CN112818567A - Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory - Google Patents

Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory Download PDF

Info

Publication number
CN112818567A
CN112818567A CN202110223708.3A CN202110223708A CN112818567A CN 112818567 A CN112818567 A CN 112818567A CN 202110223708 A CN202110223708 A CN 202110223708A CN 112818567 A CN112818567 A CN 112818567A
Authority
CN
China
Prior art keywords
monitoring
early warning
data
points
displacement
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
Application number
CN202110223708.3A
Other languages
Chinese (zh)
Other versions
CN112818567B (en
Inventor
黄磊
李娜
李永生
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Antai Data Monitoring Technology Co ltd
Original Assignee
Shenzhen Antai Data Monitoring Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Antai Data Monitoring Technology Co ltd filed Critical Shenzhen Antai Data Monitoring Technology Co ltd
Priority to CN202110223708.3A priority Critical patent/CN112818567B/en
Publication of CN112818567A publication Critical patent/CN112818567A/en
Application granted granted Critical
Publication of CN112818567B publication Critical patent/CN112818567B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Business, Economics & Management (AREA)
  • Operations Research (AREA)
  • Computational Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Development Economics (AREA)
  • Algebra (AREA)
  • Evolutionary Computation (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Databases & Information Systems (AREA)
  • Geometry (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)
  • Pit Excavations, Shoring, Fill Or Stabilisation Of Slopes (AREA)

Abstract

The invention discloses a geotechnical engineering intelligent monitoring and early warning method and device based on probability theory, wherein the method comprises the following steps: acquiring coordinates of monitoring points and coordinates of unmonitored points in a monitoring space; acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point; judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points; performing data autocorrelation judgment on the deep displacement data according to the trend structure judgment result; carrying out interpolation prediction on the unmonitored area according to the coordinates of the monitored points, the coordinates of unmonitored points and the data autocorrelation judgment result; solving the prediction error of interpolation prediction; performing conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored points according to the solved result; calculating the occurrence probability of the danger early warning; and generating a real-time early warning signal according to the danger early warning occurrence probability. The invention enlarges the range of the early warning monitoring area, reduces the monitoring cost and has good early warning effect.

