Intelligent earthwork computing system and method based on multi-source data fusion

Abstract

The invention discloses an intelligent earthwork computing system and method based on multi-source data fusion. The method comprises the following steps: acquiring a multi-angle image and high-precision coordinate data of a construction area through unmanned aerial vehicle oblique photography and RTK positioning technologies; combining fractal noise reduction and a wavelet transform algorithm to eliminate vegetation noise and extract topographic feature point cloud; a dynamic three-dimensional terrain model is constructed based on BIM platform fusion point cloud, GIS geographic information and design parameters, and excavation and filling amount deviation and geological disaster signal triggering model updating are detected in real time; optimizing an earthwork transportation path by adopting an improved ant colony algorithm, defining a dynamic weight cost function, and calculating a pavement paving area in combination with a Green's formula; and generating a visual report through the AR technology, superposing the difference between the design model and the construction state, interactively adjusting the scheme, and issuing an instruction to the unmanned transportation equipment. According to the method, high-precision calculation, multi-parameter dynamic optimization and construction whole-process closed-loop management of the earthwork volume under the complex terrain can be realized, and the method is suitable for real-time refined management and control of large-scale projects.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering construction information management, and in particular to an intelligent earthwork calculation system and method based on multi-source data fusion. Background Art

[0002] The calculation of earthwork volume is an important step in the construction of a building project. The earthwork volume must be budgeted in the design stage before construction. It is directly related to the cost estimate and scheme selection of the project. When calculating the earthwork volume, the earthwork volume calculation method needs to be used.

[0003] 1. Comparative document CN115077489A proposes a method for calculating earthwork volume using oblique photography by drones, and the protection rights are as follows: "Step 1: Collect weather information and process the weather information to obtain information on whether collection is allowed or not; Step 2: When information on collection is allowed is obtained, first use a drone to patrol and obtain surrounding images and preliminary collection location images. When information on not allowing collection is obtained, wait until the weather information meets the conditions before controlling the drone: Step 3: Process the preliminary collection location images to obtain calculation type information, and then process the surrounding images and collection location images to obtain collection warning information and preliminary analysis information: Step 4: Process the preliminary analysis information, determine that the collection location is safe, and then collect collection location information in different ways according to the size of the collection area to obtain real-time collection location parameter information. The present invention can calculate earthwork volume information with smaller errors and is more worthy of promotion and use." However, this solution will lead to large data collection errors under complex terrain or weather conditions, such as high humidity and rich vegetation.

[0004] 2. Comparative document CN111898869A proposes a method for measuring and calculating earthwork that combines oblique photography, RTK, and BIM technology. The protected rights are as follows: "When calculating the earthwork quantity, calculate the earthwork transportation volume m = AB. A positive number m indicates that there is excess earthwork that needs to be transported out of the site, and a negative number m indicates that additional earthwork needs to be transported into the site. In addition, plan the transportation route and set the priority. The first priority is the excavation unit price v w (x, y, z) and landfill unit price v t For grids with lower (x, y, z) values, the second priority is grids with lower transport unit costs (v(x, y, z)). Simulation analysis is performed in BIM software to determine the corresponding earthwork balance plan. The calculation fully considers the excavation and landfill unit costs, transport unit costs, and settlement at each grid point. Through comprehensive consideration of multiple factors, the optimal earthwork balance plan is determined. However, this technical solution cannot update project progress and earthwork changes in real time, resulting in delayed plan adjustments.

[0005] In summary, the existing earthwork calculation methods have problems such as insufficient multi-source data fusion, poor real-time performance, imprecise cost optimization, and low algorithm efficiency in complex terrain and dynamic construction environments, resulting in insufficient earthwork calculation accuracy and delayed engineering plan adjustments.

[0006] Therefore, an intelligent earthwork calculation system and method based on multi-source data fusion is proposed to solve the above problems. Summary of the Invention

[0007] The purpose of the present invention is to overcome the above-mentioned shortcomings and provide an intelligent earthwork calculation system and method based on multi-source data fusion to solve the problems raised in the background technology.

[0008] In response to the above problems, the present invention proposes an intelligent earthwork calculation system based on multi-source data fusion, comprising: a data acquisition module: configured to collect multi-angle image data and high-precision coordinate data of the construction area through an unmanned aerial vehicle equipped with oblique photography equipment and an RTK positioning device; a data processing module: used to perform fractal noise reduction on the image data and extract terrain feature point clouds through a wavelet transform algorithm; a dynamic modeling module: based on the BIM platform, it integrates point cloud data, GIS geographic information and design parameters to construct a dynamic three-dimensional terrain model; an optimization calculation module: based on the model, it calculates the excavation and filling volume, transportation route and road paving area in real time, and optimizes the earthwork scheduling plan based on the ant colony algorithm; an output module: generates visual reports and construction instructions, and supports real-time interaction among multiple terminals.

[0009] Preferably, the data acquisition module includes a meteorological sensor for real-time monitoring of wind speed, humidity and light intensity. or humidity When the alarm is triggered, data collection is paused.

[0010] In addition, the present invention also discloses a calculation method of an intelligent earthwork calculation system based on multi-source data fusion, comprising the following steps: Step 1: Obtain multi-temporal image data of the construction area through drone oblique photography, and simultaneously collect RTK positioning data; Step 2: Perform fractal noise reduction on the image data to eliminate vegetation and weather interference and extract pure terrain point cloud; Step 3: Spatially align the point cloud data with the BIM design model and GIS terrain data to construct a dynamic 3D terrain model; Step 4: Based on the dynamic model, the cut and fill volume, transportation route priority, and road paving area are calculated in real time, and the earthwork scheduling plan is optimized by combining the ant colony algorithm; Step 5: Generate construction instructions based on the optimization results, and provide real-time feedback on the project progress through a visual interface.

[0011] Furthermore, the fractal noise reduction processing in step 2 specifically includes: Calculate the local fractal dimension D of the image data using the formula: ; The side length required to cover the image area is The number of grids; If D>2.3, it is determined to be a high complexity area and wavelet transform filtering is used. The filtering function is: ; in, is the Morlet wavelet basis function, a is the scale parameter, and b is the translation parameter.

[0012] Furthermore, in the wavelet transform filtering, the scale parameter a is selected to satisfy: ; Where: J is the maximum number of decomposition layers, which is dynamically adjusted according to the image resolution.

[0013] Furthermore, the transportation route optimization in step 4 specifically includes: Define the transportation cost function, which is expressed as: ; in For Grid The excavation cost, is the transport distance, is the weight coefficient; The improved ant colony algorithm is used to iteratively solve the minimum cost path, and the pheromone update rule is: ; in, is the volatility coefficient, is the pheromone increment of the kth ant on the path (i, j).

[0014] Furthermore, the updating triggering condition of the dynamic three-dimensional terrain model in step 4 includes at least one of the following: The deviation of cut and fill exceeds the threshold ; Construction progress deviates from the plan by more than ; Geological disaster risk signals detected.

[0015] Furthermore, the method for calculating the road paving area in step 4 is: Extract the vertex coordinates of the road surface boundary polygon in the design model ; Green's formula is used to calculate the area A, which is expressed as: .

[0016] Furthermore, the visualization interface in step five integrates AR augmented reality technology and specifically includes the following steps: real-time superposition of the design model and the on-site construction status; highlighting the excavation and filling deviation areas and risk warning areas; and interactively adjusting the transportation route and resource allocation.

[0017] Furthermore, the improved ant colony algorithm introduces a dynamic weight adjustment strategy, which is expressed as follows:

[0018] in, is the attenuation coefficient, and t is the number of iterations to balance the stage optimization requirements of path cost and distance. The present invention has the following beneficial effects: 1. The present invention combines fractal noise reduction with wavelet transform to effectively eliminate interference from complex environments and greatly reduce terrain modeling errors; 2. The present invention updates the model in real time and triggers adjustments, shortening the response time to project progress deviations; 3. The present invention uses ant colony algorithm to optimize the transportation route, reducing the overall cost; 4. The present invention adopts the comprehensive integration of multi-source data of BIM, GIS and oblique photography data to support full life cycle management. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of the overall framework of the system of the present invention; Figure 2 This is the overall flow chart of the earthwork dynamic calculation method of the present invention; Figure 3 This is the detailed process of fractal noise reduction in the present invention; Figure 4 This is a flow chart of dynamic modeling and optimization calculation in the present invention; Figure 5 This is the ant colony algorithm path optimization flow chart of the present invention; Figure 6 A logic diagram for visual feedback and instruction delivery in the present invention; Figure 7 This is the logic diagram of the linkage between meteorological monitoring and data collection in the present invention. DETAILED DESCRIPTION

[0020] The present invention will be further described below with reference to the accompanying drawings and examples: See also Figures 1 to 7, an intelligent earthwork calculation system based on multi-source data fusion, including: a data acquisition module: configured to collect multi-angle image data and high-precision coordinate data of the construction area through an unmanned aerial vehicle equipped with oblique photography equipment and an RTK positioning device; a data processing module: used to perform fractal noise reduction on the image data and extract terrain feature point clouds through a wavelet transform algorithm; a dynamic modeling module: based on the BIM platform, it integrates point cloud data, GIS geographic information and design parameters to construct a dynamic three-dimensional terrain model; an optimization calculation module: calculates the excavation and filling volume, transportation route and road paving area in real time according to the model, and optimizes the earthwork scheduling plan based on the ant colony algorithm; an output module: generates visual reports and construction instructions, and supports real-time interaction among multiple terminals. Data acquisition module: Use a DJI M300 RTK drone equipped with a five-lens oblique camera, such as the Zenmuse P1, with a flight altitude of 80m, 80% heading overlap, and 70% lateral overlap; the RTK positioning device is integrated into the drone, and centimeter-level positioning data is obtained through the Qianxun location service, and the latitude, longitude, and elevation information of the image are recorded in real time; data processing module: Fractal noise reduction is performed on the collected images: Matlab is used to calculate the local fractal dimension of the image. If D>2.3, such as in dense vegetation areas, Morlet wavelet basis function is used for filtering to eliminate high-frequency noise; CloudCompare software is used to extract terrain point clouds, remove vegetation points, and detect sudden changes based on color thresholds and height mutations; dynamic modeling module: import point cloud data into Revit, perform spatial registration with the BIM design model, and use ArcGIS Engine to overlay GIS terrain data onto the model; optimization calculation module: Improved ant colony algorithm is implemented based on Python: 50 ants are simulated to initialize the volatility coefficient =0.1, pheromone increment =1 / C(x, y), iterate 100 times to solve the optimal transportation path; Output module: Use the Unity engine to develop an AR visualization interface, and use the Hololens 2 device to display the deviation between the cut and fill area and the design model in real time, with the out-of-tolerance area highlighted in red.

[0021] Preferably, the data acquisition module includes a meteorological sensor for real-time monitoring of wind speed, humidity and light intensity. or humidity When the wind speed reaches 0, the data collection is suspended and the alarm is triggered. The sensor can be installed with Davis Vantage Pro2 weather station to monitor the wind speed in real time. ,humidity If the wind speed exceeds the limit, the drone will automatically return and store the unfinished route; when the humidity exceeds the limit, the shooting will be suspended and the lens heating and defogging will be started.

[0022] In addition, the present invention also discloses a calculation method of an intelligent earthwork calculation system based on multi-source data fusion, comprising the following steps: Step 1: Obtain multi-temporal image data of the construction area through drone oblique photography, and simultaneously collect RTK positioning data; Step 2: Perform fractal noise reduction on the image data to eliminate vegetation and weather interference and extract pure terrain point cloud; Step 3: Spatially align the point cloud data with the BIM design model and GIS terrain data to construct a dynamic 3D terrain model; Step 4: Based on the dynamic model, the cut and fill volume, transportation route priority, and road paving area are calculated in real time, and the earthwork scheduling plan is optimized by combining the ant colony algorithm; Step 5: Generate construction instructions based on the optimization results, and provide real-time feedback on the project progress through a visual interface.

[0023] Furthermore, the fractal noise reduction processing in step 2 specifically includes: Calculate the local fractal dimension D of the image data using the formula: ; The side length required to cover the image area is The number of grids; If D>2.3, it is determined to be a high complexity area and wavelet transform filtering is used. The filtering function is: ; in, is the Morlet wavelet basis function, a is the scale parameter, and b is the translation parameter. The image is divided into multiple sub-regions, and the box counting method is used to calculate the D value of each sub-region; for example, for the vegetation area, when the RGB value satisfies G>100 and R<50, D≈2.5 is calculated, and it is determined to be a high complexity area. Use Matlab's cwt function, select the Morlet wavelet basis, and the scale parameter a=2 3 , decompose the image into three levels, filter out the high-frequency components and reconstruct the image.

[0024] Furthermore, in the wavelet transform filtering, the scale parameter a is selected to satisfy: ; Where: J is the maximum number of decomposition layers, which is dynamically adjusted according to the image resolution. If the image resolution is 2cm / pixel, set J=4, then a=2 4 =16, which can effectively capture small terrain features; if the resolution is 10cm / pixel, then J=2, and a=2 2 =4, avoids over-smoothing terrain details.

[0025] Furthermore, the transportation route optimization in step 4 specifically includes: Define the transportation cost function, which is expressed as: ; in For Grid The excavation cost, is the transport distance, is the weight coefficient; The improved ant colony algorithm is used to iteratively solve the minimum cost path, and the pheromone update rule is: ; in, is the volatility coefficient, is the pheromone increment of the kth ant on the path (i, j).

[0026] Furthermore, the updating triggering condition of the dynamic three-dimensional terrain model in step 4 includes at least one of the following: The deviation of cut and fill exceeds the threshold ; Construction progress deviates from the plan by more than ; Geological disaster risk signals detected. Design =10%. If the designed excavation volume is 1000m³ and the actual calculated volume is 1150m³, an update is triggered. =2 hours, then the progress lag exceeds 2 hours and triggers an early warning; the surface deformation is monitored through InSAR satellite data, and a risk area is determined when the deformation rate is greater than 5mm / day.

[0027] Furthermore, the method for calculating the road paving area in step 4 is: Extract the vertex coordinates of the road surface boundary polygon in the design model ; Green's formula is used to calculate the area A, which is expressed as: .

[0028] Furthermore, the integration of AR technology into the visualization interface in Step 5 specifically includes the following steps: real-time overlay of the design model with the on-site construction status; highlighting of areas with cut-and-fill deviations and risk warning areas; and interactive adjustment of transportation routes and resource allocation. Specifically, Unity's AR Foundation plug-in can be used to identify construction site markers using the phone's camera, overlaying the BIM model onto the real-time view. Areas with deviations greater than 5cm are displayed in translucent red. Technicians can then drag the virtual path line with gestures, and the system recalculates costs and updates the ant colony algorithm parameters in real time.

[0029] Furthermore, the improved ant colony algorithm introduces a dynamic weight adjustment strategy, which is expressed as follows:

[0030] in, is the decay coefficient, and t is the number of iterations to balance the stage optimization requirements of path cost and distance. Parameter settings: initial weight α0=0.8, β0=0.2, decay coefficient λ=0.05; after 10 iterations, α(10)=0.8×e -0.5 =0.48,β(10)=0.2×(1−e -0.5 )=0.13, gradually turning to distance optimization.

[0031] The specific implementation methods of the above embodiments are only preferred technical solutions of the present invention and should not be regarded as limitations of the present invention. The scope of protection of the present invention should be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the scope of protection of the present invention.

Claims

1. An intelligent earthwork calculation system based on multi-source data fusion, characterized in that: include: Data acquisition module: configured to collect multi-angle image data and high-precision coordinate data of the construction area through drone-mounted oblique photography equipment and RTK positioning devices; Data processing module: used to perform fractal noise reduction on image data and extract terrain feature point clouds through wavelet transform algorithm; dynamic modeling module: based on the BIM platform, it integrates point cloud data, GIS geographic information and design parameters to build a dynamic three-dimensional terrain model; optimization calculation module: based on the model, it calculates the excavation and filling volume, transportation route and road paving area in real time, and optimizes the earthwork scheduling plan based on the ant colony algorithm; output module: generates visual reports and construction instructions, and supports real-time interaction among multiple terminals.

2. The intelligent earthwork calculation system based on multi-source data fusion according to claim 1 is characterized in that: The data acquisition module includes a meteorological sensor for real-time monitoring of wind speed, humidity and light intensity. or humidity When the alarm is triggered, data collection is paused.

3. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to any one of claims 1-2, characterized in that: The following steps are involved: Step 1: Obtain multi-temporal image data of the construction area through drone oblique photography, and simultaneously collect RTK positioning data; Step 2: Perform fractal noise reduction on the image data to eliminate vegetation and weather interference and extract pure terrain point cloud; Step 3: Spatially align the point cloud data with the BIM design model and GIS terrain data to construct a dynamic 3D terrain model; Step 4: Based on the dynamic model, the cut and fill volume, transportation route priority, and road paving area are calculated in real time, and the earthwork scheduling plan is optimized by combining the ant colony algorithm; Step 5: Generate construction instructions based on the optimization results, and provide real-time feedback on the project progress through a visual interface.

4. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to claim 3 is characterized in that: The fractal noise reduction processing in step 2 specifically includes: Calculate the local fractal dimension D of the image data using the formula: ； The side length required to cover the image area is The number of grids; If D>2.3, it is determined to be a high complexity area and wavelet transform filtering is used. The filtering function is: ； in, is the Morlet wavelet basis function, a is the scale parameter, and b is the translation parameter.

5. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to claim 4 is characterized in that: In the wavelet transform filtering, the scale parameter a is selected to satisfy: ； Where: J is the maximum number of decomposition layers, which is dynamically adjusted according to the image resolution.

6. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to claim 3 is characterized in that: The transport route optimization in step 4 specifically includes: Define the transportation cost function, which is expressed as: ； in For Grid The excavation cost, is the transport distance, is the weight coefficient; The improved ant colony algorithm is used to iteratively solve the minimum cost path, and the pheromone update rule is: ； in, is the volatility coefficient, is the pheromone increment of the kth ant on the path (i, j).

7. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to claim 3 is characterized in that: The updating triggering condition of the dynamic three-dimensional terrain model in step 4 includes at least one of the following: The deviation of cut and fill exceeds the threshold ; Construction progress deviates from the plan by more than ; Geological disaster risk signals detected.

8. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to claim 3 is characterized in that: The calculation method of the pavement paving area in step 4 is: Extract the vertex coordinates of the road surface boundary polygon in the design model ; Green's formula is used to calculate the area A, which is expressed as: 。 9. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to claim 3 is characterized in that: The visualization interface in step five integrates AR augmented reality technology and specifically includes the following steps: real-time superposition of the design model and the on-site construction status; highlighting the excavation and filling deviation areas and risk warning areas; and interactively adjusting the transportation route and resource allocation.

10. The calculation method of the intelligent earthwork calculation system based on multi-source data fusion according to claim 3 is characterized in that: The improved ant colony algorithm introduces a dynamic weight adjustment strategy, which is expressed as follows: in, is the attenuation coefficient, and t is the number of iterations to balance the stage optimization requirements of path cost and distance.

Images