A MATLAB-based GUI for Performance-based Tornado Engineering (PBTE) of a Monopole, Vertical Structure with Artificial Neural Networks (ANN)

The PBTE_ANN application is a MATLAB®-based graphical user interface (GUI) that uses Artificial Neural Networks (ANN) to expedite the performance-based tornado engineering (PBTE) assessment of a monopole, vertical structure subjected to wind loads from simulated tornadoes of various intensities and sizes. Four components of PBTE are embedded: 1) simulation of structural responses, 2) approximation of fragilities, 3) hazard analysis for assessment of probabilities of structural failure, and 4) life-cycle cost assessment (LCCA). The application serves as a preliminary demonstration of machine learning in the performance-based evaluation of structures. The algorithms are based on work supported by National Science Foundation (NSF) project CMMI-1434880 led by the Principal Investigator, Associate Professor Luca Caracoglia of Northeastern University’s Civil and Environmental Engineering Department. The machine learning component of the application has been developed with support from The MathWorks, Inc. through a micro-grant awarded in 2019 by Northeastern University and Professor Miriam Leeser of the Department of Electrical and Computer Engineering. Additional information along with references that explain the algorithms in more detail are provided in the user manual. An example data set is provided to guide users.