CN107633659A - Dangerous slopes monitoring and pre-warning system and method - Google Patents
Dangerous slopes monitoring and pre-warning system and method Download PDFInfo
- Publication number
- CN107633659A CN107633659A CN201710954900.3A CN201710954900A CN107633659A CN 107633659 A CN107633659 A CN 107633659A CN 201710954900 A CN201710954900 A CN 201710954900A CN 107633659 A CN107633659 A CN 107633659A
- Authority
- CN
- China
- Prior art keywords
- early warning
- slope
- warning
- monitoring point
- rainfall
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Testing Or Calibration Of Command Recording Devices (AREA)
- Pit Excavations, Shoring, Fill Or Stabilisation Of Slopes (AREA)
- Emergency Alarm Devices (AREA)
Abstract
The invention discloses a kind of dangerous slopes monitoring and warning system, it includes:Sensor unit, the first memory cell, the second memory cell, early warning analysis unit and early warning decision unit.Wherein, sensor unit is used for displacement, soil pressure, hourly rainfall depth and the daily rainfall for detecting slope monitoring point.First memory cell is used for the detection data that storage sensors unit obtains.Second memory cell is used for the geological environment data for storing slope monitoring point.Early warning analysis unit is used to judge whether to early warning according to the data of collection.Early warning decision unit is used to mark grade and alert according to the judged result side slope dangerous situation of early warning analysis unit.The beneficial effects of the present invention are the abnormal conditions that both can effectively monitor side slope, the effect of timely early warning can be played again so that related personnel carries out the precautionary measures in advance.
Description
Technical field
The present invention relates to the technical field of slope monitoring, more particularly to a kind of dangerous slopes monitoring and pre-warning system and method.
Background technology
Existing slope monitoring technology mainly by facilities such as wiring and various sensors, machineries, is obtained and monitored in side slope
Displacement, soil pressure etc. at point, changed by relevant parameter and infer side slope state.Because the danger such as landslide, mud-rock flow occurs for side slope
The Geological Environmental Factors and rain fall of feelings and slope test point have it is close contact, and the data of existing detection fail to have
Effect gathers and utilizes these information, and when monitoring side slope exception, disaster has occurred and that mostly, it is impossible to which the early warning played in advance is made
With.
The content of the invention
To solve the above problems, the main object of the present invention is to provide a kind of dangerous slopes monitoring and warning system, the system
Both the abnormal conditions of side slope can have effectively been monitored, the effect of timely early warning can have been played again so that related personnel carries out anti-in advance
Model measure.
To achieve the above object, dangerous slopes monitoring and warning system proposed by the present invention, it includes:Sensor unit,
One memory cell, the second memory cell, early warning analysis unit and early warning decision unit.Wherein, sensor unit is used to detect
The displacement of slope monitoring point, soil pressure, the sensor unit of hourly rainfall depth and daily rainfall.First memory cell is used to store up
Deposit the detection data of sensor unit acquisition.Second memory cell is used for the geological environment data for storing slope monitoring point, the ground
Matter environmental data includes material conditions index M, weaknessization structural plane condition W and the effective freeing surface index S of slope monitoring point.
Early warning analysis unit is used for the geological environment data stored according to the detection data that sensor unit obtains with the second memory cell
Judge whether to early warning.Early warning decision unit is used to mark grade according to the judged result side slope dangerous situation of early warning analysis unit
And alert.The output end of sensor unit is connected with the input of the first memory cell.First memory cell it is defeated
Go out end, input of the output end of the second memory cell with early warning analysis unit is connected.The output end of early warning analysis unit with
Early warning decision unit is connected.
Preferably, sensor unit includes displacement transducer, soil pressure sensor and rainfall gauge.Wherein, displacement sensing
Device is used to detect displacement at slope monitoring point, and soil pressure sensor is used to detect the soil pressure at slope monitoring point, and rainfall gauge is used
The rainfall gauge of hourly rainfall depth and daily rainfall at detection slope monitoring point.
Preferably, early warning analysis unit is provided with first warning module, traditional warning module and neutral net warning module.
First warning module, traditional warning module and neutral net warning module are connected with the first memory cell, traditional early warning mould
Block, neutral net warning module are connected with the second memory module.
Preferably, displacement threshold values, soil pressure threshold values, hourly rainfall depth threshold values and daily rain amount are provided with first warning module
Threshold values is measured, in the displacement, soil pressure, hourly rainfall depth and daily rainfall at the slope monitoring point of sensor unit detection extremely
When few a certain monitor value is more than or equal to its respective threshold values, first warning module carries out first early warning.
Preferably, traditional warning module uses Classical forecast model L=R × G, wherein, L represents that slope test point is slided
The danger on slope, when L is more than the threshold values of setting, then send secondary early warning;R=R1/R1 valve+R24/R24 valves, it represents induced Landslides
The rainfall factor of generation, the R in formula1Represent the hourly rainfall depth of slope monitoring point, R1 valveRepresent the hourly rainfall depth of induced Landslides
Threshold values, R24Represent the daily rainfall of slope monitoring point, R24 valvesThe daily rainfall threshold values of induced Landslides;G=M × W × S, it is represented
Form the necessary address envirment factor on landslide.
Preferably, neutral net warning module uses multilayer neural network forecast model, and its input parameter is slope test
The hourly rainfall depth R of point1, daily rainfall R24, material conditions index M, weaknessization structural plane condition W and effective freeing surface refer to
Mark S,
Wherein, M=α FLd+ β LSA+ γ LST, the FLd in formula represent Facility-sliding strata distribution, are accumulated on LSA slopes loose
Solid matter distribution area, LST represent thickness, and α, β, γ are corresponding coefficient;
W=δ DL+ ε DLB+ θ SR, the DL in formula represent that, with accumulation bed boundary, DLB represents that accumulation horizon and interface of basement rock, SR delay
Incline roch layer interface, and δ, ε, θ are coefficient of correspondence.
