Optimal design of filament wound composite tower for 2MW horizontal axis wind turbine
- (0) Download
https://www.eduzhai.net International Journal of Composite M aterials 2013, 3(1): 15-23 DOI: 10.5923/j.cmaterials.20130301.03 A Study on Optimal Design of Filament Winding Composite Tower for 2 MW Class Horizontal Axis Wind Turbine Systems Sungjin Lim1, Changduk Kong1,*, Huynbum Park2 1Dept. of Aerospace Engineering Chosun University #375 Seosuk-dong, Dong- gu, Kwangju , 501-759, Republic of Korea 2Dept. of Defense & Science Technology, Howon University 573-718, 64 Howondae 3gil, Impi, Gunsan, Rep. of Korea Abstract A lthough the tower of the horizontal axis wind turbine system is a simp le structure in co mparison with the wind turbine rotor system, the production cost of the tower is about 20-25% of the whole wind turbine system cost. If a composite materials tower is used instead of the existing steel tower, the production cost can be reduced by using of low cost composite materials, simp le manufacturing process, easy transportation and easy assembly. Ho wever studies on the composite materia ls towers are very fe w. In this study a specific structural design procedure for 2 MW c lass glass/polyester face sheets -sand/ polyes ter core sandwich compos ite wind turbine s ystem towers is newly proposed through load cas e study, trade-off study, optima l structural design and structural analysis. Optima l tower design can minimize both weight and cost. In the structural design of the tower, three kinds of loads such as wind load, b lades, nacelle and tower weight and blade aerodynamic d rag load should be considered. Initial structural design is carried out using the netting rule and the rule of mixtu re. Then the structural safety and stability are confirmed using a commercial fin ite element code, MSC NASTRAN/PATRAN. It is confirmed that the final proposed tower meets the tower design requirements. Keywords Co mposite Materials Wind Turb ine Tower, Filament Winding, Optimal Design 1. Introduction Mankind has used the wind energy at various types of fields for several thousands of years. Even though the wind power generation has a long history and economic renewal energy technology, its past technology was a bit lower effectiveness due to relying entirely on natural wind .However recently the wind power generation system has been improved by the concentrated investments and research works on advanced technologies based on past experiences and technologies due to fossil fuel exhaustion, and become large scale wind turbines more than several MW class. Therefore both the turbine blade and the tower of the large scale wind turbine systems has become large scales. In case of the blade the composite materials are widely used to reduce the weight, but the case of the tower the application examp les of the composite materials are very few. Even though the tower is a simple structure relatively to other mechanica l structures, development of a cost effective tower is important because the tower has 20-25% of the * Corresponding author: firstname.lastname@example.org (Changduk Kong) Published online at https://www.eduzhai.net Copyright © 2013 Scientific & Academic Publishing. All Rights Reserved whole wind turbine system cost. The co mposite materials tower can reduce the production cost through use of low cost composite materials, simp le manufacturing process, easy transportation and assembly relatively to the steel tower. According to literature survey results on existing towers, Tongguang Wang studied on a high resolution tower shadow model for downwind wind turbine, P. J. Murtagh studied on an along-wind response of a wind turbine tower with blade coupling subjected to rotationally sampled wind loading, Dimos J. Po lyzois studied on the static and dynamic characteristics of mu lt i-cell jointed GFRP wind turbine towers, Jane E. Lundberg studied on load and resistance factor design of composite columns. In Korea, Hyeok-su Hong, studied on optimal design of a steel tower shell thickness of 2MW class wind turbine system through the numerical analysis of natural frequency, strength, fatigue and buckling depending on the shell thickness change and Dong-man Kim, studied on vibration characteristics of steel tower for a the large wind power turbine system considering wind loads and seismic loads. As mentioned the literature surveys on the composite materials towers are very few while studies on the steel tower are many. This work proposes a design procedure and optimal design results of the co mposite materials tower for the large 16 Sungjin Lim et al. : A Study on Optimal Design of Filament Winding Composite Tower for 