Performance Of Wells Turbine Engineering Essay

Performance Of Wells Turbine Engineering Essay ABSTRACT A Wells turbine has inherent disadvantages in comparison with conventional turbines: relative low efficiency and poor starting characteristics. In this case, the performance of wells turbine is studied on computational analysis by changing aerofoils and providing different angle of incidence for the improvement of the turbines performance. Study is based on analysing the flow of air on turbine using computational analysis at steady condition. 1 INTRODUCTION: The Ozone depletion and global warming have altered the international community and urged the need for more focus on alternative green sources of energy. Ocean wave energy is one of the renewable forms of energy which can be utilized in response to the disturbing prospect of an exhaustible source of energy. Several wave energy devices being studied under many wave energy programs make use of the principle of the oscillating water column (OWC).Potentially; the most successful device used in harnessing on wave energy has been the OWC wave energy converter. The OWC chamber, either floating or bottom standing, with the immersed end opened to the action of the sea. A reciprocating airflow is created by the action of the free surface of the water within the chamber. The conversion of this airflow into mechanical energy may be achieved by a number of devices like:- A. TAPCHAN: The TAPCHAN comprises a gradually narrowing channel with wall heights typically 3 to 5 m above mean water level. The waves enter the wide end of the channel and, as they propagate down the narrowing channel, the wave height is amplified until the wave crests spill over the walls to a reservoir which provides a stable water supply to a conventional low head turbine. The requirements of low tidal range and suitable shoreline limit the world-wide installation of this device. Fig (a) TAPERED CHANNEL[1] (TAPCHAN)http://re.emsd.gov.hk/english/other/marine/images/marine_tech_010_2.gif B. PENDULOR: The PENDULOR device consists of a rectangular box, which is open to the sea at one end. A pendulum flap is hinged over this opening, so that the action of the waves causes it to swing back and forth. This motion is then used to power a hydraulic pump and generator.http://t2.gstatic.com/images?q=tbn:ANd9GcQ7yge9ouptnhszDgsXGA_gCvAXKqbo78BeXZHFFtPB89433p0p Fig (b) PEDULOR [2] C. WELLS TURBINE: The Wells turbine is one of the most suitable air turbines for energy conversion from oscillating air flow. . A schematic view of the OWC device with a Wells turbine is shown in Fig. c. The Wells turbine is an axial flow air turbine. It consists of several symmetrical aerofoil blades set around a hub. As waves Impinge on the device, they cause the water column to rise and fall in the air chamber, which alternately compresses and depressurized the trapped air. This air is allowed to flow to and from the atmosphere through a turbine which drives an electric generator.http://www.aussiestockforums.com/forums/attachment.php?attachmentid=9213stc=1d=1180172232 Fig (c) WELLS TURBINE[3] Fig. 1.1: Schematic of the Three Major shoreline Devices The Wells turbine is a self-rectifying air turbine which is expected to be widely used in wave energy devices with the OWC (Raghunathan, 1995) .It can extract power at low airflow rate, when other turbines would be inefficient. The Wells turbines for wave power conversion have less efficiency. To increase the efficiency is the major quest all over the world, the flow of air through the wells turbine impeller is carried out in this project by using different size impeller and introducing biplane i.e. two rows of symmetrical aerofoil blades. 