A1: PREDICTION OF GEOTECHNICAL PARAMETERS USING MACHINE LEARNING
In this present study, relationships between Angle of friction using SPT N-Value, Cohesion using SPT N-Value, Plasticity Index (PI) using Liquid Limit (LL), Optimum moisture content (OMC) using Plastic limit (PL) and Maximum dry density (MDD) using Plastic Limit has been established using Machine Learning Techniques. Geotechnical data from over 300 borehole locations covering almost every district in the state of Kerala have been considered to develop models and statistical correlations. From these sets of data around 100 data were identified as being suitable for correlation between SPT N-Value and Strength parameters and another set of data were considered suitable for developing correlations between Atterberg Limits and Compaction Characteristics. The models with a higher coefficient of determination (R2) value was considered to be more appropriate for prediction and subsequently chosen to develop a Web Application. The R2 values of the models which were developed using SLR lies between 0.82-0.95. On comparison with available data, it has been observed that the predicted value are in close agreement with existing work. Using the developed algorithms, we were able to create a web application which can allow users to easily access and predict value for a given parameter. The Web Application consists of two sets. The first part consist forms to predict Angle of Friction and cohesion using SPT-N value and the second part consist of forms to predict Plasticity index and OMC from Liquid Limit and Plastic Limit respectively.
A10: LANDSLIDE HAZARD ZONATION AND MITIGATION STRATEGIES FOR PEERUMEDU TALUK
"Landslides are instant event of mass movement of earth surface down a slope, causing damage to life and property. Landslide hazard zonation mapping is an important step prior to landslide assessment planning, management and disaster mitigation. In this project, a multi-criteria analysis is used for landslide hazard zonation mapping for Peerumedu Taluk of Idukki district of Kerala. The methodology considers factors in the form of various thematic layers like slope, soil, geology, geomorphology, land use/land cover etc, were integrated in a GIS platform to delineate landslide hazard zone. A weighting rating system based on the relative importance of various causative factors were used for land hazard zonation. The hazard zonation map produced by using this technique classifies the area into relative hazard classes in which the high hazard zones correspond to high frequency of landslides. The final landslide hazard zonation map can be used for the prevention of landslide and proper planning of future infrastructure in the region. Highly susceptible landslide-prone area is used for the study of slope stability analysis and providing mitigation measures. The geotechnical parameters of soil and the terrain characteristics are used for the stability analysis of the slope, including distribution and characteristics of soils and the depth and geometry of the failures. Using GEO5 software, most geotechnical tasks can be solved by using the software output, factor of safety and comparing it with the standards, the stability of slope against sliding is analyzed. Since the slope at Elappara was found to be unstable, it is provided with various control works and mitigation measures for the improvement of soil strength. The mitigation measures are then provided for the best optimum condition with their dimensions to give the most stable condition of the slope. A cost analysis is undertaken to consider the economical feasibility of the mitigation works."
A12: SLOPE STABILITY ANALYSIS OF LANDSLIDE AREA IN DEVIKULAM TALUK
Study of slope stability analysis of landslide prone areas in Devikulam Taluk was done using geo5 software . First, in geo5 software analysis was done without mitigation measures,then with mitigation measures and based on results after analysis in geo5 software most feasible and mitigative measure with highest fos value was selected.
B6- PARAMETRIC STUDY OF MECHANICALLY STABILIZED EARTH WALL USING GEO5
"Retaining structures are usually built to support soil laterally thus preventing. catastrophes like landslides etc. A better alternative for conventional structures are Mechanically Stabilized Earth wall or MSE Wall. A mechanically stabilized earth (MSE) wall is an engineered system consisting of alternating layers of soil reinforcement and compacted backfill material fixed to the wall facing and supported on a foundation. The present study is focused on parametric analysis of single faced MSE walls and Back-to-Back MSE (BBMSE) walls using an analytical method , GEO5 MSE and a numerical model, which uses the finite element method (GEO5 FEM) to determine the factor of safety of the wall. This study mainly focuses on finding out how variations in the parameters affect the behaviour of the walls and its factor of safety. In case of single faced MSE walls, it was observed that improvement in the angle of internal friction leads to an increase in the factor of safety. It was also noted that, an increase in L/H ratio from 0.4 to 1.0 showed an increase in factor of safety by 26.4% and increase in reinforcement spacing from 0.2m to 1.5m showed a decrease in factor of safety by 37.4%. Factor of safety was also found to increase by 25.5% with an increase in characteristic strength of embedded reinforcement. From the parametric analysis, MSE walls were developed to replace the failed retaining walls at various locations in India. Parametric studies of BBMSE walls were carried out by varying the distance between side walls, by changing the backfill soil, type of reinforcement connectivity and reinforcement characteristic strength. The factor of safety and the deflection of the wall with respect to the parameters were analyzed. From the analysis of BBMSE walls, it was observed that it performed well under the improvement in soil factors, nominal amount of reinforcement, and placing of reinforcement either without spacing or overlapping gave a preferable result."