Pile foundation concrete accurate pouring control system and method based on multi-source sensing fusion

Abstract

The invention discloses a pile foundation concrete accurate pouring control system and method based on multi-source sensing fusion. The pile foundation concrete accurate pouring control system comprises a sensing layer and a control layer, wherein the sensing layer comprises a phased array ultrasonic probe, a binocular vision sensor and a laser ranging array; the modeling layer comprises a three-dimensional hole shape reconstruction module and a BIM digital modeling engine; the data layer comprises a real-time monitoring database and a multi-source data fusion center; the decision-making layer comprises a concrete dynamic evaluation model and a BIM visual platform; the execution layer comprises a self-adaptive PID controller used for adjusting the concrete pumping rate according to the decision instruction, and a three-dimensional visualization terminal used for displaying the BIM model with the tolerance zone in real time. According to the method, a BIM model containing dynamic compensation is constructed through ultrasonic three-dimensional scanning, closed-loop control is achieved by fusing multi-source sensing data, and tolerance band penetration early warning and self-adaptive PID adjustment are combined. The hidden defect discovery rate is improved, and the comprehensive construction efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of pile foundation concrete pouring technology, and in particular to a control system and method for precise pouring of pile foundation concrete based on multi-source sensor fusion. Background Technology

[0002] Traditional methods for pouring concrete for pile foundations present a series of technical problems that severely impact construction quality and efficiency. First, irregular hole shapes lead to significant discrepancies between theoretically calculated volumes and actual requirements, typically with over-pouring rates exceeding 15%. This not only increases material costs but may also pose structural safety risks. Second, the dynamic changes between pouring rate and concrete shrinkage cannot be compensated for in real-time by existing electronic metering devices, resulting in inaccurate volume control. Third, precise control of pile top elevation relies on manual experience, with errors potentially reaching ±30cm, especially during nighttime construction, severely affecting construction accuracy and safety. Finally, quality issues in concealed works, such as broken piles and necking, are often only discovered after construction is complete, with a problem detection rate of less than 40%. This not only increases rework rates but also affects the overall quality and safety of the project.

[0003] To address the aforementioned technical issues, the industry currently relies primarily on the volumetric method in the "Technical Specification for Building Pile Foundations" for monitoring concrete pouring volume. However, this method suffers from three monitoring blind spots: during the borehole formation stage, a probe cage is used to detect the borehole diameter (error ±5cm); during the pouring stage, an electromagnetic flowmeter (accuracy class 1.5) is used in conjunction with a measuring rope for verification; and in the final pouring stage, an over-pouring coefficient of 0.8~1.2m is used for control. Although some patents have proposed improved methods, most fail to fully consider the complexity of the actual borehole shape and lack an effective real-time closed-loop control system. For example, patent CN20181023567.X, "BIM-based Prediction System for Concrete Pouring Volume of Pile Foundations," uses a three-dimensional geological model to generate theoretical volumes and obtains the actual usage through a concrete mixer truck weighing system. However, in practical applications, this solution fails to fully consider the complex morphology of the borehole and lacks real-time closed-loop control measures, which significantly limits the practical application of existing technologies. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology. To achieve the above objectives, a pile foundation concrete precision pouring control system and method based on multi-source sensor fusion is adopted to solve the problems mentioned in the background technology.

[0005] A precision concrete pouring control system for pile foundations based on multi-source sensor fusion, comprising a perception layer, a modeling layer, a data layer, a decision layer, and an execution layer; The sensing layer includes a phased array ultrasonic probe, a binocular vision sensor, and a laser ranging array; The modeling layer includes a three-dimensional hole reconstruction module for processing ultrasonic point cloud data, and a BIM digital modeling engine for constructing a parametric pile foundation model with construction tolerance zone based on the IFC standard. The data layer includes a real-time monitoring database for storing multi-source sensor data, and a multi-source data fusion center that fuses visual data and laser data based on an improved DS evidence theory; The decision-making layer includes a concrete dynamic evaluation model based on a volume compensation algorithm for soil elastic parameters, and a BIM visualization platform for dynamically rendering pouring heatmaps, anomaly markers, and profile deviations using a WebGL engine. The execution layer includes an adaptive PID controller for adjusting the concrete pumping rate according to decision instructions, and a three-dimensional visualization terminal for real-time display of a tolerance-enabled BIM model.

[0006] As a further aspect of the present invention: the binocular vision sensor is used to capture the concrete flow rate at a preset frame rate, and the laser ranging array is used to monitor the elevation of the pouring surface at a preset frequency, and both are time-synchronized through a protocol.

[0007] As a further aspect of the present invention: the BIM digital modeling engine is used to perform Monte Carlo integration to calculate theoretical quantities, generate tolerance bands constructed by NURBS surface algorithm, and output the registration error between the solid model and the real-time point cloud.

[0008] As a further aspect of the present invention: the multi-source data fusion center is used to establish a spatiotemporal reference coordinate system and to map laser data onto the surface of the BIM model through an inverse projection algorithm.

[0009] As a further aspect of the present invention: the decision layer employs a hybrid neural network anomaly classifier, which triggers an audible and visual alarm and marks the anomaly type when the real-time pouring surface penetrates the tolerance zone.

[0010] As a further aspect of the present invention: the concrete dynamic compensation model integrates the ARIMA time series predictor and uses the Monte Carlo integration method to calculate the theoretical volume, the formula being: ; And the dynamic compensation formula is updated as follows: ; Then, estimate the remaining volume and pouring time, and output the remaining volume using the following formula: ; in, This is the over-irrigation coefficient. This represents the cumulative pouring volume; Based on the time forecasting algorithm, an ARIMA time series model is constructed, with the following formula: ; The remaining time is estimated as follows: .

[0011] As a further aspect of the present invention: the execution layer is connected to a Hololens AR terminal, which overlays the BIM model and real-time deviation data onto the actual construction site scene and displays PID adjustment commands.

[0012] The second aspect of the technical solution: A control method for a pile foundation concrete precision pouring control system based on multi-source sensor fusion as described in any of the above, comprising the following steps: Step S1: Generate pile hole point cloud by scanning with phased array ultrasonic probe, use improved RANSAC algorithm to remove noise, and integrate soil elastic parameters to construct BIM model with tolerance zone; Step S2: Synchronize binocular vision data and laser ranging data, fuse multi-source information through improved DS evidence theory, and map it to the BIM model in real time; Step S3: When the monitoring surface penetrates the tolerance band, the LSTM+GRU anomaly classifier is triggered to generate an early warning signal; Step S4: Predict the remaining volume based on the ARIMA model, and output the pump speed adjustment amount through the adaptive PID controller; Step S5: Dynamically render the heatmap of the pouring status, anomaly markers, and AR auxiliary commands in the WebGL engine.

[0013] Compared with the prior art, the present invention has the following technical advantages: The above technical solution employs a five-layer architecture: a perception layer, a modeling layer, a data layer, a decision-making layer, and an execution layer. The perception layer generates high-density point clouds through phased-array ultrasonic probe scanning, combining binocular vision and a laser ranging array to acquire data in real time. The modeling layer uses an improved RANSAC algorithm for noise reduction and constructs a BIM parametric model with construction tolerance zones, dynamically correcting theoretical volumes using soil elastic compensation formulas. The decision-making layer integrates multi-source data based on improved DS evidence theory and detects tolerance zone penetration behavior using an LSTM+GRU anomaly classifier. The execution layer utilizes an adaptive PID controller to adjust the pumping rate in real time and visualizes the dynamic pouring status through a WebGL engine. Through measured orifice shape and volume compensation, the volume control error was reduced from ±12% to ±1.5%. A dynamic balance equation of "orifice shape-flow rate-elevation" was established to achieve adaptive adjustment of the pouring rate. A three-dimensional digital twin with tolerance zones was developed, increasing the visualization rate of hidden engineering problems from 38% to 97%. Attached Figure Description The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the structure of the control system according to an embodiment of this application; Figure 2 This is a schematic diagram of the various layers of the control system disclosed in an embodiment of this application; Figure 3 This is a diagram of the equipment installation model disclosed in this application. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please refer to Figure 1 and Figure 2 In this embodiment of the invention, a precision concrete pouring control system for pile foundations based on multi-source sensor fusion is provided. The system includes a perception layer, a modeling layer, a data layer, a decision layer, and an execution layer. The perception layer includes a phased array ultrasonic probe, a binocular vision sensor, and a laser ranging array; The modeling layer includes a 3D hole reconstruction module for processing ultrasonic point cloud data, and a BIM digital modeling engine for building parametric pile foundation models with construction tolerance zones based on IFC standards. The data layer includes a real-time monitoring database for storing multi-source sensor data, and a multi-source data fusion center that fuses visual and laser data based on an improved DS evidence theory; The decision-making layer includes a concrete dynamic evaluation model based on a volume compensation algorithm for soil elastic parameters, and a BIM visualization platform for dynamically rendering pouring heatmaps, anomaly markers, and profile deviations using a WebGL engine. Specifically, the steps for developing a WebGL 3D engine visualization interface are as follows: Heat map of pouring progress: The completed / unpoured areas are displayed using a color gradient (red → green); Anomaly location markers: Automatically mark necking (yellow) and segregation (purple) areas in the 3D model; Dynamic profile analysis: Any cross-section shows the deviation between the actual pouring surface and the design elevation (accuracy ±2mm); The execution layer includes an adaptive PID controller for adjusting the concrete pumping rate according to decision instructions, and a 3D visualization terminal for real-time display of a tolerance-enabled BIM model.

[0016] In this embodiment, the binocular vision sensor is used to capture the concrete flow rate at a preset frame rate, and the laser ranging array is used to monitor the elevation of the pouring surface at a preset frequency, and both are synchronized in time through a protocol.

[0017] Ultrasonic 3D scanning and BIM modeling: Detailed process: Point cloud generation: The phased array ultrasonic probe (500kHz) scans to generate the original point cloud (density 2000 points per cubic meter). An improved RANSAC algorithm was used to remove noise from suspended solids in the mud. BIM model construction steps: Construct a parametric pile foundation model based on IFC standards; Then inject the key parameters; Model output: Generate a 3D BIM entity with construction tolerance zones; In this embodiment, the BIM digital modeling engine is used to perform Monte Carlo integration to calculate theoretical quantities, generate tolerance bands constructed by NURBS surface algorithm, and output the registration error between the solid model and the real-time point cloud.

[0018] In this embodiment, the multi-source data fusion center is used to establish a spatiotemporal reference coordinate system and to map laser data onto the surface of the BIM model through an inverse projection algorithm.

[0019] Specifically, dynamic comparison of multi-source data with the BIM model is performed; Among them, the real-time data fusion mechanism performs data synchronization: Establish a spatiotemporal reference coordinate system (WGS84 + engineering local coordinate system transformation); Timestamp alignment: Adopts IEEE 1588 precise time protocol (error <1ms); Specifically, the 3D alignment algorithm: The cumulative pouring volume monitored by the vision system Boolean operations on entities converted to BIM models; Laser ranging data is mapped onto the surface of the BIM model using an inverse projection algorithm. In this embodiment, the decision layer uses a hybrid neural network anomaly classifier. When the real-time pouring surface penetrates the tolerance zone, an audible and visual alarm is triggered and the anomaly type is marked.

[0020] Specifically, perform anomaly detection logic: Set the three-dimensional deviation tolerance: radial deviation: ±3cm (red warning zone); axial deviation: ±5cm (yellow warning zone). When the real-time pouring surface penetrates the tolerance body, the LSTM anomaly classifier is triggered; In this embodiment, the concrete dynamic compensation model integrates the ARIMA time series predictor and uses the Monte Carlo integral method to calculate the theoretical volume, as shown in the formula: ; And the dynamic compensation formula is updated as follows: ; Then, estimate the remaining volume and pouring time, and output the remaining volume using the following formula: ; in, This is the over-irrigation coefficient. This represents the cumulative pouring volume; Based on the time forecasting algorithm, an ARIMA time series model is constructed, with the following formula: ; The remaining time is estimated as follows: ; Specifically, visualize the output: Display a dynamic progress bar (with a 95% confidence interval) in the BIM interface; The remaining pouring height is marked on the 3D model (floating numerical labels).

[0021] In this embodiment, the execution layer is connected to the Hololens AR terminal, which overlays the BIM model and real-time deviation data onto the actual construction site and displays PID adjustment commands.

[0022] like Figure 3 As shown, the diagram is a model of the equipment installation. In this embodiment, the dynamic control example of the BIM model is specifically taken as an example of the construction of an end-bearing pile with a diameter of 2m: 1. Initial state: BIM model displays design dimensions (Including over-irrigation); Tolerance zone setting: pile top ±0.3m, pile body ±5cm; 2. Pouring process: When the vision system monitors At that time, the BIM display is as follows: The completed pouring section is green (elevation -28.3m); The remaining portion is displayed transparently, with a remaining volume of 12.8 cubic meters. 3. Exception handling: The laser system detected that the northeast side of the pile top was rising too fast (deviation +8cm). The BIM model will automatically rotate to an abnormal orientation and be marked with an orange warning box; The system executes PID control: reducing the pumping rate on this side by 15%; 4. Final irrigation control: When the remaining volume is less than 0.5 cubic meters, the BIM interface switches to millimeter-wave radar verification mode; The pouring was terminated after the 3D model was compared with the measured point cloud and the conformity rate was confirmed to be >98%. In this embodiment, the three-dimensional hole shape dynamic modeling technology is used: Ultrasonic phased array scanning generates 3D point clouds of pile holes with millimeter-level precision; A volumetric compensation algorithm that integrates soil elastic parameters (compensation coefficient formula: K=1+0.015E0.6). Employing a real-time BIM model mapping mechanism: Parametric modeling method based on IFC standard (including construction tolerance zone); Automatic association technology between point cloud data and BIM entity model (registration error <2mm); Setting up a multi-source data fusion architecture: A spatiotemporal synchronization method between a visual flow meter (200fps) and a laser array (10Hz); Improve the confidence allocation rule of the DS evidence theory (conflict factor β < 0.15). Establish an intelligent early warning system: A real-time 3D deviation comparison engine with tolerance band; LSTM+GRU hybrid neural network anomaly classifier (F1-score 0.94).

[0023] The second aspect of the technical solution: A control method for a pile foundation concrete precision pouring control system based on multi-source sensor fusion as described in any of the above, comprising the following steps: Step S1: Generate pile hole point cloud by scanning with phased array ultrasonic probe, use improved RANSAC algorithm to remove noise, and integrate soil elastic parameters to construct BIM model with tolerance zone; Step S2: Synchronize binocular vision data and laser ranging data, fuse multi-source information through improved DS evidence theory, and map it to the BIM model in real time; Step S3: When the monitoring surface penetrates the tolerance band, the LSTM+GRU anomaly classifier is triggered to generate an early warning signal; Step S4: Predict the remaining volume based on the ARIMA model, and output the pump speed adjustment amount through the adaptive PID controller; Step S5: Dynamically render the heatmap of the pouring status, anomaly markers, and AR auxiliary commands in the WebGL engine.

[0024] In this embodiment, the beneficial effects are as follows: Breakthrough in precision: Through actual measurement of hole shape and volume compensation, the volume control error was reduced from ±12% to ±1.5%; Process controllability: Establish a dynamic balance equation of "orifice shape-flow rate-elevation" to achieve adaptive adjustment of the pouring rate; Intelligent decision-making: Developing a 3D digital twin with tolerance bands increases the visualization rate of hidden engineering problems from 38% to 97%.

[0025] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention. The scope of the invention is defined by the appended claims and their equivalents, all of which should be included within the scope of protection of the invention.

Claims

1. A precision concrete pouring control system for pile foundations based on multi-source sensor fusion, characterized in that, The system includes a perception layer, a modeling layer, a data layer, a decision-making layer, and an execution layer; The sensing layer includes a phased array ultrasonic probe, a binocular vision sensor, and a laser ranging array; The modeling layer includes a three-dimensional hole reconstruction module for processing ultrasonic point cloud data, and a BIM digital modeling engine for constructing a parametric pile foundation model with construction tolerance zone based on the IFC standard. The data layer includes a real-time monitoring database for storing multi-source sensor data, and a multi-source data fusion center that fuses visual data and laser data based on an improved DS evidence theory; The decision-making layer includes a concrete dynamic evaluation model based on a volume compensation algorithm for soil elastic parameters, and a BIM visualization platform for dynamically rendering pouring heatmaps, anomaly markers, and profile deviations using a WebGL engine. The execution layer includes an adaptive PID controller for adjusting the concrete pumping rate according to decision instructions, and a three-dimensional visualization terminal for real-time display of a tolerance-enabled BIM model.

2. The pile foundation concrete precision pouring control system based on multi-source sensor fusion according to claim 1, characterized in that, The binocular vision sensor is used to capture the concrete flow rate at a preset frame rate, and the laser ranging array is used to monitor the elevation of the pouring surface at a preset frequency. Both are time-synchronized through a protocol.

3. The pile foundation concrete precision pouring control system based on multi-source sensor fusion according to claim 1, characterized in that, The BIM digital modeling engine is used to perform Monte Carlo integration to calculate theoretical quantities, generate tolerance bands constructed by NURBS surface algorithm, and output the registration error between the solid model and the real-time point cloud.

4. The pile foundation concrete precision pouring control system based on multi-source sensor fusion according to claim 1, characterized in that, The multi-source data fusion center is used to establish a spatiotemporal reference coordinate system and to map laser data onto the surface of the BIM model through an inverse projection algorithm.

5. The pile foundation concrete precision pouring control system based on multi-source sensor fusion according to claim 1, characterized in that, The decision-making layer uses a hybrid neural network anomaly classifier. When the real-time pouring surface penetrates the tolerance zone, an audible and visual alarm is triggered and the anomaly type is marked.

6. The pile foundation concrete precision pouring control system based on multi-source sensor fusion according to claim 1, characterized in that, The concrete dynamic compensation model integrates the ARIMA time series predictor and uses the Monte Carlo integral method to calculate the theoretical volume, as shown in the formula: ； And the dynamic compensation formula is updated as follows: ； Then, estimate the remaining volume prediction and pouring time, and output the remaining volume using the following formula: ； in, This is the over-irrigation coefficient. This represents the cumulative pouring volume; Based on the time forecasting algorithm, an ARIMA time series model is constructed, with the following formula: ； The remaining time is estimated as follows: 。 7. The pile foundation concrete precision pouring control system based on multi-source sensor fusion according to claim 1, characterized in that, The execution layer connects to the HoloLens AR terminal, overlays the BIM model and real-time deviation data onto the actual construction site, and displays PID adjustment commands.

8. A control method for a pile foundation concrete precision pouring control system based on multi-source sensor fusion as described in any one of claims 1 to 7, characterized in that, Includes the following steps: Step S1: Generate pile hole point cloud by scanning with phased array ultrasonic probe, use improved RANSAC algorithm to remove noise, and integrate soil elastic parameters to construct BIM model with tolerance zone; Step S2: Synchronize binocular vision data and laser ranging data, fuse multi-source information through improved DS evidence theory, and map it to the BIM model in real time; Step S3: When the monitoring surface penetrates the tolerance band, the LSTM+GRU anomaly classifier is triggered to generate an early warning signal; Step S4: Predict the remaining volume based on the ARIMA model, and output the pump speed adjustment amount through the adaptive PID controller; Step S5: Dynamically render the heatmap of the pouring status, anomaly markers, and AR auxiliary commands in the WebGL engine.

Images