KR102232918B1 - Subsidence sensor of real-time artificial intelligence based building condition diagnosis system - Google Patents

Abstract

According to the present invention, provided is a real-time artificial intelligence-based building condition diagnosis system. A subsidence detector of the real-time artificial intelligence-based building condition diagnosis system comprises: a global navigation satellite system (GNSS) module configured to receive GPS data; and a control unit configured to be operatively coupled to the GNSS module to control the GNSS module. The control unit collects the GPS data, stores the collected GPS data in a server or a data storage unit, primarily determines whether subsidence occurs based on the stored GPS data, and, when it is determined that the subsidence occurs, determines the type of subsidence and a degree of an occurrence of a building by using an artificial intelligence model in accordance with a determination result and a trend of a measurement value of the collected GPS data, thereby providing real-time artificial intelligence-based building condition diagnosis.

Description

Subsidence sensor of real-time artificial intelligence based building condition diagnosis system

The present invention relates to a building condition diagnosis system. More specifically, it is about a real-time artificial intelligence-based building condition diagnosis system.

Since the 2016 scale 5.8 Gyeongju earthquake and 2017 scale 5.4 Pohang earthquake, various earthquake-related studies are underway in Korea centered on the Geological Society, and at the government level, various information related to the Gyeongju earthquake and Pohang earthquake and earthquake preparedness Research and development related to service provision is in progress. In particular, in the case of subsidence among earthquake-related studies, it is a social issue related to the safety of people's lives because it accompanies the phenomenon of depression of the ground/building.

In general, settlement differs depending on the ground and the building. Ground subsidence can be deformed by rising or subsidence of the ground depending on various related factors.Especially in the ground caused by the accumulation of soil, or in the soft ground with a high water content such as clay, silt, organic soil, and loose sandy soil. Occurs very frequently. Building subsidence basically contains the characteristics of ground subsidence, and acts as a factor that determines the presence or absence of collapse depending on the seismic design that can withstand seismic forces.

In particular, the subsidence of a building may cause deformation of the structure itself and destruction of related facilities depending on the characteristics of the ground subsidence and the structure of the building, resulting in serious damage to the structure and auxiliary facilities, as well as loss of life in serious cases. For this reason, it can be said that the detection of settlement related to buildings is essential for the protection of structures/subsidiary facilities and human life.

This is due to frequent earthquakes after the 2016 scale 5.8 Gyeongju earthquake, requiring a device for precise subsidence of buildings in order to protect people and collapse of buildings. According to the statistics of the Meteorological Administration, the number of earthquakes with a magnitude of 2.0 or more was surveyed on average 31.8 times until 2015 as of the initial year of statistics from the Meteorological Administration, while the number of earthquakes from 2016 to the present was 169.5 on average and 5.33 times. It was investigated.

As the most fragmentary example, the Gyeongju earthquake incurred an estimated 10.6 billion won in property damage in Gyeongju, including private buildings and public facilities, and the Pohang earthquake incurred 300 billion won in property damage, including damage to houses and facilities. This means that it is no longer an earthquake safe zone in Korea. Therefore, it is the time in which a real-time measurement device capable of measuring short-term/long-term building settlement in order to prevent large-scale property damage, such as the Gyeongju and Pohang earthquakes, and to minimize the scale of damage and to ensure the right to public stability is required.

Existing devices for measurement of settlement have used various sensors such as a pressure sensor, a water pressure sensor, and a displacement sensor to measure the settlement of buildings and ground. However, existing equipment has a characteristic that it is difficult to utilize settlement information in real time at a remote location for settlement measurement. It is difficult to use settlement information and analyze it professionally.

For example, in order to perform a safety inspection of an existing building, the safety of the building (crack, subsidence, immobile subsidence, etc.) is checked through two months of inspection through two periods of building inclination surveys. However, it takes a lot of time and money to analyze a single building, and since only changes in the building are measured within a few days, it is impossible to measure the fatigue level of the building due to earthquakes depending on the initial state of the building. In particular, in the case of immobile settling, there are many difficulties in measuring due to a large amount of time and cost.

Accordingly, an object of the present invention is to provide a real-time artificial intelligence-based building condition diagnosis system in order to solve the above-described problem.

In addition, an object of the present invention is to provide a real-time artificial intelligence based building condition diagnosis system capable of transmitting settlement information in real time to a remote location.

In addition, an object of the present invention is to develop a system capable of grasping the fatigue level of a building in real time due to the effects of vibrations, earthquakes, typhoons, etc. of a building in real time.

