KR102260783B1 - Measuring device for predicting collapse of slope land and system for predicting collapse of slope land with the device - Google Patents

Abstract

The present invention relates to a slope collapse risk prediction system. An embodiment of the present invention provides the slope collapse risk prediction system, which includes: a plurality of measuring instruments installed spaced apart from each other on a slope; and a server device which receives measurement data from the plurality of measuring instruments and calculates a risk degree of a slope collapse risk, wherein the server device is configured to perform calculating a primary risk for a collapse risk based on acceleration-related measurement data applied to each measurement device among the received measurement data, and calculating a secondary risk for the collapse risk based on the received measurement data when the primary risk is greater than or equal to a preset threshold. Therefore, it is possible to evaluate and predict the risk of slope collapse with high accuracy.

Description

Measuring device for predicting collapse of slope land and system for predicting collapse of slope land with the device

The present invention relates to a system for predicting the risk of landslides or collapse on a slope or a cut slope, etc., and more particularly, to a system for predicting the risk of collapse on a slope using acceleration, acoustic waves and moisture content. will be.

Accidents in which the soil or the entire slope collapses, such as steep slopes, cut slopes, and landslides, is one of the frequent disasters. Most of the landslides and steep slope collapse detection instruments in Korea detect the collapse after it has occurred and alert the manager and the area around the dangerous area. The advanced technology of measuring instruments is developing, but analysis technology is not included, so it is built as a system for post-analysis.

Although efforts have been made to predict the collapse by detecting and measuring the risk of landslide or slope collapse in advance, it is difficult to predict the risk in real time by installing it in an actual hazardous area because the prediction system is complicated or expensive. the current situation.

Prior Document 1: Korean Patent Publication No. 2018-0057817 (published on May 31, 2018) Prior Document 2: Korean Patent Registration No. 10-2091758 (Announced on March 20, 2020)

The present invention has been devised to solve this problem, and it is an object of the present invention to provide a system that measures the acceleration to interpret the impact amount and the moving average value, and determines the grade for the risk of landslide collapse by using the moisture content and acoustic wave prediction technique. .

The present invention calculates the primary risk by measuring and interpreting the impact amount and the moving average (change value of the impact amount) of the impact amount through the acceleration measurement, further considering the slope, moisture content, and acoustic wave parameters of the measuring instrument, and artificial neural network algorithm It provides a method for real-time prediction of the risk of slope collapse by calculating the secondary risk using

According to an embodiment of the present invention, there is provided a system for predicting the risk of collapse of a slope, comprising: a plurality of measuring instruments installed at a predetermined interval on a slope; and a server device that receives measurement data from the plurality of measuring instruments and calculates a degree of risk for slope collapse risk, wherein the server device includes, on the basis of the measurement data related to acceleration applied to each measuring device among the received measurement data Calculating a primary risk for collapse risk, and calculating a secondary risk for collapse risk based on the received measurement data when the primary risk is greater than or equal to a preset threshold value It provides a system for predicting the risk of slope collapse.

According to the present invention, the primary risk is determined based on the amount of impulse caused by the acceleration and the moving average value, and then the amount of impulse, the moving average, the slope of the instrument, the moisture content, and the acoustic wave parameters are inputted into the artificial neural network algorithm as input data to enter the secondary risk. By calculating , it is possible to evaluate and predict the risk of slope collapse with high accuracy.

1 is a view showing a state in which a number of instruments are installed on a slope for predicting the risk of collapse of a slope;
2 is a view for explaining a measuring instrument according to an embodiment of the present invention;
3 is a flowchart of a method for predicting the risk of slope collapse according to an embodiment;
4 is a flowchart of an exemplary method for calculating a primary risk using acceleration;
5 is a diagram exemplarily showing variables used for calculating secondary risk and risk classification accordingly;
6 is a view for explaining the parameters of the acoustic wave;
7 is a diagram illustrating an exemplary method of calculating a secondary risk using an artificial neural network.

The above objects, other objects, features and advantages of the present invention will be easily understood through the following preferred embodiments in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed subject matter may be thorough and complete, and that the spirit of the present invention may be sufficiently conveyed to those skilled in the art.

In the drawings of the present specification, numerical values such as length, thickness, and width of components may be exaggerated for effective description of technical content. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase. The expressions 'comprising', 'consisting of', and 'consisting of' as used in the specification do not exclude the presence or addition of one or more other elements in addition to the elements mentioned in this expression.

