Table of Contents
Fetching ...

Strategic Data Center Load Shifting: Implications for Market Efficiency and Transmission Value

Aron Brenner, Line Roald, Saurabh Amin

TL;DR

The paper analyzes how large, price-responsive data-center loads interact with two-zone electricity markets through a bilevel optimization framework. By coupling upper-level flexible demand decisions with lower-level economic dispatch, it derives closed-form conditions for when decentralized load shifting aligns with or deviates from social welfare. It reveals two key phenomena: price-discontinuity-induced misalignment at generator basis changes and zero marginal value of transmission under certain strategic load shifts, even when transmission shadow prices remain positive. These results imply that conventional marginal pricing and transmission planning signals can misrepresent system value in the presence of large flexible loads, signaling a need for redesigned market coordination and planning approaches. The study uses illustrative numerical examples to demonstrate the mechanisms and provide insight into how planners should account for strategic flexible demand in capacity expansion and pricing.

Abstract

Data center electricity use may reach 12% of U.S. demand by 2030, alongside growing ability to shift workloads geographically in response to prices or carbon signals. We examine the system-level implications of such strategic flexibility using a bilevel two-zone model that couples economic dispatch with consumer cost minimization. Two market failures emerge. First, discontinuous price changes at generator capacity limits can induce flexible consumers to shift load in socially inefficient directions-for example, toward a higher-cost region to trigger a price drop elsewhere. Second, by positioning near capacity boundaries, consumers can eliminate the marginal value of transmission expansion: although shadow prices suggest additional capacity is valuable, consumers reoptimize to offset resulting flow changes, leaving dispatch and costs unchanged. We derive conditions under which these effects arise and show that conventional price signals and planning methods can misrepresent system value in the presence of large flexible loads.

Strategic Data Center Load Shifting: Implications for Market Efficiency and Transmission Value

TL;DR

The paper analyzes how large, price-responsive data-center loads interact with two-zone electricity markets through a bilevel optimization framework. By coupling upper-level flexible demand decisions with lower-level economic dispatch, it derives closed-form conditions for when decentralized load shifting aligns with or deviates from social welfare. It reveals two key phenomena: price-discontinuity-induced misalignment at generator basis changes and zero marginal value of transmission under certain strategic load shifts, even when transmission shadow prices remain positive. These results imply that conventional marginal pricing and transmission planning signals can misrepresent system value in the presence of large flexible loads, signaling a need for redesigned market coordination and planning approaches. The study uses illustrative numerical examples to demonstrate the mechanisms and provide insight into how planners should account for strategic flexible demand in capacity expansion and pricing.

Abstract

Data center electricity use may reach 12% of U.S. demand by 2030, alongside growing ability to shift workloads geographically in response to prices or carbon signals. We examine the system-level implications of such strategic flexibility using a bilevel two-zone model that couples economic dispatch with consumer cost minimization. Two market failures emerge. First, discontinuous price changes at generator capacity limits can induce flexible consumers to shift load in socially inefficient directions-for example, toward a higher-cost region to trigger a price drop elsewhere. Second, by positioning near capacity boundaries, consumers can eliminate the marginal value of transmission expansion: although shadow prices suggest additional capacity is valuable, consumers reoptimize to offset resulting flow changes, leaving dispatch and costs unchanged. We derive conditions under which these effects arise and show that conventional price signals and planning methods can misrepresent system value in the presence of large flexible loads.
Paper Structure (13 sections, 3 theorems, 14 equations, 3 figures, 1 table)

This paper contains 13 sections, 3 theorems, 14 equations, 3 figures, 1 table.

Key Result

proposition 1

Consider a positive load shift $\delta > 0$ from $A$ to $B$. Letting $p_{g_A},\,p_{g_B}$ respectively denote the pre-shift dispatch levels of marginal generators $g_A$ and $g_B$, two cases arise depending on which zone’s marginal generator changes first. Case 1. (Marginal generator at $B$ changes fi Let $\tau_1 := P_{g_B} - p_{g_B}$ be the threshold where generator $g_B$ reaches capacity. Then, fo

Figures (3)

  • Figure 1: Comparison of system generation costs (left) and flexible consumer costs (right) for a range of shifts $\delta \in [0, 250]$.
  • Figure 2: System costs (left) and flexible consumer costs (right) evaluated over a range of load shifts $\delta$ and transmission capacities $F$.
  • Figure 3: Comparison of system (left) and flexible consumer costs (right) without load shifting (blue) and with decentralized load shifting (orange) for a range of transmission capacities.

Theorems & Definitions (6)

  • definition 1: Market Clearing Problem
  • definition 2: Consumer Cost Minimization
  • definition 3: Objective Alignment
  • proposition 1
  • theorem 1: Externalities at basis changes
  • theorem 2: Zero bilevel marginal value of transmission