Preferably, the system also includes GPS unit and image unit, and wherein GPS unit is used to carry out sensor unit
Positioning, image unit are used to be monitored in real time according to the positioning side slope monitoring point of GPS unit.GPS unit is used for sensing
Device unit is positioned, and its input is connected with sensor unit.Image unit monitors according to the positioning side slope of GPS unit
Point is monitored in real time, and the output end connection of its input and GPS unit, its output end is connected with early warning decision unit.
Present invention also offers a kind of method of dangerous slopes monitoring and warning, this method comprises the following steps:
S10, displacement, soil pressure and the drop of sensor unit detection cycle slope monitoring point are set in slope monitoring point
Rainfall information, and while sensor unit is set, gather the geological environment information of slope monitoring point;Sensor unit is obtained
The detection data taken are stored in the first memory cell, by the geological environment data storage collected and the second memory cell;Its
In, the geological environment data include material conditions index M, the weaknessization structural plane condition W of slope monitoring point and effectively faced
Empty face index S;
S20, the detection data obtained according to sensor unit, judged whether to by first warning module first pre-
It is alert,
In displacement, soil pressure, hourly rainfall depth and daily rainfall at the slope monitoring point of sensor unit detection
When at least a certain monitor value is more than or equal to its respective threshold values, first warning module carries out first early warning;
S30, if having detected first early warning, according to the geological environment information collected from slope monitoring point and hour rainfall
Amount, daily rainfall information, establish Classical forecast model L=R × G and multilayer neural network forecast model respectively, by traditional pre-
Survey model to be predicted with the multilayer neural network forecast model risk L that side slope monitoring point is come down respectively, pass through biography
System warning module carries out secondary early warning respectively with neutral net warning module;Wherein, R=R1/R1 valve+R24/R24 valves, it represents triggering
Come down the rainfall factor occurred, the R in formula1Represent the hourly rainfall depth of slope monitoring point, R1 valveRepresent the hour drop of induced Landslides
Rainfall threshold values, R24Represent the daily rainfall of slope monitoring point, R24 valvesThe daily rainfall threshold values of induced Landslides;G=M × W × S, its
Represent to form the necessary address envirment factor to come down;
When the risk L that the slope monitoring point of Classical forecast model prediction comes down is more than the threshold values of setting, tradition
Warning module sends secondary early warning;
When the risk L that the slope monitoring point of multilayer neural network forecast model prediction comes down is more than the threshold values of setting
When, neutral net warning module carries out secondary early warning;
S40, according to the situation of first early warning, secondary early warning, side slope dangerous situation carries out grade mark and alarmed;Its
In, the grade of side slope dangerous situation sets I levels, II levels, III level and IV levels, when the grade of side slope dangerous situation is more than or equal to III level, then
Alarmed;
When only detecting first early warning, side slope dangerous situation is labeled as I levels;
When being detected simultaneously by first early warning with secondary early warning,
If secondary early warning is sent by traditional warning module and any one in neutral net warning module, by side slope danger
Feelings are labeled as II levels;If traditional warning module is simultaneously emitted by secondary early warning with neutral net warning module, by side slope dangerous situation mark
IV levels are designated as, and are alarmed;
After first early warning is detected, if traditional warning module continuously sends out secondary early warning with neutral net warning module,
Side slope dangerous situation is then labeled as III level, and alarmed.
Preferably, in step S30, the modeling of multilayer neural network forecast model and prediction process are as follows:
S31, to gathering hourly rainfall depth, the daily rain amount that the geological environment data of slope monitoring point obtain with sensor unit
Amount is cleaned, i.e., data normalization works, and by material conditions index M, weaknessization structural plane condition W forms effectively facing for landslide
Empty face index S, hourly rainfall depth and daily rainfall carry out data conversion, comply with the use of multilayer neural network forecast model
It is required that;
S32, the data after step S10 processing are divided into two groups in proportion, first group is used for model training, second group
Verified for modelling effect;
S33, determine that the network node, initial weight, minimum of multilayer neural network forecast model train speed, iteration time
Number, sigmoid parameter informations;
S34, carry out neural network model training according to selected output parameter, parameter values in set-up procedure 31;
S35, repeat step S32~S34, until the parameter in step S33 is optimal, stop circulation.
Preferably, this method also includes:
S50, the GPS unit being connected with sensor unit is set in slope test point, in being provided opposite to for slope test point
With the image unit of GPS unit communication connection, the view data of slope monitoring point is gathered in real time, according to first early warning, secondary pre-
Alert and slope monitoring point view data is alarmed.
Compared with prior art, dangerous slopes monitoring and warning System and method for provided by the invention is by detecting slope monitoring point
Displacement, soil pressure and rain fall, with reference to side slope Geological Environmental Factors, it is possible to achieve the dangerous situation of side slope is supervised
Survey, and send alarm in time, remind relative region personnel to carry out the precautionary measures in advance, so as to reduce because caused by geological disaster
Loss in terms of the person, property.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Structure according to these accompanying drawings obtains other accompanying drawings.
Fig. 1 is the structured flowchart of dangerous slopes monitoring and warning system first embodiment of the present invention;
Fig. 2 is the structured flowchart of dangerous slopes monitoring and warning system second embodiment of the present invention;
Fig. 3 is the structured flowchart of the early warning analysis unit of the present invention;
The object of the invention is realized, functional characteristics and advantage will be described further referring to the drawings in conjunction with the embodiments.
Embodiment
The present invention proposes a kind of dangerous slopes monitoring and warning system.
Reference picture 1, Fig. 1 are the structured flowchart of dangerous slopes monitoring and warning system first embodiment of the present invention.