Description

Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory
Technical Field
The invention relates to the technical field of geotechnical monitoring, in particular to a geotechnical engineering intelligent monitoring and early warning method and device based on probability theory.
Background
Real-time monitoring (such as deep displacement) of soil deformation indexes plays an important role in geotechnical engineering monitoring and early warning. In the existing monitoring technology, automatic monitoring based on the Internet of things is an advanced monitoring technology means. Although the monitoring method based on the Internet of things can realize real-time monitoring of deformation indexes of the rock-soil structures. However, the cost of the monitoring equipment of the internet of things is high, and in order to reduce the cost, it is difficult to densely distribute the measuring points in a large area in practical application. Due to the limited number of measuring points, soil body movement exceeding the early warning limit can occur in a large number of unobserved areas, but cannot be sensed by the sensor in real time. This often results in the early warning signal not being sent out instantaneously, resulting in loss of personnel and property.
Disclosure of Invention
The invention solves the technical problem of providing a geotechnical engineering intelligent monitoring and early warning method and device based on probability theory, which can enlarge an early warning area and can send out early warning signals in time.
In a first aspect, the invention provides a geotechnical engineering intelligent monitoring and early warning method based on probability theory, which comprises the following steps:
acquiring coordinates of monitoring points and coordinates of unmonitored points in a monitoring space;
acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point;
judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points;
performing data autocorrelation judgment on the deep displacement data according to the trend structure judgment result;
performing interpolation prediction on the unmonitored area according to the coordinates of the monitoring points, the coordinates of unmonitored points and the data autocorrelation judgment result;
solving for a prediction error of the interpolated prediction;
performing conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solved result;
calculating the danger early warning occurrence probability according to the Monte Carlo simulation result;
and generating a real-time early warning signal according to the danger early warning occurrence probability.
In one embodiment, before the step of "determining a trend structure with spatially distributed deep displacements according to the monitoring data of the monitoring points", the method further comprises:
and carrying out normalization processing on the monitoring data of the deep displacement of the soil body of the monitoring point.
In one embodiment, after the step of performing conditional random field modeling and Monte Carlo simulation on unmonitored point displacements according to the result of the solution, the method further comprises the following steps:
and carrying out inverse transformation on the Monte Carlo simulation result.
In one embodiment, the normalization processing and the monte carlo simulation result of the monitoring data are inversely transformed according to the following formula:
Figure BDA0002955954150000021
wherein λ is a vector, and the vector λ is composed of all elements of λ2Vector of (a), zi+λ2＞0，λ2Is to ensure zi+λ2Parameter constantly greater than 0, z is displacement of measured point before transformation, ziFor the displacement corresponding to the ith observation point, ztFor the displacement of the measuring point after transformation, gm (z + lambda) is the geometric mean value, lambda1I and t are natural numbers, which are the least squares of the residuals of the data set z.
In one embodiment, the method for trend structure determination includes a method by regression analysis, and the determination of the order of the regression equation includes the steps of:
determining a smaller order i, wherein i is a natural number;
decision P based on order iFAnd ScvValue if PF>0.05, reducing the order, otherwise, increasing the order to i +1, ScvIs a cross-validation index;
if the order is raised to i +1, comparing the S after the step is raisedcvValue if ScvIs raised and its PFWhen the preset value is reached, the step is continuously increased until Scv(i+1)<Scv(i) Or PF>Up to 0.05.
In one embodiment, the method of "performing data autocorrelation determination on deep displacement data according to the trend structure determination result" includes:
and performing autocorrelation structure judgment on the deep displacement data by adopting a maximum likelihood estimation method and combining a Matern equation.
In one embodiment, the Materrn equation is as follows:
Figure BDA0002955954150000031
in the formula, hijIs the space distance between any two displacement measuring points, v is a smooth parameter ranging from 0 to infinity, r is a range parameter, gamma (v) is a gamma equation, K is the distance between two displacement measuring pointsνBessel formula class II, which is order v, R (h)ij) And i and j are natural numbers which are autocorrelation equations of space distances of a plurality of displacement measuring points.
In one embodiment, the "interpolating a prediction for an unmonitored region" includes: and performing Krigin interpolation prediction on the unmonitored area.
In one embodiment, the maximum likelihood estimation method satisfies the following equation:
Figure BDA0002955954150000032
wherein W is ═ XTV-1X,Q＝I-XW-1XTV-1And X is a structure containing a trend and a measuring pointA matrix of coordinate information, V is a covariance matrix of original spatial data, I is an identity matrix, θ is a vector, θ includes (V, r, s), n is the number of observation points, p is the number of elements in θ, and p is 3 · y (I-X (X)TX)-1XT)zAnd z is the displacement value of the observation point.
In a second aspect, the invention also discloses a geotechnical engineering intelligent monitoring and early warning device based on probability theory, which comprises:
the coordinate acquisition module is used for acquiring the coordinates of the monitored points and the coordinates of the unmonitored points in the monitoring space;
the monitoring data acquisition module is used for acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point;
the trend structure judging module is used for judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points;
the data autocorrelation judging module is used for carrying out data autocorrelation judgment on the deep displacement data according to the trend structure judging result;
