International journal of advanced scientific and technical research Issue 2 volume 5, October 2012 Available online on http://www.rspublication.com/ijst/index.html ISSN 2249-9954

In past days many researchers have been worked on the stabilized Black Cotton (BC) soil to determine the Unconfined Compressive Strength (UCS) values in a conventional ways in laboratory thorough experiments, which is time consuming and requires number of persons. So we the authors of this study attempted to determine these values using Artificial Neural Networks (ANN). ANN which are simple mathematical models are inspired from the brain’s certain information-processing characteristics including the parallel processing, the ability to learn and generalize, disregard data errors and produce meaningful solutions, which fall beyond the reach of conventional digital computers. Over the last few years or so, the use of ANN has increased in many areas of engineering. In particular, ANN has been applied to many geotechnical engineering problems and has demonstrated some degree of success. A review of the literature reveals that ANN has been used successfully in pile capacity prediction, modeling soil behavior, site characterization, earth retaining structures, settlement of structures, slope stability, design of tunnels and underground openings, liquefaction, soil permeability and hydraulic conductivity, soil compaction, soil swelling and classification of soils. In the present study the BC soil has been mixed with RHA and cement with different proportions. The different combination of BC soil-RHA-Cement mixes have been tested for unconfined compressive strength immediately after preparing the samples and after (3+1) days curing period. The UCS values of mixes BC soilRHA increases increase of cement proportions and found higher values for cured samples. It is observed that UCS value is more consistent with the mix BC soil+8% cement+10% RHA compared other mixes, though little changes observed in the other mixes. Experimental results have been compared with the UCS values determined by ANN and the errors are determined. The errors found minimum of 1.18% for BC soil+ 8% cement+ 10% RHA and maximum of 9.40% for BC soil+ 4% cement+ 15% RHA. The present study deals with collection of input data base from experimental results, ANN’s training and its testing are adopted to fix the appropriate weighted matrix which in turn Prognosticates the UCS value. Experimental results have been compared with the UCS values prognosticated by using ANN. The results of this study will contribute for the prognostication of UCS, which will assist a geotechnical engineer in estimation of UCS, with minimum effort.