CN102509420B - Landslide forecast method based on deformation information of critical-sliding area - Google Patents

Abstract

The invention relates to a landslide catastrophe time forecast and prediction method for surface mine side slopes, road side slopes and mountain side slopes. According to the field monitoring data and the creep law and mechanism of landslide disasters, the method puts forward a side slope creep curve of the part impending landslide but not sliding (critical-sliding area) and four stages of deformation and development, deduces a curvilinear equation of the deformation, velocity and acceleration of the side slope in the critical-sliding area and a landslide forecast time model based on the deformation information of the critical-sliding area, and establishes a Verhulst gray model based on the deformation information of the critical-sliding area so as to determine parameters in the landslide forecast time model. The method provided by the invention has the beneficial effects that: the workload is greatly reduced in comparison with the existing landslide forecast methods while the accuracy is improved, and the judging method for forecasting the landslide disaster time is simplified, thereby being more convenient and reliable.

Description

A kind of slide prediction method based on facing the skating area deformation information

Technical field

The present invention relates to open mine side slope, road slope and the forecast of massif slope and land slide hazard prediction, particularly a kind of based on the landslide disaster creep mechanism and close on landslide but the slide prediction method of landing position (facing skating area) slope deforming information not.

Background technology

Landslide hazard prediction and warning is the advanced subject of current international landslide disaster research and environmental geology research field, is rationally to solve mankind's activity and earth's surface one of the key issue of plastid equilibrium relation naturally, has important theory significance and practical significance.

The temporal prediction of landslide disaster, can be divided into phenomenon forecast and Displacement Forecast with regard to existing theory and method.The phenomenon forecast is the directly perceived forecasting procedures of people to the experience accumulation of landslide omen reflection.According to the flip-flop of some natural geofactor, the premonitory phenomenon on landslides such as the expansion of surface cracks, surface water leakage, underground water table decline, the increase of rock noise frequency, can roughly judge the unsafe condition of side slope and the time that may destroy.Obviously, this method is insecure, and it can not provide the correct time on landslide.Utilize these premonitory phenomenons can only warn people's side slope in the hole, landslide is about to occur.Therefore, only from slope deforming information and landslide disaster mechanism, start with, could obtain Time Forecast comparatively exactly.Research shows, side slope always produced certain creep process before destroying, and illustrated and judged that the development trend of deformation different phase is the basis of landslide disaster Displacement Forecast." vegetarian rattan model " by the filial piety of Japanese scholars vegetarian rattan enlightening based on Landslide Monitoring curve and creep theory proposition once repeatedly more successfully forecast landslide disaster.In recent years, multidigit scholar and research institution have proposed landslide and rainfall intensity, the critical relation curve of duration according to rainfall data and Landslide Monitoring data.But the deformation information that these forecasting models adopt, the displacement of normally taking from slipping plane or slide mass.Along with scientific and technological development, a large amount of advanced instrument and equipments are used to the monitoring of slope deforming, wherein mostly is expensive built-in monitoring instruments, although these monitoring equipments or monitoring system can reach very high precision and dynamic, real-time monitoring effect, but follow slope instability, landslide, a lot of monitoring points are destroyed, and cause huge economic loss.

Accordingly, the present invention proposes a kind of based on the landslide disaster creep mechanism and close on landslide but the slide prediction model of landing position (facing skating area) slope deforming information not.

Summary of the invention

The object of the present invention is to provide a kind of based on the landslide disaster creep mechanism and face the directly perceived, simple, economic of skating area Monitoring of Slope Deformation information and efficiently, the slide prediction method based on facing the skating area deformation information of the tremendous economic loss that can effectively avoid expensive landslide monitoring instrument to cause with the sliding mass sliding failure.

Technical scheme of the present invention:

(1) the landslide Catastrophe Process faces skating area slope creep curve and deformation stage.

Reach Monitoring of Slope Deformation experience for many years according to the landslide disaster creep mechanism, skating area slope creep rule is faced in proposition: as shown in Figure 1, come down but landing position (facing skating area) does not occur close, side slope is subject to start distortion after certain External force interference, and initial stage speed is slow, to certain hour, speed is accelerated, and after increasing to the landslide generation, is subject to the slope system approximately, distortion slows down again gradually, and the Slow Deformation state finally tends towards stability.In the Catastrophe Process of landslide, face skating area Rock And Soil distortion and can be divided into four-stage.

