Refinement of Reliability Grid Codes in the Provision of Ancillary Services

Jul 16, 2026

Stochastic resources such as wind farms, electric vehicle aggregators, and demand-side assets increasingly provide reserve capacity in ancillary service markets, yet system operators require reliability thresholds to manage delivery uncertainty. Energinet’s P90 rule sets a minimum 90% acceptance probability, but it is a regulatory convention rather than an optimized design choice. A bilevel optimization framework endogenizes the threshold in the upper level and uses reliability-constrained bidding with chance constraints reformulated via a Weibull tail distribution. Applied to Nordic FCR-D, cost-optimal thresholds fall below P90 with up to 14.5% cost reductions, and dynamic hourly thresholds further improve outcomes by up to 2.4%.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

Introduction

Reserve procurement and reliability criteria

Transition from dispatchable to stochastic resources

Energinet P90 requirement and identified gap

Abstract—Contribution and framework overview

Bilevel optimization and reliability-constrained bidding

Weibull-based chance constraint reformulation

Nordic FCR-D results and cost trade-offs

This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. arXiv :2607 .02319v1 [ ee ss . SY] 2 Jul 2026 Refinement of Reliability Grid Codes in the Provision of Ancillary Services Torine R. Herstad, Jalal Kazempour, Senior Member, IEEE, Lesia Mitridati, and Steven A. Gabriel, Senior Member, IEEE Abstract—Stochastic resources such as wind farms, electric vehicle aggregators, and demand-side assets are increasingly participating as reserve providers in ancillary service markets. To manage delivery uncertainty, system operators impose minimum reliability thresholds on such providers. Energinet, the Danish transmission system operator (TSO), has pioneered this approach through the P90 requirement, requiring stochastic providers to make accepted reserve capacity bids available with at least 90% probability. Yet this threshold is set by regulatory convention, not optimization: no existing framework treats it as a design variable or characterizes the cost-reliability trade-off it governs. This paper closes that gap. We develop a bilevel optimization framework in which the TSO in the upper level sets the reliability threshold endogenously while providers in the lower levels respond through reliability-constrained bidding, with chance constraints reformulated analytically using a Weibull tail distribution. Applied to the Nordic frequency containment reserve for disturbances (FCR-D) market, the cost-optimal threshold lies below P90 in the studied cases, with cost reductions by up to 14.5% relative to the fixed standard. Dynamic hourly thresholds yield a further reduction of up to 2.4%, suggesting efficiency gains may increase in larger and more diverse reserve markets. Index Terms—Reserve capacity procurement, stochastic energy resources, reliability grid code, chance-constrained bilevel optimization. I. INTRODUCTION Maintaining grid stability requires transmission system operators (TSOs) to procure sufficient ancillary services, including reserves for balancing purposes 1. This reserve procurement task is becoming more complex as wind, solar, and demandside flexible assets such as electric vehicles (EVs) contribute to the reserve provision. Unlike dispatchable generation, these assets are inherently stochastic, challenging the procurement assumptions that reserve market design has historically relied upon. Early reserve procurement frameworks were designed around a fully reliable supply base of dispatchable thermal and hydro generators. In this setting, operational uncertainty arose from demand forecast errors, unplanned generation outages, and transmission contingencies. Reserve dimensioning was therefore grounded in deterministic security criteria, most notably the N-1 rule, which required sufficient upward reserve to cover the loss of the single largest generator at all times. Seminal work from this period established the theoretical and Torine R. Herstad, Jalal Kazempour, and Lesia Mitridati are with the Technical University of Denmark, Kgs. Lyngby, Denmark (e-mails: {torhe, jalal, [lemitri}@dtu.dk](lemitri}@dtu.dk)). Steven A. Gabriel is with the University of Maryland, MD, USA, the Norwegian University of Science and Technology, Trondheim, Norway, and Aalto University, Espoo, Finland (e-mail: [sgabriel@umd.edu](sgabriel@umd.edu)). 1Hereafter, we use the terms ancillary services and reserves interchangeably, although ancillary services generally encompass a broader range of services beyond reserves. computational foundations for optimal reserve sizing under these conditions, consistently arriving at the conclusion that reserves should be sourced exclusively from controllable generators and dimensioned to guarantee full coverage [1] . The large-scale integration of wind power from the early 2000s onward fundamentally altered this picture. Wind generation forecast errors introduced an additional stoch

FAQs

What problem does the paper address regarding reliability thresholds in ancillary services?

The paper addresses that reliability thresholds for stochastic reserve providers, such as Energinet’s P90 requirement, are regulatory conventions rather than optimized design variables that explicitly characterize cost–reliability trade-offs.

How does the proposed framework determine the reliability threshold?

It uses a bilevel optimization framework where the TSO sets the reliability threshold in the upper level endogenously, while providers respond in the lower levels with reliability-constrained bidding.

What do the results show for the Nordic FCR-D market?

For studied cases, the cost-optimal reliability threshold lies below P90, yielding cost reductions of up to 14.5% versus a fixed standard, and dynamic hourly thresholds provide additional reductions up to 2.4%.

Jul 16, 2026

Upload by Ava Thompson