2 M W Class Horizontal Axis Wind Turbine Systems scale wind turb ine systems. Authors express thanks for Fiber Tech Co., Ltd.'s support that provided specimen experimental test data and manufacturing cost analysis data for this work. Figure 1 shows the proposed design procedure for this work. existing 2 MW class wind turbine systems, most towers have about 80m height. However this work adopts the 100m height of tower due to considering the potentiality of application of 3-4 MW class wind turbine system tower as well as 2 MW. For easy transportation the tower is divided into 4 sections such as the top section of a constant diameter of 3m, the second section with a constant diameter of 3.4m, the third section with a constant diameter of 3.7m, the bottom section with a constant diameter of 4m. The joint parts between each sections have a truncated shape. Figere 1. Design procedure of wind turbine system tower 2.2. Load Case Analysis The wind load cases applied to the tower are classified in 3 cases such as Case 1 of the rated wind speed, Case 2 of the cut-off wind speed and Case 3 of the storm wind speed. In addition to the wind loading, the thrust load induced by the blade, the weights of nacelle, rotor blades and tower must be considered as design loads. Among them, generally the wind turbine blade has the design load at the cut-off wind speed condition with the gust. However, the tower has to endure the storm wind load as well as the cut-off wind load. Tab le 2 shows 3 load cases for designing the tower. Table 2. Load cases for tower design 2. Tower Design In the tower design, firstly the design loads are defined through the load case analysis, secondly the filament winding method is adopted for manufacturing the tower, thirdly the optimal designs on both the thickness and the lay-up angles of each tower section is performed, and finally the designed tower structure's safety is evaluated using the FE analysis. 2.1. Structural Design Requirements The tower is trad itionally made fro m steel using steel considering the design requirements, the existing structural shapes, the manufacturing method, and then the designed steel tower is redesigned using composite materia l. Table 1. Tower system specification Type Stepped shell Target 2 MW Height Diamet er Section 1 (top) 3.0 m 100m Section 2 Section 3 3.4 m 3.7 m Section 4 (bottom) 4.0 m Tower material Glass/Epoxy composite Table 1 shows the tower system specification to be designed in this work. According to the literature survey of Load case Wind speed Case I 8.0 m/s Case II 20.0m/s Case III 55.0 m/s Among 3 load cases, the load case 3 is considered as the design load case through calculating the loads applied on each tower section using the following experimental for mula . In the storm load condition, the drag load, which can g ive the bending moment to the tower, occurs on the blade. This load can be estimated by the following experimental formula (1). Pa = 0.4V 2 Fa = Pa ⋅ S = 0.4V 2S (1) where S = πd 2 / 4 is the swept area produced by rotation of the blade. Pa, Fa are means to wind load and drag load. In addition to the blade's drag, the dead weights of nacelle including generator and gear bo x and blades are applied to the top of tower. In this work, the weights of blades of 45 tons and nacelle of 180 tons including gear, shaft, generator, etc. which have been estimated in the previous study, are used. Figure 2 shows various loads applied on the tower. The wind load on the tower increases gradually along with the height. This load can be estimated using the following formula (2) defined by the regulation regarding Japanese construction standard. P =W ⋅A (2) International Journal of Composite M aterials 2013, 3(1): 15-23 17 Figure 2. Various loads applied on tower Where A is the pro jection area with a rectangular shape of tower, i.e., the wind drag area, and W is the wind pressure expressed by the formula (3). W = q⋅Cf (3) Where q is the dynamic pressure [N / m2 ] , C f is the wind coefficient. For this work C f is 0.9, 0.84, 0.78, and 0.60 for section 1, 2, 3 and 4 of the tower, respectively. q = 0.6EV02 (4) Where E is the coefficient defined by the regulat ion and 4.3, 4.0, 3.7 and 3.4 fo r section 1, 2, 3 and 4 o f the tower, and V0 is the tower design wind speed of 55m/s. Using the calculated loads above the bending mo ments can be obtained, then stresses, deflections and thicknesses of the tower using formula (5). σv = σ 2 d + 3τ 2 xy σd = (M y cosθ )2 + (M z sinθ )2 + Fx I ⋅2/d A τ xy = Fy A + M x sinθ J ⋅2/d 2 + Fz A + M x cosθ J ⋅2/d 2 (5) I ≈ πd 3t , J ≈ πd 3t , A ≈ πdt 8 4 σv ∝ 1 t Where σ v ,σ d and τ xy are equivalent, bending and shear