The uplift payment elimination through the Lagrangian relaxation of the redundant constraints

TL;DR

This paper proposes a method using Lagrangian relaxation of redundant constraints to eliminate uplift payments in power markets, ensuring stable dispatch outcomes and unchanged profits for market players.

Contribution

It introduces a novel approach to remove uplift payments by adding specific redundant constraints that do not affect the optimal dispatch or profits.

Findings

Explicit construction of redundant constraints that eliminate uplift payments.

Method extends to multi-node, multi-period markets with price-sensitive demand.

Ensures stable dispatch outcomes without profit loss for market players.

Abstract

In the presence of non-convexities, the power market may not have an equilibrium price for power that provides economic stability of the centralized dispatch outcome. In this case, to achieve an economically stable outcome, the uplift payments to the market players are introduced as part of the pricing principle. Given the general pricing principle that involves the lost profit compensation in the form of the uplift payments, we study a question if it is possible to introduce new products/services at the market and the associated prices such that the set of the optimal solutions to the centralized dispatch optimization problem is unaffected, the profit received by each market player at the centralized dispatch schedule is unchanged, and no market player has lost profit. These new products/services and the corresponding prices can be viewed as originating from the Lagrangian relaxation of redundant constraints, which dot not change the centralized dispatch schedule. For any given market price (or a pricing algorithm that sets the producer revenue as a function of its output volume) in a uninode multi-period power market with fixed load, we explicitly construct a family of the redundant constraints that do not change the maximum profit of any producer and result in zero total uplift payment. The analysis can be straightforwardly extended to a multi-node multi-period market with price-sensitive demand.