Hereinafter, the present invention will be described in detail with reference to the drawings. In describing the specific embodiments below, various specific contents have been prepared to more specifically explain and help the understanding of the invention. However, a reader having enough knowledge in this field to understand the present invention may recognize that it may be used without these various specific details. In some cases, it is mentioned in advance that parts which are commonly known in describing the invention and which are not largely related to the invention are not described in order to avoid confusion in describing the present invention.

1 schematically shows a state in which a plurality of measuring instruments are installed on a slope for predicting the risk of collapse of a slope. As shown, in order to perform the collapse risk prediction method of the present invention, a plurality of measuring instruments 10 are installed on the ground surface of the slope to be predicted. The measuring instrument 10 may be installed to be spaced apart from each other at predetermined intervals on the slope of the collapse risk prediction target area.

Each measuring instrument 10 is configured in a cylindrical case, for example, and is provided with various sensors such as an acceleration sensor and a sound wave sensor in the case. In addition, the measuring instrument 10 may further include a power supply for supplying power to these various sensors, a data processing circuit for processing data measured by the various sensors, and a transmitter for transmitting the processed data to the repeater 20 .

In this regard, Fig. 2 shows a measuring instrument 10 according to an embodiment of the present invention. Fig. 2 (a) is an external perspective view, and Fig. 2 (b) is a side view, in which a part of the interior is shown as a side cross-section. Referring to FIG. 2 , the instrument 10 according to an embodiment includes a header 11 , an upper housing 12 , a lower housing 13 , and a lower weight 14 .

The upper housing 12 and the lower housing 13 are cylindrical members having an empty space therein. A PCB board 16 is accommodated in the upper housing 12 , and various types of sensors such as an acceleration sensor and a sound wave sensor and a data processing circuit may be mounted on the PCB 16 . The acceleration sensor may measure the acceleration applied to the instrument 10 for a predetermined period of time, and the acoustic wave sensor may measure an acoustic wave (acoustic acoustic wave) inside the earth's surface. An empty space is formed inside the lower housing 13, and this space serves as a wave guide for amplifying sound waves and guiding them to the sound wave sensor.

Between the lower housing 13 and the lower weight 14, a moisture content sensor 15 for measuring the amount of moisture (moisture content) in the ground surface is interposed and installed. Since the moisture content sensor 15 is separated from the PCB board 16 , the moisture content sensor 15 and the PCB board 16 are connected by a cable 17 , and through this, the measured data of the moisture content sensor 15 is transferred to the PCB board 16 . can be transmitted The header 11 is a member that seals the upper housing 12, and a connector pin electrically connected to the PCB board 16 is formed at the upper end, and a cable is connected to the connector pin to communicate with the outside by wire.

The data processing circuit mounted on the PCB board 16 may receive measurement values from various sensors, such as an acceleration sensor, a moisture content sensor, and a sound wave sensor, and generate measurement data from the measurement values. In one embodiment, the measurement data includes at least (i) the amount of change (I) in the amount of impulse applied to the instrument 10 based on the amount of change in acceleration, (ii) a moving average of the amount of change over a predetermined time (Ima), (iii) the instrument It may include the slope of (10), (iv) moisture content inside the sloping ground surface, and (v) the number of acoustic wave peaks for a predetermined time, and each measurement data will be described later.

In one embodiment, the measuring instrument 10 and the repeater 20 may be connected by wire. One repeater 20 may be connected to each measuring instrument 10 , or several instruments 10 may be connected to one repeater 20 . The repeater 20 may transmit the measurement data received from one or more measuring instruments 10 to a server device (not shown) through any wireless communication network.

The server device may manage the landslide risk section of a certain area as a unit risk area. For example, the server device may receive the measurement data from the measuring instruments 10 installed in the danger area of one unit and calculate the risk level for the risk of collapse of the slope in the danger area.

In one embodiment, it is also possible to calculate the degree of risk for the slope collapse risk for each measuring instrument (10). For example, it may be implemented in the data processing circuit of each instrument 10 . Alternatively, in another alternative embodiment, the method for predicting the collapse risk of the present invention may be executed in the repeater 20 connected to each instrument 10, or the repeater 20 transmits the measurement data of each instrument 10 to the server device. It can also be transmitted wirelessly or by wire to the server and the prediction method of the present invention can be executed on the server device.

In the following embodiment, it is assumed that the server device receives the measurement data from one or more measuring instruments 10 and executes the method for predicting the risk of slope collapse in the server device based on this, and the embodiment of the present invention will be described.

