CN115512531B - A multi-monitoring point fusion early warning method for landslide hazards based on deformation order - Google Patents

Abstract

The invention relates to the technical field of geological disaster monitoring and discloses a deformation-based sensorAn orderly landslide hazard multi-monitoring point fusion early warning method comprises the following steps: acquiring deformation rates of monitoring sensors of a plurality of monitoring points of a landslide at preset moments; obtaining deformation rate information entropy H according to deformation rate calculation _V (t); calculating deformation direction association entropy H for monitoring deformation direction and target deformation direction _D (t); entropy H according to deformation rate information _V State of (t) and deformation direction associated entropy H _D And (t) acquiring a landslide early warning grade. According to the scheme, the deformation rate and deformation direction characteristics of different monitoring points on the same landslide body are utilized, so that the fusion utilization of multiple monitoring points is realized, and early warning and forecasting are carried out from the integral angle of the landslide, so that the omission and the deficiency of the existing early warning and forecasting method are avoided; and the deformation rate order and the deformation direction order are combined, an early warning basis is established, the early warning basis is complementary with the existing early warning model, and the reliability and the accuracy of disaster early warning and forecasting are improved.

Description

A multi-monitoring point fusion early warning method for landslide hazards based on deformation order

technical field

The invention relates to the technical field of geological disaster monitoring, in particular to a multi-monitoring point fusion early warning method for landslide disasters based on deformation orderliness.

Background technique

Landslide monitoring and early warning is a key link in the transformation of landslide disasters from "post-rescue" to "pre-prevention". How to improve the reliability of landslide early warning and forecast is the key and difficult point of landslide monitoring and early warning. The current mainstream monitoring and early warning mainly uses landslide displacement as the main parameter, and in the early warning process, the deformation time curve of a single monitoring point is mainly used for early warning. Since a single curve can only represent the deformation in a small range near the monitoring point, but cannot reflect the deformation characteristics of the entire slope, there are often false positives and negative negatives. Therefore, how to comprehensively utilize the deformation characteristics of different monitoring points on the same landslide body, achieve the fusion and utilization of multiple monitoring points, and carry out early warning and forecasting from the perspective of the overall landslide, so as to avoid the omissions and shortcomings of the existing early warning and forecasting methods, is a key issue for landslides in the new situation. Urgent issues that need to be addressed in monitoring and early warning.

Contents of the invention

The present invention intends to provide a multi-monitoring-point fusion early-warning method for landslide disasters, so as to solve the omissions and deficiencies in the early-warning of the deformation curve of a single monitoring point.

In order to achieve the above object, the present invention adopts the following technical scheme: a multi-monitoring point fusion early warning method for landslide disaster based on deformation orderliness, comprising the following steps:

S1: Obtain the deformation rate of the monitoring sensor at the preset time of multiple monitoring points of the landslide;

S2: Calculate and obtain the deformation rate information entropy H _V (t) according to the deformation rate;

S3: Obtain the main direction of landslide deformation according to the topographic features, that is, the target deformation direction;

S4: Obtain the deformation direction of each monitoring point, that is, the monitoring deformation direction;

S5: Calculate the deformation direction correlation entropy H _D (t) of the monitored deformation direction and the target deformation direction;

S6: According to the state of the deformation rate information entropy H _V (t) and the state of the deformation direction correlation entropy _HD (t), obtain the landslide warning level.

Preferably, as an improvement, the specific step of calculating and obtaining deformation rate information entropy according to the deformation rate includes: grouping according to the size of the deformation rate, and obtaining the distance L of the deformation rate group; obtaining the distance between each deformation rate in the deformation rate group The probability Pi(t) of occurrence in the current deformation rate group; calculating the deformation rate information entropy of the deformation rates at preset moments of the plurality of monitoring sensors.

Preferably, as an improvement, the deformation rate group spacing L is:

Among them, x _max and x _min are the maximum and minimum values of the deformation rate of the monitoring point respectively, and K is the number of deformation rate groups.

