Wang, Keh-Han2014-12-192014-12-19December 22012-12http://hdl.handle.net/10657/832Prediction of significant wave height is critically important to the physical and environmental impact study of coastal, estuarine or large lake environments. In this study, development of predictive models for the determination of time varying significant wave heights in Lake Okeechobee, Florida using the simplified stochastic procedure and wave energy spectrum method is presented. The stochastic procedure related models are Regression Model 1 (RM1), Regression Model 2 (RM2) and Perceptron Least Square Method (PLSM). A new wave spectrum based model, Modified Pierson-Moskowitz (MPM) Spectrum is also developed. The predicted significant wave heights from each model are compared with the Artificial Neural Network (ANN) predictions obtained by Altunkaynak and Wang (2012). The comparisons between predicted significant wave heights from each model and observed data indicate that the proposed RM1, RM2, PLSM and MPM are effective models that are acceptable for predicting significant wave height in Lake Okeechobee.application/pdfengThe author of this work is the copyright owner. UH Libraries and the Texas Digital Library have their permission to store and provide access to this work. Further transmission, reproduction, or presentation of this work is prohibited except with permission of the author(s).Significant wave heightRegression MethodPerceptron Least Square MethodPierson-Moskowitz SpectrumArtificial neural networksCivil engineering2014-12-19Thesisborn digital