As shown in figure 1, in the first embodiment of the invention, the dangerous slopes monitoring and warning system includes:Sensor unit
100th, the first memory cell 200, the second memory cell 300, early warning analysis unit 400 and early warning decision unit 500.
Wherein, sensor unit 100 is used to obtain slope monitoring point relevant information, and it includes displacement transducer 110, soil pressure
Force snesor 120 and rainfall gauge 130.Displacement transducer 110 is deployed in slope monitoring point, for detecting position at slope monitoring point
Move.Soil pressure sensor 120 is deployed in slope monitoring point, for detecting the soil pressure at slope monitoring point.Rainfall gauge 130 is disposed
In slope monitoring point, for detecting the rainfall gauge 130 of hourly rainfall depth and daily rainfall at slope monitoring point.
First memory cell 200 and displacement transducer 110, soil pressure sensor 120 and the rainfall of sensor unit 100
Meter 130 is connected, the detection such as displacement, soil pressure, hourly rainfall depth and daily rainfall obtained for storage sensors unit 100
Data.Second memory cell 300 is used for the geological environment data for storing slope monitoring point, and the geological environment data are supervised including side slope
Material conditions index M, weaknessization structural plane condition W and the effective freeing surface index S of measuring point.It should be noted that in this reality
Apply in example, geological environment data are that collection obtains when disposing the hardware facilities such as displacement transducer 110, soil pressure sensor 120.
In addition corresponding geological environment parameter can be also obtained by GIS+RG technologies.
As shown in figure 3, early warning analysis unit 400 is connected with the first memory cell 200, the second memory cell 300 respectively,
Geological environment data for being stored according to the detection data that sensor unit 100 obtains and the second memory cell 300 judge
No carry out early warning.First warning module 410, traditional warning module 420 and neutral net are provided with the early warning analysis unit 400
Warning module 430.First warning module 410, traditional warning module 420 and neutral net warning module 430 are deposited with first
Storage unit 200 is connected, and traditional warning module 420, neutral net warning module 430 are connected with the second memory module.
In the present embodiment, displacement threshold values, soil pressure threshold values, hourly rainfall depth threshold values are provided with first warning module 410
And daily rainfall threshold values, when sensor unit 100 detect slope monitoring point at displacement, soil pressure, hourly rainfall depth with
And in daily rainfall when at least a certain monitor value is more than or equal to its respective threshold values, first warning module 410 carries out first pre-
It is alert.
Traditional warning module 420 uses Classical forecast model L=R × G, wherein, L represents what slope test point came down
Danger, when L is more than the threshold values of setting, then send secondary early warning;R=R1/R1 valve+R24/R24 valves, it represents induced Landslides
The rainfall factor, the R in formula1Represent the hourly rainfall depth of slope monitoring point, R1 valveThe hourly rainfall depth threshold values of induced Landslides is represented,
R24Represent the daily rainfall of slope monitoring point, R24 valvesThe daily rainfall threshold values of induced Landslides;G=M × W × S, it represents to be formed and slided
The necessary address envirment factor on slope.Because side slope produces landslide and the rain fall and geological conditions of side slope, pass through knot
Close the collection rain fall of slope monitoring point, geological conditions build forecast model, landslide can be occurred with side slope and carry out
Effectively monitoring, to play good forewarning function.
Neural network prediction module 430 uses multilayer neural network forecast model, chooses the hour rainfall of slope test point
Measure R1, daily rainfall R24, material conditions index M, weaknessization structural plane condition W and effective freeing surface index S, as multilayer
The input parameter of neural network prediction model, so that Landslide hazard occurs as output parameter.
Specifically, in the present embodiment, specifically modeling and prediction process are as follows for multilayer neural network forecast model,
S10, data are cleaned, i.e., data normalization works, including material conditions index M, weaknessization structure noodles
The supplemental characteristic conversion work such as the processing of the factor such as the effective freeing surface index S on part W and formation landslide and rainfall.Wherein,
Material conditions index M is distributed FLd by Facility-sliding strata, bulk solid mass distribution area LSA, thickness LST for being accumulated on slope etc.
Factor determines, is embodied as:M=α FLd+ β LSA+ γ LST, α, β, γ in formula are corresponding coefficient.Weakening structure noodles part W
By accumulating bed boundary (DL), accumulation horizon and interface of basement rock (DLB) together, slow roch layer interface (SR) decision of inclining, expression formula is W=δ DL+
ε DLB+ θ SR, δ, ε, θ in formula are coefficient of correspondence.The effective freeing surface index S on landslide is formed according to different address condition research
Area, the factor being selected is different, and n takes 4, i.e., high, high, medium and low hazardous area.
S20, data are divided into two groups in proportion, one group is used for model training, and one group is used for modelling effect and verifies.
S30, determine that the network node, initial weight, minimum of multilayer neural network forecast model train speed, iteration time
The parameter informations such as number, sigmoid.
S40, neural network model training is carried out according to selected output parameter, parameter values and passed through in set-up procedure 31
Continuous adjusting parameter is revised to forecast model.
S35, repeat step S32~S34, until the parameter in step S33 is optimal, stop circulation.
Thus, after multilayer neural network forecast model is established, you can by the hourly rainfall depth R after processing1, daily rainfall
R24, material conditions index M, the data such as weaknessization structural plane condition W and effective freeing surface index S re-enter multilayer god
Through Network Prediction Model, the risk L that side slope monitoring point is come down quickly is predicted.
In the present embodiment, possibility that side slope comes down is predicted by using multilayer neural network forecast model,
And in good time re -training multilayer neural network forecast model is fed back according to the follow-up case that occurs, the later stage can be caused to reach more accurate
Prediction effect, improve the forewarning function of early warning system.
