Locational Marginal Price Decomposition Using a Fully Distributed Slack Bus Model

Abstract

The implementation of risk hedging instruments against the inherent volatile nature of locational marginal prices (LMPs) requires the decomposition of such economic signals into specific components. These components are extremely dependent on the active energy reference selection in optimal power flow (OPF) models that govern the LMP decomposition. Active power distributed slack bus models are commonly used to set the active energy reference in LMP decomposition frameworks to satisfy the financial interests of electricity market participants more equitably. However, in recent years, some important energy-related deliberative organizations have shown great interest in incorporating reactive power into market-oriented OPF models. This scenario highlights the need for research efforts focused on the formulation of more complete distributed slack bus models that address active and reactive power. In this context, this paper proposes an LMP decomposition model based on an OPF framework with a fully distributed slack bus formulation. The harmful financial impacts of conventional reactive energy reference specification strategies in the calculation of LMP components are explained from the perspective of market participants. In the proposed decomposition model, to overcome such harmful effects, active and reactive power mismatches are compensated through the conventional active power distributed slack bus and the proposed reactive power distributed slack bus, respectively. Thus, the double selection of energy reference conceives a new source of negotiation between market participants in the formulation of risk hedging instruments. Numerical simulations on the IEEE 30-bus test system corroborate the financial relevance of the proposed decomposition scheme.INDEX TERMS Electricity markets, LMP decomposition, locational marginal price (LMP), optimal power flow (OPF), reactive energy reference, reactive power distributed slack bus. NOMENCLATURE 𝐹(⋅)Total operating cost function. 𝑰 𝑖 i × i identity matrix. 𝑛, 𝑛 𝑝𝑣 , 𝑛 𝑏 , 𝑛 𝑔 Total number of buses, PV buses, branches, and generators. 𝒑, 𝒒 Vectors of net active and reactive power injections at all buses, i.e., 𝒑 = 𝒑 𝑔 − 𝒑 𝑑 and 𝒒 = 𝒒 𝑔 − 𝒒 𝑑 . 𝒑 𝑑 , 𝒒 𝑑 Vectors of active and reactive power loads at all buses. 𝑝 𝑑𝑠 , 𝑞 𝑑𝑠 Active and reactive power injections at the corresponding distributed slack buses. 𝒑 𝑓 , 𝒒 𝑓 Vectors of active and reactive power flow functions at all buses. 𝒑 𝑔 , 𝒒 𝑔 Vectors of active and reactive power injections at all buses. 𝒔 𝑓𝑙 Vector of apparent power flows at all branches. 𝜶, 𝜷 Vectors of active and reactive power participation factors. 𝜼 𝑝 , 𝜼 𝑝 , 𝜼 𝑞 , 𝜼 𝑞 Vectors of Lagrange multipliers.

Real-Time Energy Management in Smart Homes Through Deep Reinforcement Learning

“…This has emerged as a dominant trend, largely due to the billing practices of most utility companies. To break it down, homeowners are predominantly billed based on their active energy consumption, with reactive power, an equally significant component of power management, often being sidelined [37]. Such a trend is not arbitrary.…”

Optimizing Peer-to-Peer Energy Transactions: Determining the Allowable Maximum Trading Power for Participants

“…The dual variables or Lagrange multipliers corresponding to the equality constraints (11) and (12) represent the DLMPs corresponding to the MTP of sellers and buyers, respectively [35,36]. Solving the optimization problem using all of the decision variables and the Lagrange multipliers concurrently could take a considerable amount of time.…”