A real-time artificial intelligence based building condition diagnosis system according to the present invention for solving the above problems is provided. The settling detector of the temporal artificial intelligence-based building condition diagnosis system includes a GNSS module configured to receive GPS data; And a control unit operatively coupled to the GNSS module and configured to control the GNSS module. The control unit collects the GPS data and stores the collected GPS data in a server or a data storage unit, firstly determines whether subsidence has occurred based on the stored GPS data, and when it is determined that the subsidence has occurred, A real-time artificial intelligence-based building condition diagnosis may be provided by determining the type of settlement and the degree of occurrence of the building using an artificial intelligence model according to the determination result and the trend of the measured value of the collected GPS data.

According to an embodiment, if GPS data can be received, the control unit controls the received GPS data to be stored in a data storage unit inside the settlement detector, and if it is determined that communication with a remote server is possible, the GNSS It is possible to acquire precise subsidence detection data by improving GPS accuracy using module and Networked Transport of RTCM via Internet (NTRIP) correction information, and transfer the precise subsidence detection data to the server through a communication module.

According to an embodiment, the control unit may generate an orbit pattern using the GPS data, and determine whether or not there is an abnormality in the sinking sensor sensor using the generated orbit pattern. The settlement detector of the time AI-based building condition diagnosis system may further include a data storage unit configured to store the GPS data and the presence or absence of an abnormality in the settlement detector sensor. The control unit may control to display the presence or absence of an abnormality in the sinking sensor sensor on the screen of the display unit.

According to an embodiment, the controller learns a machine learning algorithm by converting the three-way vibration data of latitude, longitude, and altitude measured from the sinking sensor sensor into an orbit pattern, and when the orbit pattern is generated in a circular shape, the It can be determined that there is no abnormality in the building.

According to an embodiment, when the orbit pattern is generated in any one of an oval, a nasolabial shape, a heart shape, and a tornado shape, the controller determines that an abnormality exists in the building, and the orbit pattern is When the oval shape is generated, it is determined that the building is not balanced, and when the orbit pattern is generated in any one of the nasolabial shape, the heart shape, and the tornado shape, it may be determined that the axis of the building is shifted. .

According to an embodiment, the control unit receives sensing information in real time using a settlement detector installed inside a plurality of smart buildings and a settlement detector installed on the roof of each building, and machine learning the received sensing information. An orbit pattern may be generated by applying the algorithm, and the presence or absence of an abnormality in a building associated with the floating settlement may be determined using the generated orbit pattern. In addition, the control unit may store the received sensing information and whether there is an abnormality in the building, and provide the user with information on whether there is abnormality in the building on a screen through a display unit.

According to an embodiment, the controller may generate an orbit pattern by receiving settlement information measured using a settlement sensor sensor, and applying the settlement information to an orbit generation algorithm. In addition, the control unit generates a picture map, which is machine learning learning data, labeling the picture map, and learning a machine learning learning model, thereby generating a machine learning model for determining the measured settlement information. It may generate and determine a building abnormality symptom according to the measured settlement information using the machine learning model, and when the abnormality symptom is determined, the abnormality symptom may be notified to a manager terminal.

According to at least one embodiment of the present invention, there is an advantage that it is possible to provide a building condition diagnosis system.

In addition, according to at least one embodiment of the present invention, there is an advantage that it is possible to provide a real-time artificial intelligence based building condition diagnosis system.

On the other hand, when an earthquake occurs, the building collapses due to the earthquake damage of the building, resulting in a lot of human and material damage. Therefore, according to at least one embodiment of the present invention, it is expected that when applied to a building, it is possible to respond more quickly by diagnosing the safety level of the building through artificial intelligence analysis based on accumulated data in real time.

The features and effects of the present invention described above will become more apparent through the following detailed description in connection with the accompanying drawings, and accordingly, those of ordinary skill in the technical field to which the present invention pertains can easily implement the technical idea of the present invention. I will be able to.

1A shows the entire process of the phenomenon of liquefaction according to the present invention.
1B shows a detailed summary of the building settlement according to the present invention.
Figure 2 shows the form of the floating settlement according to the present invention and the foundation ground and load conditions.
3 shows a safety inspection procedure related to the safety evaluation of a building according to the present invention.
4 shows the structure of a building condition diagnosis system including a GNSS-based settlement detector according to the present invention.
5 is a flow chart showing the operation of the sinking detector according to the present invention.
6 shows the measured data measured by the sinking sensor according to the present invention.
7 is a flow chart showing a method for detecting abnormality using the sinking sensor of a building according to the present invention.
8 is a flowchart showing a machine learning model learning and diagnosis procedure for earthquake detection according to the present invention.
9A is an illustration of a method of converting to an orbit pattern according to the present invention.
9B shows an orbit shape according to a state and class according to the present invention.
10 shows a conceptual diagram of an installation of a settlement detector that can be installed in a building according to the present invention.

