Study Of Multi-turbine Wake Models Based On Numerical Simulation

Problems of energy resources shortage and environmental pollutions are becoming more and more prominent with the development of global economy,and rapid development of renewable energy especially for wind energy has become an important way to solve these problems.The process that turbine transforms wind energy to mechanical energy and electricity at last will cause the decrease of wake velocity and increase of turbulent intensity.In order to make full use of the space of wind farm,downstream turbines will be inevitably set in the upstream turbines’ wake region,causing wind power loss and fatigue damage of turbines.That’s why it’s really important to research the multi-turbine wake effect.In earlier stage the best way to reduce the wake effect in wind farm is to optimize its layout,and analytical wake models are often used in this work,but the existing velocity superposition models can’t consider the turbulent intensity increase in downstream wake region,which will influence the accuracy of these models.This paper focuses on the numerical simulation of multi-turbine wake,the establishment of turbulent intensity superposition models and the analysis and improvement of the existing velocity superposition models:(1)The actuator line model based on wind turbine aerodynamics including blade element momentum theory and large eddy simulation are introduced,then we use actuator line model and large eddy simulation to simulate aligned two turbines wake which is then validated with classical paper to show the accuracy of the simulation method used in this paper.After that we choose the single turbine wake model with highest precision that is used in multi-turbine models’ research through comparing with the data of single turbine wake simulation.(2)Aligned 4 turbines wake are simulated by actuator line method and large eddy simulation to study the influence of inflow velocity,turbulent intensity and spacing to multi-wake superposition effect,and we analyze the stabilized speed of downstream turbine wake velocity and intensity.We consider wind turbine wake goes into steady state from second turbine,and the higher the inflow intensity,the greater the spacing,the higher the coincidence degree between the wakes of each downstream turbine.This provides important reference to build turbulent intensity superposition model.(3)We take some examples to analyze the importance for wake velocity superposition models to consider the increase of wake turbulent intensity.Then according to velocity superposition models we propose new wake turbulent intensity superposition model with similar expression.And we filter and validate several turbulent intensity superposition models and analyze the accuracy of the newly proposed linear turbulent intensity superposition model.(4)Combining the linear turbulent intensity superposition model with the existing velocity superposition models can eliminate the error source caused by without considering the increase of wake turbulent intensity,so at this time velocity superposition models’ results are mainly related to models itself,and the accuracy and distribution law of the existing velocity superposition models can be analyzed by comparing with the numerical simulation results.Regarding to existing problems,an exponential superposition model is proposed,and the model constant is related to the spacing between aligned neighbor turbines,which further improves the applicability of the engineering wake model.(5)Lastly the existing models and the newly proposed exponential velocity superposition model and linear turbulent intensity superposition model are applied to the Lillgrund wind farm.We compare them with the measured data and LES results,and it is validated that the exponential superposition model has higher accuracy than the existing four velocity superposition models,at the same time linear turbulent intensity superposition model also has high accuracy.Thus the feasibility and accuracy of the newly proposed superposition models applied to other wind fields are verified.