Enhancing Generalizability of Predictive Models with Synergy of Data and Physics

TL;DR

This paper proposes a method that combines physics-based knowledge with machine learning to improve the generalizability and accuracy of predictive maintenance models for wind turbines, especially in non-observable parameter prediction.

Contribution

It introduces a synergy approach integrating physics principles with machine learning, enhancing model transferability across different turbines in wind energy applications.

Findings

Significant accuracy improvement in wind turbine icing prediction.

Enhanced model generalizability across different turbines.

Validation of physics-ML integration as effective in industrial settings.

Abstract

Wind farm needs prediction models for predictive maintenance. There is a need to predict values of non-observable parameters beyond ranges reflected in available data. A prediction model developed for one machine many not perform well in another similar machine. This is usually due to lack of generalizability of data-driven models. To increase generalizability of predictive models, this research integrates the data mining with first-principle knowledge. Physics-based principles are combined with machine learning algorithms through feature engineering, strong rules and divide-and-conquer. The proposed synergy concept is illustrated with the wind turbine blade icing prediction and achieves significant prediction accuracy across different turbines. The proposed process is widely accepted by wind energy predictive maintenance practitioners because of its simplicity and efficiency. Furthermore, this paper demonstrates the importance of embedding physical principles within the machine learning process, and also highlight an important point that the need for more complex machine learning algorithms in industrial big data mining is often much less than it is in other applications, making it essential to incorporate physics and follow Less is More philosophy.