Experimental study on the effect of drop size in rain erosion test and on lifetime prediction of wind turbine blades

Abstract

Rain erosion of turbine blades causes loss of production and expensive repairs in the wind energy sector. There is a common consensus, that the size of impacting rain drops has a governing effect on the added damage. However, the literature lacks systematic experimental studies on the topic. In the present paper the effects of drop sizes in Rain Erosion Tests (RET) are studied for a commercial polyurethane based top coat applied to glass fiber-epoxy specimens. The tests are conducted applying a whirling arm RET at impact velocities ranging from 90 to 150 m/s and with four different rain field setups generating mean droplet diameters of 0.76 mm, 1.90 mm, 2.38 mm and 3.50 mm respectively. The time to damage at the end of incubation is determined by inspection of photographs captured inline at regular intervals through the test.Data sets of time to damage as function of local rotor velocity are extracted for each of the four different rain fields. From this data VQ curves (V for Velocity, Q for Quantity) are presented for different representations of quantity, namely specific impacts (DNV GL), specific impacts (ASTM) and accumulated impingement. The slope of the velocity-impingement (VH) curves vary with drop size.We propose a drop size dependent empirical model for impingement (H) to damage as function of impact velocity (v), H (v) = cv−m, where the scale parameter m and the shift parameter c are functions of the drop size.The drop size-dependent impingement model is then applied for computing the expected leading edge lifetime of a virtual 15 MW IEA reference turbine at 18 different meteorological stations in Northern Europe based on 10-min time series of rain intensity and wind speeds observed during several years. The drop size dependent model predicted on average 2.35 times longer lifetime compared to models based on the standard 2.38 mm drop size. Comparing the sites, the model shows a factor 3 variation of added damage per meter of accumulated rain between sites because of differences in concurrent wind during the rain events.