The features and effects of the present invention described above will become more apparent through the following detailed description in connection with the accompanying drawings, and accordingly, those of ordinary skill in the technical field to which the present invention pertains can easily implement the technical idea of the present invention. I will be able to.

In the present invention, various modifications may be made and various embodiments may be provided, and specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to a specific embodiment, it should be understood to include all changes, equivalents, or substitutes included in the spirit and scope of the present invention.

In describing each drawing, similar reference numerals are used for similar elements.

Terms such as first and second may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another component.

For example, without departing from the scope of the present invention, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element. The term and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

Unless otherwise defined, all terms used herein including technical or scientific terms have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in the present application. Shouldn't.

Suffix modules, blocks, and parts for components used in the following description are given or mixed in consideration of only the ease of writing the specification, and do not have meanings or roles that are distinguished from each other by themselves.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement them. In describing the embodiments of the present invention below, when it is determined that a detailed description of a related known function or a known configuration may unnecessarily obscure the subject matter of the present invention, a detailed description thereof will be omitted.

Hereinafter, a real-time artificial intelligence based building condition diagnosis system according to the present invention will be described. In connection with the diagnosis of the building condition, abnormalities may occur in the building condition due to natural factors such as earthquakes and building defects during construction.

Earthquakes caused by natural factors mean the swaying of the Earth's surface, and in this process, the instantaneous displacement in the fault plane is often referred to as an earthquake. When an earthquake occurs, the ground and buildings change due to natural factors, and in the case of the ground, in particular, liquefaction occurs due to the strong shaking of the earthquake.

The liquefaction phenomenon refers to a phenomenon that rises out of the ground surface according to the components determined under the ground by a strong seismic force generated by colliding with the fault surface. In particular, the liquefaction phenomenon occurs prominently in embankment grounds made of soil or in soft grounds with high water content, such as clay, silt, organic soil, and loose sandy soil. In this regard, FIG. 1A shows the entire process of the phenomenon of liquefaction according to the present invention.

As the closest example, after the Pohang earthquake in 2017, the National Disaster and Safety Research Institute of the Ministry of Public Administration and Security conducted a Liquefaction Potential Index (LPI) survey of 212 locations in Pohang City, and 53% of them were announced as having a liquefaction risk index. I did. In other words, it means that the liquefaction phenomenon can have a great impact on the people's property damage.

Building settlement caused by earthquake refers to the displacement of the building in the downward direction due to the ground with various characteristics. Among them, there are three broad categories that affect the settlement of the building due to the ground: equal settlement, overturn settlement, and floating. It can be distinguished by subsidence.

Equal settlement generally means that the ground is compressed by an earthquake or external factors in a wide ground, and all parts of a structure or part of a structure are uniformly settled in the ground. Overturning settlement refers to settlement caused by uneven ground in a structure. Lastly, anti-freeze means subsidence due to uneven ground and soft ground.

On the other hand, Figure 1b shows a detailed summary of the subsidence of the building according to the present invention. In particular, sedimentation due to earthquakes is highly likely to cause immobile settlement, and the main factors of immobile settlement are due to various factors such as lack of support for buildings, buried ground and soft ground, seismic design, water content, etc., and identify the cause. In order to do this, it is necessary to measure the measure of settlement through short-term/long-term data collection and observation.

On the other hand, there is a method of measuring the subsidence of a building using an acceleration sensor to measure the subsidence of a building due to an earthquake. However, since the settlement measurement method using this acceleration sensor requires precise acceleration sensing information, it is difficult to measure by installing multiple sensors, and even if it is installed in a building using one sensor, it is difficult to measure the floating settlement of a building. do. In addition, since the acceleration sensor must be located in the center of the building, it is also necessary to analyze the structure of the building, so it is difficult to measure the floating settlement using the acceleration sensor.

Settlement measuring instruments measure settlement by utilizing various sensors such as pressure sensors, water pressure sensors, and displacement sensors. In particular, since the existing settlement measuring equipments are manually measured, reliability and economic feasibility may be degraded.

There are three main problems with the manual settlement instrument. First, the larger the area of the soft ground, such as the construction of a large-scale complex, the greater the distance error from the reference point, which is more likely to cause measurement errors and cumulative errors. Second, on-site conditions and weather-changing conditions lead to inability to obtain data and to deteriorate continuity. Third, an error occurs depending on the survey engineer, which is a factor in raising labor costs. In addition, there is a problem that long-term maintenance and measurement is impossible.

In order to solve this problem, it is possible to propose the development of a hall sensor-based automatic ground subsidence measuring system. The Hall sensor-based automatic ground subsidence measurement system uses a digital hall sensor and outputs real-time whether or not surrounding magnetic force is generated through magnetic force detection, and determines the amount of ground subsidence through this.