3 is a flowchart of a method for predicting the risk of slope collapse according to an embodiment of the present invention. Referring to the drawings, when the server device receives measurement data from one or more measuring instruments 10 , a primary risk is calculated based on the measured data in step S10 . In an embodiment, the server device may calculate the primary risk for the collapse risk based on the acceleration-related measurement data applied to each instrument among the measurement data, and in this case, the acceleration-related measurement data is based on the amount of change in the acceleration. ) includes the amount of change (I) of the amount of impact applied to it and the moving average (Ima) of the amount of change for a predetermined time.

Thereafter, in step S20, it is determined whether the primary risk calculated in step S10 is equal to or greater than a predetermined threshold value. If the primary risk is less than the threshold value (S20_No), since there is no risk of collapse, it is possible to return to the step S10 again without calculating the secondary risk and execute the step S10 again after a predetermined period.

In relation to the threshold, FIG. 5 is an exemplary diagram illustrating the risk criterion used when calculating the risk using the acceleration-related measurement data. The values shown in FIG. 5 are exemplary values, and it goes without saying that the reference values may vary depending on specific embodiments or implementation situations. In one embodiment, the server device determines the primary risk using the amount of change (I) of the amount of impact applied to the measuring instrument (10) and the moving average (Ima) of the amount of change for a predetermined time (S10) and threshold the primary risk Compare with value. At this time, for example, referring to FIG. 5, the threshold value of the amount of change (I) of the impulse can be set to 100 and the threshold value of the moving average (Ima) can be set to 25, and two measured values (that is, the amount of change of the impulse (I) and moving average (Ima)), if both are below each threshold, it is determined that the risk is 'safe' and returns to step S10 again, and if any of the two measured values is greater than each threshold, the risk may be more than 'interest' It may proceed to step S30.

As such, if it is determined in step S20 that the primary risk is greater than or equal to the threshold (S20_Yes), since there is a risk of slope collapse, in this case, the process proceeds to step S30 to calculate the secondary risk. In one embodiment, the server device includes the measurement data received from the measuring instrument 10, that is, (i) the amount of change in the amount of impulse (I), (ii) the moving average of the amount of change (Ima), (iii) the slope of the measuring instrument 10, The secondary risk can be calculated based on (iv) the moisture content inside the ground surface of the slope, and (v) the number of acoustic wave peaks for a predetermined time. In an embodiment, the server device may calculate the secondary risk using an artificial neural network algorithm.

As a result of calculating the secondary risk, for example, as shown in FIG. 5, the risk is classified into five levels, namely 'safe', 'interest', 'caution', 'danger', and 'alarm' level, and belongs to any of these levels. It may be a way to indicate For example, when calculating the secondary risk using an artificial neural network, the artificial neural network uses the measurement data of (i) to (v) as input data, predicts where the risk belongs to among the five levels, and outputs it as output data. (classification) It is an algorithm that solves a problem.

As such, if the secondary risk is calculated in step S30, a predetermined subsequent operation may be performed according to the calculated risk level. For example, according to the level of risk, it can be transmitted to a higher level server device such as an integrated control system to comprehensively manage the level of risk in real time.

4 shows a specific and exemplary method of calculating the primary risk by using the acceleration in FIG. 3 ( S10 ). Referring to FIG. 4 , in step S110 , variables and various parameters used in the method for predicting the risk of slope collapse of the present invention are initialized. Here, the variable and various parameters may include, for example, the amount of change (I) of the impulse amount, the moving average (Ima) of the change amount, the start time of a timer for measuring a time period, and the like, but is not limited thereto.

After initializing these variables and various parameters in step S110, the acceleration sensor of the measuring instrument 10 measures the acceleration for a predetermined time period. In general, the fundamental force involved in the risk of collapse on a slope is gravity, and rainfall influenced by gravity, falling rocks, and changes in underground rocks due to earthquakes are involved in the collapse. Free-fall motion is caused by gravity. This is called gravitational acceleration. The gravitational acceleration (9.8m/s2) is the sum of the x-axis, y-axis, and z-axis acceleration components, and is expressed in Equation 1 as follows.

--- 수식1

--- Formula 1

The meaning of the above formula is that the sum of the accelerations acting in the three axial directions is 9.8 m/s2, regardless of the position where at least an object is placed and the degree of twisting, and this is expressed as 1G. Therefore, the changed acceleration value in a place that can lead to a disaster by external forces (earthquake, weathering, gravity, etc.) becomes a very important factor in determining the safety condition.

