CN-121702477-B - Method and equipment for early warning of pre-collapse of rock slope protection network based on OTDR

CN121702477BCN 121702477 BCN121702477 BCN 121702477BCN-121702477-B

Abstract

The invention relates to the field of rock slope disaster monitoring and early warning, in particular to an early warning method and equipment for pre-collapse of a rock slope protection network based on OTDR. When falling rocks impact occurs, the invention utilizes the phenomenon that the optical fiber transmission signal generates energy attenuation, phase disturbance or scattering mode change to collect reflection curve data distributed along the optical fiber in real time, and the spatial distribution characteristics of the signal are interpreted through an algorithm to realize the positioning of the falling rocks. Meanwhile, the invention can automatically adjust the judging threshold according to the base line slope, the curve attenuation characteristic and the environmental noise of different side slope sections, thereby effectively avoiding the problems of false alarm and false omission caused by the traditional fixed threshold method when the noise is large or multi-stage attenuation exists. And further, the manual inspection frequency and the emergency maintenance cost are reduced, the risks of secondary disasters and traffic accidents are reduced, the road traffic safety is ensured, the traffic running efficiency is improved, and the method has good popularization and application values and engineering feasibility.

Inventors

CAO SHENGLIANG
ZHANG JUAN
LIU YI
LI WENSHUAI
SHI JIE
DONG CHANGSONG
HUANG YONGYI
QIU YOUQIANG

Assignees

中交第一公路勘察设计研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20260206

Claims (9)

