Extreme load estimation of an offshore wind turbine during power production P. W. SARLI

Transcription

1 Extreme load estimation of an offshore wind turbine during power production A. YAMAGUCHI P. W. SARLI T. ISHIHARA

2 Grand Renewable Energy, Tokyo, July, 214 Background 2/1 IEC614-1 DLC 1.1 (extreme load during power production) Maximum load varies with different turbulent wind field. Maximum load for the recurrence period of 5 years is used for design. Although convergence criteria is provided, resulting 5 years maximum load has large uncertainty. ty of Excee edence The load factor 1.35 in COV C.O.V. =89% 8.9%.1 IEC614-1 assumes c.o.v..1 of the extrapolated load is 5% 1 (Tarp-Johansen et al., 22) years value Probabili Tower Base Fore-aft Bending Moment (MNm)

3 Simplified model for support structure Grand Renewable Energy, Tokyo, July, 214 Based on simulation at tower base moment. Validation and the applicability to other height ht have to be investigated. t

4 Objective Grand Renewable Energy, Tokyo, July, Propose a new convergence criteria for extreme load extrapolation during power production to reduce uncertainty of the estimation 2. Investigate the applicability of simplified model for extreme load estimation to the tower.

5 Choshi offshore test wind turbine Grand Renewable Energy, Tokyo, July, 214 From manufacturer Dimension of Nacelle Blade length, twist angle and weight First mode frequency of blade Dimension and weight of tower tower moment is measured at three different level From JSCE guideline Control algorithm Aerodynamic property of blade From Measurement Frequency and damping of the first and second mode of the tower The wind turbine is build on gravity foundation and treated as "onshore" wind turbine with wave loading) MHI 24MWWindTurbine 2.4MW Wind

6 Grand Renewable Energy, Tokyo, July, 214 Dynamic simulation was carried out by using GH Bladed to calculate the tower moment at each height. Tower mome ent (MNm) measurement simulation Measurement are bin average of the maximum load in each 1 minutes over 4 months (213/9-213/12) Simulation are average of maximum of 35 different simulations. Simulation shows good agreement with measurement To ower moment (M Nm) Tower mom ment (MNm) wind speed (m/s) wind speed (m/s) wind speed (m/s)

7 Load extrapolation of Exceedance Probability Pro obability Distrib bution fitting simulation data 18mps Tower Base Fore aft Bending Moment (MNm) 18mps Wind Speed (m/s) Grand Renewable Energy, Tokyo, July, 214 Probability of Exceedence Total Load Probability of all wind speeds Tower Base Fore aft Bending Moment (MNm) Load Probability bilit of each wind speed

8 New convergence criteria Grand Renewable Energy, Tokyo, July, 214 A new convergence criteria based on the difference between simulated load and the load based on estimated distribution function. 1 1 Probability of Exceedan nce.1.1 Probability of Exceedan nce.1.1 fitting simulation data fitting simulation data Extreme Load (MNm) Extreme Load (MNm)

9 Comparison of convergence criteria Under each convergence criteria, maximum load for 5 years recurrence period was estimated. me 5 ye ear Load ( Extre 5 (MNm) % 5.% C.O.V. 89% 8.9% 5.6%.% Conventional Proposed Proposed Conventional The Uncertainty of the extrapolated load is reduced Grand Renewable Energy, Tokyo, July, 214 9

10 Applicability of simplified extrapolation coefficient Grand Renewable Energy, Tokyo, July, 214 By using the proposed convergence criteria, the maximum load with 5 year return period is calculated for different annual mean wind speed. Extrapolation coefficient was calculated for fore-aft tower top and fore-aft tower base moment. Ext trapolation C oefficient Tower Top Ishii & Ishihara 3 (21) formula Tower Base Annual Mean Wind Speed (m/s) The simplified model by Ishii and Ishihara (21) is applicable to tower base fore-aft moment, but not to tower top.

11 Applicability of simplified extrapolation coefficient Tower moment based on thrust force on the rotor (Ishii and Ishihara[21]) Tower He eight (m) Rotor Moment 8 Thrust Force Rotor Moment Thrust Moment (MNm) The thrust force based model is applicable to tower base but not to the tower top TTFBM (M MNm) TMFBM (MNm m) TBFBM (MNm) Underestimation wind speed (m/s) 35 3 Measurement 25 Simulation Ishii and Ishihara (21) wind speed (m/s) wind speed (m/s) Grand Renewable Energy, Tokyo, July,

12 Grand Renewable Energy, Tokyo, July, 214 Conclusion 12/1 A new convergence criteria was proposed for load extrapolation. The C.O.V. of the extrapolated 5-year extreme load is reduced to 5.6% from 8.9% by using the proposed criteria. The simplified model for 5-year extreme load based on the thrust force on the rotor is applicable to tower base but not to the tower top, because the moment on the rotor is dominate factor at the tower top.

13 Grand Renewable Energy, Tokyo, July, 214 Acknowledgement 13/1 This research was a part of the research project funded by NEDO (New Energy and Industrial Technology Development Organization)