Elsevier

Applied Energy

Volume 241, 1 May 2019, Pages 302-312
Applied Energy

Provision of secondary frequency regulation by coordinated dispatch of industrial loads and thermal power plants

https://doi.org/10.1016/j.apenergy.2019.03.025Get rights and content

Highlights

  • The flexibility of industrial parks is leveraged for frequency regulation.

  • The stochastic day-head optimization is transformed into deterministic.

  • An economic model predictive control is used to enhance the profit and performance.

  • The proposed hierarchical framework is verified based on a real-world power market.

Abstract

Demand responsive industrial loads with high thermal inertia have potential to provide ancillary service for frequency regulation in the power market. To capture the benefit, this study proposes a new hierarchical framework to coordinate the demand responsive industrial loads with thermal power plants in an industrial park for secondary frequency control. In the proposed framework, demand responsive loads and generating resources are coordinated for optimal dispatch in two-time scales: (1) the regulation reserve of the industrial park is optimally scheduled in a day-ahead manner. The stochastic regulation signal is replaced by the specific extremely trajectories. Furthermore, the extremely trajectories are achieved by the day-ahead predicted regulation mileage. The resulting benefit is to transform the stochastic reserve scheduling problem into a deterministic optimization; (2) a model predictive control strategy is proposed to dispatch the industry park in real time with an objective to maximize the revenue. The proposed technology is tested using a real-world industrial electrolysis power system based upon Pennsylvania, Jersey, and Maryland (PJM) power market. Various scenarios are simulated to study the performance of the proposed approach to enable industry parks to provide ancillary service into the power market. The simulation results indicate that an industrial park with a capacity of 500 MW can provide up to 40 MW ancillary service for participation in the secondary frequency regulation. The proposed strategy is demonstrated to be capable of maintaining the economic and secure operation of the industrial park while satisfying performance requirements from the real world regulation market.

Introduction

To achieve a sustainable and clean energy system of the future, governments together with the industry are putting efforts to promote the integration of renewable with an emphasis on integration of wind and solar into the power grid. The intermittency and variation of renewable by nature expose power grid to numerous operational challenges. The resulting threat on system reliability and resilience has drawn attention from legislators as well as power industry all over the world and corresponding policies have been put in place to enhance the flexibility of power systems for high penetrations of renewable. In the United States, Federal Energy Regulatory Commission (FERC) issued Order 755 to promote demand-side resources to participate in frequency regulation markets [1]. The National Grid in the United Kingdom (UK) has developed a dynamic frequency service named Enhanced Frequency Response (EFR) to improve the management of system frequency [2]. Even though China is still in an early development phase of frequency regulation market, the Chinese government has issued a notice encouraging electric energy storage resources [3] as well as large industrial loads such as steel industry loads [4] to participate in frequency regulation markets. All these orders mandate demand-side resources to be fairly compensated in frequency markets, which makes demand-side resources competitive compared to conventional generation for the provision of secondary frequency control.

Resources for frequency regulation are generally required to follow random regulation signals in a timely manner. State-of-the-art research on demand responsive resources to participate in frequency regulation is mainly focused on residential and commercial loads, i.e., electric vehicles (EVs) [5], [6], [7], [8], energy storage system [9], heating ventilation and air-conditioning (HVAC) systems [10], [11], and other thermostatically controlled loads (TCLs) [12]. The flexibility of electric vehicles (EVs) is leveraged for primary regulation in [5], secondary frequency regulation in [6] and emergency frequency regulation services in [7] respectively. Furthermore, an on-going EV vehicle-to-grid demonstration project is developed for frequency regulation in [8]. An investigation into how energy storage can fulfill the fast frequency response is considered in [9]. Experimental evaluation of frequency regulation from HVAC is verified in [10]. The potential of TCLs for frequency regulation is calculated in [11] and field experiment with TCLs to study frequency control is presented [12]. However, due to the sparseness and a limited capacity, residential loads or commercial loads have to be aggregated in a large scale for frequency control, and the requirement of an aggregator and the communication delay between the aggregator and the distributed load may affect the effectiveness to meet the regulation performance for frequency control.

