Search

CN-121977641-A - Jacket foundation concrete pouring real-time monitoring system and method based on underwater robot

CN121977641ACN 121977641 ACN121977641 ACN 121977641ACN-121977641-A

Abstract

The application provides a jacket foundation concrete pouring real-time monitoring system and method based on an underwater robot, wherein the underwater robot generates an enhanced image by combining a binocular optical camera with LED light supplementing, calculates the ratio of gap volume to CAD theoretical volume based on binocular parallax to quantify plumpness, and simultaneously adopts a model to detect defects; the underwater robot analyzes slurry uniformity and edge detection clustering slurry segregation rate through HSV space and collects temperature pH viscosity parameters, the underwater robot locks an overflow port, calculates pixel flow rate through optical flow and converts real speed through binocular depth, then filters the real speed, sends a pump frequency adjusting instruction to a pump station according to a threshold through a communication protocol, and cruises for secondary scanning after pouring of the underwater robot is finished to verify fullness and defects, floats upwards and is recovered, and a report is generated. The method and the system realize real-time monitoring and comprehensive evaluation of the concrete pouring quality of the jacket foundation and the surrounding environment state, and improve the controllability and engineering safety of underwater pouring operation.

Inventors

  • XU ZIHAO
  • LU YIKUN

Assignees

  • 广东省广星能源开发有限公司

Dates

Publication Date
20260505
Application Date
20260117

Claims (9)