The changed acceleration value can be converted into an impulse (force), and since it is precisely proportional (in F = ma, the relationship between F and a is proportional), the risk of collapse can be defined by judging by the changed value of the acceleration, that is, step (S120) ), as shown in Equation 2 below, it is possible to calculate the difference between the acceleration values of the times t1 and t2 before and after a predetermined time period, and convert this to the change amount (I) of the amount of impact applied to the measuring instrument 10 .

--- 수식2

--- Formula 2

In Equation 2 above, ax1, ay1, and az1 are the accelerations of each component at the time period start time (t1), and ax2, ay2, az2 indicate the acceleration values of each component at the time period end time (t2), respectively, and the acceleration sensor is The acceleration value of each component in Equation 2 above can be measured.

In one embodiment, the time period is, for example, 0.1 seconds. That is, the change amount (I) can be calculated according to Equation 2 above by measuring the acceleration value around 0.1 second. In another embodiment, the time period may be 0.01 seconds. That is, it is possible to measure the acceleration value around 0.01 second and calculate the change amount (I) according to Equation 2. In this case, it is also possible to obtain the change amount (I) for 0.1 second by repeating the measurement of the acceleration value every 0.01 second 10 times and obtaining the arithmetic mean.

Next, in step S130 , a moving average Ima of the change amount I is calculated using the change amount I calculated in step S120 . "Moving average (Ima) of change amount I" is an average of change amount I for a predetermined time. For example, when calculating change amount I by the difference in acceleration values for 0.1 second, this change amount ( By calculating I) 10 times, the moving average (Ima) for 1 second can be obtained.

In one embodiment, N (N is an integer greater than or equal to 3) is obtained for each predetermined time period (hereinafter referred to as “interval average”) for the change amount I, and a moving average (Ima) of the N number of interval averages is obtained. Calculate. At this time, the moving average (Ima) is calculated as the arithmetic average of the [1st to (N-1)th moving average] and the Nth interval average, and the [1st to (N-1)th moving average] ] can be calculated as the arithmetic average of the [1st to (N-2)th moving average] and the (N-1)th interval average, and this process is [1st to 2nd moving average] and the third interval average are repeated until the arithmetic average is achieved.

For example, in one embodiment, if the amount of change (I) every 0.01 seconds is obtained 10 times and the arithmetic average is obtained, an 'interval average' of the amount of change (I) for 0.1 seconds can be obtained, and this interval average is obtained 10 times. are calculated continuously. That is, 10 interval average values for 0.1 second are obtained for the amount of change (I), and the moving average (Ima) is obtained using the 10 interval average values.

At this time, in order to obtain a moving average from the 10 interval average values, first and second interval averages are calculated as the arithmetic average of the 10 interval averages (when defined as the “first to tenth” interval averages from the temporally preceding order). This is determined as the first moving average, and then the first moving average and the third section average are arithmetic averaged to determine the second moving average. Similarly, the second moving average and the fourth section average are arithmetic averaged to determine the third moving average, and by repeating this calculation, the 8th moving average and the tenth section average are arithmetic averaged to calculate the final moving average (Ima). have. By calculating the moving average (Ima) of the amount of change (I) in this way, the change trend of external forces (earthquake, weathering, gravity, etc.) on the slope can be known, so that the accuracy of the slope collapse risk prediction can be improved.

As described above, after calculating the amount of change I and the moving average Ima in steps S120 and S130, respectively, the primary risk is calculated in step S140. For example, as described above with reference to FIG. 5 , the risk level of the primary risk may be determined according to exemplary values of the impulse change amount I and the moving average Ima shown in FIG. 5 .

In an embodiment, the risk level classification of FIG. 5 may be used when calculating the secondary risk. In the table of FIG. 5, the vertical direction indicates the risk level. In FIG. 5, the risk level is classified into five levels of risk by way of example. Among the grades of the risk level, "safe" means that the current state is a stable state, and "interest" may be a situation in which a slight deformation occurs on a slope or a rockfall occurs, for example. In this state, it may be difficult to observe with the naked eye, it may be due to the movement of animals, and continuous observation and investigation are required. "Caution" of the risk level is, for example, a situation in which surrounding damage is expected due to strong wind or rain, and may be a situation in which general investigation and action are required. The "danger" of the risk level may be, for example, a situation in which it is determined that the slope is highly deformable or a rockfall occurs close to the measuring instrument 10, and may be a state that requires a detailed investigation, and the "alarm" of the risk level is serious on the slope It is a situation in which a problem may occur or the slope may collapse, and a detailed investigation and prompt response are required. Classifying the risk level into five steps is exemplary, and the classification step or the definition of each level may vary according to specific embodiments.