1.1 WAVE RESOURCEShttp://www.oceanenergy.ie/images/world-map.jpg Fig. 1.2: Global Distribution of Deep Water Wave power levels in kW/m crest length[4] Despite the climate change phenomena, the world resource for wave remains very much as shown in fig. 1.2 by Dr Tom Thorpe [5]. The highest energy waves are concentrated off the western coasts in the 40o-60o latitude range north and south. The power in the wave fronts varies in these areas between 30 and 70kW/m with peaks to 100kW/m in the Atlantic SW of Ireland, the Southern Ocean and off Cape Horn. The capability to supply electricity from this resource is such that, if harnessed appropriately, 10% of the current level of world supply could be provided [4] PERIOD AMPLITUDE POWER DENSITY VELOCITY (m/s) WAVELENGTH (sec) (m) (kW/m) (m) Storm 14 14 1700 23 320 Average 9 3.5 60 15 150 Calm 5.5 0.5 1 9 50 Fig: 1.3 Tthe nautral and technical wave energy resource for the north and west side of the UK[6] The techinical resource is dependent on the nautral conditions like the shape of the rock and location i.e beaches and gullies. The wave energy at calm sea is considered in this project. 1.2 WELLS TURBINE The monoplane Wells turbine i.e. the basic Wells turbine consists of several symmetrical aerofoil blades (NACA four digit series) set around a hub at 90 degrees with respect to the airflow. Since its an invention by Prof. A.A. Wells in 1976, most researchers have focused on improving its efficiency and its range of efficient operation. In fact, compared to other conventional air turbines (e.g. Francis turbine) the Wells turbine has a lower efficiency and a narrow operational region. Nevertheless, it can extract power at low airflow rate, when other turbines would be inefficient. Fig. 1.4: Schematic of the Monoplane Wells Turbine[7] A schematic diagram of a Wells turbine is shown in Fig. 1.4. At first sight the arrangement might seem to be unlikely means of energy conversion. However, once the blades have attained design speed the turbine is capable of producing a time-averaged positive power output from the cyclically reversing airflow with a fairly high efficiency. Wells turbine has low efficiency and poor starting characteristics. The Biplane Wells turbine: Muhammad Mamun in the Study on the Hysteretic Characteristics of the Wells Turbine in a Deep Stall Condition says the pressure drop across a mono-plane Wells turbine above is proportional to the square of the tip speed which has to be limited if transonic effects are to avoided. For wave energy devices which produce significantly larger pressure drops than the limit for a single plane turbine a biplane turbine can be used Fig. 1.5: Schematic of the Biplane Wells [7] It has certain advantages over the conventional monoplane Wells turbine as follows: I. It can operate under high loading. II. It can absorb higher wave power than the monoplane turbine if the diameter and rotational speed of the turbine are kept constant. III. The design speed is lower than that of the monoplane for the same loading. IV. It avoids the use of guide vanes and therefore the turbine would require less maintenance and repairs. [7] 1.3 PRINCIPLE OF OPERATION The principle of operation of Wells turbine is based on the classical aerofoil theory. According to the classical aerofoil theory, an aerofoil which is set at an angle of incidence ÃŽÂ ± in a fluid flow generates a lift force L normal to the free stream. The aerofoil also experiences a drag force D in the direction of the free stream (relative velocity). These lift and drag forces can be resolved into tangential (in the plane of rotation) and axial (normal to the plane of rotation) components FT and FA respectively. Fig. 1.6 Notation for determining lift, drag, and axial and tangential forces on An aerofoil[7] Resulting expression for axial and tangential forces FA = LcosÃŽÂ ± + DsinÃŽÂ ± FT = LsinÃŽÂ ± DcosÃŽÂ ± The axial force is absorbed but the turbine while the tangential force causes the turbine to rotate. For a symmetrical airfoil the direction of tangential force is the same for both positive and negative values of ÃŽÂ ±. Therefore, the direction of rotation of the rotor is independent of airflow direction. 