JP2018028436A - System and method for predicting avalanche occurrence using flying object - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately predict an occurrence of an avalanche in a region where a person cannot approach regardless of season.SOLUTION: In a method for predicting an occurrence of an avalanche using a flying object, observation of eight pieces of data including show accumulation at a place represented by initial positioning information values is started using the flying object during ploughing or after ploughing (S22) and the data is transmitted to and recorded in a main PC (S22-2). Next, an image of a land surface state is taken with a camera (S23) and the image is transmitted to the main PC so as to enable: land surface data at an observation point to be visually quantified according to classifications shown in Figure 5; and the quantified data to be recorded (S24-2). Then, the flying object measures six meteorological elements including a temperature of the observation point on the basis of an instruction from the main PC. The measured meteorological elements are transmitted to the main PC so as to enable the same to be recorded in a table of meteorological data during snow fall or after snow fall in Figure 8 (S25-2). Last, eight observation values respectively representing: the acquired snow accumulation; the land surface data; the temperature; humidity; atmospheric pressure; a wind velocity; a wind direction; and insolation, are synthesized on the basis of the table of the meteorological data during or after the snow fall and the synthesized value is recorded in the main PC (S25-3).SELECTED DRAWING: Figure 2

Description

  The present invention relates to an avalanche occurrence prediction system that accurately predicts the occurrence of an avalanche regardless of the season in an area where people cannot approach.

Conventionally, an avalanche has been predicted by installing an observation device on the surface of the earth and detecting its position change with GPS.
In addition, to measure the amount of snow, we determined the change by measuring the distance from the top of the pillar to the surface of the snow with a laser beam.

  In these systems, observation equipment must be installed in a place where an avalanche is expected to occur in advance, so it is necessary to identify dangerous locations and install them on steep slopes. It was. In addition, in the snowy season of severe winter, even if there is a machine failure or the like, it cannot be repaired, and in that case, there is a problem that an avalanche cannot be predicted. Furthermore, sensing the danger of an avalanche by the movement of snow cover often leads to an avalanche as it is, so an avalanche occurs before issuing an alarm, and there are situations that can not be predicted. Yes.

  In addition, a prediction method using a statistical method has been developed as an avalanche occurrence prediction method (such as SNOWPACK), but the accuracy is a problem because stable and timely data acquisition cannot be performed.

  In order to compensate for these shortcomings, we set the specifications of the target avalanche observation, observe multiple points of topographical conditions before snowfall, and meteorological data and weather factors that cause avalanches at regular intervals during snowfall It is necessary to get in a timely manner.

Japanese Patent Laid-Open No. 06-230101 JP 2011-149894 A

Masaaki Hanaoka, Masanori Kaneko, Yoichi Ito “Study on the altitude dependence and occurrence prediction of avalanche factors” July 21, 2009 Avalanche Landslide Research Center Naoyuki Wada, Architectural Institute of Japan Vol.74 (2009) No.642 P1787-1793 “Self-organized critical state analysis of land use in Matsumoto City”

In the conventional avalanche forecasting method, an observation device is installed at the expected avalanche occurrence point, and the avalanche occurrence is predicted by physically observing the snow movement and snow conditions. There is a problem that it is difficult to install the equipment, and it is impossible to cope with the failure of the observation equipment at the time of avalanche occurrence.
Moreover, in the avalanche occurrence prediction method using a statistical method, since the method of acquiring the weather factor related to avalanche occurrence timely has not been established, there also exists a subject that the prediction accuracy is low.

  The present invention has been made by paying attention to the above-mentioned problems. The purpose of the present invention is to determine the location of an avalanche and the area to be investigated regardless of the slope, regardless of the time. An object of the present invention is to provide a system for predicting the probability of occurrence of an avalanche and obtaining meteorological factors related to the occurrence of an avalanche at observation points in a timely manner and a method for predicting the occurrence of an avalanche.