The 1st stage: m-n section.Linearly, rate of deformation v is basicly stable for this stage distortion-time curve, and distortion acceleration a is 0, claims that this stage is initial constant speed deformation stage;

The 2nd stage: n-o section.This stage deformation velocity increases sharply, and the distortion acceleration increases, and to the o point, distortion acceleration maximum, claim that this stage is that I accelerates deformation stage;

The 3rd stage: o-p section.This stage deformation velocity still continues to increase, but be out of shape acceleration, diminishes gradually, to p point landslide, occur, now, and the deformation velocity maximum, the distortion acceleration is 0, claims that this stage is that II accelerates deformation stage;

The 4th stage: p-q section.This stage deformation velocity diminishes gradually and is returned to the constant speed deformation velocity stage, and curve is elbow down type, the distortion acceleration contrary with velocity reversal, its numerical value first increases, after reduce, claim that this stage is the deceleration deformation stage.

(2) the landslide disaster forecasting model based on facing skating area slope monitoring information.

The present invention proposes, and to face skating area slope creep curve similar in the Verhulst model of foundation in 1837 with German biomathematician P.F.Verhulst.The two is combined, propose " the landslide disaster forecasting model based on facing skating area slope monitoring information ".

$\left\{\begin{array}{c}x=\frac{m{x}_1}{n{x}_1+(m-n{x}_1)e^{m(t-t_1)}}\\ v=\frac{-m^2{x}_1(m-n{x}_1)e^{m(t-t_1)}}{{[n{x}_1+(m-n{x}_1)e^{m(t-t_1)}]}^2}\\ a=\frac{m^3{x}_1(m-n{x}_1)e^{m(t-t_1)}[(m-n{x}_1)e^{m(t-t_1)}-n{x}_1]}{{[n{x}_1+(m-n{x}_1)e^{m(t-t_1)}]}^3}\end{array}\right.---\left(1\right)$

In formula, the displacement of x-slope deforming; V-slope deforming speed; A-slope deforming acceleration; x ₁, t ₁be respectively initial displacement value and initial time; M, n are coefficients, change with different Landslides and different landslide stages, and available Grey Theory is solved.

By " the landslide disaster forecasting model based on facing skating area slope monitoring information ", when distortion acceleration a=0, slope deforming speed v maximum, the time tr that this time had both occurred for landslide:

$t_r=\frac{1}{m}\ln \left(\frac{n{x}_1}{m-n{x}_1}\right)+t_1---\left(2\right)$

(3) the Verhulst gray model based on displacement information, solve Coefficient m, n.

Because the slope deforming data of monitoring are discrete data, so Verhulst model and Grey models GM (1,1) are combined, the grey solving model of Coefficient m, n is proposed.

[m，n] ^T＝(B ^TB) ^-1B ^TY (3)

In formula, $B=\left[\begin{array}{cc}-z^{\left(1\right)}\left(2\right)& {\left(z^{\left(1\right)}\left(2\right)\right)}^2\\ -z^{\left(1\right)}\left(3\right)& {\left(z^{\left(1\right)}\left(3\right)\right)}^2\\ .& .\\ .& .\\ .& .\\ -z^{\left(1\right)}\left(n\right)& {\left(z^{\left(1\right)}\left(n\right)\right)}^2\end{array}\right],$ $Y=\left[\begin{array}{c}{x}^{\left(0\right)}\left(2\right)\\ x^{\left(0\right)}\left(3\right)\\ .\\ .\\ .\\ x^{\left(0\right)}\left(n\right)\end{array}\right];$ X ⁽⁰⁾(i)-raw data row; z ⁽¹⁾(i)-x ⁽¹⁾(i) next-door neighbour's average generation sequence; x ⁽¹⁾(i)-x ⁽⁰⁾(i) cumulative generated data row.

(4), based on model of the present invention, adopt C++ to call the Matlab engine and developed " the slide prediction model based on facing the skating area deformation information " calculation procedure.Use this program, only need lasting input face the skating area deformation measurement data, face skating area slope creep curve before and after can automatically drawing landslide disaster, solve Coefficient m, n, the forecast landslide disaster time, realize the visual and output of prog chart telogenesis fruit.