the interpolation prediction module is used for carrying out interpolation prediction on an unmonitored area according to the coordinates of the monitoring points, the coordinates of unmonitored points and the data autocorrelation judgment result;
the error solving module is used for solving the prediction error of the interpolation prediction;
the modeling Monte Carlo simulation module is used for carrying out conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solving result;
the danger probability calculation module is used for calculating the danger early warning occurrence probability according to the Monte Carlo simulation result;
and the early warning signal generation module is used for generating a real-time early warning signal according to the danger early warning occurrence probability.
The invention has the following beneficial effects: the monitoring method comprises the following steps: acquiring coordinates of monitoring points and coordinates of unmonitored points in a monitoring space; acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point; judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points; performing data autocorrelation judgment on the deep displacement data according to the trend structure judgment result; performing interpolation prediction on the unmonitored area according to the coordinates of the monitoring points, the coordinates of unmonitored points and the data autocorrelation judgment result; solving for a prediction error of the interpolated prediction; performing conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solved result; calculating the danger early warning occurrence probability according to the Monte Carlo simulation result; and generating a real-time early warning signal according to the danger early warning occurrence probability. That is to say, the invention estimates and predicts the displacement of the unmonitored region by carrying out unbiased interpolation on the real-time data, and simultaneously considers the prediction error in the calculation of the early warning index by a conditional random field modeling mode. And the real-time safety state of the monitored object is evaluated by adopting a mode of calculating the probability of danger early warning, so that the deep displacement of the soil body in the area where the measuring points are not arranged on the side slope can be predicted, the range of the early warning monitoring area is enlarged, the monitoring cost is reduced, and the early warning effect is good.
Drawings
FIG. 1 is a flow chart of the geotechnical engineering intelligent monitoring and early warning method based on probability theory.
FIG. 2 is a schematic diagram of the geotechnical engineering intelligent monitoring and early warning device based on probability theory.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and examples. It should be noted that, if not conflicting, the embodiments of the present invention and the features of the embodiments may be combined with each other within the scope of protection of the present invention.
Referring to fig. 1, the invention provides a geotechnical engineering intelligent monitoring and early warning method based on probability theory, comprising the following steps:
s1, obtaining the coordinates of the monitored points and the coordinates of the unmonitored points in the monitoring space;
the method for acquiring the coordinates of the monitored point and the coordinates of the unmonitored point can be acquired by means of site survey, a map, a coordinate measuring device, a displacement sensor with a position detection function and the like, and the acquisition mode is not particularly limited herein.
S2, acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point;
the deep displacement sensor can be arranged at a monitoring point of a rock-soil structure, and monitoring data of soil deep displacement of the monitoring point is acquired through the deep displacement sensor.
S3, judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points;
in the gaussian regression prediction (also called kriging interpolation) and the maximum likelihood estimation (i.e. the data autocorrelation structure determination method), data is required to be subjected to normal distribution, however, the acquired monitoring data is not necessarily subjected to normal distribution, which requires normalization processing on the monitoring data. Preferably, before the step of "determining a trend structure in which the deep displacement is spatially distributed according to the monitoring data of the monitoring points", the method further includes: and carrying out normalization processing on the monitoring data of the deep displacement of the soil body of the monitoring point.
In this embodiment, a Box-Cox data conversion mode is adopted to perform normalization processing on the monitoring data. Box-Cox data transformation, namely Box-Cox transformation, is a data transformation used in statistical modeling of the Box-Cox data transformation and is used for the condition that continuous response variables do not meet normal distribution. After Box-Cox transformation, the correlation of non-observable errors and predictor variables can be reduced to some extent. Wherein, the Box-Cox transformation satisfies the following formula:
Figure BDA0002955954150000051
wherein λ is a vector, and the vector λ is composed of all elements of λ2To ensure zi+λ2＞0，λ2Is to ensure zi+λ2Parameter constantly greater than 0, z is displacement of measured point before transformation, ziFor the displacement corresponding to the ith observation point, ztFor the displacement of the measuring point after transformation, gm (z + lambda) is the geometric mean value, lambda1Is the smallest residual square of the data set z.
Linear regression was performed according to the least squares method, and the regression parameters of the data were obtained by the following formula:
Figure BDA0002955954150000052
wherein, X is a matrix containing a trend structure and coordinate information of a measuring point, V is a covariance matrix of monitoring data of deep displacement of a soil body of the monitoring point, and z is a displacement value of an observation point. The order of regression equation is usually assumed in the traditional space variable prediction method, and the invention adopts P obtained by subtracting F distribution from Wald statisticFThe value determines whether a higher order regression equation is required. Generally considered as P of the regression equationF>At 0.05, the regression equation has an overfitting phenomenon, and the higher-order equation should be processed by order reduction. In addition, the accuracy of the regression equation is verified by adopting a leave-one-cross method.
Wherein the cross-validation index is obtained by the following formula:
Figure BDA0002955954150000061
in the formula (I), the compound is shown in the specification,
Figure BDA0002955954150000063
