Data-based optimal estimation of frequency bias: The case of Southwest Power Pool

TL;DR

This paper presents a data-driven method for optimally estimating time-varying frequency bias in power systems, aiming to improve regulation cost efficiency and support demand response strategies.

Contribution

It introduces a novel online parameter estimation approach for dynamic frequency bias, enhancing accuracy over traditional static assumptions and integrating physics-based models.

Findings

Improved frequency bias estimates reduce regulation costs.

Method enables better intra-dispatch load deviation estimation.

Supports demand response participation in regulation.

Abstract

In this paper, we introduce a method to optimally estimate time-varying frequency bias. Current industry practice is to assume that frequency bias is changing only on annual basis. We suggest that this improved time-dependent bias estimate can be used to reduce the cost of frequency regulation needed to meet industry standards requested by the North American Electric Reliability Corporation (NERC). Optimization of time-varying frequency bias is posed as a parameter estimation (calibration) problem whose implementation utilizes online system measurements. It is further shown how this result can be used to estimate intra-dispatch load deviations. This knowledge is needed to estimate more accurately regulation reserve needed, and to therefore reduce overall regulation cost. Methods can be introduced to give incentives to demand response to participate in frequency regulation. Overall, we show the importance of incorporating knowledge of physics-based models for data-enabled parameter estimation of physical systems.