@iitmandi.ac.in
Assistant Professor, School of Civil and Environmental Engineering, IIT Mandi
Indian Institute of Technology Mandi
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Varun Sharma, J. Dhanya, Maheshreddy Gade, and Jayalakshmi Sivasubramonian
Springer Science and Business Media LLC
Praveen Kumar, P. Priyanka, J. Dhanya, Kala Venkata Uday, and Varun Dutt
Institute of Electrical and Electronics Engineers (IEEE)
Movement of soil and associated landslides frequently occur in hilly areas. Regular monitoring, accurate prediction, and timely alerting of people about soil movements on hills susceptible to landslides are essential due to the potential destruction to life and property. A more recent strategy for predicting soil movement is the use of machine learning (ML) models. Different univariate and multivariate ML models have been proposed in the literature. However, evaluating these univariate and multivariate approaches in predicting real-world landslides have received less attention. This paper’s primary goal is to develop and compare the ability of univariate and multivariate ML models (Autoregression (AR), Seasonal Autoregressive Integrated Moving Average (SARIMA), Sequential Minimal Optimization regression (SMOreg), Multilayer Perceptron (MLP), and Long-short Term Memory (LSTM)) to predict movements at a real-world landslide. The case study used for the analysis in this paper is the Tangni landslide in India. This study makes use of weekly averaged soil movement data collected from June 2012 to December 2013 (78 weeks) at the Tangni landslide site. The dataset comprises measurements from five sensors. To calibrate the parameters in each model, we divided the collected data into a training dataset (first 62 weeks) and a test dataset (last 16 weeks). Performance analysis of the models utilized Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and R-squared values. The training results revealed that the univariate AR model demonstrated the best performance, achieving an RMSE of 0.149 degrees and an R-squared value of 0.572. The univariate SMOreg model obtained the second-best performance with an RMSE of 0.336 degrees and an R-squared value of 0.582. However, on the test dataset, the multivariate SARIMAX model outperformed the other models, achieving an RMSE of 0.351 degrees and an R-squared value of 0.769. The univariate SARIMA model also performed well with an RMSE of 0.356 degrees and an R-squared value of 0.741. The findings of this study can have significant implications in the field of landslide prediction and prevention. The results indicate that the multivariate SARIMAX model, the most accurate in predicting soil movements, can aid in developing early warning systems against landslides in hilly areas.
J Dhanya, K P Sreejaya, and S T G Raghukanth
Springer Science and Business Media LLC
J. Dhanya and S. T. G. Raghukanth
Informa UK Limited
ABSTRACT The present work aims at developing the first probabilistic fling hazard map of India and adjoined regions. First, we developed a new ANN-based ground motion prediction equation (GMPE) for fling corresponding to horizontal and vertical directions. The developed GMPE is based on permanent ground residual displacement form 556 scenario events considered consistent with the regional characteristics. The corresponding simulations are performed by suitably combining the Okada’s solutions. Developed GMPE is comparable with the existing relations and the few available data which contained fling characteristics. Further, the developed GMPE, along with the other two available prediction equations for the fling, is incorporated to represent the ground motion characteristics in the estimation of hazard using a suitable logic tree. In addition to the fling prediction equations, the evaluation of regional fling hazards requires identifying the location of all the probable seismic sources and their seismicity characteristics. In this study, we used the linear-fault model, as the fling is a near field phenomenon. We report the resultant probabilistic fling hazard map for 10%,2% 1%, and 0.5% probability in 50 Years for the region. The maps showed that the active regions in Himalayas, North-Eastern India, and Andaman experience higher values for fling than stable Peninsular India. Thus, this study develops the fling hazard map for the first time, and the results are essential in the design and rehabilitation of important structures in the region.
Dhanya J. and S. T. G. Raghukanth
Informa UK Limited
ABSTRACT The present work aims at exploring the application of nonlinear principal component analysis in dimensionality reduction and prediction of response spectra. The evaluation is performed based on log10 scaled response spectra at 91 spectral periods corresponding to 13552 records available in the NGA-West2 database. The non-linear principal component analysis performed on the data showed that 91 spectral periods can be addressed with just 3 principal components. Further, an artificial neural network (ANN) model is developed to predict these three principal components with magnitude, distance, shear wave velocity and focal mechanism as input. The inter- and intra-event residuals obtained for the response spectra predicted using the developed model are comparable with the existing ground motion prediction equations (GMPEs) from the same database. The developed model is also observed to capture all the prominent attenuation features of ground motions. Hence, the study indicates that the response spectra can be described with just three uncorrelated variables.