stresses, respectively, d is the tower diameter, I is the second area moment of inert ia, J is the polar mo ment of inertia, Fx , Fy and Fz are x, y and z directional fo rces on each section, M x , M y and M z are x, y and z directional mo ments on each section, A is the tower projection area, and θ is the wind inclination angle. As shown at the formula (5), the equivalent stress is inversely proportioned to the thickness t of tower shell. Therefore the tower thickness, which satisfies the safety factor of 1 against the yield strength, can be obtained using the formula. σ 0 v t0 t1 =σy t1 = t0 σ 0 v (6) σy where σ 0 v is the equivalent stress obtained at the initial design, t 0 and t1 are the in itially designed thickness and the modified thickness considering the safety. The formula (7) shows deformations of the tower. δt = δ Fa + δQ (7) where δ Fa = Fa L3 3EI is the tower deformation due to the blade thrust load F, L is the tower height, δQ = QL4 8EI is the tower deformation due to the distributed wind load Q . Table 3 shows the estimated wind loads on each tower section, and Table 4 shows the designed results of tower using steel. The tower weight is calcu lated using the commercial FE code, MSC.NASTRAN. The weight and the thickness of the designed tower are similar to the existing same class wind turbine tower. Figure 3 shows the designed tower configuration using steel. The tower is divided into 4 sections such as the top section of a constant diameter of 3m, the second section with a constant diameter of 3.4m, the third section with a constant diameter of 3.7m, the bottom section with a constant diameter o f 4m. The joint parts between sections have a truncated shape. The three truncated joint part sections have all the same heights of 2 m, and their upper and lower ends are met to the tower sections’ diameters. The details of the second joint part are shown at Figure 3. Table 3. Estimated wind load on each tower section Sect ion 1(top) 2 3 4 W[N/m2] 3812.46 5361.53 6305.59 7074.59 A[m2] 100 185 255 300 P[N] 381246 991883 1607925 2122200 Table 4. Tower design result using steel Section Diameter[mm] Height[mm] Thickness 1(top) 3,000 25,000 25 Mass[ton] 2 3,400 25,000 35 480 3 3,700 25,000 45 4 4,000 25,000 55 18 Sungjin Lim et al. : A Study on Optimal Design of Filament Winding Composite Tower for 2 M W Class Horizontal Axis Wind Turbine Systems Figure 3. Designed tower configuration using steel Table 5 shows technical data of the existing 2 MW class wind turbine towers,. The reason why the weight of the designed 100m tower using steel is much lighter than that of the existing similar 80m class towers is caused by the use of its each section’s different thickness and no consideration of internal ladder and platform while the existing tapered tower has the same thickness at entire walls based on the thickest thickness at bottom wall. Table 5. Technical data of existing 2MW towers using steel HAWT Rated power Height [mm] Diamet er[mm] (top/bottom) Mass[ton] MM92 2 MW 80,000 2,955/4,300 620 PMSG KBP2000M U88 2 MW 2 MW 76,900 3,000/4,200 730 80,000 3,000/4,200 740 2.3. Redesign of the Steel Tower Using Composite Materials The composite materials tower is manufactured by the filament winding method using glass/epoxy materials. The filament winding method is generally used to manufacture the shell type structures such as pressure vessel, pipe, etc.. This method is able not only to reduce manpower, manufacturing time and cost due to the manufacturing automation, but also to do mass production of large scale structures. Moreover in order to lower than manufacturing cost, the glass/epoxy face sheets-sand/epoxy core sandwich type structure is adopted. Figure 4 shows the sectional view of the sandwich type structure of the tower. The joint ring parts between sections have a smooth truncated shape. Assembly between the joint ring part and the tower section is done by the flanges with bolts co-wound on the internal tower walls. Figure 5. Conceptual design configuration of flange 2.4. Trade-off Design for Design Opti mization In this work the optimal winding pattern and thickness is determined considering proper safety factor, deflection at the tower top and weight through trade-off design. Table 7 shows the mechanical properties of used composite materials, and Table 8 shows the procedure how to determine the winding pattern through the stress analysis using FE. Among the various lay-up sequences, 0 / 90 the lay-up case has the lowest stress and displacement. Table 6 shows the stress analysis results of various lay-up sequences. Table 6. Structural analysis results accordingto various lay-up sequences Lay-up sequence