Flow Control through Surface Suction for Small Wind Turbines

2017-03-27T04:45:09Z (GMT) by Jasvipul Singh Chawla
This thesis seeks to arrive at estimates of improvement in blade aerodynamic efficiency and reduction in structural loads in small wind turbines through surface suction-based active flow control at low Reynolds numbers (<i>Re</i>). Improved aerodynamic efficiency and reduced fatigue loads help achieve lower Cost-of-Energy.<br><br> Small wind turbines typically operate off-grid, providing power at the point of consumption. Consequently, such turbines often operate in locations with poor wind speed regimes. Low wind speeds coupled with small chord length of the blades result in low operating <i>Re</i> that often lie between 10<sup>4</sup> and 10<sup>5</sup>. It is well-known that such operating <i>Re</i> imposes significant challenges in realising good aerodynamic efficiency due to the propensity for flow separation to occur. <br>     <br>    It has been established that flow separation can be avoided or delayed using active flow control. However, much of such work on active flow control has focused on high <i>Re</i> applications. This thesis focuses on the application of surface suction, a popular active flow control methodology, in low <i>Re</i> regimes seen in small wind turbines.<br><br> The approach adopted towards arriving at the aforementioned estimates of improvement in blade aerodynamic efficiency and reduction in structural loads in small wind turbines is as follows: <br>     <br>    1. Experimental characterisation of the improvement in the aerodynamic characteristics of two aerofoil profiles, NACA0012 and S814, commonly used in small wind turbine applications, in low <i>Re</i> regimes is performed. Specifically, this characterisation captures both the steady-state characteristics as well as identification of the temporal dynamics between change in the coefficient of lift (Δ<i>C<sub>L</sub></i>) and change in the coefficient of drag (Δ<i>C<sub>D</sub></i>) with the non-dimensional parameter, <i>C</i><sub>μ</sub> (defined as the ratio of the momentum of air drawn through suction and the momentum of the air flowing over the aerofoil, that is, <i>ρ</i>  <i>A<sub>slit</sub></i>  <i>u<sub>s</sub><sup>2</sup></i>) / ( <i>ρ </i> <i>A<sub>aerofoil</sub></i>  <i>U</i><sub>∞</sub><i><sup>2</sup>)</i>, where<i> ρ</i> is air density, A<sub><i>slit</i> </sub>area of the slit, <i>A<sub>aerofoil</sub></i> area of the aerofoil, <i>u<sub>s</sub></i> suction velocity and <i>U</i><sub>∞</sub> the free stream velocity): <br>    Δ<i>C<sub>L</sub></i>, Δ<i>C<sub>D</sub></i> = <i>f</i>(<i>C</i><sub>μ</sub>) <br>    Δ<i>C<sub>L</sub></i>, Δ<i>C<sub>D</sub></i> = <i>f</i>(<i>C</i><sub>μ<i>,</i></sub><i>t</i>) <br>     <br>    2. Posing a similitude argument that as long as the dimensionless parameters <i>C</i><sub>μ</sub> and <i>Re</i> remain comparable to the regimes for which the aforesaid experimental characterisation was done, the steady-state and temporal relationships between Δ<i>C<sub>L</sub></i>, Δ<i>C<sub>D</sub></i> and <i>C<sub>μ</sub></i> established could be directly used in numerical simulations for predicting the behaviour of small wind turbines. <br>     <br>    3. Using the relation Δ<i>C<sub>L</sub></i>, Δ<i>C<sub>D</sub></i> = <i>f</i>(<i>C<sub>μ</sub></i>) and steady Blade Element Momentum (BEM) theory to estimate increase in Coefficient of Power, <i>C<sub>P</sub> </i>of a small wind turbine employing surface suction on its blades working in the same <i>C<sub>μ</sub></i>,<i>Re</i> regime. <br>     <br>    4. Incorporating the dynamic map, Δ<i>C<sub>L</sub></i>, Δ<i>C<sub>D</sub></i> = <i>f(C<sub>μ,</sub>t)</i> into the dynamics of the aero-elastic simulator, FAST to formulate extended turbine dynamics. <br>     <br>    5. Utilising, appropriately, <i>C<sub>μ</sub></i> as an additional control input towards reducing fore-aft oscillations of the tower top while compensating for the said extended turbine dynamics. <br>     <br>    6. Demonstrating, through rain-flow analysis, that the reduced oscillations result in mitigation of fatigue loads on the turbine tower structure. <br>     <br>    The thesis documents that the approach indicated for increasing <i>C<sub>P</sub></i> , when applied to a small wind turbine in sub 1<i>kW</i> power output range, with 2<i>m</i> radius, NACA0012-blades, operating in steady wind speed of 7.5<i>m/s</i>, increased the expected power output from 764<i>W</i> to 1511<i>W</i> by expending 106<i>W</i> of suction power. <br>     <br>    Further, compensating for the extended dynamics of the turbine in the aero-elastic simulator, tower-top oscillations reduced for a ~ 10<i>kW</i> turbine of 2.9<i>m</i> radius, S814 blades, hub height 24<i>m</i>. Thus, for the turbine operating over 20 years in turbulent IEC III-A wind conditions, the structural damage equivalent reduced from ~ 10 to < 1.