TP24-20 Automating Structural Inspections of Concrete Cooling Towers

Observing, analyzing, and monitoring the structural condition of concrete cooling towers over time has long been a challenge for owners and operators. The structural elements of cooling towers can be obscured during operation and often inaccessible due to limited access options during brief outages. As a result, structural site conditions can be overlooked and the progression of deterioration may not be fully recognized, prompting unscheduled outages. Technology now exists to automate the cumbersome visual structural inspection process to obtain a more holistic condition assessment of concrete cooling towers. This paper presents a three-step automated methodology for imaging, analyzing, and monitoring structural deterioration in concrete cooling towers using autonomous drone imaging, machine learning for damage detection, and 3D visualization. In Step 1 drones capable of autonomous flight collect enough overlapping images to fully capture the visible portions of a cooling tower structure. These images can be stitched together to create a digital twin of a cooling tower using a process called photogrammetry. In Step 2 the collected photos are analyzed by machine learning algorithms to identify and quantify the most common structural concrete damage conditions. In Step 3 structural damage information is transferred to a 3D model and displayed in an Asset Management Portal (or 3D AMP). Having condition data available in a 3D visualization format enables engineers, owners, and operators to quickly see and understand the magnitude of structural issues. By automating the concrete cooling tower inspection process, observations can be performed at higher frequency (even during shorter outages) and fidelity allowing tracking of deterioration progress and development of customized maintenance and repairs while minimizing impact on cooling tower operations.