Landslide Prediction flow process of the present invention:

(1) the selected skating area side slope of facing, lay monitoring point and carry out data acquisition, obtains the primary monitoring data row;

(2) obtain cumulative generated data row and next-door neighbour's average generation sequence by the raw data row, bring following formula (1) into,

[m，n] ^T＝(B ^TB) ^-1B ^TY (1)

Wherein, in formula $B=\left[\begin{array}{cc}-z^{\left(1\right)}\left(2\right)& {\left(z^{\left(1\right)}\left(2\right)\right)}^2\\ -z^{\left(1\right)}\left(3\right)& {\left(z^{\left(1\right)}\left(3\right)\right)}^2\\ .& .\\ .& .\\ .& .\\ -z^{\left(1\right)}\left(n\right)& {\left(z^{\left(1\right)}\left(n\right)\right)}^2\end{array}\right],$ $Y=\left[\begin{array}{c}{x}^{\left(0\right)}\left(2\right)\\ x^{\left(0\right)}\left(3\right)\\ .\\ .\\ .\\ x^{\left(0\right)}\left(n\right)\end{array}\right];$ X ⁽⁰⁾(i) be the raw data row; z ⁽¹⁾(i) be x ⁽¹⁾(i) next-door neighbour's average generation sequence; x ⁽¹⁾(i)-x ⁽⁰⁾(i) be cumulative generated data row;

Obtain Coefficient m, n;

(3) Coefficient m of above-mentioned steps, n are brought into to following formula (2):

$\left\{\begin{array}{c}x=\frac{m{x}_1}{n{x}_1+(m-n{x}_1)e^{m(t-t_1)}}\\ v=\frac{-m^2{x}_1(m-n{x}_1)e^{m(t-t_1)}}{{[n{x}_1+(m-n{x}_1)e^{m(t-t_1)}]}^2}\\ a=\frac{m^3{x}_1(m-n{x}_1)e^{m(t-t_1)}[(m-n{x}_1)e^{m(t-t_1)}-n{x}_1]}{{[n{x}_1+(m-n{x}_1)e^{m(t-t_1)}]}^3}\end{array}\right.$

Wherein, x ₁for initial displacement value t ₁for initial time;

Obtain slope deforming displacement x, slope deforming speed v and be slope deforming acceleration a;

(4) the slope deforming displacement x obtained according to step 3, slope deforming speed v and be slope deforming acceleration a, then by step 2 obtain Coefficient m, n brings formula (3) into:

$t_r=\frac{1}{m}\ln \left(\frac{n{x}_1}{m-n{x}_1}\right)+t_1---\left(3\right)$

Obtain the time t that landslide occurs _r.

Annotate: along with the increase of primary monitoring data, the Coefficient m solved by gray model, n can change, so the slope deforming incipient stage extends monitoring periods relatively, if distortion is accelerated to shorten monitoring periods, up-to-date Monitoring Data is more, and the slide prediction precision is higher.

The invention has the beneficial effects as follows: existing other slide prediction method workload greatly reduces, and has improved precision simultaneously, has simplified the determination methods of landslide disaster Time Forecast, convenient and reliable.

The accompanying drawing explanation

Fig. 1 is that the present invention's Catastrophe Process that comes down faces skating area slope creep curve synoptic diagram;

Fig. 2 is the invention process example sliding mass zone and monitoring point schematic diagram;

Fig. 3 is the invention process example sliding mass and faces monitoring point, skating area distortion schematic diagram.

Fig. 4 is that the invention process example faces monitoring point, skating area H1 deformation information and slide prediction result schematic diagram.

Fig. 5 is that the invention process example faces monitoring point, skating area H2 deformation information and slide prediction result schematic diagram.

Embodiment

Below in conjunction with specific embodiment, technical scheme of the present invention is described further.

Embodiment: ore deposit, the Hebei landslide monitoring data of take are carried out Time Prediction of Landslide as basis, verify the reliability of the adjustment model of the present invention and precision.

(1) selected landslide, lay monitoring point

On the stope of Mou Kuang North, Hebei on September 7 in 2004, dish 68m step is found micro-cracks, by careful geologic examination, has determined potential zone, landslide.For further monitoring and the Study of Landslides Catastrophe Process, in position, potential landslide, upper and lower three steps have been laid G4, nine monitoring points of H1～H8 (G4 is original GPS monitoring point).As shown in Figure 2, H5～H8 is positioned on sliding mass, and with the landslide failure, and G4, H1～H4 are positioned at the trailing edge in sliding mass crack, except H3, on landslide, occur to destroy after two months, and all the other 4 monitoring points are not all damaged.