Early warning decision unit 500 is used to mark grade simultaneously according to the judged result side slope dangerous situation of early warning analysis unit 400
Alert.The output end of sensor unit 100 is connected with the input of the first memory cell 200.First memory cell
The input of 200 output end, the output end of the second memory cell 300 with early warning analysis unit 400 is connected.Early warning analysis list
The output end of member 400 is connected with early warning decision unit 500.
When first warning module 410 sends first early warning, early warning decision unit 500 starts, and by the beginning of side slope dangerous situation
Beginning grade mark is I levels.
After first warning module 410 sends early warning, if follow-up traditional warning module 420 and neutral net warning module
430 non-early warning, then side slope dangerous situation mark is remained I levels by early warning decision unit 500, and carries out continuing observation;It is if traditional pre-
Any one in alert module 420 or neutral net warning module 430 have issued secondary early warning, then early warning decision unit 500 will
The grade mark of side slope dangerous situation is II levels, and notifies the administrative staff of relative region;If traditional warning module 420 and neutral net
The early warning simultaneously of warning module 430, then the grade mark of side slope dangerous situation is IV levels by early warning decision unit 500, and to relative region
Administrative staff alarmed.
After it is II levels that early warning decision unit 500 is by the grade mark of side slope dangerous situation, if in subsequently observation process is continued,
Model in early warning analysis unit 400 continuously sends out pre-warning signal, or the model of non-early warning sends pre-warning signal before, then early warning
Grade is risen to III level by decision package 500, and is alarmed to the administrative staff of relative region.
In addition, when early warning decision unit 500 changes the grade of side slope dangerous situation every time, the pipe of relative region can be notified in time
Reason personnel, in order to which administrative staff make a policy in time, and the precautionary measures are carried out in advance.
Dangerous slopes monitoring and warning system of the present invention is by detecting displacement, soil pressure and the rainfall feelings of slope monitoring point
Condition, with reference to side slope Geological Environmental Factors, it is possible to achieve the dangerous situation of side slope is monitored, and sends alarm in time, reminds phase
Close regional personnel and carry out the precautionary measures in advance, so as to reduce the loss because in terms of the person, property caused by geological disaster.Pass through
Marked using multilayer modes of warning, and to warning grade, can not only improve the early warning effect of side slope dangerous situation, while be easy to phase
The warning grade that the administrative staff for closing area mark according to side slope dangerous situation makes corresponding emergency plan, with the measure of improving the precaution
Validity.
As shown in Fig. 2 in the second embodiment of the present invention, the system also includes GPS unit 600 and image unit 700,
Wherein GPS unit 600 is used to position sensor unit 100, and image unit 700 is used for the positioning according to GPS unit 600
Side slope monitoring point is monitored in real time.GPS unit 600 is used to position sensor unit 100, its input and sensing
Device unit 100 connects.Image unit 700 is monitored in real time in the positioning side slope monitoring point according to GPS unit 600, and its is defeated
Enter end to be connected with the output end of GPS unit 600, its output end is connected with early warning decision unit 500.
In the present embodiment, the image information of slope monitoring point particular location can be obtained by image unit 700, and should
Image information returns to early warning decision unit 500, so that the administrative staff of relative region can pass through the shooting list
Member 700 carrys out the actual conditions of more intuitive understanding side slope.
Present invention also offers a kind of method of dangerous slopes monitoring and warning, this method comprises the following steps:
S10, displacement, soil pressure and the drop of sensor unit detection cycle slope monitoring point are set in slope monitoring point
Rainfall information, and while sensor unit is set, gather the geological environment information of slope monitoring point;Sensor unit is obtained
The detection data taken are stored in the first memory cell, by the geological environment data storage collected and the second memory cell;Its
In, the geological environment data include material conditions index M, the weaknessization structural plane condition W of slope monitoring point and effectively faced
Empty face index S;
S20, the detection data obtained according to sensor unit, judged whether to by first warning module first pre-
It is alert,
In displacement, soil pressure, hourly rainfall depth and daily rainfall at the slope monitoring point of sensor unit detection
When at least a certain monitor value is more than or equal to its respective threshold values, first warning module carries out first early warning;
S30, if having detected first early warning, according to the geological environment information collected from slope monitoring point and hour rainfall
Amount, daily rainfall information, establish Classical forecast model L=R × G and multilayer neural network forecast model respectively, by traditional pre-
Survey model to be predicted with the multilayer neural network forecast model risk L that side slope monitoring point is come down respectively, pass through biography
System warning module carries out secondary early warning respectively with neutral net warning module;Wherein, R=R1/R1 valve+R24/R24 valves, it represents triggering
Come down the rainfall factor occurred, the R in formula1Represent the hourly rainfall depth of slope monitoring point, R1 valveRepresent the hour drop of induced Landslides
Rainfall threshold values, R24Represent the daily rainfall of slope monitoring point, R24 valvesThe daily rainfall threshold values of induced Landslides;G=M × W × S, its
Represent to form the necessary address envirment factor to come down;
When the risk L that the slope monitoring point of Classical forecast model prediction comes down is more than the threshold values of setting, tradition
Warning module sends secondary early warning;
When the risk L that the slope monitoring point of multilayer neural network forecast model prediction comes down is more than the threshold values of setting
When, neutral net warning module carries out secondary early warning;
S40, according to the situation of first early warning, secondary early warning, side slope dangerous situation carries out grade mark and alarmed;Its
In, the grade of side slope dangerous situation sets I levels, II levels, III level and IV levels, when the grade of side slope dangerous situation is more than or equal to III level, then
Alarmed;
When only detecting first early warning, side slope dangerous situation is labeled as I levels;
When being detected simultaneously by first early warning with secondary early warning,
If secondary early warning is sent by traditional warning module and any one in neutral net warning module, by side slope danger
Feelings are labeled as II levels;If traditional warning module is simultaneously emitted by secondary early warning with neutral net warning module, by side slope dangerous situation mark
IV levels are designated as, and are alarmed;
After first early warning is detected, if traditional warning module continuously sends out secondary early warning with neutral net warning module,
Side slope dangerous situation is then labeled as III level, and alarmed.