  The avalanche occurrence prediction system according to the present invention includes a means for acquiring terrain information of an area before snowfall in an area where the occurrence of an avalanche is to be predicted, and subdividing the area into observation areas. Means for recording position information (latitude, longitude, altitude, surface condition) of the observation point, means for acquiring the acquired position information, snow amount information acquired after snowfall, and surface condition during or after snowfall; A means to acquire and quantify six meteorological elements (hereinafter referred to as “synthetic”), such as temperature, humidity, barometric pressure, wind speed, wind direction, and solar radiation, in the amount of snow and the ground surface, and observe the synthesized numerical values. Means for recording as numerical values at the time of observation at the site, means for graphing and visualizing the synthesized numerical values after snowfall that changes every moment, and avalanche predicting means for predicting avalanches by changing the synthesized numerical values Do And wherein the door.

  In addition, the flying object of the avalanche occurrence prediction system according to the present invention is characterized in that a fixed point stays in a configuration in which a balloon and a driving unit are combined, and a long-time stay is performed using the buoyancy of the balloon.

  Furthermore, there is a method for predicting the occurrence of avalanches using flying objects, including a pre-snow terrain data acquisition step for acquiring terrain data in a snow avalanche occurrence prediction survey area before a snowfall (which means that there is no snow). Set the observation area by dividing the occurrence prediction survey target area into meshes, specify the observation point, record the latitude, longitude, and altitude of the observation point as the initial position information and the means of staying during or after snowfall The step of acquiring snow amount information and the meteorological information of the observation point recorded at the observation point position information initial value by the flying object provided, the step of acquiring and recording the snow amount information and the meteorological information momentarily, and the snow amount The information and the weather information are synthesized and digitized to include a step of calculating a critical point of occurrence of an avalanche and a step of visualizing the calculated result.

  The avalanche occurrence prediction system according to the present invention is a more accurate avalanche occurrence prediction system by changing the weight of the ground state and the numerical coefficient of the weather element based on the prediction by the avalanche occurrence prediction means and the actual result. It can be.

These are the block diagrams in connection with the data acquisition before snowfall. These are the block diagrams in connection with the data acquisition after snowfall. These are the flowchart figures regarding data acquisition and processing before snowfall. These are the flowchart figures regarding the data acquisition and processing after snowfall. Is a set value of the ground surface state. These are the image figures which set an observation area from an observation area. Is a data input table before snowfall. Is a data input table during or after snowfall. These are graphs of synthesized values. These are the structures of the flying object scheduled to be used in the present invention.

  The avalanche prediction system according to the present invention subdivides the area for each mesh (hereinafter referred to as “observation area”) in a region where it is desired to predict the occurrence of an avalanche, and means for acquiring the topographic information of the area before snowfall. , Means for recording position information (latitude, longitude, altitude, surface condition) at any point in the observation area, and the acquired position information, snow amount information acquired after snowfall, and surface condition during or after snowfall , A means for acquiring and quantifying six meteorological elements (hereinafter referred to as “compositing”), such as temperature, humidity, atmospheric pressure, wind speed, wind direction, and solar radiation, in the amount of snow and the surface condition, and the composition To record the numerical values as observation values at the observation point, to graph and visualize the synthesized values after the falling snow, and to predict an avalanche by the change of the synthesized values avalanche Characterized by comprising a measuring means.

  The avalanche occurrence prediction system according to the present invention is a more accurate avalanche occurrence prediction system by changing the weight of the ground state and the numerical coefficient of the weather element based on the prediction by the avalanche occurrence prediction means and the actual result. It can be.

  Embodiments of an avalanche occurrence prediction system will be described below with reference to the accompanying drawings. Fig. 1 is a configuration diagram in which a flying object is suspended above the observation area before snowfall, and the surface condition of the observation point is observed and recorded on the main PC. A configuration diagram is shown in which the ground surface state and weather data are acquired and recorded in the main PC.

Embodiments will be described with reference to the accompanying drawings 3 and 4. First, the observation area is specified (S11), and the observation area is divided into meshes as shown in Fig. 6 based on the numerical map of the Geographical Survey Institute of the Ministry of Land, Infrastructure, Transport and Tourism, and the observation area is set. Set (hereinafter referred to as observation point) (S12), and record the latitude, longitude, and altitude of the observation point as position information (S13). For example, the area of observation area is preferably 50m square.