(2) Monitoring of Slope Deformation result and landslide developing stage

From year mid-May in September, 2004 to 2005 each monitoring point the distortion achievement as shown in Figure 3.With the present invention, coming down, to face skating area slope creep curve corresponding for Catastrophe Process, and this landslide is faced the skating area creep and developed for 4 stages and be: 1. initial constant speed deformation stage: at the beginning of in September, 2004～12 month, slope deforming is slow, landslide trailing edge and survey limit and the drawing crack seam occurs; 2. I accelerates deformation stage: at the beginning of 2004 12 months～in early March, 2005, rate of deformation increases; 3. II accelerates deformation stage: in early March, 2005～late March, and rate of deformation continues to increase, to landslide generation in evening March 26; 4. deformation stage slows down: in by the end of March, 2005～mid-May, near landslide but the position of landing does not occur, deformation velocity diminishes gradually and is returned to the initial constant speed deformation velocity stage.

(3) slide prediction result

According to the present invention " the landslide disaster forecasting model based on facing skating area slope monitoring information ", this ore deposit Time Prediction of Landslide formula of deriving.As shown in table 1, obtain coefficient row m, n and slide prediction time t _r.Wherein come down in Catastrophe Process, the slope displacement of measuring point H1, H2, speed, accelerating curve are as shown in Figure 4.

Table 1 slide prediction outcome table

(4) result relatively

The actual time that landslide occurs is evening on March 26th, 2005, and the 186th day, the result of prediction and distortion composes curve and actual conditions were more identical, illustrate that model of the present invention has higher forecast precision.

Claims (1)

1. the slide prediction method based on facing the skating area deformation information, is characterized in that, specifically comprises the following steps:

(1) the selected skating area side slope of facing, lay monitoring point and carry out data acquisition, obtains the primary monitoring data row;

(2) obtain cumulative generated data row and next-door neighbour's average generation sequence by the raw data row, bring following formula (1) into:

[m，n] ^T＝(B ^TB) ^-1B ^TY (1)

Wherein, in formula $B=\left[\begin{array}{cc}-z^{\left(1\right)}\left(2\right)& {\left(z^{\left(1\right)}\left(2\right)\right)}^2\\ -z^{\left(1\right)}\left(3\right)& {\left(z^{\left(1\right)}\left(3\right)\right)}^2\\ .& .\\ .& .\\ .& .\\ -z^{\left(1\right)}\left(n\right)& {\left(z^{\left(1\right)}\left(n\right)\right)}^2\end{array}\right],$ $Y=\left[\begin{array}{c}{x}^{\left(0\right)}\left(2\right)\\ x^{\left(0\right)}\left(3\right)\\ .\\ .\\ .\\ x^{\left(0\right)}\left(n\right)\end{array}\right];$ X ⁽⁰⁾(i) be the raw data row; z ⁽¹⁾(i) be x ⁽¹⁾(i) next-door neighbour's average generation sequence; x ⁽¹⁾(i)-x ⁽⁰⁾(i) be cumulative generated data row,

Solution formula 1 obtains Coefficient m, n;

(3) Coefficient m of above-mentioned steps, n are brought into to slope distortion, speed, accelerating curve equation, as shown in formula (2):

$\left\{\begin{array}{c}x=\frac{m{x}_1}{n{x}_1+(m-n{x}_1)e^{m(t-t_1)}}\\ v=\frac{-m^2{x}_1(m-n{x}_1)e^{m(t-t_1)}}{{[n{x}_1+(m-n{x}_1)e^{m(t-t_1)}]}^2}\\ a=\frac{m^3{x}_1(m-n{x}_1)e^{m(t-t_1)}[(m-n{x}_1)e^{m(t-t_1)}-n{x}_1]}{{[n{x}_1+(m-n{x}_1)e^{m(t-t_1)}]}^3}\end{array}\right.$

Wherein, x ₁for initial displacement value t ₁for initial time;

Obtain slope deforming displacement x, slope deforming speed v and be slope deforming acceleration a;

(4) the slope deforming displacement x obtained according to step 3, slope deforming speed v and be slope deforming acceleration a, then by step 2 obtain Coefficient m, n brings formula (3) into:

$t_r=\frac{1}{m}\ln \left(\frac{n{x}_1}{m-n{x}_1}\right)+t_1---\left(3\right)$

Obtain the time t that landslide occurs _r.

Images