Coordinated operation of electricity and natural gas systems from day-ahead to real-time markets

https://doi.org/10.1016/j.jclepro.2020.124759Get rights and content

Highlights

  • A framework for the coordination of electric power and gas systems is implemented.

  • Coordination across three energy markets is evaluated.

  • Coordination substantially reduces out-of-merit order dispatch in the electricity system.

  • Shaped flows nominations may become increasingly valuable for reducing unserved gas.

Abstract

Power systems worldwide are becoming more reliant on energy from natural gas, wind, and solar, posing potential reliability and coordination challenges from the tighter coupling of these infrastructure systems. This paper proposes a framework for the market-based coordination of electricity and natural gas system operations. The proposed framework includes a power system model that accounts for flexibility in the commitment of power plants with short start-up and shut-down times, coupled with a dynamic gas model that simulates when gas cannot be delivered to generators. The capabilities of the framework are illustrated using real-world electric power and gas systems, including scenarios around wind and solar penetrations and the analysis of time-variant, “shaped flow” gas nominations. Our results indicate that coordination between power and gas systems improves total gas delivery and reduces out-of-merit order dispatch in the electricity system, and that shaped flows may reduce unserved gas in systems with high penetrations of wind and solar energy sources. Coordination can have mixed effects on carbon-dioxide emissions, with emissions increasing with coordination for current systems during high load weeks but decreasing for systems with high renewable penetrations, particularly during periods of high variability.

Introduction

The availability of low-cost natural gas in the United States—driven largely by the shale gas revolution—has led many electric power systems to become more reliant on natural gas. This trend has been amplified by an increasing penetration of renewable energy sources, as natural gas-fired power plants are frequently used to offset the uncertainty and variability associated with wind and solar power generation. In the United States, the power sector accounted for 35.5% of total natural gas demand in 2018, up from just 22.3% in 2000 (U.S. Energy Information Administration U.S. Energy Information Administration (EIA), 2019b, U.S. Energy Information Administration (EIA), 2019a. The share of generation from natural gas increased from 14.2% to 31.5% over the same period, while the share from renewable energy—largely driven by increases in wind and solar—has increased from 8.8% to 17.4%, with these trends expected to persist into the future (U.S. Energy Information Administration U.S. Energy Information Administration (EIA), 2019b, U.S. Energy Information Administration (EIA), 2019a.

The increasing reliance on natural gas for power system operations poses coordination and reliability challenges. For example, under severe weather conditions both electricity demand and natural gas consumption for heating tend to increase, with fuel shortages leaving gas power plants without contracts for firm gas transportation unable to fulfill their power generation schedules (Craig et al., 2020; Saldarriaga-C. and Salazar, 2016; Shahidehpour et al., 2005). Addressing the issue will require increased coordination between the two systems across different decision-making levels, including planning and operations.

At the planning level, the objective is to optimize the location, capacity, and timing of investment decisions associated with generation or production, transmission, and storage assets in an integrated system. Previous research has explored co-planning of electric power and natural gas systems based on deterministic approaches, e.g., bi-level programming (Zeng et al., 2017) and mixed integer linear (Unsihuay-Vila et al., 2010) and nonlinear (Chaudry et al., 2014; Qiu et al., 2015a) programming. Additionally, stochastic (Qiu et al., 2015b) and robust (Shao et al., 2017) optimization approaches have been proposed to address the uncertainty presented in this type of planning problem, such as uncertainties in gas demand, electricity demand, and gas prices.

At the operational level, the objective is to improve reliability and minimize the operational costs associated with natural gas and electricity supply, natural gas supply contracts, and load shedding or unserved natural gas (Chaudry et al., 2008). Operational coordination can be addressed using either central-planning or market-based approaches. With central-planning—which has been the most common research approach—the operation of the two systems is optimized simultaneously. This includes approaches using deterministic optimal power flow coupled with steady state natural gas systems (Martinez-Mares and Fuerte-Esquivel, 2012; Seungwon An et al., 2003), deterministic integrated unit commitment and/or economic dispatch formulations based on steady state (Cong Liu et al., 2009; Li et al., 2008) or dynamic (Chiang and Zavala, 2016; Correa-Posada and Sanchez-Martin, 2015; Zlotnik et al., 2017) gas models, as well as stochastic (Chen et al., 2018; Qadrdan et al., 2014), robust (He et al., 2017), and interval (Bai et al., 2016; Qiao et al., 2017) optimization models that address the uncertainty associated with electrical load, renewable power forecast, and outages of generation and transmission assets. Some of these studies demonstrated that central-planning coordination benefits (economically) both electricity and natural gas networks (Chiang and Zavala, 2016; Seungwon An et al., 2003). In other words, the economic objectives of the electricity and gas system operators do not compete. Also, it was illustrated that natural gas systems with higher linepacks (Correa-Posada and Sanchez-Martin, 2015; Ordoudis et al., 2019) and gas power generators with fuel switching capabilities (Li et al., 2008; Shahidehpour et al., 2005) could facilitate the coordinated operation of the gas and electricity systems. For example, higher linepacks in the gas system and gas power generator with fuel switching capabilities could reduce unserved gas and electricity demands while reducing total operational cost.

