Optimal Design Of Airfoil Blunt Trailing Edge And Research On Blade Output Characteristics Of NREL Phase VI Wind Turbine Under Open Ice Conditions

The weather conditions such as low ambient temperature,high moisture content,freezing rain and snow usually lead to the wind turbine blade icing,changing the aerodynamic shape of blade thus reducing the wind energy utilization.The optimizing the geometry shape of blade in icing environment can effectively improves the aerodynamic performance of wind turbine.The shape of glaze ice is more complex,irregular and angular than rime ice.Therefore,the study on aerodynamic performance calculation and optimization design of blunt trailing-edge profile for airfoil,and output performance of blade under glaze ice condition has great engineering value of application for making full use of wind energy.The main research work and results are as follows:(1)The optimization model in which the biases of nodes,weights between nodes and the parameter of the link switch are some of design variables and the training error is minimized is established.The test error is introduced to interfere with the global optimal particle selection process,and the improved Quantum Particle Swarm Optimization(QPSO)algorithm,whose the updating modes of the potential well centre of the non-optimal particles are modified by social learning and the optimal particle position is updated by Lévy flight and greedy algorithm,combined with binary particle swarm optimization algorithm is used to solve the optimization model.The new neural network(SLLQPSO-BPNN)is established after training by Back Propagation(BP)algorithm.The lift and drag coefficients of S809 and NACA64618 airfoils with glaze ice at different angles of attack are predicted and compared with those of BP neural network.The results show that the information exchange between particles,population diversity,the ability of escaping from local optimal solution and the convergence speed of the improved QPSO algorithm are enhanced.The mean absolute error and mean relative error between the prediction values of SLLQPSO-BPNN and the experimental results are reduced,and the average relative error is less than 4%,the linear correlation coefficient is closer to 1,which indicates that SLLQPSO-BPNN is more accurate and effective.(2)The parametric control equations of the asymmetric blunt trailing-edge profile are established by introducing the trailing-edge thickness and its distribution parameters into mixed function of index.Taking the control parameters as the design variables and maximizing lift coefficient and lift-drag ratio before and after ice as the design objectives,the aerodynamic coefficients of airfoil with and without glaze ice are calculated by computational fluid dynamics(CFD)method and BP neural network respectively,and the S809 airfoil is optimized by using improved QPSO algorithm.The S809BT(BT represents blunt trailing-edge)airfoil with blunt trailing-edge thickness of2.621% times chord length and distribution ratio of 1:28.40 is obtained,and its aerodynamic performance and flow field characteristics before and after glaze ice are studied by CFD method.The results show that the flow field and pressure distribution of S809 BT airfoil are significantly improved,and the pressure difference between suction and pressure surfaces increases,which makes the lift coefficient and lift-drag ratio of airfoil increase obviously,improving the aerodynamic performance of the airfoil.(3)The fourth-order polynomial function is used to fit the distribution of twist angle along the blade span,and the parametric control equation of spanwise twist angle distribution is established.Based on S809 BT airfoil,the NREL Phase ? blunt trailingedge blade is constructed by using coordinate transformation equation.Taking the control parameters of twist angle distribution and maximizing wind energy utilization coefficient under rated conditions as design variables and design objective respectively,the improved QPSO algorithm combined with the Blade Element Momentum theory is used to optimize the twist angle distribution along the span-wise.The output characteristics of the sharp and blunt trailing edge blades before and after optimization are analyzed under glaze ice conditions.The results show that the wind energy utilization coefficient of the blunt trailing-edge blade is significantly improved before and after the ice.