Structural health monitoring of concrete dams using least squares support vector machines

This study presents a least squares support vector machine (LSSVM) based displacement prediction model for health monitoring of concrete dams. LSSVM is a novel machine learning technique. The model can produce similar good generalization performance and learns faster than the basic support vector machines in engineering problems. The advantages such as high prediction accuracy, fast training speed of the LSSVM model are verified using monitoring data of a real concrete dam. Results are also compared with that of the multiple linear regression and stepwise regression models for dam health monitoring.