This paper is published in Volume-8, Issue-2, 2022
Area
Computer Science & Engineering
Author
Gopi Edimudi, Katika Arshad, Mohammed Muzammil
Org/Univ
Jain (Deemed-to-be University), Bengaluru, Karnataka, India
Pub. Date
13 April, 2022
Paper ID
V8I2-1259
Publisher
Keywords
Predicting the Power of CCPP, Machine Learning

Citationsacebook

IEEE
Gopi Edimudi, Katika Arshad, Mohammed Muzammil. Predicting the power of Combined Cycle Power Plant, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Gopi Edimudi, Katika Arshad, Mohammed Muzammil (2022). Predicting the power of Combined Cycle Power Plant. International Journal of Advance Research, Ideas and Innovations in Technology, 8(2) www.IJARIIT.com.

MLA
Gopi Edimudi, Katika Arshad, Mohammed Muzammil. "Predicting the power of Combined Cycle Power Plant." International Journal of Advance Research, Ideas and Innovations in Technology 8.2 (2022). www.IJARIIT.com.

Abstract

With the rising share of renewables in the energy mix, there is a need for alternative flexible power capacity to mitigate its intermittency. The emerging role of the aggregator is important in untapping the potential of new flexibility sources. An aggregator bundles the flexibility of distributed generation and demand response, and offers the collective resources to wholesale electricity markets. This technical aggregation is referred to as a Virtual Power Plant (VPP). The main aim of this paper is to work towards a model to estimate flexibility characteristics of aggregates of DR (demand response) and DG (distributed generation) to better provide smart grid services from those aggregates in existing and future wholesale markets. This knowledge is the missing connection to ensure capabilities of real-time response into energy markets with a fixed time control loop. Traditionally this has been done using detailed information on device makeup of a virtual power plant (VPP), i.e with detailed knowledge of flexible device state and their performance capabilities. This low level detailed knowledge requirement presents barriers such as performance and security of private information. Therefore, this paper will investigate the applicability of machine learning techniques using only currently available market information to black box estimate aggregate flexibility characteristics, specifically capacity available during run-time and longevity, the amount of time a power deviation can be maintained.