Dangerous slopes monitoring and pre-alarming method of the present invention is by detecting displacement, soil pressure and the rainfall feelings of slope monitoring point
Condition, with reference to side slope Geological Environmental Factors, it is possible to achieve the dangerous situation of side slope is monitored, and sends alarm in time, reminds phase
Close regional personnel and carry out the precautionary measures in advance, so as to reduce the loss because in terms of the person, property caused by geological disaster.Pass through
Marked using multilayer modes of warning, and to warning grade, can not only improve the early warning effect of side slope dangerous situation, while be easy to phase
The warning grade that the administrative staff for closing area mark according to side slope dangerous situation makes corresponding emergency plan, with the measure of improving the precaution
Validity.
Specifically, in step S30, the modeling of multilayer neural network forecast model and prediction process are as follows:
S31, to gathering hourly rainfall depth, the daily rain amount that the geological environment data of slope monitoring point obtain with sensor unit
Amount is cleaned, i.e., data normalization works, and by material conditions index M, weaknessization structural plane condition W forms effectively facing for landslide
Empty face index S, hourly rainfall depth and daily rainfall carry out data conversion, comply with the use of multilayer neural network forecast model
It is required that;
S32, the data after step S10 processing are divided into two groups in proportion, first group is used for model training, second group
Verified for modelling effect;
S33, determine that the network node, initial weight, minimum of multilayer neural network forecast model train speed, iteration time
Number, sigmoid parameter informations;
S34, carry out neural network model training according to selected output parameter, parameter values in set-up procedure 31;
S35, repeat step S32~S34, until the parameter in step S33 is optimal, stop circulation;
In the present embodiment, by first establishing multilayer neural network forecast model, then by the hourly rainfall depth R after processing1、
Daily rainfall R24, material conditions index M, the data such as weaknessization structural plane condition W and effective freeing surface index S re-enter
After the multilayer neural network forecast model, you can the risk L that side slope monitoring point is come down is predicted.
Preferably, this method also includes:
S50, the GPS unit being connected with sensor unit is set in slope test point, in being provided opposite to for slope test point
With the image unit of GPS unit communication connection, the view data of slope monitoring point is gathered in real time, according to first early warning, secondary pre-
Alert and slope monitoring point view data is alarmed.
In the present embodiment, by image unit obtain slope monitoring point particular location image information, so as to so that
The administrative staff of relative region can be side slope danger by the image unit come the actual conditions of more intuitive understanding side slope
The early warning of feelings provides auxiliary foundation.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the scope of the invention, it is every at this
Under the inventive concept of invention, the equivalent structure transformation made using description of the invention and accompanying drawing content, or directly/use indirectly
It is included in other related technical areas in the scope of patent protection of the present invention.
Claims (10)
1. a kind of dangerous slopes monitoring and warning system, it is characterised in that it includes:
Sensor unit, for detecting the displacement of slope monitoring point, soil pressure, hourly rainfall depth and the sensor of daily rainfall
Unit;
First memory cell, the detection data obtained for storing the sensor unit;
Second memory cell, for storing the geological environment data of slope monitoring point;Wherein, the geological environment data include side
Material conditions index M, weaknessization structural plane condition W and the effective freeing surface index S of slope monitoring point;
Early warning analysis unit, for detection data and second memory cell storage obtained according to the sensor unit
Geological environment data judge whether to early warning;
Early warning decision unit, for marking grade according to the judged result side slope dangerous situation of the early warning analysis unit and sending report
Alert information;
The output end of the sensor unit is connected with the input of first memory cell;First memory cell it is defeated
Go out end, input of the output end of second memory cell with the early warning analysis unit is connected;The early warning analysis list
The output end of member is connected with the early warning decision unit.
2. dangerous slopes monitoring and warning system as claimed in claim 1, it is characterised in that the sensor unit includes:With
The displacement transducer of displacement at detection slope monitoring point, for detecting the soil pressure sensor of soil pressure at slope monitoring point,
And for detecting the rainfall gauge of hourly rainfall depth and daily rainfall at slope monitoring point.
3. dangerous slopes monitoring and warning system as claimed in claim 1, it is characterised in that the early warning analysis unit is provided with just
Secondary warning module, traditional warning module and neutral net warning module;First warning module, traditional warning module and nerve
Network Warning module is connected with first memory cell, and traditional warning module, neutral net warning module are and institute
State the connection of the second memory module.
4. dangerous slopes monitoring and warning system as claimed in claim 3, it is characterised in that be provided with the first warning module
Displacement threshold values, soil pressure threshold values, hourly rainfall depth threshold values and daily rainfall threshold values, when the slope monitoring of sensor unit detection
At least a certain monitor value is more than or equal to its respective valve in displacement, soil pressure, hourly rainfall depth and daily rainfall at point
During value, the first warning module carries out first early warning.
5. dangerous slopes monitoring and warning system as claimed in claim 3, it is characterised in that traditional warning module is using biography
Unite forecast model L=R × G, wherein,
L represents the danger that slope test point comes down, and when L is more than the threshold values of setting, then sends secondary early warning;
R=R1/R1 valve+R24/R24 valves, it represents the rainfall factor that induced Landslides occur, the R in formula1Represent the hour of slope monitoring point
Rainfall, R1 valveRepresent the hourly rainfall depth threshold values of induced Landslides, R24Represent the daily rainfall of slope monitoring point, R24 valvesInduced Landslides
Daily rainfall threshold values;
G=M × W × S, it represents the necessary address envirment factor for forming landslide.
