Assessing the wave heights of the ocean using neural networks and fuzzy logic

Abstract

Soft computing techniques like neural networks are popular in the field of oceanography. To obtain the spectrum of oceanic parameters particularly the ocean wave heights are selected for the coastal regions of Tamil Nadu. The paper maintains keenness in applying the computer knowledge to ocean's welfare. The data sets are trained, based on the actual wave periods and heights. Such effort of measuring is achieved with the calibration of the tonal variation and the predicted wave table with the actual measurements and assess the error corrections to the predicted wave table with the actual measurements of the sea coast. This data is required to the planners of offshore structures, boat authorities, navigators, fisherman and the engineers producing oil from the offshore platforms. Once the training set is obtained, it is verified with actual height and necessary pruning has been introduced after that. The predicted and actual values of wave height reached, the inferred profile and the different stages applied are kept as reference block, by applying the root mean square corrections the data has been validated. A tolerance level of 7 to 10 percent error is envisaged in the measurements. The model is found successful.