Different from residential and commercial loads, energy-intensive industrial loads are advantageous to provide this frequency regulation ancillary service in that: (1) Industrial loads can provide a large amount of ancillary service due to an enormous power capacity (a series of industrial loads’ power capacity can be up to 1000 MW); (2) Industrial loads are equipped with advanced infrastructures which enable a central controller to dispatch demand-responsive assets to follow regulation signals; (3) Industrial loads have substantial thermal mass, which allows for instantaneous power change following a regulation signal without significantly impacting the quality of electricity service; (4) Industrial loads are operated continuously with high cyclicality, thereby offering reliable sources for frequency regulation; (5) Industrial loads have self-owned thermal power plants, which can be coordinated with demand responsive loads to further leverage the flexibility from industrial loads for up and down regulation reserve.

Over the last decade, extensive research has been conducted on demand response for frequency regulation in power systems. Ref. [13] demonstrated that industrial loads with a large storage capacity or fuel-switching capability are able to provide frequency regulation service. These loads include data centers, industrial electrolysis, industrial cement, pulp mills, arc furnaces, steel rolling loads and electric boilers. In [14], data center loads are regarded as a source of dynamic flexibility for primary regulation and a joint power management strategy of a data center and plug-in electric vehicles for frequency regulation is studied in [15]. To avoid the discreteness of crushers in the cement industry and pulp in the paper industry, a coordinated framework with energy storage is proposed in [16] that improves frequency regulation performance. Among the various types of industrial loads, industrial electrolysis is demonstrated to be able to provide more flexibility to support system operation. Minute-to-minute regulation of electrolysis aluminum loads is achieved in [17] by manually controlling the ratio of the voltage-regulating transformer. A frequency feedback control framework is proposed to allow aluminum smelter loads (ASLs) for system frequency control in an isolated power system by controlling the generator’s excitation voltage [18], [19] or the saturation reactor of the ASL [20]. Based on this architecture, experimental verification of ASLs for frequency regulation is reported in [21], [22] based on the real-world isolated power system. The flexibility of industrial loads, especially industrial electrolysis loads is proven for frequency control while the flexibility of up reserves has not been fully investigated.

Apart from flexibility, industrial regulation resources are required to determine their optimal regulation capacity over a multi-hour operation within the constraints of market timelines [23]. The challenge is to determine the optimal regulation reserve considering the uncertainty from market prices and regulation signals in the day-ahead market. To manage this uncertainty, the research explored the stochasticity [24], [25], [26] of the resources or managed the risk robustly in a deterministic framework [27], [28], [29]. In [24], a stochastic algorithm is developed for frequency control with a consideration of the randomness of prices and the regulation signal. Reference [25] proposed a bidding algorithm for the aggregator to participate in the day-ahead market based on stochastic optimization. Based on stochastic programming with a set of possible price curves, [26] proposes an optimal bidding strategy to maximize regulation revenues. The uncertainty of the regulation signal is verified by the worst-case trajectories in [27]. Based on this robust optimization framework, the regulation resources succeed in following the uncertain signal reliably for most of the time. A robust optimization framework that models the regulation signal as energy constrained is proposed in [28], [29]. Furthermore, experimental verification is conducted to verify the effectiveness of the robust framework in [30], [31]. However, the aforementioned work is based on the premise that regulation resources respond quickly to follow the regulation signal and the regulation capacity is small. Since industrial loads have an enormous regulation capability, the uncertainty of the regulation signal will have a more significant influence on the performance of the industrial loads for frequency regulation.