the residual value vector of the monitoring data is equal to the displacement observed quantity of each point minus the corresponding trend structure value, and n is the total number of the observed points. Generally considered as cross-validation index ScvThe larger the regression equation, the better the fit. That is, the present embodiment is realized by a regression analysis method by determining a trend structure in which deep displacements are spatially distributed.
Specifically, the method for determining the trend structure of the spatial distribution of deep displacement comprises the following steps:
determining a smaller order i, wherein i is a natural number;
decision P based on order iFAnd ScvValue if PF>0.05, reducing the order, otherwise, increasing the order to i +1, ScvIs a cross-validation index;
if the order is raised to i +1, comparing the S after the step is raisedcvValue if ScvIs raised and its PFWhen the preset value is reached, the step is continuously increased until Scv(i+1)<Scv(i) Or PF>Up to 0.05.
S4, performing data autocorrelation judgment on the deep displacement data according to the trend structure judgment result;
in this embodiment, a maximum likelihood estimation method is used in combination with a matern equation to perform data autocorrelation determination on the deep displacement data, that is, a matern clustering method is used to estimate parameters of the matern clustering by a maximum likelihood estimation method.
Wherein the Materrn equation is as follows:
Figure BDA0002955954150000062
in the formula, hijIs the space distance between any two displacement measuring points, v is a smooth parameter ranging from 0 to infinity, r is a range parameter, gamma (v) is a gamma equation, K is the distance between two displacement measuring pointsνBessel formula of the second kind, R (h), of order vij) An autocorrelation equation for the spatial distance of several displacement points, R (h)ij) Used as a parameter for interpolation prediction in the next step S5, i and j are both natural numbers.
The Materrn equation is governed by v and r, which is a spatial data autocorrelation equation with a flexible form. The autocorrelation distance of the data can be generally considered as R (h)ij) H is equal to 0.05sijThe value is obtained.
In one embodiment, the maximum likelihood estimation method satisfies the following equation:
Figure BDA0002955954150000071
wherein W is ═ XTV-1X,Q＝I-XW-1XTV-1X is a matrix containing a trend structure and coordinate information of measurement points, V is a covariance matrix of original spatial data, I is a unit matrix θ is a vector, θ contains (V, r, s), n is the number of observation points, p is the number of elements in θ, and p is 3 · y (I-X (X)TX)-1XT)zAnd z is the displacement value of the observation point, and L (theta | y) is used as a parameter for estimating the cluster in the Materrn equation.
The method well avoids the problem that a large amount of observation data is needed due to the traditional moment method-based judgment mode. Furthermore, in the moment estimation, the form of the autocorrelation equation is usually assumed, which causes a large prediction error in the case of a limited observation point. The prediction algorithm of the invention adopts a maximum likelihood estimation method and combines a Materrn equation, so that the autocorrelation structure of the spatial displacement data can be judged under the limited observation data. Since the method does not need to assume the form of an autocorrelation equation, the accuracy is higher than that of the traditional moment method.
S5, carrying out interpolation prediction on the unmonitored area according to the coordinates of the monitoring points, the coordinates of unmonitored points and the data autocorrelation judgment result;
the interpolation prediction may be a prediction method such as a kriging interpolation prediction, and in this embodiment, the interpolation prediction is a kriging interpolation prediction. It is understood that gaussian regression prediction is also called kriging interpolation prediction. The displacement of the unmeasured point in the space can be assumed as a space random variable, and the prediction and estimation of the deep displacement of the unmeasured region adopt a linear unbiased interpolation estimation mode, and the equation is as follows:
Figure BDA0002955954150000072
zkr,j＝μk(x0)+(z-μk0)Tβ(j)；
in the above formula, l is a vector having all values of 1. Beta is a(j)Is an interpolation weighting coefficient containing all monitoring points to unknown points,
Figure BDA0002955954150000073
lambda is the lagrange multiplier and,
Figure BDA0002955954150000074
is one n × neIs based on n observation points and neAnd (4) a non-measured point. Mu.sk0And the trend structure value of the observation point. Mu.sk(x0) About a spatial coordinate point x0The trend structure of (1) is obtained from the formula. z is a radical ofkr,jIs the predicted value of the jth untested point in space. The prediction error corresponding to each point is as follows:
Figure BDA0002955954150000081
in the formula, σeThe prediction error value of each measuring point is contained in one vector.
Wherein, the formula of the covariance equation is as follows:
Figure BDA0002955954150000082
where Δ x, Δ y, Δ z are the intervals between any two points on the x, y, z axes, respectively, and R () is obtained from the matern equation of step S4.
S6, solving the prediction error of the interpolation prediction;
s7, performing conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solved result;
the displacement of unmeasured points in the monitoring space of the monitoring area can be assumed as a space random variable, and Monte Carlo simulation is carried out by adopting a space random field theory.
According to random field theory, random variables with spatial autocorrelation properties obey the functional relationship shown below:
Figure BDA0002955954150000083
in the formula zjAnd mujThe displacement value of the jth point in space and its corresponding trend structure, respectively, σ is the standard deviation of the data set,
Figure BDA0002955954150000084
is the jth element of the field standard gaussian random field for the ith simulation (i.e., corresponding to the jth point in space).
Figure BDA0002955954150000085
The method can be obtained by a Cholesky matrix decomposition method:
R＝LLT (3)；
ε(i)＝Lsi；
in the above formula, L is the Cholesky factorization factor of the autocorrelation matrix R, and si is an neX 1 independent standard gaussian random number vector.
When a conditional random field is considered, the autocorrelation matrix of the monitoring data of the deep displacement of the soil body of the monitoring point is considered as the autocorrelation matrix after being constrained by a known point, and the solving formula is as follows:
Figure BDA0002955954150000091
Rcond＝D-1/2VcondD-1/2；
