Multivariate Data Analysis of Maximum Stress Concentration Factors in FRP-Retrofitted Two-Planar KT-Joints under Axial Loads for Offshore Renewables

Abstract

With growing concerns about the danger of global climate change and worldwide demand for energy, the interest in the investigation and construction of renewable energy technologies has increased. Fixed platforms are a type of support structure for wind turbines composed of different types of tubular joints. These structures are under different kinds of cyclic loadings in ocean environmental conditions, which must be designed and reinforced against fatigue. In the present paper, the relationships between the parameters in DKT-joints reinforced with FRP under axial loads are investigated using several models, under 16 axial loading cases, with different nondimensional parameters and different FRP materials, and orientations were generated in ANSYS (total 5184) and analyzed. The four loading conditions that cause the maximum stress concentration factors were selected. After analyzing the 1296 reinforced models, relevant data were extracted, and possible samples were created. The extracted data were used in a multivariate data analysis of maximum stress concentration factors. The Pearson correlation coefficient is utilized to study the relationship between parameters and subsequently to make predictions. To reduce the number of variables and to group the data points into clusters based on certain similarities, hierarchical and non-hierarchical classifications are used, respectively.