Search

CN-122024449-A - Method and system for monitoring stability of rock slope in real time based on three-dimensional laser scanner and deep sensor

CN122024449ACN 122024449 ACN122024449 ACN 122024449ACN-122024449-A

Abstract

The invention discloses a method and a system for monitoring stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor, and relates to the technical field of geological disaster monitoring and early warning; the method comprises the steps of obtaining multi-time-phase point cloud data and a deep monitoring sequence, constructing point cloud frames and a deep sequence window to bind and output a point cloud frame index set and a deep abnormal event index set, converting abnormal space coordinates according to abnormal measuring point depths, drilling hole orifice coordinates and drilling track parameters and projecting the abnormal space coordinates to a slope model to form an abnormal mapping relation, aggregating and diffusing according to slope unit numbers to generate a risk hot zone, judging a destabilizing mode based on mapping weights, triggering characteristic quantities and normal consistency parameters, and outputting a hot zone mode result set.

Inventors

  • LI MIAN
  • XU YANG
  • TANG XIN

Assignees

  • 重庆安全技术职业学院

Dates

Publication Date
20260512
Application Date
20260305

Claims (10)

  1. 1. A method for monitoring stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor is characterized by comprising the following steps: Acquiring multi-time phase point cloud data output by a three-dimensional laser scanner and a deep monitoring sequence output by a deep sensor, constructing a binding relation between a point cloud frame and a deep sequence window, and outputting a point cloud frame index set and a deep abnormal event index set when the deep sequence window triggers an abnormality; based on depth of abnormal measuring points corresponding to the deep abnormal event index, obtaining abnormal space coordinates by combining drill hole orifice coordinates and drill track parameters, projecting the abnormal space coordinates to the slope model to obtain a slope unit number, generating an abnormal mapping table and outputting an abnormal mapping relation set; based on the abnormal mapping relation set and the slope model, mapping and aggregating deep abnormal event indexes according to the slope unit numbers, generating unit risk scores according to mapping confidence parameters and triggering characteristic quantities, and performing diffusion aggregation on the unit risk scores according to the slope adjacency relation to obtain a risk hot zone set comprising a hot zone number, a hot zone unit set, a hot zone risk score and a hot zone associated deep abnormal event index set; the mapping weight parameters in the abnormal mapping table are summed according to the sloping surface unit numbers to obtain unit weight parameters, the weighted sum is used for obtaining a hot zone representative normal vector and calculating a hot zone normal consistency parameter, the triggering duty ratio parameter is counted according to the triggering channel name and the maximum value of the ratio of the triggering characteristic quantity to the corresponding threshold value is taken to obtain an intensity indication parameter, the dominant parameter is generated and corrected, the toppling dominant parameter is corrected, the destabilizing mode label and the mode confidence score are determined, and the result is written into a hot zone mode result set and output.
  2. 2. The method for monitoring the stability of the rock slope in real time based on the three-dimensional laser scanner and the deep sensor according to claim 1, wherein the method for acquiring the result set of the hot zone mode comprises the following steps: traversing the risk hot zone set according to the hot zone number to construct a hot zone geometric data packet and a hot zone event data packet, summarizing unit weight parameters according to an abnormal mapping table, calculating a hot zone normal consistency parameter, a channel trigger duty parameter set and an intensity indication parameter set according to characteristic summarization rule parameters, generating a displacement dominant parameter, a shearing dominant parameter, a dumping dominant parameter and a pore pressure dominant parameter according to a dominant fusion parameter set, correcting the dumping dominant parameter according to a normal consistency threshold parameter and a normal correction parameter set, determining a destabilization mode label according to a mode judgment threshold parameter set, calculating a mode confidence score according to a mode confidence score parameter, and writing the mode result set of the hot zone.
  3. 3. The method for monitoring the stability of the rock slope in real time based on the three-dimensional laser scanner and the deep sensor according to claim 2, wherein the steps of constructing a hot zone geometry data packet, a hot zone event data packet and parameter calculation include: The method comprises the steps of reading a unit normal vector corresponding to a slope unit number from a slope model according to a hot area unit set to form a hot area geometric data packet, reading a trigger channel name, a trigger feature quantity, a threshold value and a trigger time stamp from an abnormal event table according to a hot area related deep abnormal event index set to form a hot area event data packet, summing mapping weight parameters in an abnormal mapping table according to the slope unit number to obtain a unit weight parameter, calculating a hot area representative normal vector according to a weighted sum rule, calculating a hot area normal consistency parameter based on an included angle of each unit normal vector and the hot area representative normal vector, counting a displacement change rate trigger duty ratio parameter, a strain change rate trigger duty ratio parameter, an inclination angle change rate trigger duty ratio parameter and a pore pressure change rate trigger duty ratio parameter according to the trigger channel name, and calculating a hot area displacement intensity indication parameter, a hot area strain intensity indication parameter, a hot area dumping indication parameter and a hot area pore pressure increase indication parameter according to a maximum rule.
