Publications

Abstract : In this paper, we have reported an effective adoption of machine learning (ML) algorithms for wireless sensor networks (WSN) deployed for real-time monitoring and early warning of landslides. The WSN system performs real-time monitoring of landslide triggering parameters such as rainfall, soil moisture, pore-water-pressure (PWP), and movements. The challenges involved in landslide early warning and the data driven context for overcoming these challenges are being presented. Linear models were applied on real-time data from the WSN system and the data from laboratory experiments to discover the interrelationships existing between rain, soil, and moisture, PWP and construct the feature-set required for forecasting slope stability conditions. Two algorithms Current-PWP and 24-PWP based on the support vector regression (SVR) method is implemented for efficacious nowcast and 24 h forecast of the slope stability conditions for early warning notification of landslides. Comparison of the results from the Current-PWP and 24-PWP algorithms with the actual sensor measurements recorded by the WSN system revealed mean absolute error (MAE) values 4.2k and 4.7 kPa, respectively. The results section elaborates 1) on relevance of the Current-PWP algorithm towards ensuring the reliability of early warning WSN systems, constrained by data availability, 2) application of the 24-PWP algorithm to yield an additional duration of 24 h for landslide preparedness, 3) the empirical appraisal of the regression model from laboratory experiments for estimation of soil moisture from PWP, and vice versa, 4) the efficacy of multi-level decision making for the preclusion of false warnings, and 5) the timely early warnings issued from the proffered WSN system.