Aerodynamic optimization of variable-section J-shaped vertical axis wind turbine using feature-parameter microsegment dimension reduction method

This study aims to improve the adaptability of vertical axis wind turbines (VAWTs) in urban wind fields and reduce the cost of 3-D optimization. It focuses on 2-D to 3-D optimization of J-shaped blades. The goal is to enhance kinetic energy capture during startup and improve moment and power performance under different tip speed ratios (TSR). Based on the improved blade-element momentum (BEM) theory, 2-D J-shaped blades were developed for multi-environment applications. They were extended into 3-D forms using variable-section lofting. The 2-D optimization used the Northern Goshawk Optimization–Radial Basis Function (NGO-RBF) neural network and the Multi-Objective Harris Hawks Optimization (MOHHO) algorithm. The objectives were average power coefficient (CPA) and maximum moment coefficient (Cm-max). Maximum thickness (δ) and length ratio (lo/c) were design variables. For 3-D blades, moment (M) and average power (Pb) were used as performance indicators. Integrating the 2-D designs produced symmetric tapered and diffused tips and asymmetric tapered blades. CFD simulations revealed aerodynamic behavior under various TSRs. Optimized 2-D blades enhanced drag effects at low TSR and reduced drag loss while improving lift at high TSR. Among the 3-D designs, the symmetric tapered tip performed best. It increased minimum moment by 7.9% at low TSR and average power by 24.2% at high TSR. The 3-D flow results also confirm the conclusions from the 2-D optimization mechanisms. This study provides an effective design method for VAWT blades in urban parallel wind conditions. It also lays the foundation for future research on adaptive intelligent blades for complex wind environments.