  4. 4. A method for monitoring rock slope stability in real time based on a three-dimensional laser scanner and a depth sensor according to claim 3, wherein the determining of the destabilizing mode label and mode confidence score comprises: According to the intensity fusion weight parameter and the duty fusion weight parameter in the dominant fusion parameter set, carrying out weighted summation on the hot zone displacement intensity indication parameter and the displacement change rate triggering duty parameter to obtain a displacement dominant parameter, carrying out weighted summation on the hot zone strain intensity indication parameter and the strain change rate triggering duty parameter to obtain a shearing dominant parameter, carrying out weighted summation on the hot zone dumping indication parameter and the inclination angle change rate triggering duty parameter to obtain a dumping dominant parameter, carrying out weighted summation on the hot zone pore pressure jump indication parameter and the pore pressure change rate triggering duty parameter to obtain a pore pressure dominant parameter, correcting the dumping dominant parameter according to a normal punishment coefficient parameter when the hot zone normal consistency parameter is larger than a normal consistency threshold value parameter, comparing the order of the pore pressure dominant parameter, the dumping dominant parameter, the shearing dominant parameter and the displacement dominant parameter with corresponding thresholds to determine a destabilizing mode label, and calculating a mode confidence score according to a first position dominant value, a secondary position dominant value and a hot zone risk score according to a mode confidence score parameter.
  5. 5. The method for monitoring the stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor according to claim 1, wherein the method for acquiring the risk hot zone set comprises the following steps: The method comprises the steps of reading an abnormal mapping relation set from an abnormal mapping table according to a point cloud frame index, reading a trigger time stamp, a trigger feature quantity and a threshold value from an abnormal event table according to a deep abnormal event index, reading a unified time stamp from the point cloud frame index table according to the point cloud frame index, determining a fusion reference time stamp according to the unified time stamp maximum value in the abnormal mapping relation set, screening an abnormal mapping relation with the unified time stamp time difference not larger than a fusion time window length parameter, calculating an abnormal strength parameter as a ratio of the trigger feature quantity to the threshold value according to the screened abnormal mapping relation, normalizing according to a preset strength upper limit parameter to obtain a normalized abnormal strength parameter, obtaining a confidence weight parameter according to the mapping confidence parameter, obtaining a one-low confidence map to obtain a preset low confidence weight parameter according to the high confidence map, multiplying the normalized abnormal strength parameter by the confidence weight parameter, accumulating the mapping weight parameter according to a slope unit number to obtain a unit risk score, and summarizing the deep abnormal event index set corresponding to the slope unit number.
  6. 6. The method for monitoring the stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor according to claim 5, wherein the method for acquiring the risk hot zone set further comprises: The method comprises the steps of establishing a slope surface adjacency list based on a unit center coordinate corresponding to a slope surface unit number in a slope surface model and a unit adjacency distance threshold parameter, taking a matched slope surface unit number in an abnormal mapping relation set as a diffusion starting point, carrying out layer-by-layer diffusion to a diffusion radius parameter limiting layer number along the slope surface adjacency list, taking an adjacency layer number by a path length parameter, normalizing the ratio of the path length parameter to the diffusion radius parameter and calculating a distance attenuation weight according to the path length parameter, reading a trigger time stamp according to a deep abnormal event index, obtaining a time difference parameter by solving an absolute difference according to a unified time stamp corresponding to a diffusion starting point cloud frame index, normalizing the ratio of the time difference parameter to a fusion time window length parameter and calculating a time attenuation weight according to the normalized time difference parameter; The method comprises the steps of obtaining a unit risk score of a candidate slope unit number by multiplying a mapping weight parameter, a distance attenuation weight and a time attenuation weight, meanwhile updating a deep abnormal event index set corresponding to the candidate slope unit number, screening the unit risk score according to a hot zone risk threshold parameter, carrying out communication clustering according to a slope adjacency list to obtain a hot zone unit set and a hot zone number, obtaining the maximum unit risk score in the hot zone unit set by the hot zone risk score, obtaining a union of the deep abnormal event index set corresponding to the hot zone unit set by the hot zone associated deep abnormal event index set, and filtering according to the minimum hot zone unit number parameter to obtain a risk hot zone set.
  7. 7. The method for monitoring the stability of the rock slope in real time based on the three-dimensional laser scanner and the deep sensor according to claim 1, wherein the method for acquiring the abnormal mapping relation set comprises the following steps: Traversing the deep abnormal event index set, reading a trigger time stamp, a trigger channel name, a trigger feature quantity and a threshold value comparison result from the abnormal event table, determining a point cloud frame index associated with the deep abnormal event index according to the binding relation table, and acquiring a deep sequence window data packet; Grouping deep sequence window data packets according to a measuring point number field, comparing a channel numerical value field corresponding to a trigger channel name with a threshold value in a threshold value comparison result, counting measuring point threshold value according to preset continuous threshold value exceeding times parameters, determining an abnormal measuring point number, reading an abnormal measuring point depth parameter from a measuring point depth field, and writing the abnormal measuring point depth parameter into an abnormal event table; Calculating an abnormal space coordinate based on an abnormal measuring point depth parameter, a drilling hole orifice coordinate and a drilling track parameter, acquiring a slope grid unit set from a slope model based on an associated point cloud frame index, performing a projection matching process according to projection mode parameters to determine a matched slope unit number, calculating a mapping distance parameter, a track deviation upper limit parameter and a point cloud resolution parameter, and marking a mapping confidence parameter according to a confidence threshold parameter; Writing the point cloud frame index, the deep abnormal event index, the abnormal space coordinates, the matched slope unit numbers and the mapping confidence parameters into an abnormal mapping table and outputting the abnormal mapping table as an abnormal mapping relation set.
