Bridge, Chuo-ku, Tokyo	0.34
Green	Komagome, Toshima-ku, Tokyo	0.37
Green	Nishi-Ikebukuro, Toshima-ku, Tokyo	0.41
Green	Nishi-Ikebukuro, Toshima-ku, Tokyo	0.41
Green	Suehiro, Okegawa, Saitama	0.37
Green	Senju, Adachi-ku, Tokyo	0.33
Green	Senju, Adachi-ku, Tokyo	0.33
Yellow	Negishi, Taito-ku. Tokyo	0.34
Yellow	Ueno Sakuragi, Taito-ku, Tokyo	0.52
Yellow	Takada, Toshima-ku, Tokyo	0.58
Yellow	Takadanobaba, Shinjuku-ku, Tokyo	0.58
Yellow	Nakamachi, Kodaira City, Tokyo	0.34
Green	Matsudo, Matsudo, Chiba	0.34
Green	Kashiwa, Kashiwa, Chiba	0.41
Yellow	Kashiwa, Kashiwa, Chiba	0.50
Gray	Senju, Adachi-ku, Tokyo	0.33

Table 6.

Judgment value of aggregated data and classification by V/Vs.

5. Conclusion and recommendation

The summary of this study is shown below.

• Shows the specific procedure for configuring the risk assessment prediction system and operates it.

• The app was actually published to show the usefulness of the app.

• Land dependence was observed in comparison with the values obtained from the hazard MAP.

• It is difficult to build data covering the entire country, like the scattered data of bowling pin diagrams. However, in Japan, it is possible to develop such an application because seismic information covering the entire country is disclosed based on geological layers and past earthquakes, as seen on the JSHIS website.