The advantage of such a system is that it can measure surface subsidence and multiple floors at the same time without a separate reference point, and has various advantages such as siege, construction speed, economy, and operational management.

However, the automatic ground subsidence measuring system uses magnetic force to measure the degree of subsidence, so there is a disadvantage in that the error is large due to the difference in the magnetic force of the building. In addition, in the case of an automatic settlement measuring instrument, the factor of error is large depending on the capabilities of the survey engineer, and there is a characteristic that settlement information cannot be used in real time from a remote place for settlement measurement. In addition, when the measurement information is analyzed using the GPS information measured by the settlement detector, changes in GPS information due to wind and changes in GPS information due to life vibration can be observed. Changes can occur.

Here, anti-freeze is a phenomenon that unevenly subsides in various parts of the structure as the foundation ground of the structure subsides, and can be referred to as anti-freeze.

Article 52 (Safety Evaluation of Structures or Similar Facilities) Standards on the Rules on Occupational Safety and Health Standards (abbreviation: Safety and Health Regulations), Paragraph 2) Cracks, twists, etc. in structures or similar facilities If there is a risk of collapse due to its own weight, snowfall, wind pressure, or other loads, the safety evaluation of the building shall be carried out.

Meanwhile, the safety inspection procedure related to the safety evaluation of the building is the same as that of FIG. 3. In this regard, Figure 3 shows the safety inspection procedure associated with the safety evaluation of the building according to.

In order to measure the immovable settlement of a building as shown in FIG. 3, since it takes a long time to investigate, and the collapse of the building may occur within the survey period, a technology capable of quickly determining the collapse of the building is required. In addition, in the case of a high-risk building, it may be very difficult to enter the building because the construction manager must enter the building and inspect the building using the inspection equipment.

In addition, since such a building safety measurement method only measures the degree of inclination, it is difficult to measure the fatigue level of a building in real time due to instantaneous earthquakes and typhoons.

The present invention installs two subsidence detectors in a building, measures equal settlement, overturn settlement, and floating settlement accurately and quickly through it, and based on this, data using artificial intelligence technology based on information from settlement detectors installed in a number of buildings. In addition to monitoring, it minimizes human damage by accurately grasping whether or not floating settlement occurs, the degree of occurrence, etc., and by notifying the building manager in advance of the collapse of the building so that quick action can be taken.

In the case of immobile settlement, it is most important to understand the size, direction, and trend/hardness/deformation over time. In order to measure this, the laser level (PR3) is used to measure the level of horizontality (floating settlement), and through this, the process of moving the structure to a dangerous state entails deformation and measures whether it is a progressive deformation.

The present invention relates to a system capable of accurately measuring the floating settlement due to an earthquake by using artificial intelligence technology for real-time precise floating settlement detection of a building.

The building condition diagnosis system according to an embodiment of the present invention includes a GNSS-based settlement detector, a settlement information measurement server, and an artificial intelligence based floating settlement detection model based on sensing information.

The GNSS building subsidence detector measures the subsidence of a building due to earthquake damage in real time and transmits measurement data. The GNSS building settlement detector measures precise settlement information by improving GPS accuracy by using the Global Navigation Satellite System (GNSS) module and Networked Transport of RTCM via Internet (NTRIP) correction information.

In addition, the GNSS building settlement detector may further include LTE, Ethernet, and LTE cave modules, and through this, the GNSS building settlement detector may transmit real-time settlement measurement data and transmit building settlement information in real time.

In addition, Ntrip (Network Transport of RTCM via Internet Protocol), which is included in the GNSS building settlement detector, transmits GPS correction signals through the Internet network. Existing DGPS transmits a GPS position error correction signal through a medium-wave signal medium, which may cause problems in terms of transmission efficiency and accuracy. Here, RTCM (Radio Technical Committee for Maritime Service) refers to a GPS correction signal transmission data format.

Ntrip correction information according to an embodiment of the present invention can be downloaded through the Internet, and by using this correction information, the error of GPS by using Ntrip correction information from the sensing information measured by the GPS sensor of the building in which the settlement detector is installed. To obtain precise GPS information.

On the other hand, the building condition diagnosis system according to the embodiment of the present invention installs and operates a building sinking detector to check the state of the building in real time, and when an immovable settlement occurs, it notifies the relevant person in real time. It is a system that can minimize damage. In particular, the present invention may be composed of a building settlement detector, settlement information collection server, artificial intelligence diagnosis server, and administrator notification web/app.

The GNSS building sinking detector according to an embodiment of the present invention can be located on the roof of a building, receives GPS information, and uses Ntrip correction information based on this to reduce GPS errors, thereby reducing accurate building latitude, longitude, and altitude information. Measure The settlement information collection server refers to a system that can store the data measured by the settlement detector in real time.