  8. 8. The method for monitoring the stability of the rock slope in real time based on the three-dimensional laser scanner and the deep sensor according to claim 1, wherein the method for acquiring the point cloud frame index set and the deep abnormal event index set comprises the following steps: Establishing a device time offset table, wherein the device time offset table records the time offset of the three-dimensional laser scanner and the time offset of the deep sensor, the initial value of the device time offset table is written in by a device registration process, and the device registration process records a device identifier, a communication delay measurement result and an initial clock difference measurement result; Triggering a point cloud frame warehousing process every time a three-dimensional laser scanner generates a frame of point cloud data, receiving the original acquisition time of the point cloud frame in the point cloud frame warehousing process, generating a uniform timestamp by combining with a device time offset table, generating a point cloud frame index according to the uniform timestamp, and writing the point cloud frame index into a point cloud frame index table; the deep sensor outputs a deep monitoring sequence according to a preset sampling period, wherein the deep monitoring sequence comprises in-hole displacement data, strain data, inclination angle data and hole pressure data, and the deep sequence warehousing process receives deep sampling original time, generates a unified time stamp by combining with an equipment time offset table and writes the unified time stamp into a deep sequence index table; The method comprises the steps of taking a point cloud frame index table as a main index, generating a deep sequence window for each point cloud frame index, limiting the deep sequence window by a preset time window length parameter, determining a window starting point and a window ending point according to uniform time stamps corresponding to the point cloud frame indexes, searching a deep monitoring sequence ranging from the window starting point to the window ending point in the deep sequence index table to form a deep sequence window data packet, writing the deep sequence window data packet into a binding relation table, executing an abnormality detection process when the binding relation table is generated or updated, and outputting a point cloud frame index set and a deep abnormal event index set.
  9. 9. The method for monitoring the stability of the rock slope in real time based on the three-dimensional laser scanner and the deep sensor according to claim 8, wherein the method for executing the abnormality detection process when the binding relation table is generated or updated comprises the following steps: Taking a deep sequence window data packet as input, and calculating a characteristic quantity set in a window according to abnormal triggering condition parameters, wherein the characteristic quantity set in the window comprises a displacement change rate in the window, a strain change rate in the window, an inclination change rate in the window and a pore pressure change rate in the window; The method comprises the steps of comparing a characteristic quantity set in a window with a threshold value parameter set, judging whether a deep sequence window meets an abnormal trigger condition by combining a continuous superthreshold frequency parameter, distributing deep abnormal event indexes for abnormal events when the abnormal trigger condition is met, writing comparison results of the deep abnormal event indexes, a trigger time stamp, a trigger channel name, the trigger characteristic quantity and the threshold value into an abnormal event table, adding point cloud frame indexes meeting the abnormal trigger condition into the point cloud frame index set, and adding corresponding deep abnormal event indexes into the deep abnormal event index set.
  10. 10. A system for monitoring stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor, for realizing the method for monitoring stability of a rock slope in real time based on the three-dimensional laser scanner and the deep sensor as set forth in any one of claims 1 to 9, comprising: The multi-source acquisition time synchronization module is used for acquiring multi-time phase point cloud data output by the three-dimensional laser scanner and a deep monitoring sequence output by the deep sensor, constructing a binding relation between a point cloud frame and a deep sequence window, and outputting a point cloud frame index set and a deep abnormal event index set when the deep sequence window triggers an abnormality; The system comprises a point cloud frame index module, a space mapping module, an abnormal space coordinate, an abnormal mapping relation set, a slope unit number and a slope unit number, wherein the point cloud frame data volume corresponds to the point cloud frame index, and the slope unit number is calculated by combining a drilling hole opening coordinate and a drilling track parameter based on the depth of an abnormal measuring point corresponding to the deep abnormal event index; the fusion positioning module is used for carrying out mapping aggregation on the deep abnormal event indexes according to the slope unit numbers based on the abnormal mapping relation set and the slope model, generating unit risk scores according to the mapping confidence parameters and the triggering characteristic quantity, and carrying out diffusion aggregation on the unit risk scores according to the slope adjacency relation to obtain a risk hot zone set comprising a hot zone number, a hot zone unit set, a hot zone risk score and a hot zone associated deep abnormal event index set; the unstability mode judging module sums the mapping weight parameters in the abnormal mapping table according to the sloping surface unit numbers to obtain unit weight parameters, weights and sums the unit weight parameters to obtain a hot zone representative normal vector and calculate a hot zone normal consistency parameter, counts the triggering duty ratio parameters according to the triggering channel names and takes the ratio maximum value of the triggering characteristic quantity and the corresponding threshold value numerical value to obtain an intensity indication parameter, generates a leading parameter and corrects the dumping leading parameter, determines an unstability mode label and a mode confidence score, writes the label and the confidence score into a hot zone mode result set and outputs the label and the confidence score.