6. dangerous slopes monitoring and warning system as claimed in claim 3, it is characterised in that the neutral net warning module is adopted
With multilayer neural network forecast model, its input parameter is the hourly rainfall depth R of slope test point1, daily rainfall R24, material bar
Part index M, weaknessization structural plane condition W and effective freeing surface index S,
Wherein, M=α FLd+ β LSA+ γ LST, the FLd in formula represent Facility-sliding strata distribution, and LSA represents to accumulate on slope loose
Solid matter distribution area, LST represent thickness, and α, β, γ are corresponding coefficient;
W=δ DL+ ε DLB+ θ SR, the DL in formula represent that with accumulation bed boundary DLB represents accumulation horizon and interface of basement rock, the slow rocks that incline of SR
Bed boundary, δ, ε, θ are coefficient of correspondence.
7. the dangerous slopes monitoring and warning system as described in claim 1~6 any one, it is characterised in that also include:
GPS unit, for being positioned to the sensor unit;
Image unit, monitored in real time for the positioning side slope monitoring point according to the GPS unit;
The input of the GPS unit is connected with the sensor unit, and its output end is connected with the image unit input;
The output end of the image unit is connected with the early warning decision unit.
A kind of 8. method of dangerous slopes monitoring and warning, it is characterised in that comprise the following steps:
S10, displacement, soil pressure and the rainfall of sensor unit detection cycle slope monitoring point are set in slope monitoring point
Information, and while sensor unit is set, gather the geological environment information of slope monitoring point;Sensor unit is obtained
Detection data are stored in the first memory cell, by the geological environment data storage collected and the second memory cell;Wherein, institute
Stating geological environment data includes material conditions index M, weaknessization structural plane condition W and the effective freeing surface of slope monitoring point
Index S;
S20, the detection data obtained according to sensor unit, first early warning is judged whether to by first warning module,
In displacement, soil pressure, hourly rainfall depth and daily rainfall at the slope monitoring point of sensor unit detection at least
When a certain monitor value is more than or equal to its respective threshold values, first warning module carries out first early warning;
S30, if having detected first early warning, according to the geological environment information that is collected from slope monitoring point and hourly rainfall depth,
Daily rainfall information, Classical forecast model L=R × G and multilayer neural network forecast model are established respectively, pass through Classical forecast mould
Type is predicted with the multilayer neural network forecast model risk L that side slope monitoring point is come down respectively, by traditional pre-
Alert module carries out secondary early warning respectively with neutral net warning module;Wherein, R=R1/R1 valve+R24/R24 valves, it represents induced Landslides
The rainfall factor of generation, the R in formula1Represent the hourly rainfall depth of slope monitoring point, R1 valveRepresent the hourly rainfall depth of induced Landslides
Threshold values, R24Represent the daily rainfall of slope monitoring point, R24 valvesThe daily rainfall threshold values of induced Landslides;G=M × W × S, it is represented
Form the necessary address envirment factor on landslide;
When the risk L that the slope monitoring point of Classical forecast model prediction comes down is more than the threshold values of setting, traditional early warning
Module sends secondary early warning;
When the risk L that the slope monitoring point of multilayer neural network forecast model prediction comes down is more than the threshold values of setting,
Neutral net warning module carries out secondary early warning;
S40, according to the situation of first early warning, secondary early warning, side slope dangerous situation carries out grade mark and alarmed;Wherein, side
The grade of slope dangerous situation sets I levels, II levels, III level and IV levels, when the grade of side slope dangerous situation is more than or equal to III level, is then reported
It is alert;
When only detecting first early warning, side slope dangerous situation is labeled as I levels;
When being detected simultaneously by first early warning with secondary early warning,
If secondary early warning is sent by traditional warning module and any one in neutral net warning module, by side slope dangerous situation mark
It is designated as II levels;If traditional warning module is simultaneously emitted by secondary early warning with neutral net warning module, side slope dangerous situation is labeled as
IV levels, and alarmed;
, will if traditional warning module continuously sends out secondary early warning with neutral net warning module after first early warning is detected
Side slope dangerous situation is labeled as III level, and is alarmed.
9. dangerous slopes monitoring and pre-alarming method as claimed in claim 8, it is characterised in that in the step S30, multilayer nerve
Network Prediction Model building process is as follows:
S31, hourly rainfall depth, the daily rainfall obtained to the geological environment data for gathering slope monitoring point with sensor unit enter
Row cleaning, i.e. data normalization are worked, and material conditions index M, weaknessization structural plane condition W are formed into the effective freeing surface on landslide
Index S, hourly rainfall depth and daily rainfall carry out data conversion, and complying with the use of multilayer neural network forecast model will
Ask;
S32, the data after step S10 processing are divided into two groups in proportion, first group is used for model training, and second group is used for
Modelling effect is verified;
S33, determine the network node of multilayer neural network forecast model, initial weight, minimum training speed, iterations,
Sigmoid parameter informations;
S34, carry out neural network model training according to selected output parameter, parameter values in set-up procedure 31;
S35, repeat step S32~S34, until the parameter in step S33 is optimal, stop circulation.
