Multi-turbine Associative Model for Wind Turbine Performance Monitoring

Abstract

Comparing the performance parameters of a set of wind turbines in a single region will provide insights that prevent raising unnecessary alarm, while confirming actual faults. Wind turbines operating in a wind farm experience similar operating and environmental conditions that could indicate either normality for that group or failures that manifest in those conditions. These norms and failures are an orthogonal set of data rich in information that can be utilized in performance monitoring algorithms to supply better prediction accuracy and low false positives. In this paper, we describe the use of an associative model (AM) for fault detection in a population. An associative model maps system parameters to an identical set of virtual parameters. The AM-based approach can be used to capture the underlying correlation of an observable system, such as performance parameters of a set of wind turbines in a wind farm. The residuals between the model output and the input can then be used to detect anomalies and isolate faults.