Preferably, as an improvement, the probability Pi(t) that each deformation rate in the deformation rate group appears in the current deformation rate group is:

Among them, N _i (t) is the number of monitoring sensors whose deformation rate falls within the i grouping interval at the preset time, and N is the total number of monitoring sensors.

Preferably, as an improvement, the deformation rate information entropy H _V (t) is specifically:

where i represents the grouping of the ith deformation rate.

Preferably, as an improvement, the grouping according to the magnitude of the deformation rate specifically includes: determining the number of groups according to the number of monitoring sensors, and grouping the deformation rates according to the number of groups and the magnitude of the deformation rate.

Preferably, as an improvement, the deformation direction correlation entropy _HD (t) is specifically:

Among them, i represents the i-th monitoring point, p _i is the probability of the correlation coefficient, and N is the total number of monitoring sensors.

Preferably, as an improvement, the calculation of the deformation direction correlation entropy _HD (t) of the monitoring deformation direction and the target deformation direction specifically includes: expressing the monitoring deformation direction sequence and the target deformation direction sequence according to the azimuth angle; Perform normalization processing with the target deformation direction sequence; calculate the gray correlation degree between the normalized monitoring deformation direction sequence and the target deformation direction sequence; calculate the probability distribution of the correlation coefficient according to the gray correlation degree; calculate the probability distribution of the correlation coefficient according to the gray correlation degree Deformation direction correlation entropy _HD (t).

Preferably, as an improvement, the warning levels include: blue warning, yellow warning, orange warning and red warning.

Preferably, as an improvement, the deformation rate information entropy H _V (t) states include SH _V -1, SH _V -2 and SH _V -3, wherein the SH _V -1 represents the deformation rate information entropy H _V (t) develops steadily or fluctuates up and down around the central axis with time, and there is no obvious increase or decrease trend. The SH _V -2 indicates that the change of deformation rate information entropy H _V (t) has an obvious downward trend over time , the SH _V -3 means that the deformation rate information entropy H _V (t) keeps a downward trend over time, and gradually approaches zero;

The deformation direction associated entropy _HD (t) states SH _D -1, SH _D -2, SH _D -3 represent, wherein, the SH _D -1 represents the change of deformation direction associated entropy _HD (t) with time Develop steadily or fluctuate up and down around the central axis, without obvious increase or decrease trend, said SH _D -2 means that the change of deformation direction correlation entropy _HD (t) has a clear upward trend over time, and said SH _D -3 means The deformation direction correlation entropy H _D (t) keeps increasing with time, and gradually approaches -ln(1/N), where N is the number of monitoring sensors on the same landslide mass;

When the state of the deformation rate information entropy H _V (t) is SH _V -1 and the state of the deformation direction correlation entropy _HD (t) is SH _D -1, the warning level is blue warning; the deformation rate information entropy When the H _V (t) state is SH _V -1 and the deformation direction correlation entropy _HD (t) is SH _D -2, or the deformation rate information entropy H _V (t) state is SH _V -2 and the deformation direction correlation When the entropy _HD (t) state is SH _D -1, the warning level is yellow warning; the deformation rate information entropy H _V (t) state is SH _V -1 and deformation direction correlation entropy _HD (t) state When SH _D -3, the deformation rate information entropy H _V (t) state is SH _V -2 and the deformation direction correlation entropy H _D (t) state is SH _D -2 or the deformation rate information entropy H _V (t) When the state is SH _V -3 and the deformation direction correlation entropy H _D (t) state is SH _D -1, the warning level is orange warning; the deformation rate information entropy H _V (t) state is SH _V -2 and deformation direction correlation entropy H _D (t) state is SH _D -3, the deformation rate information entropy H _V (t) state is SH _V -3 and deformation direction correlation entropy _HD (t) state is SH When _D -2 or when the deformation rate information entropy H _V (t) state is SH _V -3 and the deformation direction correlation entropy _HD (t) state is SH _D -3, the warning level is red warning.