In the table of FIG. 5 , the horizontal direction indicates five pieces of measurement data. Among the measurement data, “angle change” is the inclination of the measuring instrument 10 and indicates how much the measuring instrument 10 is tilted compared to the first installed state. An accelerometer or gyro sensor can be used to calculate the angle change (slope). When an acceleration sensor is used, the angle of inclination with respect to each axis can be obtained using a trigonometric function from the acceleration value of each component of the acceleration vector. For example, as shown in Equation 3 below, the angle of inclination with respect to each axis is compared with the angle at the time of initial installation of the measuring instrument 10 to obtain the intermediate inclination ф of the measuring instrument 10 .

--- 수식3

--- Equation 3

In Equation 3 above, θx, θy, and θz are the currently measured angles, and θxi, θyi, and θzi are the angles when the instrument is initially installed.

Thereafter, the long-term alarm risk is calculated using the slope increment (ф) calculated in Equation 3 in step S170. At this time, for example, the long-term alarm risk may be determined according to the risk standard shown in FIG. 5 . For example, if the slope increment ф is less than 10 degrees, it may be determined as a 'safe' level, and if it is between 11 and 30 degrees, it may be determined as a 'interest' level, respectively.

Acoustic acoustic waves (also simply referred to as 'acoustic waves' or 'elastic waves') among the measurement data may be measured by the acoustic wave sensor of the measuring instrument 10 . Before the occurrence of landslides and slope collapses, the amount and frequency of acoustic waves tend to increase rapidly, and since a certain range around the sensor is measured, it is useful for capturing preliminary signs of landslides in a wider range than the existing measurement methods. can be used

In general, an aura occurs in which the number of occurrences of a destructive sound increases just before the collapse of a slope, and at this time, the largest acoustic wave energy is generated. For example, FIG. 6 shows exemplary characteristic parameters of the acoustic wave. As shown in FIG. 6 , characteristic parameters such as maximum amplitude, rise time, frequency, number of peaks, duration Td, and energy E may be extracted from one measured acoustic wave waveform. As shown in FIG. 6 , when measuring acoustic waves, a voltage greater than the voltage level of background noise is generally set as a threshold value At, and a waveform exceeding the threshold value At is detected as an effective acoustic wave. The maximum amplitude of the acoustic wave means the maximum value among the amplitudes of the acoustic wave, the rise time is the time taken from the start of the acoustic wave to the maximum amplitude, and the number of peaks is the number of peaks in the acoustic wave per minute. to be. The frequency may be calculated as the number of peaks per unit time. The duration Td is the time from the start to the extinction of the acoustic wave, and the energy E may mean an area of an envelope of the acoustic wave. In an embodiment of the present invention, the number of peaks is used as the measurement data. That is, the number of acoustic wave peaks measured for a predetermined time (eg, 1 minute) is used as measurement data.

7 shows an artificial neural network algorithm for calculating secondary risk according to an embodiment. Referring to the drawings, in an embodiment, the artificial neural network includes an input layer (Lin), one or more hidden layers (Lh1, Lh2), and an output layer (Lout). The input layer Lin has five input nodes each having the above-described measurement data of (i) to (v) as input variables. In the illustrated embodiment, the artificial neural network has two hidden layers Lh1 and Lh2, and each hidden layer is composed of four hidden nodes. The output layer Lout has five output nodes, and each output node corresponds to five risk levels of reassurance, concern, caution, danger, and warning for slope collapse risk, respectively.

The arrow connecting each input node and the hidden node, the arrow connecting each hidden node and the hidden node, and the arrow connecting each hidden node and the output node mean weight and bias. And weight and bias are determined by learning by learning data. In one embodiment, according to the risk level classification criterion of FIG. 5 , past measurement data and correct answers (ie, risk level) for each measurement data may be prepared and learned in advance as learning data.

