Structural parameters multi-objective optimisation and dynamic characteristics analysis of large-scale wind turbine towers

Abstract

To optimise the structural parameters of wind turbine tower, a multi-objective optimisation model is established. The minimum mass and the minimum tower top displacement are selected to be the optimisation objectives. A non-linear mathematical model is established to map the relationship between the tower structure parameters and stress, top displacement, and natural frequency based on BP artificial neural network theory (BP-ANN). The NSGA-II algorithm is employed to solve the optimisation problem. Furthermore, the tower top vibration displacement, velocity and acceleration characteristics are analysed in the ANSYS and BLADED software based on the structural parameters before and after optimisation, respectively. The results show that tower structural parameters can be further optimised by employing multi-objective optimisation algorithm. In addition, in the iterative optimisation calculation, it has obvious advantages for using the established BP-ANN model. Changes of tower structural parameters will also change the real-time state of structure aero-elasticity, which also affects the vibration characteristics of the tower in operation process.