This scheme uses the deformation rate and deformation direction characteristics of different monitoring points on the same landslide to achieve the integration and utilization of multiple monitoring points, and carry out early warning and forecasting from the overall perspective of the landslide, so as to avoid the omissions and deficiencies of the existing early warning and forecasting methods; When entering the stage of accelerated deformation, it is the overall movement of the landslide that controls the surface, and the control effect of the local microtopography is obviously weakened. At this time, the deformation of the surface tends to be orderly. According to the orderly evolution process, the deformation and failure stage of the landslide can be quantitatively described , so that the landslide early warning can be carried out according to the order. This scheme uses the monitoring sensor to obtain the deformation rate of the landslide, and quantitatively describes the deformation stage of the landslide through the orderly evolution process of the deformation rate. At the same time, the main sliding direction is selected as the target direction. According to the degree of correlation between the deformation direction and the main sliding direction obtained by the monitoring points, the order of the deformation direction is represented, and combined with the order of the deformation rate and the order of the deformation direction, an early warning basis is established, which is complementary to the existing early warning model. Improve the reliability and accuracy of disaster early warning and forecasting; use two important parameters of landslide deformation: deformation rate and deformation direction, with fewer parameters, which can describe the stage of landslide deformation and failure and reduce early warning errors.

Description of drawings

Fig. 1 is a schematic diagram of calculating deformation rate according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of deformation direction calculation according to an embodiment of the present invention.

Fig. 3 is an example landslide according to the embodiment of the present invention. Fig. 3 is a time-series evolution curve of deformation rate information entropy.

Fig. 4 is a schematic diagram of calculation of a combined displacement direction according to an embodiment of the present invention.

Fig. 5 is a diagram showing the variation of deformation rate information entropy H _V (t) with time according to the embodiment of the present invention.

Fig. 6 is a diagram showing the variation of deformation direction correlation entropy _HD (t) with time according to the embodiment of the present invention.

Detailed ways

The following is further described in detail through specific implementation methods:

Example:

A multi-monitoring-point fusion early-warning method for landslide hazards based on deformation orderliness, which specifically includes the following steps:

S1: Obtain the deformation rate of the monitoring sensor at the preset time of multiple monitoring points of the landslide;

The complete landslide monitoring data is divided into initial deformation stage, constant velocity deformation stage and accelerated deformation stage. In the process of landslide deformation and failure, the deformation rate is an important parameter to characterize the stable state of the landslide. When the displacement sensing equipment installed on the surface of a landslide reaches a certain scale and meets the statistical conditions, the overall deformation of the landslide can be analyzed statistically. The rate order is analyzed. The specific process is as follows:

Extract the deformation rate of different monitoring sensors at a certain moment, the specific process is shown in Figure 1;

Firstly, on the time series curve of landslide deformation, according to a fixed period of time (usually calculated in days or hours), the time displacement curve is divided into several small segments on the time axis, and when calculating the deformation rate at time t, extract (t- △t, t) the deformation increment △S in the time period, then the deformation rate at time t is:

S2: Use the deformation rate to group the deformation rate. The number of groups adopts the empirical formula proposed by the American scholar Sturgess, namely:

K＝1+3.322×log(N) (2)

In the formula: K is the number of deformation rate groups, and N is the number of monitoring sensors on the same landslide. If there are decimals in the above calculation results, they shall be processed according to the rounding rules.

S21: Calculate the group spacing according to the number of deformation rate groups, namely:

In the formula: L is the distance between the deformation rate groups, x _max and x _min are the maximum and minimum values of the deformation rate of the monitoring points, respectively;

S22: According to the rate grouping results, count the number of sensors whose deformation rate is in the interval in each interval, and calculate the probability in the interval, namely:

In the formula: P _i (t) is the probability of deformation rate appearing in the i-th group interval at time t, N _i (t) is the number of sensors whose deformation rate falls in the i-th group interval at time t, and N is the total number of sensors.

S23: Introduce the information entropy theory to quantitatively calculate the order of deformation rate at a certain moment:

S24: Calculate the orderliness at any time t in the entire time period from 0 to t, and then obtain the evolution process curve of deformation rate orderliness with time.

The following is an example of obtaining the evolution curve of deformation rate order over time. For example, there are 20 deformation monitoring points on a landslide, calculated in weeks, and the deformation rate of the first 10 weeks from the calculation week is counted. , calculate the deformation rate of each day according to formula (1), and obtain the maximum and minimum deformation rate statistics as follows:

In the formula, t ₀ represents the current week, t ₁ represents the week before the calculation time, and so on.