10. dangerous slopes monitoring and pre-alarming method as claimed in claim 8, it is characterised in that also include:
S50, the GPS unit that is connected with sensor unit is set in slope test point, slope test point be provided opposite to and GPS
Unit communications connection image unit, in real time gather slope monitoring point view data, according to first early warning, secondary early warning and
The view data of slope monitoring point is alarmed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710954900.3A CN107633659B (en) | 2017-10-13 | 2017-10-13 | Dangerous slopes monitoring and pre-warning system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710954900.3A CN107633659B (en) | 2017-10-13 | 2017-10-13 | Dangerous slopes monitoring and pre-warning system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107633659A true CN107633659A (en) | 2018-01-26 |
CN107633659B CN107633659B (en) | 2019-07-23 |
Family
ID=61105305
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710954900.3A Active CN107633659B (en) | 2017-10-13 | 2017-10-13 | Dangerous slopes monitoring and pre-warning system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107633659B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108364435A (en) * | 2018-02-22 | 2018-08-03 | 海南师范大学 | A kind of common natural calamity early warning system in international tourism island |
CN108597166A (en) * | 2018-03-22 | 2018-09-28 | 广东长亨石业有限公司 | A kind of stone pit safety monitoring system and method |
CN109003422A (en) * | 2018-08-02 | 2018-12-14 | 北京大学深圳研究生院 | Monitoring data processing method and landslide forecasting procedure for landslide |
CN109596177A (en) * | 2019-01-08 | 2019-04-09 | 昆山高新轨道交通智能装备有限公司 | Railway slope Life cycle on-line monitoring system and method |
CN110189505A (en) * | 2019-06-14 | 2019-08-30 | 成都理工大学 | A kind of method for early warning of level rock sliding surface rock landslip and its application |
CN110285904A (en) * | 2019-06-12 | 2019-09-27 | 泉州装备制造研究所 | A kind of slope of highway pressure detecting and prior-warning device |
CN110390800A (en) * | 2019-06-06 | 2019-10-29 | 北京市地质研究所 | Net formula disaster monitoring and early-warning system |
CN110795812A (en) * | 2019-08-13 | 2020-02-14 | 福建工程学院 | Landslide prediction method and system based on big data analysis |
CN111539568A (en) * | 2020-04-22 | 2020-08-14 | 深圳市地质局 | Safety monitoring system and method based on unmanned aerial vehicle and three-dimensional modeling technology |
CN113012396A (en) * | 2021-01-28 | 2021-06-22 | 深圳市协润科技有限公司 | Monitoring and early warning system and method for slope micro-deformation |
CN113114935A (en) * | 2021-04-07 | 2021-07-13 | 中煤科工集团重庆研究院有限公司 | Vibration identification method based on video image |
CN114111905A (en) * | 2021-11-26 | 2022-03-01 | 中国电建集团成都勘测设计研究院有限公司 | High slope construction scaffold safety monitoring system |
CN114991226A (en) * | 2022-06-07 | 2022-09-02 | 浙江天成项目管理有限公司 | Foundation pit displacement automatic monitoring system |
CN115273404A (en) * | 2022-06-27 | 2022-11-01 | 长春丰泽水文气象仪器有限公司 | Mountain torrent early warning monitoring method, system, terminal and storage medium |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101944274A (en) * | 2010-10-09 | 2011-01-12 | 张春阳 | Landside and mud-rock flow alarm |
CN202075836U (en) * | 2011-05-19 | 2011-12-14 | 赵红 | Geological deformation and slip early warning system |
CN102867110A (en) * | 2012-08-22 | 2013-01-09 | 广东电网公司电力科学研究院 | Rainstorm disaster risk evaluation method for foundation slope of transmission line tower |
CN103700054A (en) * | 2013-12-10 | 2014-04-02 | 中国地质大学武汉 | Sudden-onset geological disaster emergency plan digitization system |
KR20150048413A (en) * | 2013-10-28 | 2015-05-07 | 대한민국(산림청 국립산림과학원장) | Apparatus for calculating landslide warning standard rainfall using tank model and method thereof |
CN104751615A (en) * | 2015-02-11 | 2015-07-01 | 西南科技大学 | Debris flow evolution process based hierarchy early-warning method |
CN104951993A (en) * | 2014-09-04 | 2015-09-30 | 国网山东省电力公司应急管理中心 | Comprehensive monitoring and early warning system based on meteorology and power grid GIS and method thereof |
CN105957311A (en) * | 2016-06-01 | 2016-09-21 | 中国水利水电科学研究院 | Adaptive expansion slope stability intelligent monitoring early warning system |
CN106023530A (en) * | 2016-06-07 | 2016-10-12 | 长安大学 | Heavy rain type diluted debris flow monitoring, forecast and early warning device and method |
CN106021875A (en) * | 2016-05-11 | 2016-10-12 | 兰州大学 | Multi-scale debris flow risk assessment method for earthquake disturbance area |
-
2017
- 2017-10-13 CN CN201710954900.3A patent/CN107633659B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101944274A (en) * | 2010-10-09 | 2011-01-12 | 张春阳 | Landside and mud-rock flow alarm |
CN202075836U (en) * | 2011-05-19 | 2011-12-14 | 赵红 | Geological deformation and slip early warning system |
CN102867110A (en) * | 2012-08-22 | 2013-01-09 | 广东电网公司电力科学研究院 | Rainstorm disaster risk evaluation method for foundation slope of transmission line tower |
KR20150048413A (en) * | 2013-10-28 | 2015-05-07 | 대한민국(산림청 국립산림과학원장) | Apparatus for calculating landslide warning standard rainfall using tank model and method thereof |
CN103700054A (en) * | 2013-12-10 | 2014-04-02 | 中国地质大学武汉 | Sudden-onset geological disaster emergency plan digitization system |
CN104951993A (en) * | 2014-09-04 | 2015-09-30 | 国网山东省电力公司应急管理中心 | Comprehensive monitoring and early warning system based on meteorology and power grid GIS and method thereof |
CN104751615A (en) * | 2015-02-11 | 2015-07-01 | 西南科技大学 | Debris flow evolution process based hierarchy early-warning method |
CN106021875A (en) * | 2016-05-11 | 2016-10-12 | 兰州大学 | Multi-scale debris flow risk assessment method for earthquake disturbance area |