The artificial intelligence diagnosis server according to an embodiment of the present invention is equipped with artificial intelligence models such as life vibration diagnosis, fatigue diagnosis of buildings due to earthquakes, and floating settlement diagnosis using settlement information measured by a settlement detector. Use the information to perform the analysis.

The manager notification web/app according to an embodiment of the present invention may perform a function of notifying the manager of the state of the building for each location analyzed through the artificial intelligence model.

As an example, GPS information is measured in real time by a GNSS building settlement detector installed on a roof, and transmitted to a settlement information collection server. The collected information is transmitted to the collected data analysis artificial intelligence server, and the real-time collected data is analyzed and the result is expressed in a manager notification app/web to determine the safety level of the building against the effects of natural disasters (earthquakes, typhoons, etc.) in real time. can do. The safety level of a building can be judged by dividing into floating settlement, cracking, slope, settlement degree, equal settlement, overturn settlement, and differential settlement.

Meanwhile, the GNSS building subsidence detector according to an embodiment of the present invention utilizes a subsidence detection control board and a GNSS module to perform precise building subsidence detection.

4 shows the structure of a building condition diagnosis system including a GNSS-based settlement detector according to the present invention.

Referring to FIG. 4, the building condition diagnosis system includes a settlement detector 100 including a GNSS module 110 and a control unit 120. As an example, the controller 120 may be configured as a control board, and the settlement detector 100 may be a GNSS-based settlement detector. The sinking detector 100 may further include a data storage unit 130. The GNSS module 110 of the sinking detector 100 may be communicatively coupled with the GPS satellite 200.

The GNSS module 110 may be configured to receive GPS data. The controller 120 may be configured to be operatively coupled with the GNSS module 110 to control the GNSS module 110. The controller 120 may collect GPS data and store the collected GPS data in a server or a data storage unit. The controller 120 may primarily determine whether subsidence has occurred based on the stored GPS data. In addition, when it is determined that the subsidence has occurred, the control unit 120 may be configured to determine the subsidence type and the degree of occurrence of the building using an artificial intelligence model according to the trend of the determination result and the measured value of the collected GPS data. have.

Accordingly, the control unit 120 may 1) collect data from the device, that is, the sinking sensor, and 3) store the collected data in the server or the data storage unit 130, while the device, that is, the sinking detector is configured. . In addition, the control unit 120 may 4) determine by a first determination algorithm based on the stored data, and 5) determine a part requiring in-depth analysis based on the determination result by the first determination algorithm through machine learning. On the other hand, referring to FIG. 5 with respect to the specific operation of the sinking detector 100 is as follows.

5 is a flow chart showing the operation of the sinking detector according to the present invention. The sinking detector according to an embodiment of the present invention first determines whether GPS reception is possible after power is applied (S110), receives GPS data (S120), and acquires settlement data using the received GPS data, and stores internal storage. It is stored in (S130). At this time, the settlement data stored in the internal storage may be overwritten on the oldest data when the capacity of the internal storage is full. Thereafter, the proposed settlement detector may determine whether Ethernet is possible (S140) and receive correction data (S150). Thereafter, the GNSS module correction may be performed (S160), and the settlement detection data information may be transmitted to a remote server (S170). Finally, it is determined whether the power is terminated (S180). While repeatedly performing such a process, the GNSS building settlement detector according to an embodiment of the present invention may accumulate settlement detection data.

In this regard, if GPS data can be received, the controller 120 may control the received GPS data to be stored in a data storage unit inside the settlement detector. On the other hand, if it is determined that communication with a remote server is possible, the controller 120 can acquire precise subsidence detection data by improving GPS accuracy using GNSS module and Networked Transport of RTCM via Internet (NTRIP) correction information from GPS data. have. In addition, the control unit 120 may transmit the precise subsidence detection data to the server through the communication module.

The building condition diagnosis system according to an embodiment of the present invention is to accurately measure GPS information of a building, that is, latitude, longitude, and altitude information, through a monitoring system using a GNSS building settlement detector, and monitor it in real time through the Internet.

In addition, by using the artificial intelligence model to determine the condition of the building, such as the accurate floating settlement of the building, more quickly by using the GPS information measured through this, according to the trend of the measured value, it is possible to take a quicker action, thereby minimizing personal injury.

As shown in FIG. 5, the GNSS building settlement detector may measure the corrected GPS information through a wired/wireless communication module. Through this GPS information, it is possible to measure a measurement value with an error of less than 2cm.

Meanwhile, referring to FIG. 4, the GNSS building settlement detector 100 generates an orbit pattern using the received GPS information, and uses the generated orbit pattern to check the presence or absence of an abnormality in the current settlement detector. It may include a control unit 120 to determine.