As described above, each of the input nodes of the input layer Lin corresponds to each measurement data (ie, the amount of change in the amount of impact, the moving average of the amount of change, the slope, the moisture content, and the number of elastic wave peaks). In an embodiment, the measurement value of each measurement data may be directly input to each input node. However, in an alternative embodiment, the measurement data is pre-processed and then input to each input node. For example, such preprocessing may be, for example, unifying measurement units or normalizing so that the mean is 0 and the standard deviation is 1. The effect of the unit difference, the effect of the difference of the maximum and minimum values, the effect of the average value difference, the effect of the distribution difference, etc. are irrelevant to the essence of the data, and in order to learn the weights and biases in the artificial neural network, It is desirable to remove the factors that hinder learning while leaving only the essence of

Meanwhile, in an embodiment, rainfall information and geological characteristics may be applied to an artificial neural network model. For example, as rainfall information, weights may be given to gradually increase the risk according to the rainfall of 30 mm, 40 mm, 60 mm, and 80 mm per hour. In this case, for example, according to the rainfall information, at least one input data value among five input data can be input to the input node by increasing or decreasing . Similarly, in the case of geological properties, weights to increase the risk gradually depending on which of the soil (sandy soil, cohesive soil), ripping rock (weathered rock), soft rock (blast rock), and hard rock (blast rock) on which the measuring instrument 10 is installed , and even in this case, for example, the input data value of at least one of the five input data may be increased or decreased according to the geological characteristics and input to the input node.

In one embodiment, each output node of the output layer Lout corresponds to five risk levels of relief, concern, caution, danger, and warning for the risk of slope collapse, for example, depending on which output node has the highest output value. One of the five risk levels is determined. In an alternative embodiment, the risk level may be output as a probability by applying a Softmax activation function to the rear end of the output layer Lout. In this case, the sum of all output values of the five output nodes becomes 1, and the largest output value becomes the secondary risk level calculated for the input data, and the output value at this time means the probability of the risk level.

As described above, those of ordinary skill in the art to which the present invention pertains can understand that various modifications and variations are possible from the description of this specification. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the following claims as well as the claims and equivalents.

10: Instrument 11: Header
12: upper housing 13: lower housing
14: lower weight 15: moisture content sensor
16: PCB board 17: cable
20: repeater

Claims (5)

As a slope collapse risk prediction system,
A plurality of measuring instruments installed at a predetermined interval on a slope; and
A server device that receives measurement data from the plurality of measuring instruments and calculates a degree of risk for slope collapse risk;
The server device, the step of calculating a primary risk for the collapse risk based on the acceleration-related measurement data applied to each measurement among the received measurement data, and the reception if the primary risk is greater than or equal to a preset threshold configured to execute a step of calculating a secondary risk for a collapse risk based on one measurement data,
The measurement data includes (i) the amount of change in the amount of impact applied to the instrument based on the change in acceleration (I), (ii) a moving average of the amount of change over a predetermined time (Ima), (iii) the inclination of the instrument, (iv) ) including the moisture content in the interior of the inclined land surface, and (v) the number of acoustic wave peaks for a predetermined time,
The server device is configured to calculate the secondary risk using an artificial neural network algorithm,
The artificial neural network algorithm includes: an input layer having five input nodes, each using the measurement data of (i) to (v) as input variables; one or more hidden layers composed of a plurality of nodes; and an output layer having a plurality of output nodes corresponding to each level of the risk level for the slope collapse risk;
When the input variable is input to the artificial neural network algorithm, rainfall information and geological characteristics are reflected in the input variable, and the input value of at least one of the input variables of (i) to (v) is increased or decreased according to the rainfall information. And, the slope collapse risk prediction system, characterized in that configured to increase or decrease the input value of at least one of the input variables of the (i) to (v) according to the type of the ground on which the measuring instrument is installed. The method of claim 1 , wherein each meter comprises:
an acceleration sensor for measuring the acceleration applied to the instrument for a predetermined time;
a moisture content sensor that measures the amount of moisture inside the sloped surface;
a sound wave sensor that measures an acoustic wave inside the inclined surface; and
Slope collapse risk prediction system comprising a; data processing circuit for generating measurement data from the measurement values of the acceleration sensor, the moisture content sensor, and the sound wave sensor. delete The method of claim 1,
Slope collapse risk prediction system, characterized in that the server device calculates the primary risk based on the amount of change (I) of the impact amount and the moving average (Ima) in the step of calculating the primary risk. The method of claim 1,
The hidden layer of the artificial neural network algorithm consists of at least four nodes,
The output layer is a slope collapse risk prediction system, characterized in that it has five output nodes corresponding to each level of the five risk levels of relief, interest, caution, danger, and warning for the slope collapse risk.

Images