  1. 1. The jacket foundation concrete pouring real-time monitoring method based on the underwater robot is characterized by comprising the following steps of: the underwater robot is deployed to an offshore pile foundation area through a mother ship crane and an umbilical cable, self-tests a binocular optical camera, a sonar system, a propeller, an auxiliary sensor and an infrared auxiliary imaging module, and then is led into a CAD pile foundation model for path pre-planning, so as to obtain a pre-planned grid obstacle avoidance path; after the underwater robot fuses GPS and INS positioning and locks a pile foundation through a forward-looking sonar, submerging to a target water depth according to a pre-planned grid obstacle avoidance path at a gradient speed to finish posture adjustment and hovering near an overflow port; the underwater robot generates an enhanced image by combining a binocular optical camera with LED light filling, calculates the ratio of the gap volume to the CAD theoretical volume based on binocular parallax to quantify the plumpness, and simultaneously adopts a model to detect the defect; The underwater robot analyzes slurry uniformity and edge detection cluster slurry segregation rate through HSV space and collects temperature pH viscosity parameters; The underwater robot locks the overflow port, calculates the pixel flow rate through the optical flow and converts the real speed through binocular depth, and then filters the real speed to send a pump frequency adjusting instruction to the pump station through a communication protocol according to a threshold value; the underwater robot builds a seabed topography model through a multi-beam sounding system, calculates the depth volume area and gradient factor of the scouring pit to generate a risk index, transmits the risk index to a ground control station in real time, and triggers corresponding response measures according to the judging level corresponding to the risk index; and cruising the underwater robot after pouring, secondarily scanning and verifying the plumpness and the defects, and floating up for recycling to generate a report.
  2. 2. The method of claim 1, wherein the underwater robot is deployed to an offshore pile foundation area by a mother ship crane and an umbilical cable, self-inspected, and then introduced into a CAD pile foundation model for path pre-planning after a binocular optical camera, a sonar system, a propeller, an auxiliary sensor and an infrared auxiliary imaging module, to obtain a pre-planned grid obstacle avoidance path, comprising: after the sea condition parameters are confirmed to reach the standards through a weather station and matched detection equipment, the underwater robot is hoisted and launched through a crane and connected with an umbilical cable; The binocular optical camera performs distortion calibration, the forward-looking sonar performs circumferential scanning verification, the bottom sonar performs precision verification, the propeller performs parameter adjustment, the auxiliary sensor performs zero point verification, and the infrared auxiliary imaging module performs functional self-checking; And the control station is imported into the CAD model to plan the grid obstacle avoidance path by adopting a unified coordinate system, so as to obtain a pre-planned grid obstacle avoidance path.
  3. 3. The method of claim 1, wherein after the underwater robot fuses the GPS and INS positioning and locks the pile foundation by forward-looking sonar, submerging to a target water depth at a gradient speed according to a pre-planned grid obstacle avoidance path to complete posture adjustment and hover against an overflow port, comprising: After the GPS and the INS are initialized and fused, calibrating positioning accuracy based on a unified coordinate system, and closing the GPS when the depth exceeds a threshold value; The forward looking sonar extracts a locking pile foundation through the characteristics, compares the real-time positioning data with a pre-planned path, and dynamically adjusts the submergence direction; Gradient submerging is carried out after gesture adjustment according to the pre-planned path, depth fluctuation is kept, the deflection angle is monitored in real time through IMU data, and when the deflection angle exceeds a set threshold or deviates from the pre-planned path, a compensation mechanism is triggered, and propulsion torque is adjusted; The underwater robot transmits video data to a ground control station, and infrared auxiliary imaging is switched when sonar signals are attenuated; and approaching to a pre-planned hovering point, realizing hovering stability by adopting PID control, and collecting video and uploading a log.
  4. 4. The method of claim 1, wherein generating the enhanced image by a binocular optical camera in combination with LED light filling comprises: after the binocular optical cameras collect images, estimating underwater global background light and underwater transmissivity based on a preset underwater dark channel defogging model, and recovering to generate a foggy image; The multi-scale fusion weight is used for processing the color restoration factors and then is maintained through a guiding filtering edge; And carrying out self-adaptive stretching on the image subjected to the guided filtering treatment to obtain a final enhancement frame.
  5. 5. The method of claim 1, wherein calculating gap volume based on binocular disparity and quantifying fullness while detecting defects using a model based on a CAD theoretical volume ratio comprises: Obtaining depth data of the jacket annular gap through binocular parallax, and integrating the depth data to obtain the actual total volume V actual of the jacket annular gap, wherein the actual total volume is the maximum volume which can be filled in the annular gap; defining the CAD theoretical volume as the designed total volume V design of the jacket annular gap, and obtaining the actual slurry filling volume V filled by integrating and calculating the depth data of the slurry filling area through binocular parallax analysis; calculating the ratio of the unfilled gap volume to the plumpness to quantify the plumpness, wherein the plumpness directly reflects the filling completion degree of the annular gap; Judging whether the plumpness error is in an allowable range by detecting the continuous discharge state of the slurry and the result without bubble and cavity defects and combining the plumpness calculation result; triggering an alarm when the confidence of the model detection defect exceeds a threshold; Comparing the collected temperature, pH and viscosity parameters with a preset qualification threshold, comprehensively judging whether the quality of the slurry meets the standard according to the uniformity detection result and the segregation rate detection result, and triggering a slurry quality alarm if any parameter exceeds the preset threshold or the comprehensive detection result does not meet the requirement.
  6. 6. The method according to claim 1, wherein the locking the overflow port and calculating the pixel flow rate by the optical flow and filtering after the binocular depth conversion true speed sends the pump frequency adjusting instruction to the pump station by the communication protocol according to the threshold value comprises: The model locks the overflow port area, and optical flow calculation eliminates outlier points to select quantile pixel values; the center depth of the binocular depth map is combined with the pixel size focal length to convert the real speed; Filtering to process the flow velocity value, continuously sending a pressurizing instruction at a low speed, and monitoring at a high speed normally without improving pump stopping and early warning at a medium speed.
  7. 7. The method of claim 1, wherein constructing a seabed terrain model by a multi-beam sounding system and calculating a scout pit depth volume area and gradient factor generation risk index comprises: Constructing model registration by multi-beam sounding voxel filtering; Taking the theoretical seabed height corresponding to the outer wall of the pile body as a reference, subtracting the actual height of the terrain model, calculating the depth of the flushing pit, summing the depths of the flushing pits to obtain the volume of the flushing pit, and simultaneously determining the maximum depth of the flushing pit, the contour surrounding area of a region with the depth larger than a preset depth threshold value and the local gradient factor; the risk index is calculated and generated, the risk index is divided into three judging levels of safety, attention and danger according to a preset threshold value, the judging levels and the risk index are synchronously transmitted to a ground control station, an emergency response instruction is triggered when the risk level exists, and the taking of a block stone throwing and filling or mud sucking backfilling measure is prompted.
  8. 8. The method of claim 1, wherein the post-casting cruise secondary scan verifies post-defect float recovery and report generation, comprising: Cruising along a pre-planned cruising path curve, and checking plumpness and defects through multi-beam and optical scanning leakage threshold values; maintaining a safe distance during floating, moving to a specified recovery area according to a pre-planned recovery route, flushing the underwater robot after recovery, and checking the corrosion degree; Registering the terrain model in the post-processing stage, generating a slurry flow heat map, optimizing a working path, generating a model analysis report containing slurry quality detection, fullness detection, scouring risk assessment and path execution duplication, and completing data backup.
  9. 9. Jacket foundation concrete placement real-time monitoring system based on underwater robot, its characterized in that, the system includes: The equipment deployment and path pre-planning module is used for deploying the underwater robot to an offshore pile foundation area through a mother ship crane and an umbilical cable, completing self-inspection of equipment including a binocular optical camera and a sonar system, importing a CAD pile foundation model, and carrying out path pre-planning to obtain a pre-planned grid obstacle avoidance path; The positioning submerging and hovering module is used for controlling the underwater robot to submerge to the target water depth at a gradient speed according to a pre-planned grid obstacle avoidance path by means of GPS and INS fusion positioning and forward-looking sonar locking pile foundation, and achieving stable hovering by abutting against an overflow port after posture adjustment is completed; The image enhancement and fullness quantification and defect detection module is used for generating an enhanced image by combining a binocular optical camera with LED light filling, calculating the annular gap volume of the jacket based on binocular parallax, quantifying the fullness according to the ratio of the annular gap volume to the CAD theoretical volume, and detecting the defect by adopting a model; The slurry quality parameter monitoring module is used for analyzing slurry uniformity through HSV space, acquiring slurry segregation rate by combining an edge detection clustering algorithm, and synchronously acquiring temperature, pH and viscosity parameters of the slurry; The overflow speed monitoring and pump station control module is used for locking an overflow port area, calculating pixel flow speed through optical flow, obtaining real overflow speed through binocular depth conversion, and sending a pump frequency adjusting instruction to a pump station through a communication protocol according to a preset threshold after filtering treatment; The seabed topography modeling and scouring risk assessment module is used for constructing a seabed topography model through the multi-beam sounding system, calculating depth, volume, area and gradient factors of a scouring pit, generating a risk index, transmitting the risk index to a ground control station in real time, and triggering corresponding response measures according to a judging level corresponding to the risk index; And the cruising verification and recovery and report generation module is used for controlling the underwater robot to cruise after pouring is finished, scanning and verifying the plumpness and the defects for the second time, completing the floating recovery operation and generating an analysis report containing monitoring key data.