CN105957311A (en) * | 2016-06-01 | 2016-09-21 | 中国水利水电科学研究院 | Adaptive expansion slope stability intelligent monitoring early warning system |
CN106023530A (en) * | 2016-06-07 | 2016-10-12 | 长安大学 | Heavy rain type diluted debris flow monitoring, forecast and early warning device and method |
Non-Patent Citations (1)
Title |
---|
赵 霞: "宁夏隆德县防汛抗旱现状及对策", 《农田水利》 * |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108364435A (en) * | 2018-02-22 | 2018-08-03 | 海南师范大学 | A kind of common natural calamity early warning system in international tourism island |
CN108597166A (en) * | 2018-03-22 | 2018-09-28 | 广东长亨石业有限公司 | A kind of stone pit safety monitoring system and method |
CN109003422A (en) * | 2018-08-02 | 2018-12-14 | 北京大学深圳研究生院 | Monitoring data processing method and landslide forecasting procedure for landslide |
CN109596177A (en) * | 2019-01-08 | 2019-04-09 | 昆山高新轨道交通智能装备有限公司 | Railway slope Life cycle on-line monitoring system and method |
CN110390800A (en) * | 2019-06-06 | 2019-10-29 | 北京市地质研究所 | Net formula disaster monitoring and early-warning system |
CN110390800B (en) * | 2019-06-06 | 2021-04-20 | 北京市地质研究所 | Network type disaster monitoring and early warning system |
CN110285904A (en) * | 2019-06-12 | 2019-09-27 | 泉州装备制造研究所 | A kind of slope of highway pressure detecting and prior-warning device |
CN110189505A (en) * | 2019-06-14 | 2019-08-30 | 成都理工大学 | A kind of method for early warning of level rock sliding surface rock landslip and its application |
CN110189505B (en) * | 2019-06-14 | 2021-03-23 | 成都理工大学 | Rock stratum sliding surface rock landslide early warning method and application thereof |
CN110795812B (en) * | 2019-08-13 | 2023-01-31 | 福建工程学院 | Landslide prediction method and system based on big data analysis |
CN110795812A (en) * | 2019-08-13 | 2020-02-14 | 福建工程学院 | Landslide prediction method and system based on big data analysis |
CN111539568A (en) * | 2020-04-22 | 2020-08-14 | 深圳市地质局 | Safety monitoring system and method based on unmanned aerial vehicle and three-dimensional modeling technology |
CN113012396A (en) * | 2021-01-28 | 2021-06-22 | 深圳市协润科技有限公司 | Monitoring and early warning system and method for slope micro-deformation |
CN113114935A (en) * | 2021-04-07 | 2021-07-13 | 中煤科工集团重庆研究院有限公司 | Vibration identification method based on video image |
CN113114935B (en) * | 2021-04-07 | 2022-08-19 | 中煤科工集团重庆研究院有限公司 | Vibration identification method based on video image |
CN114111905A (en) * | 2021-11-26 | 2022-03-01 | 中国电建集团成都勘测设计研究院有限公司 | High slope construction scaffold safety monitoring system |
CN114991226A (en) * | 2022-06-07 | 2022-09-02 | 浙江天成项目管理有限公司 | Foundation pit displacement automatic monitoring system |
CN115273404A (en) * | 2022-06-27 | 2022-11-01 | 长春丰泽水文气象仪器有限公司 | Mountain torrent early warning monitoring method, system, terminal and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN107633659B (en) | 2019-07-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107633659B (en) | Dangerous slopes monitoring and pre-warning system and method | |
WO2022000528A1 (en) | Bridge scouring multi-source monitoring system and monitoring method, and scouring depth evaluation method therefor | |
CN103743441B (en) | The polynary of a kind of side slope safety is coupling in line monitoring system and method | |
CN104316108B (en) | Method for establishing and analyzing online monitoring system for 500kv power transmission tower in mountain environment | |
JP5148589B2 (en) | A method for evaluating the safety of bridge structures by vibration measurements. | |
CN108154651A (en) | A kind of device and method for landslide disaster Monitoring and forecasting system in real-time | |
CN104281920A (en) | Tailing pond layered index safety assessment and early-warning method and system | |
CN109493569A (en) | Come down method for early warning, device, computer equipment and storage medium | |
CN103424099A (en) | Slope monitoring and early warning system and method based on deformation data | |
CN109615118A (en) | Based on big data hazards control Informatization Service integrated control system and method | |
CN112435443A (en) | Geological disaster monitoring and early warning system based on big data analysis | |
CN105989697A (en) | Multi-source sensor-based landslide monitoring and early-warning device | |
CN108280969A (en) | A kind of high slope surface deformation monitoring and warning system and its method for early warning | |
CN105783756A (en) | Fiber-grating-based power transmission line iron tower deformation on-line monitoring device and method | |
CN112053093A (en) | Geological disaster landslide real-time monitoring management system based on big data | |
CN113155186A (en) | Dam safety monitoring management equipment and system thereof | |
CN114298549A (en) | Water conservancy safety monitoring system and method based on big data | |
de Santiago et al. | A comparative study of models to predict storm impact on beaches | |
Ciavola et al. | Prediction of storm impacts on beach and dune systems | |
CN108205874A (en) | Geo-hazard early-warning method based on multi-parameter linkage, live master station and system | |
CN116824807B (en) | Multi-disaster early warning and alarming method and system | |
CN114997548B (en) | Water movement geological seepage real-time early warning method based on genetic neural network | |
CN103759868B (en) | A kind of bridge lateral based on stress proportion couples real time evaluating method | |
CN107270863A (en) | It is a kind of to assess early warning system and the method for early warning that vertical deep foundation pit deformation influences on embankment | |
CN114399889A (en) | Geological disaster early warning system for rock-soil slope |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |