Dam Defect Mapping
Drone-based inspection, digital defect mapping and precision aerial survey for one of the largest dams in NSW
Project Overview
One of the largest dams in NSW has a rock wall that is 747m long and 114m high, with a rock-fill embankment and clay core, and a 300m long concrete chute spillway located at the eastern end of the dam wall. Routine monitoring of the dam requires inspection of the spillway structure to identify the presence of any defects. Conventional inspection of the spillway involves personnel working at heights to manually map the location and extent of identified defects. Diodrone was engaged by WaterNSW to develop a digital engineering solution for inspection and defect mapping over the spillway structure.
Diodrone Solution
Diodrone developed a drone based reality capture solution which would provide a detailed defect map for the spillway structure, a precision aerial survey for the embankment to assist with routine displacement monitoring, and a reality model of stilling basin rock exposures to assist with geotechnical assessment.
Project Scope
The use of drone-based reality capture eliminated working at heights and created an inspection data set significantly better than conventionally available.
SPILLWAY INSPECTION
A small multirotor drone was used to capture RGB (visual) and thermal IR imagery over the extent of the spillway structure. Images were then processed in photogrammetry software to develop an ultra-high-resolution 3D model.
Photogrammetry processing achieved an average resolution of 2mm per pixel and an accuracy of 7mm RMSE when compared to a network of 5no. precise check points.
A digital inspection was then carried out, referencing the 3D model and thermal IR photographs, to develop a detailed defect map which marked out the extent of cracks, patch repairs, pitting, spalls, debris, vegetation, and thermal IR anomalies.
Each class of defect was presented as a separate layer in a web-based 3D GIS as well as detailed in a slab-by-slab defect report which included CAD and PDF defect maps.
DELIVERABLES
- Web-based 3D GIS
- Vectors (linework) for various classes of inspection features
- CAD and PDF outputs for each concrete slab including chute and training walls
- Factual Inspection Report
Digital Inspection and Defect Mapping over 13,000m2 of concrete using ultra-high-resolution 3D modelling and thermal IR.
Precision Aerial Survey
A small multirotor drone equipped with RTK direct georeferencing was utilised to capture imagery for a precision aerial survey over the dam embankment. Photogrammetry processing of the imagery achieved sub-pixel reprojection error and the resulting point cloud output achieved an accuracy of 15mm RMSE when compared to a network of 17no. check points established throughout the survey area.
The high-density point cloud was gridded into a digital elevation model (DEM) from which 1m contours were developed. With future surveys of the area, change detection analysis of the point cloud and DEM can be carried out to assist the WaterNSW survey team with modelling and monitoring displacements of the structure.
Photogrammetry outputs achieved an accuracy of 15mm RMSE when compared to a network of high-precision check points.
3D Modelling
Imagery from the spillway inspection and the precision aerial survey was combined to form a data set of more than 10,000 photographs, which covered the entire dam structure at 20mm resolution or less. Extensive oblique photography was included in the data set to ensure steep, vertical, and overhanging surfaces were captured.
Photogrammetry software was then used to reconstruct a 3D model for the entire dam, which achieved an accuracy of 15mm RMSE when compared to a network of checkpoints established throughout the project area. The 3D model included an extremely high level of detail over rock exposures on the abutments and at the base of the spillway, for use in geological mapping and geotechnical assessment.
High-resolution 3D textured mesh of the site, suitable for digital inspection, geological modelling, and general site visualisation.