• Energy Research

Abstract The ϕ-shape Darrieus wind turbines have great potential in application due to their omni-directionality and structural advantages. However, to achieve a higher aerodynamic performance, the design of such turbine needs attentive optimization to fit the surrounding wind variation. In this paper, a performance optimization of the shape of ϕ-shape Darrieus wind turbine with a given range of inlet wind speed is carried out. By involving a heuristic search algorithm, Covariance Matrix Adaptation Evolutionary Strategy (CMAES), into Double Multiple Streamtube model (DMST), three geometrical variables of the rotor: the equatorial radius ( R ), the ratio of radius over half-height ( β ) and the blade number ( B ) are modified according to the fitness function that was specially built to satisfy the inlet wind range requirements. Moreover, to validate the optimization output, a 3D CFD simulation is conducted as a comparison. The result shows that this program can present an entirely optimized model under the given range of inlet wind speed, with a 12.5% improved C p at the optimal velocity compared with the baseline. Verification from CFD method shows a satisfactory agreement for the optimized model compared with the DMST output, indicating that this algorithm could provide a reliable reference for the shape selection of ϕ-shape Darrieus turbines under a certain inlet wind condition.

Abstract Using the random forest (RF) algorithm, this study presented a key parameter to characterize the mean wake of H-rotor VAWTs while modelling the wake. First, the RF algorithm was used to establish the regression relationship between the average wake velocity distribution and the rotor features. Next, the feature crosses method was combined with the RF algorithm to analyze the interaction and importance of the inputs. It was found that the normalized importance of a synthetic feature in wake modelling occupied a considerable significance, reaching 0.884 out of 1. The RF wake model with this parameter as the only input feature could successfully reconstruct the wake. It was found that this feature may reflect the ability of incident wind passing through the operating rotor and played a decisive role in the wake velocity distribution, including initial velocity deficit and wake recovery rate. The universality of this parameter was proved through cases analysis of wind turbines under different sizes and operating conditions. The study of the wake field is important for the modelling of the H-rotor VAWT wake field, and hence affects the optimal configuration of the wind farm.

Abstract The maximum lift-to-drag coefficient of an airfoil directly affects the aerodynamic performance of wind turbine. Machine learning methods are known for being really effective in helping to predict this parameter in a faster and more accurate way. So far, the majority of related studies have focused on the use of artificial neural networks to make this prediction, but this model has issues with its poor interpretation and the confidence level of its results was unclear. In this paper, a novel framework is proposed, involving the Gaussian process regression and a hybrid feature mining process. The aim is to use the new framework to evaluate the maximum lift-to-drag ratio of given airfoils under a turbulent flow condition, where the Reynolds number is around 100,000. The feature mining process here designed contains a hybrid feature pool that comprises various geometric characters, and a hybrid feature selector that can assist the prediction performance and make it better. Based on the airfoil dataset of the University of Illinois at Urbana-Champaign that contains a total of 1432 profiles, a comparative analysis was conducted. The results showed that the current framework can provide a more accurate estimate than parallel models in both single-point and interval aspects of view. Noticeably, the model reached an overall precision of 95.2% and 94.1% on training and testing sets, respectively. Moreover, the simplicity and the confidence reference from the model output were further illustrated with a case study, which also verified that how it can serve real engineering application.

AbstractWith the development of urbanization and the application of renewable energy, wind turbine is becoming an important approach for wind energy reservation and utilization. This study provides a numerical investigation on understanding the surface pressure distribution, flow characteristics and dynamic responses of a parked straight‐bladed vertical axis wind turbine (VAWT), which is helpful for its design. Together with the two‐way coupling method between simulation platforms such as STAR‐CCM+ and ABAQUS, the SST k‐ω turbulence model is used to obtain the surface pressure and surrounding flow of the VAWT, and the finite element method is used to obtain the dynamic responses of its structural components. The results show that the contours of the pressure distribution on the windward surface of the VAWT are similar even under a few different conditions, and the deformation of the VAWT can lead to changes in surface pressure; the turbulent flow characteristics and the wake effect become more obvious as the wind velocity increases; the blades and support arms of the VAWT need to be reinforced during the design, and the effect of the parked condition on the dynamic responses of the VAWT can be neglected. The two‐way coupling method as well as the numerical simulation results is expected to provide references for the design of VAWTs subjected to coming wind action.

Abstract Short-term wind speed forecast is of great importance to wind farm regulation and its early warning. Previous studies mainly focused on the prediction at a single location but few extended the task to 2-D wind plane. In this study, a novel deep learning model was proposed for a 2-D regional wind speed forecast, using the combination of the auto-encoder of convolutional neural network (CNN) and the long short-term memory unit (LSTM). The 12-hidden-layer deep CNN was adopted to encode the high dimensional 2-D input into the embedding vector and inversely, to decode such latent representation after it was predicted by the LSTM module based on historical data. The model performance was compared with parallel models under different criteria, including MAE, RMSE and R2, all showing stable and considerable enhancements. For instance, the overall MAE value dropped to 0.35 m/s for the current model, which is 32.7%, 28.8% and 18.9% away from the prediction results using the persistence, basic ANN and LSTM model. Moreover, comprehensive discussions were provided from both temporal and spatial views of analysis, revealing that the current model can not only offer an accurate wind speed forecast along timeline (R2 equals to 0.981), but also give a distinct estimation of the spatial wind speed distribution in 2-D wind farm.

Abstract The vertical axis wind turbine (VAWT) is regarded as an important device to utilize the renewable offshore wind energy to supplement the existing power systems. Hence, the demand for higher wind energy conversion makes the research focus on the blade optimization of wind turbines. This paper attempts to propose a novel VAWT structure with V-shaped blade to improve the power outputs at moderate tip speed ratios. The feasibility of the Reynolds-Averaged Navier-Stokes SST k - ω turbulence model applied on the VAWT was verified against available experiments at first. Then a comprehensive investigation on the aerodynamic performance of such V-shaped VAWT was carried out using the SST k- ω model. The results indicated that the maximum enhancement in power coefficient obtained in the optimal V-shaped blade was about 24.1 % . In addition to the great improvement of the power efficiency, the V-shaped blade was proven to alleviate the damage caused by lateral loads to the wind turbine. Besides, the flow structures over the blade surface were studied to reveal the mechanism of dynamic stall with the reason of power increase explained. Moreover, it was found that the V-shaped blade could effectively suppress the flow separation and delay the dynamic stall in the middle of the blade, and the undesirable blade tip effect would not be more serious comparing to that of the conventional straight blade. It was finally concluded that the current work could be practically applied to the design and optimization of the VAWT blades.