In addition, the GNSS building settlement detector 100 may further include a data storage unit 130 for storing the received sensing information and the presence or absence of an abnormality in the settlement detector, and a display unit for providing the user with an abnormality in the settlement detector.

In this regard, the controller 120 may generate an orbit pattern using GPS data, and determine whether or not there is an abnormality in the settlement sensor sensor using the generated orbit pattern. The data storage unit 130 may be configured to store GPS data and whether there is an abnormality in the sinking sensor sensor. In this regard, the sinking sensor sensor may be a sensor of a corresponding sinking sensor or another sinking sensor equipped with the control unit 120. The display unit may be configured to display on the screen whether there is an abnormality in the sinking sensor sensor.

The data inference unit may be included in the control unit 120 or separately from the control unit 120. The data inference unit may learn a machine learning algorithm by converting vibration data in three directions (latitude, longitude, altitude) measured from a sinking sensor sensor into an orbit pattern using the seq2seq algorithm.

The controller 120 may learn a machine learning algorithm by converting the three-way vibration data of latitude, longitude, and altitude measured from the sinking sensor sensor into an orbit pattern.

When the orbit pattern is generated in a circular shape, the controller 120 may determine that there is no abnormality in the building.

Meanwhile, when the orbit pattern is generated in any one of an oval shape, a nasolabial shape, a heart shape, and a tornado shape, the controller 120 may determine that an abnormality exists in the building. In addition, when the orbit pattern is generated in the elliptical shape, the control unit 120 may determine that the building is out of balance. In addition, when the orbit pattern is generated in any one of the nasolabial shape, the heart shape, and the tornado shape, the control unit 120 may determine that the axis of the building is shifted.

Another embodiment of the present invention relates to a method for diagnosing floating settlement using a settlement detector of a building. Specifically, in the floating settlement diagnosis method, the step of receiving the sensed information in real time using a settlement detector installed inside a smart building and a settlement detector installed on the roof, and applying the received sensing information to a machine learning algorithm. Generating a pattern and determining the presence or absence of an abnormality in the current building (floating settlement, etc.) using the generated orbit pattern, determining the presence or absence of the received sensing information and the state of the building, and the received sensing It may include storing information and whether there is abnormality in the building, and providing information about whether there is abnormality in the building to a user.

Specifically, the control unit 120 may perform a step of receiving sensing information sensed in real time using a settlement detector installed inside a plurality of smart buildings and a settlement detector installed on the roof of each building. The controller 120 may generate an orbit pattern by applying the received sensing information to a machine learning algorithm, and may perform a step of determining whether or not there is an abnormality in a building associated with the floating settlement using the generated orbit pattern.

The controller 120 may perform the step of determining whether the received sensing information and the state of the building are abnormal. The controller 120 may perform the step of storing the received sensing information and whether the building is abnormal. The control unit 120 may perform the step of providing information on whether the building is abnormal through a display unit to a user.

On the other hand, Figure 6 shows the measured data measured by the sinking sensor according to the present invention. As shown in FIG. 6, GPS information can be measured in real time, and such data can be converted into an orbit pattern. Specifically, FIGS. 6 (a), (b) and (c) show changes in latitude, longitude and altitude with time, respectively.

When the shape of the orbit pattern for determining the presence of an abnormality is generated as an orbit pattern, which is measured when equal settlement, floating settlement, or conduction settlement occurs, it may take a specific shape. On the other hand, the nasolabial shape and the heart shape of the orbit pattern are only one example.

The orbit pattern is generated by an algorithm, and a picture map is generated using the orbit pattern. The artificial intelligence model learns such a picture map, and through the learning, it diagnoses the state of the measurement information of various settlement detectors.

On the other hand, Figure 7 is a flow chart showing the presence or absence of an abnormality detection method using the sinking detector of the building according to the present invention. 1A to 7, the control unit 120 may perform an operation S210 of receiving the measured settlement information using a settlement sensor sensor. The controller 120 generates an orbit pattern by applying the settlement information to an orbit generation algorithm, applies the received sensing information to a machine learning algorithm to generate an orbit pattern, and uses the generated orbit pattern. A step (S220) of determining whether there is an abnormality in the building associated with the floating settlement may be performed. The control unit 120 may perform a step (S230) of storing the received sensing information and the presence or absence of an abnormality in the state of the building. The control unit 120 may perform the step (S240) of providing information on whether or not the building is abnormal to the user through a display unit.

Meanwhile, FIG. 8 is a flowchart showing a procedure for learning and diagnosing a machine learning model for seismic detection according to the present invention.

