Engineering Commons^™

Convolution And Autoencoders Applied To Nonlinear Differential Equations, Noah Borquaye

Electronic Theses and Dissertations

Autoencoders, a type of artificial neural network, have gained recognition by researchers in various fields, especially machine learning due to their vast applications in data representations from inputs. Recently researchers have explored the possibility to extend the application of autoencoders to solve nonlinear differential equations. Algorithms and methods employed in an autoencoder framework include sparse identification of nonlinear dynamics (SINDy), dynamic mode decomposition (DMD), Koopman operator theory and singular value decomposition (SVD). These approaches use matrix multiplication to represent linear transformation. However, machine learning algorithms often use convolution to represent linear transformations. In our work, we modify these approaches to …

Characteristics Of Refractivity And Sea State In The Marine Atmospheric Surface Layer And Their Influence On X-Band Propagation, Douglas Matthew Pastore

Electronic Theses and Dissertations

Predictions of environmental conditions within the marine atmospheric surface layer (MASL) are important to X-band radar system performance. Anomalous propagation occurs in conditions of non-standard atmospheric refractivity, driven by the virtually permanent presence of evaporation ducts (ED) in marine environments. Evaporation ducts are commonly characterized by the evaporation duct height (EDH), evaporation duct strength, and the gradients below the EDH, known as the evaporation duct curvature. Refractivity, and subsequent features, are estimated in the MASL primarily using four methods: in-situ measurements, numerical weather and surface layer modeling, boundary layer theory, and inversion methods.

The existing refractivity estimation techniques often assume …