(1) According to the above information, it can be known that N=20, correspondingly according to formula (2), the number of groups K=5.322 can be obtained, and K=5.

The histograms of deformation rates within 10 weeks from t ₀ to t ₉ were respectively established, and t ₀ was taken as an example for illustration.

(2) First, according to formula (3) and Table 1, calculate the distance between the histogram groups at time t ₀ , L ₀ :

(3) Then the grouping rules at time t ₀ can be established:

[1.2～1.6), [1.6～2.0), [2.0～2.4), 2.4～2.8), [2.8～3.2]

(4) According to the above grouping rules, according to the formula (4) to count the frequency of the speed in each interval, P ₁ (t ₀ ), P ₂ (t ₀ ), P ₃ (t ₀ ), P ₄ (t ₀ ), P ₅ (t ₀ ), obtained by calculation: P ₁ (t ₀ )=0.15, P ₂ (t ₀ )=0.14, P ₃ (t ₀ )=0.36, P ₄ (t ₀ )=0.21, P ₅ (t ₀ )=0.14.

(5) Calculate the deformation rate information entropy at time t ₀ according to formula (5):

(6) Calculate the deformation rate information entropy at time t ₀ , t ₁ , t ₂ , t ₃ , t ₄ , t ₅ , t ₆ , t ₇ , t ₈ , and t ₉ to obtain The evolution curve is specifically shown in Figure 3, and Figure 3 is the evolution process curve of the deformation rate information entropy over time.

S3: Obtain the main direction of landslide deformation, that is, the target deformation direction, according to the topographic features.

For typical geological disasters, a main slip direction is usually determined during the investigation process, and this main slip direction can be used as our target direction.

S4: Obtain the deformation direction of each monitoring point, that is, the monitoring deformation direction;

According to the degree of correlation between the deformation direction and the main sliding direction obtained at each monitoring point, the order of the deformation direction is expressed. Obviously, the greater the degree of correlation, the stronger the order of the deformation direction system. According to the gray theory, this correlation can be obtained by gray Represented by the correlation coefficient.

S51: In the process of orderly analysis, only the direction of the plane is considered, and according to the time interval determined when the deformation rate is solved in S1, through vector calculation, the combined direction of the deformation plane is obtained, and it is expressed by the spatial azimuth.

The deformation increment directly acquired on the sensor includes the deformation increment ΔSx in the x direction and the deformation increment ΔSy in the y direction. Combined with Figure 4 (schematic diagram for calculating the combined displacement direction), the combined displacement direction angle α can be obtained by calculation:

When ΔS _x >0 and ΔS _y ≥0, it is the first quadrant, and the resultant displacement direction angle α=α;

When ΔS _x <0 and ΔS _y ≥0, it is the second quadrant, and at this time the sum displacement direction angle α=180°-α;

When ΔS _x <0 and ΔS _y <0, it is the third quadrant, and at this time the sum displacement direction angle α=180°+α;

When ΔS _x >0 and ΔS _y <0, it is the fourth quadrant, and at this time the sum displacement direction angle α=360°-α;

When ΔS _x =0 and ΔS _y >0, then α=90°;

When ΔS _x =0 and ΔS _y <0, then α=270°;

S52: Deformation direction sequence expression: Let x=(x ₁ ,x ₂ ,…,x _N ) be the deformation direction sequence formed by each monitoring point on the disaster body at a certain time t, y=(y ₁ ,y ₂ ,…, y _N ) is the target deformation direction sequence, where y ₁ =y ₂ =,...,y _N =y*, y* is the main sliding direction of the landslide. The deformation direction is represented by the azimuth angle. Since the azimuth angle ranges from 0 to 360°, the difference between the deformation direction of each monitoring point and the main sliding direction cannot be completely processed according to the ordinary sequence, as shown in Figure 2.