Referring to FIGS. 1A to 8, the controller 120 may perform an operation S310 of receiving measured settlement information using a settlement sensor. The controller 120 may apply the settlement information to an orbit generation algorithm to generate an orbit pattern (S320). The control unit 120 generates a picture map, which is machine learning learning data (S330), labeling the picture map (S340), and performs learning on a machine learning learning model (S350), and the measured settlement Generating a machine learning model for determining information (S360) may be performed. The control unit 120 may perform the step S370 of determining a sign of a building abnormality according to the measured settlement information using the machine learning model. When the abnormal symptom is determined, the controller 120 may perform a step S380 of notifying the abnormal symptom to the manager terminal.

Therefore, as shown in FIG. 8, it is possible to determine whether there is an abnormality in the settlement information measured in real time by using the settlement sensor information learned through a machine learning model.

Meanwhile, FIG. 9A is an illustration of a method of converting to an orbit pattern according to the present invention. In addition, Figure 9b shows the orbit shape according to the state and class according to the present invention. Referring to FIG. 9B, the orbit pattern may have various shapes as shown in FIG. 9B. Meanwhile, the shape of FIG. 9B is only an exemplary embodiment, and the shape may be modified depending on the exemplary embodiment.

As described above, the settlement detector may be installed in a building as shown in FIG. 10.

Referring to FIG. 10, in the case of an apartment, such a building may have the shape of such a building, and a settlement detector may be installed on the roof. In this case, the values of x ₁ to x ₄ , y ₁ to y ₄ , and z ₁ to z ₄ may be measured through a settling detector.

In such a building structure, information of latitude, longitude, and altitude is measured through a settlement detector. In this case, the method of determining the type of settlement is as follows.

Referring to FIG. 10, the initial values measured by the sinking detector at each of the four corner points are k _i1 (x _i1 , y _i1 , z _i1 ), k _i2 (x _i2 , y _i2 , z _i2 ), k _i3 (x _{Assume that i3} , y _i3 z _i3 ) and k _i3 (x _i3 , y _i3 , z _i3 ).

Current measurements for each corner point are k ₁ (x ₁ , y ₁ , z ₁ ), k ₂ (x ₂ , y ₂ , z ₂ ), k ₃ (x ₃ , y ₃ , z ₃ ), k ₄ ( Suppose x ₄ , y ₄ , z ₄ ).

At this time, the equal settlement by altitude can be determined by the following algorithm.

|x ₁ | = (x _i1 -x ₁ ) represents the value. That is, it means the difference between the initial value and the current value. |x ₂ -x ₁ | means the difference in latitude between the location k ₂ and the location k _{1 , and |y 2} -y ₁ | means the difference in the longitude between the location k ₂ and the location k _1. Also, |z ₂ -z ₁ | means the difference in elevation between the position k ₂ and the position k _1.

On the premise of this, the average amount of change for each latitude, longitude, and altitude is as follows.

Latitude x = (|x _i1 -x ₁ |+|x _i2 -x ₂ |+...+|x _i4 -x ₄ |)/4, longitude longitude y = (|y _i1 -y ₁ |+| y _i2 -y ₂ |+...+|y _i4 -y ₄ |)/4, altitude Altitude z = (|z _i1 -z ₁ |+|z _i2 -z ₂ |+|z _i3 -z ₃ | It can be expressed as +|z _i4 -z _{4 |)/4.}

Latitude x refers to measuring the difference in how much current latitude information has changed compared to the latitude information measurement value during initial installation. Longitude y refers to measuring the difference in how much current hardness information has changed compared to the hardness information measurement value at the time of initial installation. Altitude z refers to measuring the difference in how much the current altitude information has changed compared to the altitude information measured during initial installation.

The value of ∝ to be described below is a constant value for the error, and the value may vary slightly depending on latitude, longitude, and altitude, and this value can be determined by experimental values.

If the latitude x and longitude y values for equal settlement are less than ∝ and the altitude z is greater than ∝, it can be determined as floating settlement. That is, if latitude x <∝ and longitude y <∝ and altitude z> ∝, it is determined as floating settlement.

Also, |z ₁ -z ₂ | If> ∝ and (|z ₁ |> ∝ && |z ₂ |> ∝), it can be judged as equal settlement.

If Latitude x <∝ and (Altitude z> ∝ && (|z ₁ |+|z ₃ |)/2 <∝ && (|z ₂ |+|z ₄ |)/2> ∝) Latitude x <∝, longitude y <∝, (Altitude z> ∝ and (|z ₂ |+|z ₄ |)/2 <∝, and (|z ₁ |+| If z ₃ |)/2> ∝ ), it can be judged to be overturned.

Floating sinking (|x ₂ -x ₁ | <|x _i2 -x _i1 |), (|x ₄ -x ₃ | <|x _i4 -x _i3 |), ((|z ₂ -z ₁ | <| If z _i2 -z _i1 |) and (|z ₄ -x ₃ | <|z _i4 -z _i3 |), it can be determined as immobile settlement.

