WAVELET-BASED TIME DELAYED ADAPTIVE NEURO FUZZY INFERENCE SYSTEM FOR PREDICTING NONLINEAR BEHAVIOR OF SMART CONCRETE STRUCTURE EQUIPPED WITH MR-DAMPERS

摘要

The present invention relates to a wavelet-based time delayed adaptive neuro-fuzzy inference system to predict a nonlinear behavior of a concrete structure equipped with MR-dampers, including: a first step of collecting data from the structure; a second step of performing modeling to describe the nonlinear behavior; a third step of inputting the data collected in the first step to the model modeled in the second step to be repeatedly modeled; a fourth step of checking a difference between a value predicted in the third step and an actual value using a root-mean-square error, and returning to the second step in case the error is larger than an allowed error or proceeding the next step in case the error is smaller than the allowed error; and a fifth step of testing validity of the model. The wavelet-based time delayed adaptive neuro-fuzzy inference system of the present invention can define a behavior of a physical system by combining the rules of fuzzy, language, and number and can predict the nonlinear behavior of the concrete structure efficiently under various levels of impact loads.