  The flying object stays in the sky above the designated observation area (S14), and after obtaining the GPS position information, the flying object transmits it to the main PC system and registers it as the initial value of the flying object position information (S15-2).

  Next, the flying object acquires a current state photograph of the target area (S16) and transmits it to the main PC system (S16-2).

In the main PC, the altitude and surface condition of the observation point are visually digitized according to the classification of the ground surface setting values (Fig. 5) (S17-2) and entered into the data entry table before snowfall in Fig. 7, Save as the initial value of the target point (S17-3).
Thereby, the process before snowfall is completed.

During or after snowfall, the flying object is retained at the latitude, longitude, and altitude of the initial position information, and observation is started (S21).
By adjusting the position of the flying object during or after snowfall to the latitude, longitude, and altitude before snowfall, the accuracy of acquired data can be increased.

  The flying object acquires data on the amount of snow at the observation point using laser waves (S22), and sends it to the main PC for recording (22-2).

  Further, the ground surface state is photographed by the camera (S23), transmitted to the main PC, and the ground surface data at the observation point is visually digitized and recorded according to the classification of FIG. 5 (24-2).

Next, according to the command from the main PC, the altitude is lowered to near the surface of the observation point (the latitude and longitude are not changed at this time) (S25), and the temperature, humidity, atmospheric pressure, wind speed, wind direction, and solar radiation Meteorological elements are measured. The altitude at this time is better as it is closer to the ground surface.
The observed meteorological elements are transmitted to the main PC and recorded in the data table during or after snowfall in FIG. 8 (S25-2).

After that, based on the data table at the time of snowfall or after snowfall in FIG. 8, the eight observation numerical values of the obtained snow cover amount, ground surface data, temperature, humidity, atmospheric pressure, wind speed, wind direction and solar radiation are synthesized and recorded on the main PC ( S25-3). Thereby, the first information acquisition is completed.
Thereafter, data acquisition of S21 to S25-3 is repeated for the set number of observations. The time required for one observation is preferably within 60 minutes.

  Here, compositing means synthesizing position information, altitude difference, ground surface data, and six meteorological elements obtained by a flying object as one state space quantity of fluctuation.

The synthesis method is performed according to the following concept and calculation formula.
State space is to capture each observed element as one variable. That is, it is considered that the variable can be explained by p variables φ1, φ2,..., Φp, and this is summarized as a composite variable with m components. If the coefficient acting on the variable is Wki, and the synthesized variable is Qm, it can be expressed as follows.
(2-1)
This is the first concept used for principal component analysis that summarizes m components when considering p variables and n individuals. Next, in order to make each variable easier-to-understand information, it is summarized in a low dimension and converted into a composite variable by Biplot. By doing this, it is possible to express how much density the variables applied to each observation point are applied as a load.
Here, when each variable is centered on an average of 0 and an observable n × p data matrix is φ and its rank is r,
(2-2)
The method of describing it as an expression of the singular value decomposition is the basic idea of the Bipot of synthesis theory.
B is a matrix having an orthonormal vector as a column vector, A′A = B′B = I, A ′, B ′: transposed matrix, I (unit matrix). Biplot approximates this φ with rank 2 to summarize in a lower dimension,
(2-3)
At this time, the right side is placed as follows.
(2-4)
A and B are matrices having orthonormal vectors as column vectors, Dλ is a diagonal matrix of diagonal elements λ, and (2) in the notation A (2) is a rank for placing in two dimensions.
F represents the first and second principal component scores of each observation point as an individual, and G represents the principal component coefficient.
This is the first concept used for principal component analysis that summarizes m components when considering p variables and n individuals. Next, in order to make each variable easier-to-understand information, it is summarized in a low dimension and converted into a composite variable by Biplot. By doing this, it is possible to express how much density is applied to the variables applied to each observation area.

  This concept and calculation method are applied to the present invention, the observation point is the snow cell volume, ground surface data, temperature, humidity, atmospheric pressure, wind speed, wind direction, solar radiation, and the eight observed numerical values are the corresponding cells in the table shown in FIG. Is inserted into the numerical value, and eight individual values are synthesized and combined into one numerical value.