S. Jayalakshmi, J. Dhanya, S. T. G. Raghukanth, and P. M. Mai
Springer Science and Business Media LLC
J. Dhanya, S. Jayalakshmi, and S. T. G. Raghukanth
Springer Singapore
Maheshreddy Gade, Partha Sarathi Nayek, and J. Dhanya
Springer Science and Business Media LLC
S. Sangeetha, J. Dhanya, and S. T. G. Raghukanth
Informa UK Limited
ABSTRACT The present study focuses on developing a 3D crustal velocity model and applying it to perform ground motion simulations for North East India. The study region encompasses area between 89°E to 97°E longitude and 22°N to 30°N latitude. The calibration of the material property is based on 48 shear wave profiles available for the region along with the geotectonic features reported in the literature. The 3D material model arrived from the study is implemented in a finite element framework for ground motion simulations. The developed model is validated using the strong motion data available for four events; 1988 Mw 7.2 India–Bangladesh earthquake, 2011 Mw 6.3 India–Myanmar earthquake, 2013 Mw 5.5 Assam earthquake and Mw 5.2 Bhutan earthquake. The simulations are able to capture the prominent features of the recorded data up to 2Hz. Hence, the developed model can be implemented for estimating ground motions (< 2Hz) in the north eastern region of India. The simulated results can be used to estimate region-specific hazard and the displacement-based design of structures.
J. Dhanya and S. T. G. Raghukanth
Current Science Association
The present work aims to ascertain the deterministic tsunami hazard map for maximum wave height along the Indian coastline due to subduction events in the Sumatra region. The region between 15°S–30°N lat. and 50°E–115°N long. was modelled in using Geoclaw, which discretizes and solves the shallow-water wave equation using adaptive finite volume algorithm. The developed model was suitably validated for the available tidal gauge and altimeter data for the 2004 Mw 9.12 Sumatra earthquake with input source characteristics from the SRCMOD data. Further, a sensitivity study based on slip variability and location was conducted which revealed that near-field stations were highly sensitive at all locations while far-field stations were more sensitive towards source locations. Furthermore, 25 non-Gaussian slip fields were generated for the maximum possible event (Mw 9.12) for the region and placed suitably along the active Sumatra subduction region. Then the wave heights from all the simulations were assembled to determine the deterministic tsunami hazard values with respect to maximum wave heights for the coastal regions of India and adjoining regions. The results will find application in the design of structures along the coastline of the study region.
J. Dhanya and S. T. G. Raghukanth
Springer Science and Business Media LLC
J. Dhanya and S. T. G. Raghukanth
Springer Science and Business Media LLC
J. Dhanya and S. T. G. Raghukanth
Springer Science and Business Media LLC
S. Jayalakshmi, J. Dhanya, S.T.G. Raghukanth, and P. Martin Mai
Elsevier BV
J. Dhanya and S. T. G. Raghukanth
Springer Science and Business Media LLC
Anjali Dhabu, J. Dhanya, and S. T. G. Raghukanth
Springer Singapore
J. Dhanya, Dwijesh Sagar, and S. T. G. Raghukanth
Springer Singapore
J. Dhanya and S. T. G. Raghukanth
Current Science Association
The present study aims at developing a model for simulating ground motion for earthquakes in the Sumatran region where one of the most devastating earthquakes took place in 2004 with a moment magnitude (M w ) of 9.1. With advancements in instrumentation, the three-dimensional material properties, topography and bathymetry of the region are available in the global database. These parameters are used as inputs in Spectral Finite Element Method to simulate ground motions. The model is first validated with the IGCAR broadband velocity data for 2012 M w 8.6 Sumatra Earthquake. Due to favourable comparison, our model is also used to generate ground displacement characteristics of M w 9.1 event. The source uncertainties are accounted by using three finite fault slip models available in the global database. The simulated time histories showed that the ground motion is sensitive to input slip models. The peak ground displacement (PGD) and ground residual displacement (GRD) in both horizontal and vertical directions are presented as contour plots. PGD obtained from various slip models in the epicentral region is of the order of 14–22 m in horizontal direction and 7–16 m in vertical direction. GRD in the epicentral region is of the order of 6–17 m in East–West (E–W) 4–17 m in the North–South (N–S) directions. The vertical uplift obtained from various slip models is around 2–8 m. The developed model can be used to simulate ground motion time histories, which can be further used in hazard analysis, tsunami simulations, etc.
J. Dhanya and S. T. G. Raghukanth
Springer Science and Business Media LLC
J. Dhanya, Maheshreddy Gade, and S. T. G. Raghukanth
Springer Science and Business Media LLC