As described above, immobile settlement, overturning settlement, and equal settlement may be determined through the above determination algorithm.

In the above, a real-time artificial intelligence-based building condition diagnosis system according to various embodiments of the present invention has been described. Meanwhile, the technical effects of the real-time artificial intelligence-based building condition diagnosis system according to various embodiments of the present invention are as follows.

According to an embodiment of the present invention, there is an advantage that it is possible to provide a building condition diagnosis system.

In addition, according to at least one embodiment of the present invention, there is an advantage that it is possible to provide a real-time artificial intelligence based building condition diagnosis system.

On the other hand, when an earthquake occurs, the building collapses due to the earthquake damage of the building, resulting in a lot of human and material damage. Therefore, according to at least one embodiment of the present invention, it is expected that when applied to a building, it is possible to respond more quickly by diagnosing the safety level of the building through artificial intelligence analysis based on accumulated data in real time.

The features and effects of the present invention described above will become more apparent through the following detailed description in connection with the accompanying drawings, and accordingly, those of ordinary skill in the technical field to which the present invention pertains can easily implement the technical idea of the present invention. I will be able to.

In the present invention, various modifications may be made and various embodiments may be provided, and specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to a specific embodiment, it should be understood to include all changes, equivalents, or substitutes included in the spirit and scope of the present invention.

According to the software implementation, not only the procedures and functions described in the present specification, but also design and parameter optimization for each component may be implemented as a separate software module. The software code can be implemented with a software application written in an appropriate programming language. The software code may be stored in a memory and executed by a controller or a processor.

Claims (7)

In the settlement detector of the real-time artificial intelligence-based building condition diagnosis system,
A GNSS module configured to receive GPS data; And
And a control unit configured to be operatively coupled with the GNSS module to control the GNSS module,
The control unit,
Collecting the GPS data and storing the collected GPS data in a server or a data storage unit,
Firstly determine whether or not settlement has occurred based on the stored GPS data,
When it is determined that the subsidence has occurred, a control unit configured to determine the subsidence type and the degree of occurrence of the building using an artificial intelligence model according to the trend of the determination result and the measured value of the collected GPS data,
The control unit,
If GPS data can be received, the received GPS data is controlled to be stored in the data storage unit inside the settlement detector,
When it is determined that communication with a remote server is possible, precise settlement detection data is obtained from the GPS data by improving GPS accuracy using the GNSS module and Networked Transport of RTCM via Internet (NTRIP) correction information,
A settlement detector for transmitting the precise settlement detection data to the server through a communication module. delete The method of claim 1,
The control unit,
An orbit pattern is generated using the GPS data, and an abnormality of the sinking sensor sensor is determined using the generated orbit pattern,
The settlement detector further comprises a data storage unit configured to store the GPS data and the presence or absence of an abnormality in the settlement sensor sensor,
The control unit,
A sinking sensor for controlling to display the presence or absence of an abnormality in the sinking sensor sensor on a screen of the display unit. The method of claim 1,
The control unit,
The machine learning algorithm is trained by converting the three-way vibration data of latitude, longitude, and altitude measured from the sinking sensor sensor into an orbit pattern.
When the orbit pattern is generated in a circular shape, it is determined that there is no abnormality in the building. The method of claim 4,
The control unit,
If the orbit pattern is generated in any one of an oval, a nasolabial shape, a heart shape, and a tornado shape, it is determined that an abnormality exists in the building,
If the orbit pattern is generated in the elliptical shape, it is determined that the balance of the building is not correct,
When the orbit pattern is generated in any one of the nasolabial shape, a heart shape, and a tornado shape, it is determined that the axis of the building is shifted. The method of claim 1,
The control unit,
It receives sensing information in real time by using a settlement detector installed inside a plurality of smart buildings and a settlement detector installed on the roof of each building,
Applying the received sensing information to a machine learning algorithm to generate an orbit pattern, and using the generated orbit pattern to determine whether there is an abnormality in a building associated with the floating settlement,
Store the received sensing information and whether there is an abnormality in the building in a data storage unit,
A sinking detector that provides information on whether or not the building is abnormal on a screen through a display unit. The method of claim 1,
The control unit,
Receive the measured settlement information using the settlement sensor sensor,
Applying the settlement information to an orbit generation algorithm to generate an orbit pattern,
Generate a picture map, which is machine learning learning data, label the picture map, perform training on a machine learning learning model, and generate a machine learning model to determine the measured settlement information,
Using the machine learning model to determine the signs of building abnormalities according to the measured settlement information,
When the abnormality symptom is determined, the subsidence detector notifies the abnormality symptom to the manager terminal.

Images