Inclusion of frequency nadir constraint in the unit commitment problem of small power systems using machine learning

M. Rajabdorri, B. Kazemtabrizi, M. Troffaes, L. Sigrist, E. Lobato

Sustainable Energy, Grids and Networks Vol. 36, pp. 101161-1 - 101161-10

Summary:

As the intention is to reduce the amount of thermal generation and to increase the share of clean energy, power systems are increasingly becoming susceptible to frequency instability after outages due to reduced levels of inertia. To address this issue frequency constraints are being included in the scheduling process, which ensure a tolerable frequency deviation in case of any contingencies. In this paper, a method is proposed to integrate the non-linear frequency nadir constraint into the unit commitment problem, using machine learning. First, a synthetic training dataset is generated. Then two of the available classic machine learning methods, namely logistic regression and support vector machine, are proposed to predict the frequency nadir. To be able to compare the machine learning methods to traditional frequency constrained unit commitment approaches, simulations on the power system of La Palma island are carried out for both proposed methods as well as an analytical linearized formulation of the frequency nadir. Our results show that the unit commitment problem with a machine learning based frequency nadir constraint is solved considerably faster than with the analytical formulation, while still achieving an acceptable frequency response quality after outages.

Spanish layman's summary:

En este artículo se presenta un método para integrar la restricción no lineal de minima frecuencia en un problema de despacho economico. Primero, se genera un conjunto de datos de entrenamiento sintético para posteriormente estimar el valor de minima frecuencia mediante logistic regression y support vector machine.

English layman's summary:

In this paper, a method is proposed to integrate the non-linear frequency nadir constraint into the unit commitment problem, using machine learning. First, a synthetic training dataset is generated. Then, logistic regression and support vector machine, are proposed to predict the frequency nadir.

Keywords: Data-driven method; Mixed integer linear programming; Frequency constrained unit commitment; Machine learning

JCR Impact Factor and WoS quartile: 4,800 - Q1 (2023)

DOI reference: https://doi.org/10.1016/j.segan.2023.101161

Published on paper: December 2023.

Published on-line: September 2023.

Citation:
M. Rajabdorri, B. Kazemtabrizi, M. Troffaes, L. Sigrist, E. Lobato, Inclusion of frequency nadir constraint in the unit commitment problem of small power systems using machine learning. Sustainable Energy, Grids and Networks. Vol. 36, pp. 101161-1 - 101161-10, December 2023. [Online: September 2023]