Estimation of Inframe Fill Stability Using Echo State Neural Network

In regions of high seismicity, infilled frames are commonly used for low and medium-height buildings. "Infilled frame" is a composite structure. It is formed by one or more infill panels surrounded by a frame. Infilled frame also refers to the situation in which the frame is built first and then infilled with one or more masonry panels. The primary function of masonry was either to protect the inside of the structure from the environment or to divide inside spaces. The presence of masonry infills helps the overall behavior of structures when applying lateral forces. The lateral stiffness and the lateral load capacity of the structure largely increase when masonry infills are considered to interact with their surrounding frames. In this paper, ANSYS 14 software is used for analyzing the infill frames. Echo state neural network (ESNN) has been used to supplement the estimation of stress values of the proposed infill frame model. The number of nodes or reservoirs in the hidden layer for ESNN algorithm varies depends upon the accuracy of estimation required. Exact number of reservoirs is fixed based on the trial and error method, through which the accuracy of estimation by the ESNN is achieved.