in the above formula, VcondTo consider the covariance matrix in the conditional case, its matrix diagonal elements,
Figure BDA0002955954150000092
the prediction error of each unmonitored point corresponding to the kriging interpolation method. D is a radical based on neN of unmonitored pointse×neOpposite angleMatrix of
Figure BDA0002955954150000093
N in (1)eAnd (4) the components. When a conditional random field simulation is performed, R in the formula (3) is replaced with Rcond(ii) a While, mu in the formula (2)jAnd
Figure BDA0002955954150000094
respectively replacing the predicted values mu by the Kelimen interpolationkr,jAnd corresponding prediction error
Figure BDA0002955954150000095
In the algorithm, a regression kriging method is adopted to obtain mukr,j。
In this embodiment, after the step of performing conditional random field modeling and monte carlo simulation on the displacement of the unmonitored point according to the result of the solution, "the method further includes:
and carrying out inverse transformation on the Monte Carlo simulation result.
Wherein, the inverse transformation of the Monte Carlo simulation result satisfies the following formula:
Figure BDA0002955954150000096
wherein λ is a vector, and the vector λ is composed of all elements of λ2Vector of (a), zi+λ2＞0，λ2Is to ensure zi+λ2Parameter constantly greater than 0, z is displacement of measured point before transformation, ziFor the displacement corresponding to the ith observation point, ztFor the displacement of the measuring point after transformation, gm (z + lambda) is the geometric mean value, lambda1I and t are natural numbers, which are the least squares of the residuals of the data set z.
S8, calculating the danger early warning occurrence probability according to the Monte Carlo simulation result;
based on the monte carlo simulation, the risk pre-warning occurrence probability can be calculated by the following formula:
Figure BDA0002955954150000101
in the formula, PfProbability of occurrence of danger warning, NTFor number of simulations, disiF is a specified early warning threshold value. Wherein the deep displacement values of the whole region comprise real-time deep displacement values of the monitored points and predicted deep displacement values of unmonitored points.
And S9, generating a real-time early warning signal based on the danger early warning occurrence probability.
The early warning signal can be sent to a mobile phone or a computer of a user, and therefore the site can be monitored in real time.
Referring to fig. 2, the present invention also discloses a geotechnical engineering intelligent monitoring and early warning device based on probability theory, which includes:
the coordinate acquisition module 1 is used for acquiring the coordinates of the monitored points and the coordinates of the unmonitored points in the monitoring space;
the monitoring data acquisition module 2 is used for acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point;
the trend structure judging module 3 is used for judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points;
the data autocorrelation judging module 4 is used for carrying out data autocorrelation judgment on the deep displacement data according to the trend structure judging result;
the interpolation prediction module 5 is used for carrying out interpolation prediction on an unmonitored area according to the coordinates of the monitored points, the coordinates of unmonitored points and the data autocorrelation judgment result;
an error solving module 6, configured to solve a prediction error of the interpolation prediction;
the modeling simulation module 7 is used for carrying out conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solving result;
the danger probability calculation module 8 is used for calculating the danger early warning occurrence probability according to the Monte Carlo simulation result;
and the early warning signal generating module 9 is used for generating a real-time early warning signal according to the danger early warning occurrence probability.
Preferably, the device further comprises a data normalization processing module and a data inverse conversion module, wherein the data normalization processing module is used for performing normalization processing on the monitoring data of the soil deep displacement of the monitoring point. And the data inverse conversion module is used for carrying out inverse conversion on the Monte Carlo simulation result.
In summary, the monitoring method of the present invention includes the following steps: acquiring coordinates of monitoring points and coordinates of unmonitored points in a monitoring space; acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point; judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points; performing data autocorrelation judgment on the deep displacement data according to the trend structure judgment result; performing interpolation prediction on the unmonitored area according to the coordinates of the monitoring points, the coordinates of unmonitored points and the data autocorrelation judgment result; solving for a prediction error of the interpolated prediction; performing conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solved result; calculating the danger early warning occurrence probability according to the Monte Carlo simulation result; and generating a real-time early warning signal according to the danger early warning occurrence probability. That is to say, the invention estimates and predicts the displacement of the unmonitored area by carrying out unbiased interpolation on the real-time data, simultaneously considers the prediction error in the calculation of the early warning index by a conditional random field modeling mode, and evaluates the real-time safety state of the monitored object by a mode of calculating the probability of danger early warning, therefore, the deep displacement of the soil body of the area where the side slope is not provided with the measuring points can be predicted, thereby enlarging the range of the early warning monitoring area, reducing the monitoring cost and having good early warning effect.
The geotechnical engineering intelligent monitoring and early warning method based on probability theory provided by the invention is introduced in detail, the principle and the implementation mode of the invention are explained by applying specific examples, and the description of the examples is only used for helping to understand the method and the core idea of the invention. Meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In summary, the present disclosure is only an embodiment of the present disclosure, and not intended to limit the scope of the present disclosure, and all equivalent structures or equivalent flow transformations made by using the present disclosure and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present disclosure, and should not be construed as limiting the present disclosure.