Description

Method and system for monitoring stability of rock slope in real time based on three-dimensional laser scanner and deep sensor Technical Field The invention relates to the technical field of geological disaster monitoring and early warning, in particular to a method and a system for monitoring stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor. Background Rock slopes widely exist in mining, traffic engineering, hydraulic engineering and mountain infrastructure construction, and the stability state of the rock slopes is directly related to engineering structure safety and operation safety. With the development of slope monitoring technology, a three-dimensional laser scanner is adopted to periodically scan the slope surface to acquire multi-time-phase point cloud data, so that the method has become an important means for acquiring slope morphological change, local deformation characteristics and surface layer geometric information. Meanwhile, the deep sensor can continuously collect deep monitoring data such as in-hole displacement, strain, inclination angle, hole pressure and the like, and is used for reflecting the slope internal structure response and the deep evolution process. In the prior engineering practice, three-dimensional laser scanning and deep sensing monitoring are commonly deployed in a combined way so as to obtain slope surface information and deep information at the same time, thereby improving the integrity and reliability of slope stability monitoring. However, in actual rock slope monitoring, three-dimensional laser scanning data mainly reflects slope surface deformation, and deep sensing data mainly reflects deep evolution information such as hole displacement, strain, inclination angle, hole pressure and the like. Because the two types of data have differences in time reference, index mode and space expression caliber, the prior art generally adopts a step analysis or simple corresponding mode to carry out comprehensive judgment, lacks a unified data link and a consistent judgment caliber, and is difficult to form a continuous and unified monitoring result. Further, in the existing rock slope monitoring, although the surface deformation can be observed through three-dimensional laser scanning, the deep evolution process is difficult to synchronously explain with the surface observation result, and the method is particularly characterized in that the deep abnormal event is difficult to determine the corresponding slope position, difficult to judge whether the corresponding slope position corresponds to the current slope change or not, and difficult to further judge the instability mode. Therefore, a method and a system for monitoring a rock slope in real time, which can uniformly and spatially correlate deep abnormal events with slope deformation information and realize uniform discrimination, are needed. In view of the above, the present invention provides a method and a system for monitoring the stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor to solve the above problems. Disclosure of Invention In order to overcome the defects in the prior art and achieve the purposes, the invention provides a method for monitoring the stability of a rock slope in real time based on a three-dimensional laser scanner and a deep sensor, which comprises the following steps: Acquiring multi-time phase point cloud data output by a three-dimensional laser scanner and a deep monitoring sequence output by a deep sensor, constructing a binding relation between a point cloud frame and a deep sequence window, and outputting a point cloud frame index set and a deep abnormal event index set when the deep sequence window triggers an abnormality; based on depth of abnormal measuring points corresponding to the deep abnormal event index, obtaining abnormal space coordinates by combining drill hole orifice coordinates and drill track parameters, projecting the abnormal space coordinates to the slope model to obtain a slope unit number, generating an abnormal mapping table and outputting an abnormal mapping relation set; based on the abnormal mapping relation set and the slope model, mapping and aggregating deep abnormal event indexes according to the slope unit numbers, generating unit risk scores according to mapping confidence parameters and triggering characteristic quantities, and performing diffusion aggregation on the unit risk scores according to the slope adjacency relation to obtain a risk hot zone set comprising a hot zone number, a hot zone unit set, a hot zone risk score and a hot zone associated deep abnormal event index set; the mapping weight parameters in the abnormal mapping table are summed according to the sloping surface unit numbers to obtain unit weight parameters, the weighted sum is used for obtaining a hot zone representative normal vector and calculating a hot zone normal consistency parameter