2 AIMS AND OBJECTIVES OF THE PROJECT Simulation of air flow through wells turbine impeller by means of numerical method using a CFD (Computational fluid dynamics) called FLUENT and check the flow process of different parameters and the factors affecting the differences. Since wells turbine is a low efficiency turbine to increase the efficiency of Wells turbine is the other aim. Path followed to meet the requirements is first calculation of efficiency theoretically. Simulating a modified design by using different angle of incidence and making biplane i.e. two rows. Comparing the results of different model and selecting the suitable design. 3 LITERATURE REVIEW: 3.1 Types of CFD PROCESS USED: Commercial CFD code: FLUENT, Star-CD, FLOW-3D, CFX/AEA, etc. Research CFD code: Self-developed Public domain software (PHI3D, HYDRO, and WinpipeD, etc.) Other CFD software includes the Grid generation software (e.g. Gridgen, Gambit) and flow visualization software (e.g. Tecplot, FieldView) Commercial CFD code FLUENT is used in this project. 3.2 General working on CFD Table3.1: CFD working layout The Processes shown in the table 3.1 is divided into pre- process and post- process viz. GAMBIET AND FLUENT General sequence of GAMBIT operations Initial setup Solver selection, Mesh size, Defaults, etc. Geometry creation (ACIS, IGES or Mesh import) Create full geometry Decompose into mesh-able sections Meshing Local meshing: Edge and Boundary layers Global meshing: Face and/or Volume Mesh examination Zone assignment Continuum and Boundary attachment Mesh export General sequence of FLUENT operations Selection of appropriate models. Turbulence, combustion, multiphase, etc. Define material properties Fluid Solid Mixture Prescribe operating conditions Prescribe boundary conditions at all boundaru zones Provide and initial solution Set up solver controls Set up convergence monitors 3.3 Grid generation: Grid generation is one of the key elements in Computational Fluid Dynamics (CFC). It has now become a fairly common tool for use in the numerical solution of partial differential equations on arbitrarily shaped regions. The numerical solution of partial differential equations requires some discretization of the field into a collection of points (nodes) or elemental volumes (cells). The differential equations are approximated by a set of algebraic equations on this collection, and this system of algebraic equations is then solved to produce a set of discrete values which approximates the solution of the partial differential system over the field. The practice of discretizing the physical domain into a finite number of elements is called as grid generation. 3.4 Grid topologies Generally, the governing equations may be transformed into finite element, finite difference, or finite volume equations. The cell types supported by FLUENT are followed as: triangular and quadrilateral cells in 2D are accepted, and in 3D, tetrahedral, hexahedral, wedge, and pyramid cells can be used FIG3.2: Different types of grids Structured versus Unstructured Grids The section presents a brief description of grid generation. The grid generation techniques available at present fall into two categories, namely: a) structured grid generation and b) unstructured grid generation. The structured grid generation techniques are based on the transformation of the complex physical domain into a simple computational domain, which is often chosen to be rectangular in shape (quadrilateral and hexahedron). The unstructured grid generations have been used with FEM (finite element method) procedure only, whereas structured grids have general applicability.[7] . 