Compressive st ress(max) Tensile st ress(max) [0/90]50S 30.6[MPa] 14.2[MPa] [45/90]50S [0/90/45]50S 36.0[MPa] 42.7[MPa] 17.0[MPa] 20.5[MPa] Analysis result Figure 4. Glass/epoxy face sheets-sand/epoxy core sandwich type structure of tower Even though the best lay-up sequence case is 0 / 90 , the minimu m possible winding by the manufacture Fiber Tec Co. is 35 instead of 0 . Therefore the practical winding angle becomes 35 / 90 instead of 0 / 90 . Table 8 shows maximu m co mpressive and tensile stresses and tower top deflections depending on the lay-up ratio of 35 / 90 . The Case 3 has the lowest compressive stress, while it has the higher tensile stress and deflection than the Case 6. Therefore the Case 6 is determined as a final best lay-up International Journal of Composite M aterials 2013, 3(1): 15-23 19 sequence. Figure 6 shows the trend of maximu m compressive and tensile stresses and tower top deflections depending on the lay-up ratio of 35 / 90 . Table 7. Mechanical propert ies of used materials Mat erial Prop ert y E1[N/mm2] E2[N/mm2] G12[N/mm2] ν Xt[N/mm2] Xc[N/mm2] Yt[N/mm2] Yc[N/mm2] Density Ply thickness[mm] Glass/Polyester UD 17000 500 6500 0.3 233 128 215 108 2000E-9 kg/mm3 0.38 Polyester Sand 5600 5300 1930 0.37 96.5 54.6 83.8 53.8 1700E-9 kg/mm3 Table 8. St ruct ural analysis result s depending on various lay-up rat io Case Lay-up ratio(±35 to 90)〫 Compressive st ress(max) [MP a] Tensile st ress(max) [MP a] Deflect ion (max) [mm] 1 10% 90% 568 2 20% 80% 307 3 30% 70% 219 4 40% 60% 269 5 50% 50% 355 6 60% 40% 230 7 70% 30% 469 8 80% 20% 435 631 6.71×104 193 3.98×104 143 2.93×104 248 2.44×104 320 2.13×104 89 1.90×104 365 1.88×104 323 1.83×104 Figure 6. Max stresses (tensile and compressive) and tower top deflection versus lay-up ratio (w/o core) 20 Sungjin Lim et al. : A Study on Optimal Design of Filament Winding Composite Tower for 2 M W Class Horizontal Axis Wind Turbine Systems Fi gure 7. Loading and geomet rical boundary condit ions of tower for FE analysis Table 9. Thickness design cases of 4 sect ions Section 1st section 2nd section 3rd section 4th section Case thickness[mm] thickness[mm] thickness[mm] thickness[mm] 1 28.08 37.44 46.8 59.28 2 31.2 43.68 53.04 62.4 3 37.44 46.8 59.28 68.87 4 62.4 74.88 84.24 93.6 5 84.24 90.48 93.6 106.8 6 84.24 93.6 99.84 109.2 7 84.24 93.6 102.96 110.6 8 84.24 93.6 106.08 112.4 9 84.24 93.6 106.08 113.8 10 84.24 93.6 106.08 115.44 Table 10. Stress analysis results including maximum stresses and top deflection of 10 design cases St ress Case Compressive st ress(max) Tensile st ress(max) Deflect ion (max ) 1 5 80 [MP a] 5 16 [MP a] 67100[mm] 2 5 73 [MP a] 4 50 [MP a] 41100[mm] Figure 8. stress analysis results including stress distribution and deformed shape of the tower at Case 6 lay-up sequence Figure 7 and 8 show the loading and geo metrical boundary conditions and the stress analysis results including stress distribution and deformed shape of the tower, respectively. According to the stress analysis results of Fig. 8, the tower top deflection is very large. Therefore by adapting the sandwich structure with sand/epoxy core and increasing its thickness, the top deflection is reduced and the structural stability enhanced. The thickness optimization is performed by gradual increasing through checking both the structural stability and the deflection. The lay -up thickness allocation having the symmetric[±35], and[core] sequence at each section has 30%, 20% and 50%, respectively. Table 9 shows 10 design cases for various thickness at 4 tower sections, Table 10 and Fig. 9 shows the stress analysis results including maximu m stresses and top deflection of 10 design cases. Among all 10 design cases, the case 10 is selected as a final design result that satisfies both the structural safety and stability and the top deflection. The final design results are shown at Tab le 11. The structural stability means the structural buckling safety due to compressive loads occurred by the tower bending moment or the tower dead weight including weights of blades, nacelles and power generation system. 