CORRUGATED STEEL CULVERT PIPE DETERIORATION

This research provides the basis for developing a comprehensive plan for inspection, cleaning, condition assessment and prediction of remaining service life of CSCP (Corrugated Steel Culvert Pipe). Inspection frequency guidelines were developed that rate CSCPs at three levels. The rating categories are based on the following factors: corrosion and erosion, bed load, pH, and culvert size, age and importance, and are ranked according to increasing need; e.g., annual inspections are recommended for Category III (older pipes with reported problems). A four condition state assessment system based upon the Caltrans system was developed, including quantifiable section losses, specific surface features, and a prescribed response associated with each condition state. A Markov deterioration model was used to predict the future condition state of new CSCP in urban and rural settings. These improvements will be addressed in the next phase of the research project within the context of a Culvert Information Management System (CIMS). In addition, the Weibull distribution will replace the Markov model for predicting the remaining service life of CSCP in the next phase of this project. The proposed CIMS will be capable of analyzing decisions to inspect, rehabilitate/replace, or do nothing at both project and network levels. At the project level this will be achieved by comparing inspection and/or rehabilitation/replacement costs with risks and costs associated with failure. At the network level, the associated costs will be optimized to meet annual maintenance budget allocations by prioritizing CSCPs needing inspection and rehabilitation/replacement. CIMS will also be used to estimate the required annual budgetary allocation for a stipulated planning horizon to maintain or improve the aggregate condition state of the CSCP network, or to maintain or improve the total highway CSCP network asset value, thereby meeting the GASB-34 requirements. The optimum sequential path in the annual decision making process may then be determined using a combination of operations research tools. A framework for real time and automated monitoring of the condition of culverts based on the identification of internal defects via video inspection was developed. An innovative approach of judiciously extracting image frames from the video and analyzing the frames to locate and categorize major defects was developed. Each frame is preprocessed to enhance contrast using an adaptive scheme and reduced dimensionality in pixel-space by implementing region based processing. The preprocessing is followed by a two-step image segmentation process, which implements a background elimination procedure in the first step and shape detection in the second step. Fuzzy clustering is used as the underlying segmentation model. Defect shape and depth information after post-processing are used as input to an automated condition state assessment methodology. A simple formulation based on both the damage area and depth is then utilized to assess the condition of culverts based on a 4-point condition assessment scale. The proposed framework was demonstrated with a test example. Future research would entail consolidating the concept by extensive testing and integration for real time application.