| Popis: |
Pitching turbine blades into the wind increases the thrust coefficient, $C_{T}$ , which increases the power generated by the wind turbine. However, excessive $C_{T}$ increments beyond rotor mean wind speed $C_{T}$ -equivalent, tend to cause overexertion and increased loads. Consequently, the rated operational lifetime of the turbine is reduced. This study uses a high-fidelity 2-D Gaussian wake model and an augmented version of Frandsen’s turbulence intensity (TI) model to simulate a hexagonally deployed wind plant (WP) operation. Turbines’ axial-induction factor $\alpha $ is optimised using Particle Swarm Optimisation (PSO) and Genetic Algorithm (GA), to maximise WP power and annual energy production (AEP), with constrains on individual turbine $C_{T}$ values to remain within rotor wind speed equivalent based on turbine’s thrust curve. At a $5D$ minimum turbine-to-turbine (T-2-T) separation distance, results show that $C_{T}$ constraints on individual turbines increased the wind speed range of healthy operations by up to 66.67% considering extreme loads. AEP gains reduced from 11.91% and 13.25% (optimised without constraints), to approximately 7.59% and 5.74% (with constraints), when compared to the corresponding $5D$ Base case (non-optimised and unconstrained), using PSO and GA, respectively. The study also shows that WP power maximisation can increase turbulence intensity levels within the WP especially if turbines are tightly deployed. The outcome of this study has implications for new wind farm layouts and wind plant power optimization. |
