Mitsumasa Iino

Examination of simplified load model on fatigue analysis of small wind turbine blade using aeroelastic analysis

In this research, probabilistic fatigue load estimation by the aeroelastic modeling was carried out for a small wind turbine with 1m diameter. The results were compared with the simplified load model which is a deterministic method, and quantitative evaluation of the difference between two methods was presented. An aeroelastic simulation code for 5 bladed wind turbines is used for aeroelastic modeling. This code is based on NRELs FAST and modified for 5 bladed turbines by authors. On the fatigue analysis, wind condition is set to be IEC 61400-2 Ed.2 DLC1.1 which is assumed to be equivalent condition of simplified load model. From the time series results from aeroelastic analysis, the load amplitude and cycles are counted. Then the cycles are weighted by wind speed probability distribution correspond to IEC 61400-2 Class I to IV. The result from aeroelastic modelling was similar to one from simplified load model in Class I but different in other Classes. The reason is that the fatigue load in high wind speed, which is not encompassed with simplified load model, is dominant for fatigue load with target turbine’s model regardless of IEC Class. Therefore it is important to consider probability of wind speed over all operating condition in fatigue load estimation.

J056034 Probabilistic Fatigue Analysis of Small Wind Turbine Blade Using Aeroelastic Analysis

In this research, we aim to reveal, 1. The difference of fatigue load from aeroelastic model and SLM 2. The reasons of difference between aeroelastic model and SLM 3. The applicability of the idea that operation around Vdesign can represent the lifetime fatigue load which is the background of SLM For fatigue load design of small wind turbines, IEC61400-2 Ed.2[1] provides two fatigue load calculation methods; Simplified load model(SLM) and aeroelastic modeling. Although aeroelastic modeling includes more realistic effect, most of small wind turbines below 1kW in Japan are designed using SLM because of the complicated procedure of aeroelastic modeling. So, we aims to reveal how SLM is applicable for these “micro wind turbines” and how these turbines designed for SLM are safety or danger sided or how much the cost of energy increase or decrease using SLM instead of aeroelastic model. Moreover, we wish to establish more practical and theoretical load design method affordable for micro wind turbine manufacturers. In this research, fatigue load design is focused. Fatigue load estimation by the aeroelastic modeling was carried out for AURA1000 135W small wind turbine with 1m diameter In this research. The results were compared with the simplified load model (SLM). And the reasons of difference between two methods are discussed. An aeroelastic simulation code for 5 bladed wind turbines based on NREL’s FAST[2] is used for aeroelastic modeling. Four annual wind speed distributions are considered; Class I to Class IV defined in IEC61400-2 Ed.2. The result of fatigue load showed that SLM is more sensitive to annual average wind speed than the aeroelastic modeling. And the difference indicated that SLM’s assumption cannot be applied for the researched turbine In addition, the idea of representative operational range around Vdesign which is used in SLM is applied for the result of Aeroelastic model. It is revealed that wider range than the assumption of SLM in most cases. Especially in Class IV, the ranges were unrealistic value.

Proposal of Grid-Connected Small Wind Turbine Simulator Based on Aeroelastic Dynamic Modeling

Under a current system, small wind turbines grid-connected are not evaluated efficiently. It leads to problems such as price increase. We developed an experimental simulator for small wind turbine with grid-connection for the evaluation. The requirements of this system are to simulate torque and electric power in a laboratory according to various wind conditions. In this study we use NREL’s FAST, which calculates wind power based on BEM (Blade Element and Momentum theory) for controlling the simulator. For the simulation, an AC motor was connected to a generator of a real wind turbine. The same torque as wind calculated by FAST was given on the connecting shaft by the motor. The simulator enabled the generator to produce electric power like the wind turbine in the field. The results by this simulator well agreed with results only by FAST at constant wind speed. When the wind speed changes we must consider each moment of inertia of each experiment part such as the connecting shaft. We estimated the momentum of inertia of this system and considered the influence on testing results such as rotational speeds. Moreover by comparing the tests in the field the accuracy of this simulation is verified.Copyright