1. An OTDR-based rock slope protection network pre-collapse early warning method is characterized by comprising the following steps: S1, continuously arranging a plurality of sensing optical fibers parallel to the ground on a main beam of a protection network to be early-warned, and acquiring a light reflection curve sequence of the protection network to be early-warned in real time; s2, carrying out smooth noise reduction treatment on the light reflection curve sequence and outputting a smooth sequence; S3, calculating a multi-scale slope characteristic of the smooth sequence, and generating a self-adaptive adjustment dynamic threshold according to the multi-scale slope characteristic, wherein the multi-scale slope characteristic comprises a global average slope, a global variance and a global standard deviation; S4, obtaining abnormal candidate points of the smooth sequence according to the self-adaptive adjustment dynamic threshold value; S5, continuously clustering the abnormal candidate points into a plurality of abnormal segments, performing false positive detection, and outputting a plurality of effective abnormal segments; S6, calculating confidence scores of the effective abnormal sections, and outputting current collapse early warning results of the protection network to be early warned according to the confidence scores; the generation of the adaptive adjustment dynamic threshold comprises the following steps: s31, calculating local slopes of all windows; , Wherein, the For a starting position i and a window size of Is used to determine the local slope of the window of (c), As window scale, y is the relative intensity of light power at position i in the sequence of light reflection curves; s32, calculating local multiscale slope characteristics of each window; wherein, the initial position is i, and the window scale is The local multiscale slope of the window of (2) is characterized by: Local continuous slope average : , Local variance : , Local standard deviation : , Wherein ω is the window dimension, and wherein, The slope value corresponding to the kth position under the condition that the window scale is omega; s33, normalizing the local variance, wherein the expression is as follows: , wherein localVarNorm is the normalized local variance, ɛ is the set local variance correction factor, and globalVar is the global variance; s34, generating an adaptive adjustment dynamic threshold, wherein the expression is as follows: , Wherein, the For adaptive adjustment of the dynamic threshold, the factor is the set detection sensitivity, As a local variance suppression coefficient avgSlope is the global average slope.
2. The method for early warning of pre-collapse of a rock slope protection network based on OTDR according to claim 1, wherein the step S2 comprises the following steps: s21, calculating the local signal-to-noise ratio of the light reflection curve sequence; S22, determining a Gaussian filter radius according to the local signal-to-noise ratio; s23, carrying out smooth noise reduction processing on the light reflection curve sequence according to the Gaussian filter radius, and outputting a smooth sequence.
3. The method for early warning of pre-collapse of a rock slope protection network based on OTDR according to claim 2, wherein the expression of the Gaussian filter radius is: , where r is the Gaussian filter radius, f () is the radius calculation function, For a preset maximum gaussian filter radius and minimum gaussian filter radius, SNR is the local signal-to-noise ratio, Is the set signal to noise ratio threshold.
4. The method for early warning of pre-collapse of a rock slope protection network based on OTDR according to claim 1, wherein the calculation formula of the multi-scale slope characteristics is: global average slope avgSlope: , global variance globalVar: , Global standard deviation globalStd: , Wherein slope [ i ] is the slope at the i position, X [ i ] is the distance of the i position, For the relative intensity of the optical power at the i position in the smoothing sequence, N is the number of windows, Is the average slope of the smoothed sequence.
5. The method for early warning of pre-collapse of a rock slope protection network based on OTDR according to claim 4, wherein the judging of the abnormal candidate points in S4 includes self-adaptive adjustment dynamic threshold judgment and/or standard score statistic judgment; the expression of the adaptive adjustment dynamic threshold judgment is as follows: , The expression of the standard score statistic judgment is as follows: , , wherein z [ i ] is the standard score statistic of the i position, Is a preset standard fraction threshold.
6. The method for early warning of pre-collapse of a rock slope protection network based on OTDR according to claim 5, wherein S6 comprises the following steps: S61, calculating abnormal segment parameters of each effective abnormal segment, wherein the abnormal segment parameters comprise a start-stop position, a maximum loss amplitude, an average slope in the segment and local variance consistency; s62, calculating a confidence score of the effective abnormal section according to the abnormal section parameters; s63, outputting the current collapse early warning result of the protection network to be early warned according to the abnormal section parameters and the confidence score, wherein the collapse early warning result comprises the following steps of: The first-stage early warning judgment, namely judging the first-stage warning if the comprehensive loss amplitude is more than or equal to a first loss amplitude threshold value, the effective abnormal section continuity is more than or equal to middle, the confidence score is more than or equal to a first confidence threshold value, and otherwise, entering the second-stage early warning judgment; a second-level early warning judgment, namely judging the device to be a second-level alarm if the comprehensive loss amplitude is more than or equal to a second loss amplitude threshold value, the continuity of the effective abnormal section is more than or equal to middle, the confidence score is more than or equal to a first confidence threshold value, and otherwise, entering a third-level early warning judgment; three-level early warning judgment, namely judging three-level warning if the first loss amplitude threshold value is set to be greater than or equal to the comprehensive loss amplitude, the continuity of the effective abnormal section is greater than or equal to low, the confidence score is greater than or equal to the second confidence threshold value, and judging the warning as invalid abnormality if the confidence score is greater than or equal to the second confidence threshold value; And the effective abnormal section continuity is scored according to the abnormal section continuous length and the abnormal gap proportion, and the continuity is classified into high, medium and low according to a preset continuity scoring threshold.
7. The OTDR-based rock slope protection network pre-collapse early warning method according to claim 6, wherein the confidence score expression is: Wherein Score is a confidence Score, , , As the weight coefficient of the light-emitting diode, For the magnitude of the maximum loss to be the same, For the consistency of the local variance, , As the average slope in the segment, To be the average length of the effective anomaly segments, Maximum length of effective anomaly segment.
8. The OTDR-based rock slope protection network pre-collapse early warning method according to claim 1, wherein the false positive detection in S5 comprises: Setting a plurality of sampling points back before the starting point of each abnormal segment, calculating the preposed slope of the sampling points, outputting the abnormal segment as an effective abnormal segment if the preposed slope is less than or equal to 0, and eliminating if not.
9. The rock slope protection network pre-collapse early warning device based on the OTDR is characterized by comprising at least one processor and a memory in communication connection with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the rock slope protection network pre-collapse early warning method based on the OTDR according to any one of claims 1 to 8.

