Home // IARIA Congress 2024, The 2024 IARIA Annual Congress on Frontiers in Science, Technology, Services, and Applications // View article
Authors:
Swayam Mittal
Vishwaas Narasinh
Nikhil Kulkarni
Remish Minz
Nilanjan Chakravortty
Prateek Mital
Keywords: wind farm, anomaly detection, power prediction, machine learning, clustering, physics-informed learning.
Abstract:
In this paper, we present a robust power prediction model for wind turbines. Our model leverages error detection in the sensor data, clustering-based imputation of filtered erroneous or missing data, and a Physics-Informed Neural Network (PINN). We introduce data preprocessing steps, including the detection and filtering of erroneous data and clustering-based data imputation. We demonstrate that these preprocessing steps, along with the PINN framework, improve power prediction accuracy in the presence of erroneous sensor data.
Pages: 41 to 48
Copyright: Copyright (c) IARIA, 2024
Publication date: June 30, 2024
Published in: conference
ISBN: 978-1-68558-180-0
Location: Porto, Portugal
Dates: from June 30, 2024 to July 4, 2024