Key words: machine learning; explainable; grid frequency; stability

Download 15,56 Mb. Pdf ko'rish
bet	312/693
Sana	13.05.2024
Hajmi	15,56 Mb.
	#228860

1 ... 308 309 310 311 312 313 314 315 ... 693

Bog'liq
Тўплам

Namangan Institute of Engineering and Technology nammti.uz

Key words: machine learning; explainable; grid frequency; stability

Introduction. The functionality of a power system depends on the supply and demand for
electricity being balanced [1]. Significant power disparities put the supply security at risk, which in
severe circumstances may result in widespread blackouts [2]. As though Electricity markets are
utilised to balance supply and demand, usually every hour; hence, extra control techniques are
required. to achieve power system balancing over shorter time horizons. The grid of power the main
objective of these control strategies is frequency [3]. That reflects power disparities resulting from
departures from its 50-point reference value (or 60) Hz, which need to stay modest to provide
steady grid functioning. As we shift to renewable energy sources, there are more and more different
causes for grid frequency variations. Regular deterministic frequency fluctuations are caused by the
market-based dispatch of generators [4], and other influences on the grid frequency may come from
major social gatherings [5], local laws [6, 7], and other factors. Modern power networks face
significant obstacles when integrating wind and solar power generation on a broad scale because
these energy sources lack inherent inertia [8] and because of their fluctuating nature [9, 10], which
can result in extra power imbalances and frequency variations. Because electricity system stability

Namangan Institute of Engineering and Technology
nammti.uz
10.25.2023
Pg.291
is so important, it is critical to comprehend and separate these various effects on the grid frequency.
Moreover, control measures are required to stabilise the frequency [1]. Research on the
relationship between these control requirements and outside influences is still ongoing [11].
In this article, we present how eXplainable Artificial Intelligence (XAI) may be used to better
comprehend the factors that influence frequency stability and control, as well as the hazards
involved. We demonstrate applications of our technique to frequency stability indicators [1],
deterministic frequency deviations [6], and frequency control activation in Germany [11]
throughout Europe using operational data from key European power systems.

Download 15,56 Mb.

1 ... 308 309 310 311 312 313 314 315 ... 693

Download 15,56 Mb.

Pdf ko'rish