Performance Investigation and Adaptive Neuro-Fuzzy Prediction of Building Integrated Straight-Bladed Vertical Axis Wind Turbine

Abstract

This paper presents the performance investigation and adaptive neuro-fuzzy prediction of a building integrated straight-bladed vertical axis wind turbine (VAWT). An experiment was conducted with the VAWT integrated on the building rooftop. The coefficient of power of the VAWT was predicted using adaptive neuro-fuzzy inference system (ANFIS). The input variables for the model development include the rotational speed, angular velocity, and tip speed ratio, while coefficient of power is the output. In the fuzzy logic of the fuzzy inference system (FIS), the parameter of the membership function is adjusted by the neural network in ANFIS. MATLAB/Simulink was used to implement this intelligent algorithm and the performance was investigated using root mean square error (RMSE) and coefficient of determinant (R2). In addition, the ANFIS technique precision was evaluated against the results of the experiment. The result obtained indicates that the maximum coefficient of power (Cpmax) was obtained at about Y = 250 mm above the building rooftop. Furthermore, it was also established that the developed ANFIS model is very effective and reliable in predicting the performance of building integrated straight-bladed VAWT.