3.5 Types of structured grid In FLUENT, both single-block and multi-block structured meshes are acceptable, as well as hybrid meshes containing quadrilateral and triangular or hexahedral, tetrahedral, pyramid, and wedge cells Multiple Block Sometimes, it is possible to combine several structured computational meshes together to fit the physical domain. Multi- locking has the advantage of the speed of a structured solver, without as many mapping constraints apparent in single block meshes. Single Block In this technique, one computational grid is mapped to fit the whole physical domain. For even moderately complex shapes, it may be practically impossible to define a transformation which will map the outer surface of the computational domain to the required physical shape, while ensuring that the resulting grid has desirable attributes of smoothness.[7] 3.6 Mesh quality The quality of mesh plays a significant role in the accuracy and stability of the numerical simulations. The attributes associated with mesh quality are density of node, cell shape, smoothness and flow-field dependency. In many cases, poor resolution in critical regions can dramatically alter the flow characteristics. 3.7 The Capabilities of FLUENT This section provides a brief introduction to FLUENT and an explanation of its capabilities [10].FLUENT used in this project is a commercial code and a state-of-the-art computer program for modelling single and multiphase flows, heat and mass transfer, chemical reaction phenomena, and etc. in complex geometries. This code includes following components; FLUENT, the flow solver; GAMBIT, the pre-processor for geometry modelling and mesh generation; pre-PD, and etc. FLUENT solver utilizes a finite-volume, pressure-based, multiphase space marching method (SIMPLE algorithm), for solving the governing integral equations for conservation of mass and momentum, and for energy and other scalars such as turbulence and chemical species. It has the following modelling capabilities: à ¢Ã¢â€š ¬Ã‚ ¢ Flows in 2D or 3D geometries using triangular/tetrahedral, quadrilateral/hexahedral, or mixed (hybrid) grids that include prisms (wedges) or pyramids à ¢Ã¢â€š ¬Ã‚ ¢ In compressible or compressible flows à ¢Ã¢â€š ¬Ã‚ ¢ Steady-state or transient analysis à ¢Ã¢â€š ¬Ã‚ ¢ Laminar and turbulent flows à ¢Ã¢â€š ¬Ã‚ ¢ Newtonian or non-Newtonian flow à ¢Ã¢â€š ¬Ã‚ ¢ Convective heat transfer, including natural or forced convection à ¢Ã¢â€š ¬Ã‚ ¢ Coupled conduction/convective heat transfer à ¢Ã¢â€š ¬Ã‚ ¢ Radiation heat transfer à ¢Ã¢â€š ¬Ã‚ ¢ Inertial (stationary) or non-inertial (rotating) reference frame models à ¢Ã¢â€š ¬Ã‚ ¢ Multiple moving reference frames, including sliding mesh interfaces and mixing planes for rotor/stator interaction modelling à ¢Ã¢â€š ¬Ã‚ ¢ Chemical species mixing and reaction, including combustion sub models and surface deposition reaction models à ¢Ã¢â€š ¬Ã‚ ¢ Arbitrary volumetric sources of heat, mass, momentum, turbulence, and chemical species à ¢Ã¢â€š ¬Ã‚ ¢ Flow through porous media à ¢Ã¢â€š ¬Ã‚ ¢ One-dimensional fan/heat-exchanger performance models à ¢Ã¢â€š ¬Ã‚ ¢ Two-phase flows, including cavitations à ¢Ã¢â€š ¬Ã‚ ¢ Free-surface flows with complex surface shapes FLUENT can provides a number of boundary conditions, including: à ¢Ã¢â€š ¬Ã‚ ¢ Velocity or Pressure Driven Inlets/Outlets à ¢Ã¢â€š ¬Ã‚ ¢ Stationary or Moving Walls, with or without Friction à ¢Ã¢â€š ¬Ã‚ ¢ Periodic Boundary Conditions à ¢Ã¢â€š ¬Ã‚ ¢ Symmetry Boundary Conditions à ¢Ã¢â€š ¬Ã‚ ¢ Pressure Far-filed Boundary Conditions à ¢Ã¢â€š ¬Ã‚ ¢ Outflow Boundary Conditions à ¢Ã¢â€š ¬Ã‚ ¢ Inlet/Outlet Vent Boundary Conditions à ¢Ã¢â€š ¬Ã‚ ¢ Intake/Exhaust Fan Boundary Conditions As the Well turbine has a complex geometry for modelling, a large number of modelling capabilities are required of the CFD code for the turbine. FLUENT can incorporates all of these capabilities, and is most suitable for modelling the Wells turbine.