Short-term wind speed forecast for Urla wind power plant: A hybrid approach that couples weather research and forecasting model, weather patterns and SCADA data with comprehensive data preprocessing

Abstract

Short-term wind speed forecast model that uses both supervisory control and data acquisition (SCADA) based data and weather research and forecasting (WRF) model outputs for Urla wind power plant (WPP) has been proposed in this study. Two different WRF models were run to gather atmospheric variables from four surrounding grids of Urla WPP and calculate weather patterns affecting Urla WPP. After detecting outliers in the SCADA data by coupling of k-mean and isolation forest (IF) methods, statistical methods were used for data treatment and the outputs of WRF models were used for missing data imputation. The effect of each data type and data preprocessing techniques on the model was evaluated separately. The best model performance was achieved with 0.9085 [Formula: see text], and 0.81 MAE in the dataset which includes each data type and each data preprocessing was applied on. Otherwise, the dominant weather pattern affecting Urla WPP was found to be purely advective and the best result was achieved in this pattern.

Review of AI-Based Wind Prediction within Recent Three Years: 2021–2023

“…Ammar et al [47] detected outliers using interquartile range and then corrected them using the last observation carried forward method. Özen et al [48] coupled k-means with the isolation forest method to detect outliers in SCADA data and used the output of the WRF model to impute the missing data.…”