Claims (10)

1. A geotechnical engineering intelligent monitoring and early warning method based on probability theory is characterized by comprising the following steps:
acquiring coordinates of monitoring points and coordinates of unmonitored points in a monitoring space;
acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point;
judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points;
performing data autocorrelation judgment on the deep displacement data according to the trend structure judgment result;
performing interpolation prediction on the unmonitored area according to the coordinates of the monitoring points, the coordinates of unmonitored points and the data autocorrelation judgment result;
solving for a prediction error of the interpolated prediction;
performing conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solved result;
calculating the danger early warning occurrence probability according to the Monte Carlo simulation result;
and generating a real-time early warning signal according to the danger early warning occurrence probability.
2. The method for geotechnical engineering intelligent monitoring and early warning based on probability theory as claimed in claim 1, wherein before the step of 'determining trend structure of deep displacement distributed in space according to monitoring data of the monitoring points' further includes:
and carrying out normalization processing on the monitoring data of the deep displacement of the soil body of the monitoring point.
3. The geotechnical engineering intelligent monitoring and early warning method based on probability theory as claimed in claim 2, characterized in that after the step of performing conditional random field modeling and Monte Carlo simulation on unmonitored point displacement according to the solved result, the method further comprises:
and carrying out inverse transformation on the Monte Carlo simulation result.
4. The geotechnical engineering intelligent monitoring and early warning method based on probability theory as claimed in claim 2 or 3, characterized in that, the monitoring data normalization processing and Monte Carlo simulation result inverse transformation should satisfy the following formula:
Figure FDA0002955954140000021
wherein λ is a vector, and the vector λ is composed of all elements of λ2Vector of (a), zi+λ2＞0，λ2Is to ensure zi+λ2A parameter constantly greater than 0, z is the displacement of the measured point before conversion, ziFor the displacement corresponding to the ith observation point, ztFor the converted displacement of the measured point, gm (z + lambda) is the geometric mean value, lambda1I and t are natural numbers, which are the least squares of the residuals of the data set z.
5. The geotechnical engineering intelligent monitoring and early warning method based on probability theory as claimed in claim 1 or 2, wherein said 'performing interpolation prediction on unmonitored region' includes: and performing Krigin interpolation prediction on the unmonitored area.
6. The geotechnical engineering intelligent monitoring and early warning method based on probability theory as claimed in claim 1 or 2, wherein the trend structure determination method includes determination by regression analysis, and the determination method of regression equation order includes the following steps:
determining a smaller order i, wherein i is a natural number;
decision P based on order iFAnd ScvValue if PF>0.05, reducing the order, otherwise, increasing the order to i +1, ScvIs a cross-validation index;
if the order is raised to i +1, comparing the S after the step is raisedcvValue if ScvIs raised and its PFWhen the preset value is reached, the step is continuously increased until Scv(i+1)<Scv(i) Or PF>Up to 0.05.
7. The geotechnical engineering intelligent monitoring and early warning method based on probability theory as claimed in claim 1 or 2, wherein the method of 'making data autocorrelation judgment on deep displacement data according to the trend structure judgment result' includes:
and performing autocorrelation structure judgment on the deep displacement data by adopting a maximum likelihood estimation method and combining a Matern equation.
8. The geotechnical engineering intelligent monitoring and early warning method based on probability theory according to claim 7, characterized in that the Materrn equation is as follows:
Figure FDA0002955954140000022
in the formula, hijIs the space distance between any two displacement measuring points, v is a smooth parameter ranging from 0 to infinity, r is a range parameter, gamma (v) is a gamma equation, KνBessel formula class II, which is order v, R (h)ij) And i and j are natural numbers which are autocorrelation equations of space distances of a plurality of displacement measuring points.
9. The geotechnical engineering intelligent monitoring and early warning method based on probability theory as claimed in claim 8, wherein said maximum likelihood estimation method satisfies the following formula:
Figure FDA0002955954140000031
wherein W is ═ XTV-1X,Q＝I-XW-1XTV-1X is a matrix containing a trend structure and coordinate information of measuring points, V is a covariance matrix of original space data, I is a unit matrix, theta is a vector, theta contains (V, r and s), n is the number of observation points, p is the number of elements in theta, and p is 3-y (I-X (X-X)TX)-1XT) And z is the displacement value of the observation point.
10. The utility model provides a geotechnical engineering intelligent monitoring early warning device based on probability theory which characterized in that includes:
the coordinate acquisition module is used for acquiring the coordinates of the monitored points and the coordinates of the unmonitored points in the monitoring space;
the monitoring data acquisition module is used for acquiring monitoring data of the deep displacement of the soil body of the monitoring point according to the coordinates of the monitoring point;
the trend structure judging module is used for judging a trend structure of deep displacement distributed in space according to the monitoring data of the monitoring points;
the data autocorrelation judging module is used for carrying out data autocorrelation judgment on the deep displacement data according to the trend structure judging result;
the interpolation prediction module is used for carrying out interpolation prediction on an unmonitored area according to the coordinates of the monitoring points, the coordinates of unmonitored points and the data autocorrelation judgment result;
the error solving module is used for solving the prediction error of the interpolation prediction;
the modeling simulation module is used for carrying out conditional random field modeling and Monte Carlo simulation on the displacement of the unmonitored point according to the solving result;
the danger probability calculation module is used for calculating the danger early warning occurrence probability according to the Monte Carlo simulation result;
and the early warning signal generation module is used for generating a real-time early warning signal according to the danger early warning occurrence probability.
CN202110223708.3A 2021-03-01 2021-03-01 Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory Active CN112818567B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110223708.3A CN112818567B (en) 2021-03-01 2021-03-01 Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110223708.3A CN112818567B (en) 2021-03-01 2021-03-01 Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory

Publications (2)

Publication Number Publication Date
CN112818567A true CN112818567A (en) 2021-05-18
CN112818567B CN112818567B (en) 2023-06-16

Family

ID=75862454

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110223708.3A Active CN112818567B (en) 2021-03-01 2021-03-01 Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory

Country Status (1)

Country Link
CN (1) CN112818567B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116295604A (en) * 2023-01-04 2023-06-23 中铁十一局集团有限公司 Intelligent dust real-time monitoring and control system

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102619209A (en) * 2012-04-25 2012-08-01 重庆大学 Geotechnical engineering cataclysm dynamic real-time intelligent early warning method based on displacement monitoring
CN103578229A (en) * 2013-11-15 2014-02-12 鞍钢集团矿业公司 Mine side slope deformation monitoring and early warning system and early warning method thereof
CN104766129A (en) * 2014-12-31 2015-07-08 华中科技大学 Subway shield construction surface deformation warning method based on temporal and spatial information fusion
CN104778369A (en) * 2015-04-20 2015-07-15 河海大学 Method and system for decision making and early warning based on ground subsidence monitoring
US20170293048A1 (en) * 2016-04-09 2017-10-12 Powerchina Huadong Engineering Corporation Limited Response Surface Method for Identifying The Parameters of Burgers Model for Slope Soil
CN110309525A (en) * 2019-03-22 2019-10-08 北京北科安地科技发展有限公司 A kind of side slope geometric distortion and destroy trend calculation method
CN110837669A (en) * 2019-10-25 2020-02-25 中国地质大学（武汉） Landslide uncertain model dynamic construction method based on multi-source heterogeneous data fusion
CN111144651A (en) * 2019-12-26 2020-05-12 杭州鲁尔物联科技有限公司 Geological disaster prediction method, device and equipment
CN111210073A (en) * 2020-01-06 2020-05-29 四川省公路规划勘察设计研究院有限公司 Landslide disaster prediction method and device