[10] 4 ANALYSIS OF TASK 4.1 Theoretical calculation: The dimension used in this project is of prototype obtained from others experimental work, the model is designed and simulated by using the two different models shown in the table below. a [8] b[9] Table 4.1 Dimension of wells turbine The theoretical calculation of efficiency is done using the above two different dimension, the method used to calculate the efficiency is shown below. CALCULATION FOR EFFICIENCY: Similarly, CALCULATION AT 4 DEGREE ANGLE OF ATTACK: At ÃŽÂ ±= 4 degree The table below shows a calculated efficiency at different angle of attack calculated using the format shown above. ÃŽÂ ±(degree) ÃŽÂ ±(radians) W(relative velocity) à Ã¢â‚¬ °(rads/sec) Re Cl Cd ÃŽÂ · 4.00 0.07 143.36 42.45 1221641.45 0.40 0.01 21.86 5.00 0.09 114.74 42.39 977760.01 0.50 0.01 23.81 6.00 0.10 95.67 42.32 815255.46 0.60 0.01 25.18 7.00 0.12 82.06 42.24 699252.15 0.70 0.01 26.02 8.00 0.14 71.85 42.14 612312.29 0.80 0.01 26.44 9.00 0.16 63.92 42.03 544748.24 0.90 0.01 26.73 10.00 0.17 57.59 41.91 490747.45 1.00 0.01 26.78 11.00 0.19 52.41 41.77 446611.02 1.10 0.01 26.96 12.00 0.21 48.10 41.62 409873.06 1.20 0.02 26.97 13.00 0.23 44.45 41.46 378826.41 1.30 0.02 26.89 14.00 0.24 41.34 41.29 352251.70 1.40 0.02 26.90 15.00 0.26 38.64 41.10 329254.75 1.30 0.02 23.38 Table 4.2 : Efficiency at different angle Using the values of efficiency and the angle of attack from the above table (4.2) a direct relation between efficiency and the angle of attack is obtained which can be seen in the graph below (fig 4.3). Usig a Trendline option in Microsoft Excel an equation of direct relation between angle of attack and efficiency is obtained. The equation shown in the graph is a sixth order equation which is difficult to differentiate to obtained the angle at which the efficiency will be maximum,so a 2nd order equation is obtained from trendline option. Differentiating the equation gives the value of an angle at which the efficiency is max. From this procedure 12 degree is the calculated angle obtained at which the efficiency is max. Fig 4.3: Efficiency Vs. Angle of Attack y = -0.00026 + 0.00785 0.10484 + 0.70883 2.70462 + 6.5617x + 17.369 when x = 12 y = ÃŽÂ · = 25.98 % After substituting the value on angle obtained for maximum efficiency a difference between the two values is found and it is due to the R squared value. More closer the value of R square to unity more accurate results can be obtained. Equations obtained from Microsoft Excel at different orders are shown below:- Order 2 y = -0.12842 + 1.8889x + 20.336 R ² = 0.8797 Order 3 y = -0.00383 0.05482 + 1.4909x + 20.851 R ² = 0.8848 Order 4 y = -0.00524 + 0.13263 1.22762 + 5.2125x + 17.578 R ² = 0.9636 Order 5 y = -0.0015 + 0.02864 0.26813 + 0.86162 + 0.6489x + 20.649 R ² = 0.9869 Order 6 y = -0.00026 + 0.00785 0.10484 + 0.70883 2.70462 + 6.5617x + 17.369 R ² = 0.9945 Similarly using the dimension in table 4.1 (b) the calculated efficiency is show below ÃŽÂ ±(degree) ÃŽÂ ±(radians) W(relative velocity) à Ã¢â‚¬ °(rads/sec) Re Cl Cd ÃŽÂ · 4.00 0.07 143.36 78.24 916231.00 0.40 0.01 53.72 5.00 0.09 114.74 78.13 733319.93 0.50 0.01 58.49 6.00 0.10 95.67 78.00 611441.53 0.60 0.01 61.87 7.00 0.12 82.06 77.85 524439.06 0.70 0.01 63.93 8.00 0.14 71.85 77.67 459234.17 0.80 0.01 64.96 9.00 0.16 63.92 77.47 408561.14 0.90 0.01 65.68 10.00 0.17 57.59 77.24 368060.55 1.00 0.01 65.80 11.00 0.19 52.41 76.99 334958.23 1.10 0.01 66.24 12.00 0.21 48.10 76.72 307404.76 1.20 0.02 66.28 13.00 0.23 44.45 76.42 284119.78 1.30 0.02 66.07 14.00 0.24 41.34 76.10 264188.75 1.40 0.02 66.10 15.00 0.26 38.64 75.76 246941.04 1.30 0.02 57.46 TABLE 4.4: Efficiency at different angles Similarly in this case a graphical representation of Angle of Attack Vs. Efficiency is obtained which can be seen below and the equation represents a direct relation between efficiency and angle of attack. Fig 4.5 : Efficiency Vs. Angle of Attack Order 6 y = -0.00066 + 0.02925 0.62024 + 6.85633 42.0712 + 139.39x 138.43 R ² = 0.9945 when x =12 y = ÃŽÂ · = 64.24% Similarly using the order 2 equation to find the angle at which the efficiency will be maximum. The calculate angle using the same procedure as above is 12 degree at which the efficiency is maximum. . 