Description

Method and equipment for early warning of pre-collapse of rock slope protection network based on OTDR Technical Field The invention relates to the field of rock slope disaster monitoring and early warning, in particular to an early warning method and equipment for pre-collapse of a rock slope protection network based on OTDR. Background The road rock slope is subjected to geological disasters such as falling rocks and collapse due to factors such as complex geological structure, weathering effect, rainfall erosion and the like, so that road traffic safety is seriously threatened. The disaster has the characteristics of strong burst and large impact energy, and is more remarkable particularly in steep slopes or joint development areas. In engineering, a flexible protective net is often adopted as a core protective measure, and disaster influence is reduced in a blocking mode, an energy consumption mode and the like. However, in the existing protection system, the real-time monitoring perception of the operation state of the protection network is lacking, so that a remarkable dead zone exists in the safety management. On one hand, the lack of a real-time monitoring means cannot sense and feed back the dynamic information of the falling rocks of the side slope in real time, including the occurrence frequency, scale and specific position of the falling rocks, so that the management and maintenance unit is difficult to evaluate the actual risk level and disaster evolution trend, and on the other hand, due to the lack of continuous monitoring data on the stress state, deformation condition and interception effect of the protection network, engineering personnel cannot accurately judge the current safety margin of the protection network, and the residual service life of the protection network is difficult to predict. The lack of monitoring exposes the potential failure risk of the protection system, prevents accurate maintenance decision based on data, possibly causes secondary disasters after the protection failure, and finally threatens the road traffic safety and the operation benefit. Therefore, a rock slope protection network pre-collapse early warning method and equipment capable of achieving both complex scene adaptability and intelligent anomaly identification are needed. Disclosure of Invention The invention aims to solve the problem that the current situation of falling rocks of a side slope cannot be monitored and fed back in real time in the prior art, and provides a rock slope protection network pre-collapse early warning method and equipment based on OTDR. In order to achieve the above object, the present invention provides the following technical solutions: an OTDR-based rock slope protection network pre-collapse early warning method comprises the following steps: S1, continuously arranging a plurality of sensing optical fibers parallel to the ground on a main beam of a protection network to be early-warned, and acquiring a light reflection curve sequence of the protection network to be early-warned in real time; s2, carrying out smooth noise reduction treatment on the light reflection curve sequence and outputting a smooth sequence; S3, calculating a multi-scale slope characteristic of the smooth sequence, and generating a self-adaptive adjustment dynamic threshold according to the multi-scale slope characteristic, wherein the multi-scale slope characteristic comprises a global average slope, a global variance and a global standard deviation; S4, obtaining abnormal candidate points of the smooth sequence according to the self-adaptive adjustment dynamic threshold value; S5, continuously clustering the abnormal candidate points into a plurality of abnormal segments, performing false positive detection, and outputting a plurality of effective abnormal segments; And S6, calculating the confidence score of the effective abnormal section, and outputting the current collapse early warning result of the to-be-early warning protection network according to the confidence score. As a preferred embodiment of the present invention, the step S2 includes the steps of: s21, calculating the local signal-to-noise ratio of the light reflection curve sequence; S22, determining a Gaussian filter radius according to the local signal-to-noise ratio; s23, carrying out smooth noise reduction processing on the light reflection curve sequence according to the Gaussian filter radius, and outputting a smooth sequence. As a preferred embodiment of the present invention, the gaussian filter radius is expressed as: , where r is the Gaussian filter radius, f () is the radius calculation function, For a preset maximum gaussian filter radius and minimum gaussian filter radius, SNR is the local signal-to-noise ratio,Is the set signal to noise ratio threshold. As a preferred embodiment of the present invention, the calculation formula of the multi-scale slope characteristic is: global average slope avgSlope: , global variance globalV