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102619209A (en) * 2012-04-25 2012-08-01 重庆大学 Geotechnical engineering cataclysm dynamic real-time intelligent early warning method based on displacement monitoring
CN103578229A (en) * 2013-11-15 2014-02-12 鞍钢集团矿业公司 Mine side slope deformation monitoring and early warning system and early warning method thereof
CN104766129A (en) * 2014-12-31 2015-07-08 华中科技大学 Subway shield construction surface deformation warning method based on temporal and spatial information fusion
CN104778369A (en) * 2015-04-20 2015-07-15 河海大学 Method and system for decision making and early warning based on ground subsidence monitoring
US20170293048A1 (en) * 2016-04-09 2017-10-12 Powerchina Huadong Engineering Corporation Limited Response Surface Method for Identifying The Parameters of Burgers Model for Slope Soil
CN110309525A (en) * 2019-03-22 2019-10-08 北京北科安地科技发展有限公司 A kind of side slope geometric distortion and destroy trend calculation method
CN110837669A (en) * 2019-10-25 2020-02-25 中国地质大学（武汉） Landslide uncertain model dynamic construction method based on multi-source heterogeneous data fusion
CN111144651A (en) * 2019-12-26 2020-05-12 杭州鲁尔物联科技有限公司 Geological disaster prediction method, device and equipment
CN111210073A (en) * 2020-01-06 2020-05-29 四川省公路规划勘察设计研究院有限公司 Landslide disaster prediction method and device

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
WANG H ET AL.: "A relationship-based and object-oriented software for monitoring management during geotechnical excavation", 《ADVANCES IN ENGINEERING SOFTWARE》 *
刘亚等: "基于时间序列分析法的滑坡变形特征研究", 《四川地质学报》 *
容静等: "基于方差补偿自适应Kalman滤波的ARMA与PSO-SVM模型变形预测", 《大地测量与地球动力学》 *
蔡泽宏: "基于滑坡监测数据的预警判据研究", 《福建建筑》 *
霍冬冬等: "多源数据融合在岩质滑坡监测预警中的应用", 《四川理工学院学报(自然科学版)》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116295604A (en) * 2023-01-04 2023-06-23 中铁十一局集团有限公司 Intelligent dust real-time monitoring and control system
CN116295604B (en) * 2023-01-04 2024-02-06 中铁十一局集团有限公司 Intelligent dust real-time monitoring and control system

Also Published As

Publication number Publication date
CN112818567B (en) 2023-06-16

Similar Documents

Publication Publication Date Title
Ryu et al. Kriging interpolation methods in geostatistics and DACE model
CN107315874B (en) Sensor layout method for simultaneously acquiring local deformation and overall modal information of structure
CN108629452A (en) A kind of Weather Risk decision-making technique based on multi-mode multi-parameter DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM
KR20100116502A (en) Pridiction method for monitoring performance of plant instruments
CN104008304B (en) A kind of weary information multisensor neutral net entropy evaluation of uncertainty in measurement method
CN108519631B (en) Precipitation intensity prediction method
CN112818567A (en) Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory
CN115858609A (en) Electric vehicle charging pile state monitoring method, fault identification method and electronic equipment
CN114936504A (en) Equipment residual life prediction method and system based on Bayesian multi-source data fusion
CN107977730A (en) A kind of wind measurement method of multisensor Data Fusion technology
CN116957422B (en) Ecological environment evaluation method based on convolution self-coding and sharp point mutation model
CN112883335B (en) Real-time slope stability evaluation method combining pore water pressure
Pama et al. A new approach for estimating voltage collapse point based on quadratic approximation of PV-curves
Kyung et al. Maximum likelihood estimation for directional conditionally autoregressive models
JP2009289154A (en) Apparatus and method for determining monitoring requirement, and monitoring and diagnostic apparatus
Fuglstad et al. Non-stationary spatial modelling with applications to spatial prediction of precipitation
CN113723006B (en) LS-SVM (least squares-support vector machine) -based single-station earth change magnetic field modeling prediction method and system
Mehtätalo Localizing a predicted diameter distribution using sample information
Even-Tzur Variance factor estimation for two-step analysis of deformation networks
CN112749470B (en) Layout optimization fitting method for structural deformation sensor
CN110717290B (en) Method for drawing dam contour line based on three-time interpolation method
CN110145695B (en) Heating pipeline leakage detection method based on deep confidence network information fusion
CN112182739A (en) Aircraft structure non-probability credible reliability topological optimization design method
Sheikhhasan A comparison of interpolation techniques for spatial data prediction
KR102518804B1 (en) Method for interpolating rain gauge data considering rainfall characteristics

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