Design of Green Power Clouds for Intelligent Virtual Power Plants

Abstract

Traditional virtual power plants (VPPs) combine power from distributed energy resources (DER) to supply energy to users. However, they fall short of net-zero goals because of neglecting carbon footprints during power aggregation. This paper proposes a novel intelligent virtual power plant framework (iVPPF) to address this gap. iVPPF comprises a central iVPP (iVPP<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathrm {C}}$ </tex-math></inline-formula>) and several regional iVPPs (iVPP<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathrm {n}}$ </tex-math></inline-formula>), n = E, S, M, and N. These iVPPn are geographically distributed systems for intelligently managing iVPPs in four regions: east, south, middle, and north, respectively, while the iVPPC is responsible for dispatching power across iVPPn. We built iVPPC and iVPPn on individual green power clouds, which can provide abundant computing resources and realize intelligence through AI technologies for iVPPF. We also design universal computing devices called cyber-physical agents (CPAs) to collect essential data on manufacturing, carbon footprint, and energy usage for iVPPn. iVPPn can intelligently control DERs based on the collected data. Also, iVPPF can empower enterprises to participate in power balancing services offered by Taipower, thereby enhancing the flexibility of the overall power grid. Furthermore, we integrate iVPPF with the I4.2-GiM framework, offering intelligent carbon and energy management capabilities to achieve the net-zero goal. The testing results show that iVPPF can significantly reduce energy usage (up to 25.6%) and carbon emissions (up to 509 kg) through power dispatch. Thus, the proposed iVPPF promises to contribute economic benefits for businesses and the pursuit of net-zero emissions. Note to Practitioners—This paper proposes an intelligent virtual power plant framework <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$({i} \text { VPPF})$ </tex-math></inline-formula> consisting of a central coordinator <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$({i}\text {VPP}_{\text {C}})$ </tex-math></inline-formula> and distributed regional managers (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${i}\text {VPP}_{\text {n}}$ </tex-math></inline-formula> for East, South, Middle, and North). Leveraging green power clouds, both <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${i} \text { VPP}_{\text {C}}$ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${i}\text {VPP}_{\text {n}}$ </tex-math></inline-formula> harness AI for intelligent management and power dispatch across regions. We detail the system architecture and showcase practical applications, including scenarios like dispatching and aggregating for demand response, using the IEEE 13-node test feeder. Additionally, we explore the design of green power clouds and cyber-physical agents (CPAs).

Topics & Concepts