Minimizing elastic interaction in multi-material wind turbine blade joints using hybrid constitutive modelling and genetic algorithm

A generalized sequence-dependent constitutive model is formulated and coupled with genetic algorithm (GA) to minimize the effect of bolt-on-bolt elastic interaction (EI) of a large size wind turbine blade joint during the process of joint preloading prior to turbine operation. The composite blade flange is bolted to a steel hub flange on the turbine tower side. A Python-based tool is developed for big data sets of elastic interaction among the 54-bolt wind turbine blade-to-hub connection. Trial tightening sequence and patterns include (a) clockwise sequential tightening of six bolts at each apex of three rotated triangles, (b) clockwise sequential tightening of nine bolts at each apex of two rotated triangles, (c) separately tightening one pair of bolts that are located 180° diagonally across each other and repeat 26 more times in a clockwise direction. A genetic algorithm is used to identify feasible tightening sequences that minimize the scatter of bolt preload and achieve the desired level of a uniform joint preload.