When the azimuth x does not exceed y*+180°, the angle difference between x and y* can be directly subtracted from the two, and when x exceeds y*+180°, the difference between x and y* should actually be After x-360°, make difference with y*. Then the x deformation sequence can be divided into two parts, namely:

S53: Deformation direction normalization processing: normalize the deformation direction observation sequence x=(x ₁ ,x ₂ ,…,x _N ) and target direction sequence y=(y ₁ ,y ₂ ,…,y _N ) processing, the result is as follows:

In the formula: x _i is the deformation direction value of the i-th monitoring point at time t, and x _min and x _max are the maximum and minimum values of the direction of the monitoring point at time t, respectively. Obviously, for the target direction sequence y, since the target direction of each monitoring point is the main sliding direction of the landslide, the result after dimensionless is all 1, that is: y′ _i =1.

S54: Calculation of gray correlation degree: the gray correlation coefficient of sequence x' and y' at time t is expressed as:

In the formula: Δ(min)=min(|x′ _i -y′ _i |), Δ(max)=max(|x′ _i -y′ _i |), 0<ρ<1 is the resolution coefficient, by adjusting This value can change the significance of the correlation coefficient, here ρ=0.5.

S55: Calculate the probability distribution of the correlation coefficient:

S56: Calculate the deformation direction correlation entropy H _D (t) of the monitoring deformation direction and the target deformation direction, and the deformation direction correlation entropy _HD (t) is specifically:

In the formula, p _i (t) represents the correlation coefficient probability at time t.

S6: According to the state of the deformation rate information entropy H _V (t) and the state of the deformation direction correlation entropy _HD (t), obtain the landslide warning level.

Formula (5) is the final expression of the orderliness of the deformation rate. From the expression and its analysis process, it can be seen that the information entropy of the rate decreases with the increase of the orderliness. When the deformation rates of all monitoring points tend to be uniform When the value is , the deformation rate information entropy tends to zero, and at this time it can be considered that the landslide is about to occur. Therefore, the deformation rate approaching zero can be used as an early warning criterion for the order of the deformation rate.

For the order of the deformation direction, formula (9) is used to calculate the gray relational entropy. Obviously, for the gray relational entropy determined in formula (9), the larger _HD is, the stronger the order of the deformation direction is. When all the deformation When the directions all tend to one direction, ξ _i is a constant, at this time p _i =1/N, the gray relational entropy reaches the maximum, and its maximum value is -ln(1/N). Therefore, the early warning criterion can be determined by comparing the gray relational entropy of the deformation direction obtained actually with the maximum gray relational entropy-ln(1/N).

When a landslide is about to occur, its deformation rate and direction tend to be the same. Therefore, a single early warning of the deformation rate and deformation direction is likely to cause false positives and false positives. Therefore, it is necessary to combine the two. The mutual complementation and confirmation of landslides can improve the accuracy of landslide early warning and forecasting.

Firstly, three states of deformation rate information entropy and deformation direction correlation entropy are defined, among which deformation rate information entropy states SH _V -1, SH _V -2, SH _V -3 represent, deformation direction correlation entropy states are represented by SH _D -1, SH _D -2 and SH _D -3 indicate that each state is specifically defined as follows:

The change of deformation rate information entropy H _V (t) over time is shown in Figure 5 (schematic diagram of the order division of deformation rate).

SH _V -1 means that the deformation rate information entropy H _V (t) develops steadily or fluctuates around the central axis with time, and there is no obvious increase or decrease trend;

SH _V -2 means that the change of deformation rate information entropy H _V (t) with time has an obvious downward trend (the rate of change in more than three consecutive time intervals is basically the same or increases);

SH _V -3 means that the deformation rate information entropy H _V (t) keeps a downward trend with time, and gradually approaches zero.

The variation of the deformation direction correlation entropy _HD (t) over time is shown in Figure 6 (schematic diagram of the order division of the deformation direction).

SH _D -1 means that the deformation direction correlation entropy _HD (t) develops steadily or fluctuates around the central axis with time, without obvious increase or decrease trend;

SH _D -2 means that the deformation direction correlation entropy _HD (t) has an obvious upward trend over time (the rate of change in more than three consecutive time intervals is basically the same or increases);

SH _D -3 means that the deformation direction correlation entropy _HD (t) keeps increasing with time, and gradually approaches -ln(1/N).