In addition to the day-ahead regulation reserve schedule, a real-time regulation strategy to track the regulation signal while maintaining economic and secure operation of the industrial loads is in a critical need. A large number of literature have explored the strategy based on model predictive control (MPC) for real-time regulation control. In [32], an MPC strategy is presented for tracking the frequency signal by groups of EVs, controllable loads, and cogeneration power plants. A framework based on a decentralized MPC is proposed in [33] to coordinate the generator and EVs in a three-area interconnected power system. However, economic factors during the regulation process or the correlation between day-ahead optimal reserve scheduling and real-time regulation have not been considered. The economic MPC (EMPC) with an objective to directly reflect process economics is proposed in [34]. The application of EMPC is utilized in the chemical process [35], and the power management [36], but a few kinds of literature report its application for frequency regulation.

To bridge the gap, this paper proposes a hierarchical control framework built upon the work in [30] to explore industrial parks for participation in frequency regulation. In the proposed framework, it considers the slow dynamic process of industrial loads and thermal power plants together with the fact that the random regulation signal has a more severe impact on the profits of the industrial parks. Different from [30] which assumes HVAC can quickly track the regulation signal, energy and production loss and the regulation penalty are incorporated during the regulation process in this study. The main contributions of this study are as follows:

(1) a coordinated scheme of the self-owned thermal power plant and the industrial loads is proposed to enhance the flexibility of industrial parks for frequency regulation. By complementary the large regulation capability of thermal power plants with the fast regulation ability of industrial loads, the industrial park is able to provide a high-quality regulation service in the power market.

(2) a hierarchical control framework is proposed to enable industrial parks to fully achieve their flexibility to provide ancillary service for frequency regulation. In the day-ahead ancillary market, the stochastic regulation signal is replaced by the specific extremely trajectories. Furthermore, the extremely trajectories are achieved by the day-ahead predicted regulation mileage. The resulting benefit is to transform the stochastic reserve scheduling problem into a deterministic optimization. In the real-time dispatch, the economic model predictive control with a cost function of the regulation process is proposed to maximize the potential revenue during regulation. In addition, the control scheme considers the correlation of the day-ahead regulation reserve schedule and real-time regulation operation.

Section snippets

Power system topology of the industrial park

An industrial park is a zone area composed of energy-intensive industrial consumers, e.g., industrial electrolysis and the steel industry. The annual energy consumption of these industrial loads is up to 14.49MWHr per ton so that industrial parks must utilize self-owned thermal power plants for part of their electricity supply while the bulk grid provides additional electricity through the transmission line. The typical topology of an industrial park is shown in Fig. 1.

The potential flexibility

Objective function

In this section, the regulation reserve of industrial parks is optimized with an objective to maximize the potential revenue. We denote the operation time horizon, the set of thermal power plants and the set of ASLs by H={1,...,h},I={1,...,i}, and J={1,...,j}, respectively. Each hour is divided into multiple time slots as T={1,...,t}. Each time slot corresponds to the day-ahead calculate time step. For an industrial park, the total daily income denoted as Incomeday, is calculated by:maxIncomeday

Electrolysis loads model

Electrolysis loads all utilize the same process whereby cells are placed in series and heated by a large direct current. The equivalent circuit of aluminum smelter loads (ASLs) is shown in Fig. 6. The ASL is equal to the series connection of a counter electromotive force E and resistance R.

In our previous works, the dynamic model of ASLs has been obtained through field experiments [22] as shown in Fig. 7. The state space model with state variable ΔPASL(s) and the control variable ΔPASLref is

Simulation system

In this section, we validate the framework's performance in a typical industrial park with ASLs and thermal power plants. We consider that the industrial park has one series of aluminum loads with a capacity of 500 MW, and two self-owned thermal power plants with a capacity of 600 MW. The industrial park power system is connected to the bulk power system through a tie-line. The detailed parameters of the industrial park are listed in Table 2.