  In the present invention, since the avalanche occurrence prediction accuracy is improved by continuously acquiring data for a long time, it is desirable that the flying object has a function of being able to stay for a long time.

  In order to stay in the space for a long time, a structure that can secure energy according to the staying time is required.

  Therefore, as shown in FIG. 10, the flying object used in the present invention is attached to a support body combined with a long cube with a balloon filled with a gas lighter than air, and a central axis passing through the center of gravity of the aircraft at the lower part of the aircraft. And six rotor units disposed at equal intervals from front to back and left and right, and a control means for driving and controlling each rotor unit, so that buoyancy can be generated by a balloon and moved by the rotor unit. did. As a result, energy for movement was saved, and a flying object capable of staying for a long time was obtained.

  In addition, since the flying object has a fixed-point aerial function, the error adjustment of the acquired data is reduced, the memory of the computer can be saved and the calculation speed can be improved, and the avalanche occurrence can be predicted more quickly and accurately. Therefore, it is desirable that the flying object has a fixed point regression function that automatically adjusts the position of the stagnant flight and that the tip always points in a certain direction.

As described above, in order to visually grasp the information obtained and repeatedly synthesized for each observation point by a flying object that has stayed at a fixed point for a long time, the synthesized values are continuously recorded. As a phenomenon, the point at which the correlation gradually disappears and the correlation disappears is regarded as a critical point, and this point is considered as the time of avalanche occurrence (self-organized critical state theory). Visualize by graphing.
FIG. 9 is a diagram showing a collapsed point as a result of arranging the synthesized data based on the above theory.

  By graphing the change in the synthesis value of the observation point, it is possible to visually grasp whether the current state of the observation point is stable or moving toward collapse. In addition, this system can accurately grasp the state of the observation point from time to time by observing for a long time, so the snow condition in the target area is constantly monitored, and the state of the observation point approaches the critical state. If it is, an alarm that an avalanche is expected can be issued.

  Since the avalanche occurrence prediction system according to the present invention uses a flying object to determine the surface condition and weather elements, it can be predicted even in places where people are not accessible, and it can also be used during periods of snowfall, snowfall, snowmelt, etc. Regardless of the type of snow avalanche, it is possible to continuously acquire snow cover data and meteorological elements, and accurately predict the occurrence of avalanches, making it possible to take appropriate avalanche countermeasures including land use.

Claims (4)

Pre-snow terrain data acquisition means for acquiring the topographic data of the avalanche occurrence prediction survey area before the snowfall (which means that there is no snow) using a flying object;
An avalanche occurrence prediction survey area is divided into meshes, an observation area is set, an observation point is specified, a means for recording the latitude, longitude, and altitude of the observation point as an initial position information value, and a means of staying during or after snowfall Means for acquiring snow amount information of the point recorded at the initial value of the observation point position information and meteorological information of the observation point, and means for acquiring and recording the snow amount information and meteorological information moment by moment, and a snow cover An avalanche occurrence prediction system using a flying object, characterized by comprising means for calculating the critical point of an avalanche by combining quantity information and meteorological information, and means for visualizing the calculation result   An avalanche occurrence prediction system using a flying object according to claim 1, wherein the flying object is provided with a fixed point retention system.   An avalanche occurrence prediction system using the flying object according to claim 1, which uses a fixed point staying system and a flying object capable of staying for a long time. A method for predicting the occurrence of an avalanche using a flying object,
Pre-snow terrain data acquisition step for acquiring terrain data of the target area of the avalanche occurrence prediction survey before snowfall (which means that there is no snow),
Avalanche occurrence prediction survey area is divided into meshes, observation areas are set, the observation point is specified, the latitude, longitude and altitude of the observation point are recorded as initial position information, and the means of staying during or after snowfall A step of acquiring snow amount information of the point recorded in the initial value of the observation point position information and weather information of the observation point, and a step of acquiring and recording the snow amount information and the meteorological information momentarily, and a snow cover A method for predicting avalanche occurrence using a flying object, characterized in that it includes a step of calculating a critical point of avalanche generation and a step of visualizing the calculation result by synthesizing and digitizing quantity information and weather information