By contrast, in market-based approaches the two systems are optimized or simulated separately, with coordination occurring via the exchange of information such as prices, gas demand from generators, gas availability from the gas network, etc. This coordination can occur at a range of timescales—including day-ahead (DA), intra-day (ID), and real-time (RT)—which evolve as demand and generation uncertainties are reduced over time. Some of the proposed market-based approaches are based on deterministic unit commitment and/or economic dispatch formulations using steady state (Cui et al., 2016; Yang et al., 2019) or dynamic (Pambour et al., 2018; Zhao et al., 2019) gas models. Market-based approaches based on stochastic unit commitment formulations and steady state (or simplified versions of the dynamic state) gas models have been proposed to deal with uncertainty in renewable power generation (Alabdulwahab et al., 2015; Ordoudis et al., 2019). Regarding the market-based coordination mechanisms, the focus has been either on just (1) natural gas demand from the power system operator and gas supply (availability) from the gas system operator (Alabdulwahab et al., 2015) or (2) both gas supply and availability as well as marginal prices for gas supply and demand (Ordoudis et al., 2019; Zhao et al., 2019). Moreover, most of existing electricity and gas system coordination studies consider a 24 h operation time frame with hourly resolution (Chiang and Zavala, 2016; Cui et al., 2016; Ordoudis et al., 2019) or 15 min resolution (Zhao et al., 2019).

Although steady state gas models can be appropriate for long-term planning (Guerra et al., 2016; Üster and Dilaveroğlu, 2014), these models do not capture the dynamics of linepack storage, underestimating operational flexibility of the gas system (Chiang and Zavala, 2016; Zlotnik et al., 2017). Dynamic gas models capture linepack dynamics, making them better suited for studying the operational coordination of electric power and natural gas systems (Chaudry et al., 2008; Chiang and Zavala, 2016; Ríos-Mercado and Borraz-Sánchez, 2015).

Despite the prospect of more optimized coordination between electric power and gas operations offered by a central planning perspective, currently electric power and gas markets in the United States are cleared separately and interact with limited information exchange. This makes a market-based approach perspective more realistic for capturing the operational coordination of today’s electric power and natural gas systems (Ordoudis et al., 2019; Zhao et al., 2019). Natural gas-fired power plants can purchase their fuel based on contracts (long-term or short-term) or the spot gas market, which involve uncertainties associated with gas prices and availability. Note that “primary” firm transportation contracts specify a total maximum daily offtake (maximum delivery quantity) and the firm contract holders exercise their contractual rights by nominating their planned daily offtake below or at that maximum the day before delivery.

One current challenge with executing market-based coordination is that electricity market and gas nominations cycles are currently not aligned. U.S. Federal Energy Regulatory Commission (FERC) Order 809 was proposed to address this issue by creating three ID gas nomination cycles and shifting them to better match the timing of power system markets.2 An additional issue is that outside the RT market, natural gas generators are typically required to submit a steady, non-varying quantity of gas offtakes for each hour over the course of a nomination cycle, known as a “ratable” flow. The use of ratable flows may discourage gas generators from flexible operations, since they cannot be guaranteed gas delivery above their ratable flow level and will be subject to risk in the spot market. In contrast, moving to time-variant gas nominations (“shaped flow”) at the DA and ID market levels has been proposed as a mechanism of improving gas and power sector coordination (Peress and Karas, 2017).

In this study we propose and evaluate a novel framework for the operational coordination of electric power and gas systems using a dynamic gas model and a unit commitment/economic dispatch power system model. Unlike previous studies, which typically focus on coordination in a single market (Alabdulwahab et al., 2015; Cui et al., 2016; Zhao et al., 2019), we include coordination across three market levels (DA, ID, and RT) in order to realistically account for flexibility and commitment constraints in the power sector as well as wind and solar forecast improvements from DA to ID and then RT. Furthermore, we test our framework using real network and operational data from the Colorado power and gas system, making this work one of the first to explore coordination on a model of a real system. We also compare results from our coordination framework using a 2018 system and a projected 2026 system in order to study how the value of coordination changes in a future scenario with increased renewable generation. Finally, we quantify the potential benefits of moving from ratable to shaped flows, the value of which has not been studied extensively, particularly for systems with higher renewable penetrations. Specifically, the novelties of this study are: (i) a market-based coordination framework is proposed and implemented for the DA, ID, RT coordination of real interdependent gas and electricity systems, (ii) the proposed framework is used to assess the value of shaped flow gas nominations as a coordination mechanism. Moreover, the results from the assessment of shaped flow gas nominations can inform energy policy makers on the development and implementation of regulations to facilitate power and gas systems coordination during planning and operation.