4.2 Gambiet (Pre Processing):- The figure below shows an impeller of wells turbine designed with blades at 0 degree angle of incidence and using the dimension from the table 4.1 (a). Fig 4.6: Impeller of wells turbine Creating a model using gambiet and then meshing the geometry for which meshing size is selected based on the Reynolds number. Since the Reynolds number lies in the transational flow at the angle in which the efficiency is maximum,using turbulence boundary layer formula: à °Ã‚ Ã¢â‚¬ ºÃ¢â‚¬ ¦ =0.00269 The thicknes of boundary layer is 0.003 m. The mesh size comes to be 0.001m to get three elements in one layer to get fine meshing. In case of 3-Dimensional model the mesh elemet used is Tet/Hybrid. Checking the meshing quality the Aspect Ratio lies between 1 to 4. Boundary conditions is given for impeller is moving wall and interfaces is decided so that the fluid can be rotated within this volume. The mesh is exported for post processing in Fluent 4.3 Fluent (Post Processing) Steps used in fluent is as follows: Step 1 Opening the case file Step 2 Defining the grid interfaces Step 3 Grid check Step 4 Defining model as viscous and using K-epsilon (2 equation ) Step 5 Defining boundary condition In boundary condition fluid within the impeller is made to rotate at 40 rads/sec. The impeller is a moving wall rotating relative to cell zone at 0 rads/sec. Inlet velocity is 10 m/sec and the turbulence method selected is intensity and hydraulic cylinder. Step 6 Solution is converged after ilteraion 5 RESULTS AND DISCUSSION: The results shown below contains pressure contours, velocity vectors and pathlines at different cros-section of the models designed using the dimension from table 4.1 (a). Model with blades at 0 degree angle of incidence and inlet flow from top Model with blades at 0 and 2 degree(+) angle of incidence and inlet flow from top Model with blades at 0 and 2 degree(+) angle of incidence and inlet flow from bottom Model with blades at 2 degree(+) angle of incidence and inlet flow from top Model with blades at 2 degree(+) angle of incidence and inlet flow from bottom Biplane models Model with blades at 0 degree angle of incidece and inlet from top Model with blades at 0 degree angle of incidence and inlet flow from bottom Model with blades at 2(+)and 2(-) degree angle of incidence and inlet flow from top Model with blades at 2(+)and 2(-) degree angle of incidence and inlet flow from bottom Model with blades at 0 degree angle of incidence and inlet flow from top Model with blades at 0 and 2 degree(+) angle of incidence and inlet flow from top Model with blades at 0 and 2 degree(+) angle of incidence and inlet flow from bottom Model with blades at 2 degree(+) angle of incidence and inlet flow from top Model with blades at 2 degree(+) angle of incidence and inlet flow from bottom Biplane models Model with blades at 0 degree angle of incidece and inlet from top Model with blades at 0 degree angle of incidence and inlet flow from bottom Model with blades at 2(+)and 2(-) degree angle of incidence and inlet flow from top Model with blades at 2(+)and 2(-) degree angle of incidence and inlet flow from bottom Comparing the above graphical results under a range of 0-400 for comparison except the last two model. The table below shows the value of dynamic pressure (max) in Pascals of above design. From the table it can be seen that the introduction of two rows provides a better result in terms of dynamic pressure. After giving the installatoin angle the maximun dynamic pressure obtained is 1176 pascals by which we can say that the two rows impeller with and an installaition angle is better than the single rows . Assumptions: Various assumptions made to carry out the simulation is as follows: PATHLINES OF PARTICLES ON IMPELLER: AOA 0 INLET FROM TOP AOA 0 AND 2 DEGREE INLET AT TOP AOA 0 AND 2 DEGREE DEGREE INLET AT BOTTOM AOA 2 DEGREE INLET AT TOP AOA 2 DEGREE INLET AT BOTTOM TWO ROWS AOA 0 DEGREE INLET AT TOP AOA 0 DEGREE INLET AT BOTTOM AOA +2 -2 DEGREE INLET AT TOP AOA +2 -2 DEGREE INLET AT BOTTOM The results shown below is of the dimension used from table 4.1 (b). modelling of the wells turbine is divided into two parts theoretical and practical