Description

Jacket foundation concrete pouring real-time monitoring system and method based on underwater robot Technical Field The invention relates to the technical field of information, in particular to a jacket foundation concrete pouring real-time monitoring system and method based on an underwater robot. Background The field of underwater engineering takes a vital role in modern marine development, especially in offshore pile foundation construction, the quality of which is directly related to engineering safety and long-term stability. The importance of this field goes without saying that any quality problem can lead to catastrophic consequences, as the pile foundation serves as a basis for supporting the offshore installation. However, the current monitoring means for underwater pile foundation construction is difficult to meet the requirements in the complex marine environment, and the limitations of the prior art are broken through, so that the reliability and the safety of engineering are guaranteed. In the prior art, there is a general problem of insufficient adaptation to the complexity of the underwater environment. Many methods often cannot accurately acquire key information in the construction process when facing situations of low seawater energy, large ocean current interference, changeable seabed topography and the like. The limitation is not only reflected in real-time judgment of construction quality, but also in insufficient perceptibility of changes of surrounding environment, so that potential risks in construction are difficult to discover and cope with in time. The technical difficulty of the deeper level is how to realize accurate monitoring and management of the grouting process in an underwater environment. Grouting is used as a core link in pile foundation construction, the fullness degree of the grouting directly influences the stability of the pile foundation, but due to the fact that the visualization and perception conditions of the underwater environment are limited, whether grouting is uniform, whether gaps or defects exist or not and other problems are difficult to accurately identify. Further, the monitoring difficulty is closely related to dynamic environmental changes in the construction process, for example, the seabed is flushed by ocean currents to possibly form pits, so that stability around a pile foundation is affected, and the prior art cannot comprehensively evaluate grouting quality and environmental changes at the same time. In particular, in an actual business scenario, constructors often face the dilemma that when grouting operation is performed, whether slurry completely fills each corner inside a pile foundation or not cannot be intuitively judged, whether local hollows or uneven conditions exist or not, meanwhile, changes of sea bed topography can occur silently in the construction process, for example, the formation of a flushing pit can weaken the supporting force of the pile foundation, and the changes are often discovered after the fact, so that engineering hidden danger is increased. Therefore, how to realize accurate monitoring of grouting quality in a complex underwater environment and evaluate the change of the seabed environment in real time becomes a key problem for guaranteeing the construction safety and stability of offshore pile foundations. Disclosure of Invention The invention provides a jacket foundation concrete pouring real-time monitoring method based on an underwater robot, which mainly comprises the following steps: the underwater robot is deployed to an offshore pile foundation area through a mother ship crane and an umbilical cable, self-tests a binocular optical camera, a sonar system, a propeller, an auxiliary sensor and an infrared auxiliary imaging module, and then is led into a CAD pile foundation model for path pre-planning, so as to obtain a pre-planned grid obstacle avoidance path; after the underwater robot fuses GPS and INS positioning and locks a pile foundation through a forward-looking sonar, submerging to a target water depth according to a pre-planned grid obstacle avoidance path at a gradient speed to finish posture adjustment and hovering near an overflow port; the underwater robot generates an enhanced image by combining a binocular optical camera with LED light filling, calculates the ratio of the gap volume to the CAD theoretical volume based on binocular parallax to quantify the plumpness, and simultaneously adopts a model to detect the defect; The underwater robot analyzes slurry uniformity and edge detection cluster slurry segregation rate through HSV space and collects temperature pH viscosity parameters; The underwater robot locks the overflow port, calculates the pixel flow rate through the optical flow and converts the real speed through binocular depth, and then filters the real speed to send a pump frequency adjusting instruction to the pump station through a communication protocol according to a threshold value;