Development of the USACE Automated SMART Gate System for Lock Gates: Detection of Barge Impact Events Using Statistical Process Control

Inland navigation locks are vital to the U.S. economy and due to their advanced age are expensive to maintain for proper operation. The U.S. Army Corps of Engineers is developing structural health monitoring technologies to provide decision support information that will be used to prioritize maintenance and assist in operational decisions. The Structural Monitoring and Analysis in Real Time of Lock Gates (SMART Gate) system will provide a suite of structural health monitoring capabilities for miter lock gates. Miter gates are the most common gate type in U.S. locks and are impacted by barges frequently and damage can accumulate over time from even minor impacts. Barge impacts into opened (recessed) gates often go unnoticed, especially for mild to moderate impacts. In this paper, an application of statistical process control to detect barge impacts to lock miter gates is described. As part of the SMART Gate project, the downstream miter gates at Lock 27 on the Mississippi River, near St. Louis, were instrumented with accelerometers and water pressure gages in 2014. Measurements were taken over the course of several lockage events to monitor dynamic gate response from door swing and wave action. Measurements included data taken during a barge impact event. These measurements were subsequently used to train and validate a barge impact detection algorithm. This algorithm consists of an autoregressive time series model and an x-bar chart to measure anomalies of model prediction error. After calibrating the algorithm to mitigate false-positives, it was found that barge impact was positively identified. This algorithm was programmed directly into affordable field data acquisition hardware for real-time detection and alerting the lock operator. Upon one final validation exercise, the algorithm will be employed on all future SMART Gate systems. doi: 10.12783/SHM2015/77