The income of the industrial park is equal to its

Conclusions

This paper presents a hierarchical framework to coordinate thermal power plants and industrial loads in the industrial park for frequency regulation. The framework consists of the day-ahead regulation reserve optimization and a real-time model predictive controller. By utilizing the mileage feature of the regulation signal, the random regulation signal is transformed into a deterministic optimization, and the regulation reserve of each operating hour is optimized. The utilization of economic

References (40)

  • Federal Energy Regulatory Commission (FERC). FERC Order 755: Frequency Regulation Compensation in the Organized...
  • National id National Grid. Enhanced frequency regulation response: invitation to tender for pre-qualified parties:...
  • National Energy Administration of China. Notice on the promotion of electric energy storage to participate in the...
  • National Development and Reform Commission of China, National Energy Administration of China. Notice on further...
  • H. Liu et al.

    EV dispatch control for supplementary frequency regulation considering the expectation of EV owners

    IEEE Trans Smart Grid

    (2018)
  • D.F. Nicholas et al.

    Day ahead optimization of an electric vehicle fleet providing ancillary services in the Los Angeles Air Force Base vehicle-to-grid demonstration

    Appl Energy

    (2018)
  • J. Kondoh et al.

    An evaluation of the water heater load potential for providing regulation service

    IEEE Trans Power Syst

    (2011)
  • Y. Lin et al.

    Experimental evaluation of frequency regulation from commercial building HVAC systems

    IEEE Trans Smart Grid

    (2015)
  • S. Li et al.

    Integrated power management of data centers and electric vehicles for energy and regulation market participation

    IEEE Trans Smart Grid

    (2014)
  • X. Zhang et al.

    Demand response of ancillary service from industrial loads coordinated with energy storage

    IEEE Trans Power Syst

    (2018)
  • Cited by (24)

    • Comprehensive frequency regulation control strategy of thermal power generating unit and ESS considering flexible load simultaneously participating in AGC

      2023, Journal of Energy Storage
      Citation Excerpt :

      With the increasing complexity of the modern power grid, for example, large-scale grid-connected renewable energy cannot provide enough frequency modulation backup capacity [1–4], and the start-up time of traditional units is up to 30 s, the frequency problem caused by modern power grid cannot be tracked and adjusted quickly and in real time, which causes great trouble to the frequency of power grid [5–7].

    • A gas-thermal inertia-based frequency response strategy considering the suppression of a second frequency dip in an integrated energy system

      2023, Energy
      Citation Excerpt :

      IESs can also take advantage of the gas system's resistance to energy fluctuations to provide power support for the fast-dynamic electrical systems [12]: 1) energy fluctuations brought by flexible net-load requirements can be accommodated by gas generation [13]; 2) energy fluctuations brought by strikes to networks can be prevented considering the underground gas storages [14]. The inertial resources in IESs are often utilized in the frequency response: 1) thermal inertia is considered to provide frequency response with the help of fast-responding combined heat and power (CHP) units [15]; 2) coordinated dispatch of industrial loads and thermal power plants are considered to provide frequency regulation through ancillary services [16]; 3) gas turbines are considered to improve primary frequency response through control strategies [17]. Current works focused on exclusive use of thermal and gas systems, while the efficiency of frequency response will be largely improved with their coordinate utilization given the simultaneous existence of thermal inertia and gas storage in IESs.

    • Distributional two-level synergetic unit commitment considering three scheduling states

      2022, Electric Power Systems Research
      Citation Excerpt :

      Given the small fluctuation, the system can be restored to a new stable state level by the PFR [7–9]. In terms of the larger deviations, SFR can restore the system to rated operating conditions by releasing available power reserves [10], this process can be performed by automatic generation control (AGC) dispatching. When the uncertainties of load and RESs are high, the dispatch capability may deviate greatly from the actual load curve, or even exceed the maximum correction range of the control.

    View all citing articles on Scopus

    This work was supported in part by the Natural Key R&D Program of China (2017YFB0902900), in part by the Natural Science Foundation of Hubei Province, China (No. 2018CFA080) and in part by the National Natural Science Foundation of China (No. 51707136).

    View full text