3 4 80 [MP a] 3 97 [MP a] 32100[mm] 4 3 50 [MP a] 3 00 [MP a] 11563[mm] 5 2 40 [MP a] 2 22 [MP a] 10023[mm] 6 2 27 [MP a] 1 20 [MP a] 9412[mm] 7 1 30 [MP a] 1 15 [MP a] 6555[mm] 8 1 19 [MP a] 1 03 [MP a] 3526[mm] 9 1 06 [MP a] 71[MPa] 1860[mm] 10 88.8[MPa] 57.6[MPa] 1530[mm] Table 11. Design result of composite tower using sandwich structure Section Diameter[mm] Height[mm] Thickness Mass[ton] 1(top) 3,000 25,000 84.24 [(±35)32/(90)22/core]S 2 3,400 3 3,700 25,000 93.6 [(±35)36/(90)24/core]S 270 25,000 106.08 [(±35)40/(90)27/core]S 4 4,000 25,000 115.44 [(±35)43/(90)29/core]S International Journal of Composite M aterials 2013, 3(1): 15-23 21 Figure 9. Max stresses (tensile and compressive) and top deflection versus thickness(w core) The weight of the final proposed 100m height composite tower is 270ton. It is much lighter than the existing 80m height steel tower having 450~700ton. Figure 10 shows spanwise stress distribution on outer skin and deformation of the proposed composite tower. Figure 10. Spanwise stress distribution on outer skin and deformation of proposed composite tower 22 Sungjin Lim et al. : A Study on Optimal Design of Filament Winding Composite Tower for 2 M W Class Horizontal Axis Wind Turbine Systems Figure 11 shows the buckling mode shapes and the buckling load factors, and Fig. 12 the Campbell d iagram to check the resonance possibility at the constant operating rotational speed of 20rp m. Figure 11. Tower buckling mode shapes and its load factors Figure 12. Campbell diagram of proposed tower In order to investigate the cost effectiveness of the proposed composite tower, its manufacturing cost is compared with the e xisting steel tower's manufacturing cost. Table 12 shows the comparison result between the manufacturing costs of the proposed composite tower and the existing steel tower. This manufacturing cost data are provided by a wind turbine system tower manufacturing company, FIBERTEC Co. Ltd. The cost contains materials price, labor cost and other costs. Table 12. Comparison of manufacturing cost between composite and steeltowers (provided by FIBRETECH. Co. ltd.) Mat erial Composite St eel Manufacturing Cost per each (USD) 237,000 USD 317,000USD 3. Conclusions This work proposes a composite materials tower for 2 MW class horizontal wind turbine system using the filament winding manufacturing process. The tower is designed by the following methods; Firstly, the glass/epoxy face sheets-sand/epoxy core sandwich type structure is adopted considering the manufacturing cost, and the structure is optimized through minimizing the weight together with structural stability and s afety . Secondly, the base design winding angle has the min imu m manufacturing capable angle of ±35° instead of the theoretical optimal angle of 0° having the best strength and the combination angle of 90° having the secondary strength. The thicknesses of each tower section are optimized through reducing stress, deformat ion and weight by the thickness variation. Thirdly, it is confirmed that the final proposed composite materials tower is safe through investigation of stresses, deformations, structural stability and natural frequencies, and much lighter than the existing same c lass steel towers. Finally, the proposed tower will be validated by the International Journal of Composite M aterials 2013, 3(1): 15-23 23 structural tests of the prototypes to be manufactured. ACKNOWLEDGMENTS and Industrial Aerodynamics 89, 2001, pp. 873-892  P.J.M urtagh, "Along-wind response of a wind turbine tower with blade coupling subjected to rotationally sampled wind loading", Engineer ing Structures 27, 2005, pp. 1209-1219 Authors express full acknowledg ment for ASM E’s  Dimos J. Polyzois. "Static and dunamic characteristics of publication permission of this material as orig inal publisher ASME Turbo Expo 2013, Paper Nu mber GT2013-94124 in multi-cell jointed GFRP wind turbine towers", Composite Structures, 2009, pp. 34-42 International Journal of Co mposite Materials.  Jane E.Lundberg, Theodore V.Galambos "Load and resistance factor design of composite columns" Structural Safety, Vol.18, No.2/3, pp. 169-177. REFERENCES  C. Kong, J. Bang, Y. Sugiyama, "Structural investigation of composite wind turbine blade considering various load case and fatigue life" Journal of Energy, Vol.30, 2005, pp. 2101-2114.  Desire Le Gourieres "Wind power plants theory and design" Pergamon Press, UK, 1998.  Tongguang Wang. "A high resolution tower shadow model for downwind wind turbine", Journal of Wind Engineering  Hyeoksoo Hong. “Research for 2M W Wind Turbine Tower Shell Design Optimization", The Korean Society For New and Renewable Energy, 12.Vol.2, No.4 2006, pp. 19-26  Kim Dongman. " Vibration Analysis of M W Class Wind Turbine Tower Considering Earthquake Base Excitation", Korea Wind Energy Association(KWEA), 2009  Spera. D.A "Wind turbine technology", ASM E Press, 1994, pp .139-194  Hani M . Negm. " Structural design optimization of wind turbine towers", Computers and Structures 74, 2000, pp. 649-666
... pages left unread,continue reading
Free reading is over, click to pay to read the rest ... pages
0 dollars，0 people have bought.
Reading is over. You can download the document and read it offline
0people have downloaded it