Section snippets

Methods

This section provides an overview of the coordination structure. First, we describe the modeling behind both the power system and the gas network. Then, we describe the coordination framework.

Case study

We test our coordination framework on a case study using data from Colorado’s power and natural gas systems. The PLEXOS model for the power system is adapted from a model developed previously for the Western Electricity Coordinating Council (Brinkman et al., 2016). This model is representative of the 2018 Colorado grid and includes 2 regions, 1476 buses, and 1841 transmission lines, along with just under 21 GW of generating capacity.

Because the value of coordination is likely to increase with

Results

We use the following terminology to describe our results at different levels of iteration between the power and gas models, as illustrated in Fig. 1:

  • Power system only: results from the first iteration of the power system model, before any communication with the gas network.

  • Co-simulation: results after simulating gas offtakes from the power system model in the gas network; reflects curtailed gas but has not yet reoptimized the power system in response to gas constraints.

  • Coordination: results

Conclusions

In this study, we examine the benefits of coordinating natural gas and power system operations through communication between models of the two infrastructure systems. Our study finds substantial potential benefits to coordination between the two sectors. Coordination substantially reduces curtailed gas generation, reducing the need for potentially costly out-of-market operator interventions or the possibility of unserved electricity demand. From the gas system perspective, we find that much of

CRediT authorship contribution statement

Omar J. Guerra: Conceptualization, Methodology, Software, Data curation, Writing - original draft. Brian Sergi: Conceptualization, Methodology, Software, Data curation, Writing - original draft, Investigation, Visualization. Michael Craig: Conceptualization, Methodology, Software, Data curation, Writing - original draft, Investigation, Visualization. Kwabena Addo Pambour: Conceptualization, Methodology, Supervision, Writing - original draft. Carlo Brancucci: Conceptualization, Methodology,

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research was supervised and coordinated by the Joint Institute for Strategic Energy Analysis (www.JISEA.org), ensuring it met institutional standards for objectivity, evidence, and responsiveness to the study scope. We would like to thank the following organizations for their support: Midcontinent Independent System Operator, Hewlett Foundation, Environmental Defense Fund, American Gas Association, American Electric Power, and Kinder Morgan. Support does not imply endorsement; the analysis

References (41)

  • Z. Qiao et al.

    An interval gas flow analysis in natural gas and electricity coupled networks considering the uncertainty of wind power

    Appl. Energy

    (2017)
  • R.Z. Ríos-Mercado et al.

    Optimization problems in natural gas transportation systems: a state-of-the-art review

    Appl. Energy

    (2015)
  • H. Üster et al.

    Optimization for design and operation of natural gas transmission networks

    Appl. Energy

    (2014)
  • Z. Yang et al.

    Coordination of integrated natural gas and electrical systems in day-ahead scheduling considering a novel flexible energy-use mechanism

    Energy Convers. Manag.

    (2019)
  • Q. Zeng et al.

    A bi-level programming for multistage co-expansion planning of the integrated gas and electricity system

    Appl. Energy

    (2017)
  • A. Alabdulwahab et al.

    Coordination of interdependent natural gas and electricity infrastructures for firming the variability of wind energy in stochastic day-ahead scheduling

    IEEE Trans. Sustain. Energy

    (2015)
  • Seungwon An et al.

    Natural gas and electricity optimal power flow

  • C. Barrows et al.

    Multi-operator production cost modeling

    IEEE Trans. Power Syst.

    (2019)
  • G. Brinkman et al.

    Low Carbon Grid Study: Analysis of a 50% Emission Reduction in California

    (2016)
  • R. Chen et al.

    Clearing and pricing for coordinated gas and electricity day-ahead markets considering wind power uncertainty

    IEEE Trans. Power Syst.

    (2018)
  • Cited by (12)

    • A robust multi-objective joint scheduling of integrated electricity and gas grids considering high penetration of wind and solar units and flexible loads towards achieving a sustainable operation

      2023, International Journal of Hydrogen Energy
      Citation Excerpt :

      A distributed planning method based on the multi-agent concept is suggested to separately consider the profit of stakeholders in electricity-gas network and energy hubs [15]. A new framework is designed for market-based coordination of electricity and natural gas systems under the uncertainty of wind and solar radiation, where the results show positive effects of the mentioned approach on greenhouse gases emission [16]. The above studies have focused on the deterministic state, while many researchers attempt to analyze the effect of uncertainties on the system operation.

    • Optimal gas-electric energy system decarbonization planning

      2022, Advances in Applied Energy
      Citation Excerpt :

      For comprehensive recent reviews on this previous work, we refer the reader to He et al. (2018) [20], Farrokhifar et al. (2020) [21], and Huang et al. (2020) [22]. Using models for co-optimized gas and electric flows, several researchers have investigated the value of coordination and optimal control of the two integrated networks [23–26]. The fidelity of simulation used for optimizing control varies from convexified steady-state simulations [27,28] to transient flow models [29,30].

    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    View full text