Based on the state definitions of the above two entropy values, the early warning criterion matrix is constructed as follows:

This scheme uses the deformation rate and deformation direction characteristics of different monitoring points on the same landslide to achieve the integration and utilization of multiple monitoring points, and carry out early warning and forecasting from the overall perspective of the landslide, so as to avoid the omissions and deficiencies of the existing early warning and forecasting methods; When entering the stage of accelerated deformation, it is the overall movement of the landslide that controls the surface, and the control effect of the local microtopography is obviously weakened. At this time, the deformation of the surface tends to be orderly. According to the orderly evolution process, the deformation and failure stage of the landslide can be quantitatively described , so that the landslide early warning can be carried out according to the order. This scheme uses the monitoring sensor to obtain the deformation rate of the landslide, and quantitatively describes the deformation stage of the landslide through the orderly evolution process of the deformation rate. At the same time, the main sliding direction is selected as the target direction. According to the degree of correlation between the deformation direction and the main sliding direction obtained by the monitoring points, the order of the deformation direction is represented, and combined with the order of the deformation rate and the order of the deformation direction, an early warning basis is established, which is complementary to the existing early warning model. Improve the reliability and accuracy of disaster early warning and forecasting; use two important parameters of landslide deformation: deformation rate and deformation direction, with fewer parameters, which can describe the stage of landslide deformation and failure and reduce early warning errors.

What is described above is only an embodiment of the present invention, and common knowledge such as specific technical solutions and/or characteristics known in the solutions will not be described here too much. It should be pointed out that for those skilled in the art, some modifications and improvements can be made without departing from the technical solutions of the present invention. In the present invention, unless otherwise specified and limited, the terms "installation", Terms such as "connected", "connected" and "fixed" should be understood in a broad sense, for example, they can be fixedly connected, detachably connected, or integrally connected; they can be directly connected or indirectly connected through an intermediary, It can be a connection between two components. Those of ordinary skill in the art can understand the specific meanings of the above terms in the present invention according to specific situations. The scope of protection required by this application shall be based on the content of the claims, and the specific implementation methods and other records in the specification may be used to interpret the content of the claims.

Claims (8)

1. A multi-monitoring point fusion early warning method for landslide disaster based on deformation orderliness, is characterized in that, comprises the following steps: S1: Obtain the deformation rate of the monitoring sensor at the preset time of multiple monitoring points of the landslide; S2: Calculate and obtain the deformation rate information entropy H _V (t) according to the deformation rate; S3: Obtain the main direction of landslide deformation according to the topographic features, that is, the target deformation direction; S4: Obtain the deformation direction of each monitoring point, that is, the monitoring deformation direction; S5: Calculate the deformation direction correlation entropy H _D (t) of the monitored deformation direction and the target deformation direction; S6: Obtain the landslide warning level according to the state of the deformation rate information entropy H _V (t) and the state of the deformation direction correlation entropy H _D (t); The specific steps of calculating and obtaining the deformation rate information entropy according to the deformation rate include: grouping according to the size of the deformation rate, and obtaining the deformation rate group spacing L; obtaining the probability that each deformation rate in the deformation rate group appears in the current deformation rate group Pi(t); Calculate the deformation rate information entropy of the deformation rate at the preset moment of multiple monitoring sensors; The calculation of the deformation direction correlation entropy H _D (t) of the monitoring deformation direction and the target deformation direction specifically includes: expressing the monitoring deformation direction sequence and the target deformation direction sequence according to the azimuth angle; normalizing the monitoring deformation direction sequence and the target deformation direction sequence Gray correlation degree calculation is performed on the monitoring deformation direction sequence and the target deformation direction sequence after normalization processing; the correlation coefficient probability distribution is calculated according to the gray correlation degree; the deformation direction correlation entropy H _D (t is calculated according to the correlation coefficient probability distribution ). 2. A kind of landslide disaster multi-monitoring point fusion early warning method based on deformation orderliness according to claim 1, is characterized in that: described deformation rate group distance L is:

Among them, x _max and x _min are the maximum and minimum values of the deformation rate of the monitoring point respectively, and K is the number of deformation rate groups. 3. A kind of landslide hazard multi-monitoring point fusion early warning method based on deformation orderliness according to claim 1, is characterized in that: the probability Pi that each deformation rate occurs in the current deformation rate group in the deformation rate group (t) is:

Among them, N _i (t) is the number of monitoring sensors whose deformation rate falls within the i grouping interval at the preset time, and N is the total number of monitoring sensors. 4. A kind of landslide hazard multi-monitoring point fusion early warning method based on deformation orderliness according to claim 3, is characterized in that: described deformation rate information entropy H _V (t) is specifically:

where i represents the grouping of the ith deformation rate. 5. A kind of landslide disaster multi-monitoring point fusion early warning method based on deformation orderliness according to claim 1, is characterized in that: described grouping according to the size of deformation rate specifically comprises: according to monitoring sensor quantity, determines grouping number, The deformation rates are grouped according to the number of groups and the size of the deformation rate. 6. A kind of landslide hazard multi-monitoring point fusion early warning method based on deformation orderliness according to claim 1, is characterized in that: described deformation direction correlation entropy H _D (t) is specifically:

Among them, i represents the i-th monitoring point, p _i is the probability of the correlation coefficient, and N is the total number of monitoring sensors. 7. A multi-monitoring point fusion early warning method for landslide hazards based on deformation orderliness according to claim 1, characterized in that: said early warning levels include: blue early warning, yellow early warning, orange early warning and red early warning. 8. A kind of landslide hazard multi-monitoring point fusion early warning method based on deformation orderliness according to claim 7, is characterized in that: described deformation rate information entropy H _V (t) state comprises SH _V -1, SH _V -2 and SH _V -3, wherein, said SH _V -1 means that the deformation rate information entropy H _V (t) develops steadily with time or fluctuates up and down around the central axis, without obvious increasing or decreasing trend, said SH _V -2 indicates that the change of deformation rate information entropy H _V (t) has an obvious downward trend with time, and the SH _V -3 indicates that the change of deformation rate information entropy H _V (t) with time continues to maintain a downward trend, and gradually close to zero; The deformation direction associated entropy _HD (t) states SH _D -1, SH _D -2, SH _D -3 represent, wherein, the SH _D -1 represents the change of deformation direction associated entropy _HD (t) with time Develop steadily or fluctuate up and down around the central axis, without obvious increase or decrease trend, said SH _D -2 means that the change of deformation direction correlation entropy _HD (t) has a clear upward trend over time, and said SH _D -3 means The deformation direction correlation entropy H _D (t) keeps increasing with time, and gradually approaches -ln(1/N), where N is the number of monitoring sensors on the same landslide mass; When the state of the deformation rate information entropy H _V (t) is SH _V -1 and the state of the deformation direction correlation entropy _HD (t) is SH _D -1, the warning level is blue warning; the deformation rate information entropy When the H _V (t) state is SH _V -1 and the deformation direction correlation entropy _HD (t) is SH _D -2, or the deformation rate information entropy H _V (t) state is SH _V -2 and the deformation direction correlation When the entropy _HD (t) state is SH _D -1, the warning level is yellow warning; the deformation rate information entropy H _V (t) state is SH _V -1 and deformation direction correlation entropy _HD (t) state When SH _D -3, the deformation rate information entropy H _V (t) state is SH _V -2 and the deformation direction correlation entropy H _D (t) state is SH _D -2 or the deformation rate information entropy H _V (t) When the state is SH _V -3 and the deformation direction correlation entropy H _D (t) state is SH _D -1, the warning level is orange warning; the deformation rate information entropy H _V (t) state is SH _V -2 and deformation direction correlation entropy H _D (t) state is SH _D -3, the deformation rate information entropy H _V (t) state is SH _V -3 and deformation direction correlation entropy _HD (t) state is SH When _D -2 or when the deformation rate information entropy H _V (t) state is SH _V -3 and the deformation direction correlation entropy _HD (t) state is SH _D -3, the warning level is red warning.

Images