برنامه ریزی غیرخطی عدد صحیح مختلط و رویکرد مبتنی بر شبکه های عصبی مصنوعی برای خدمات جانبی در بازارهای برق رقابتی انتقال

Ancillary services represent a good business opportunity that must be considered by market players. This paper presents a new methodology for ancillary services market dispatch. The method considers the bids submitted to the market and includes a market clearing mechanism based on deterministic optimization. An Artificial Neural Network is used for day-ahead prediction of Regulation Down, regulation-up, Spin Reserve and Non-Spin Reserve requirements. Two test cases based on California Independent System Operator data concerning dispatch of Regulation Down, Regulation Up, Spin Reserve and Non-Spin Reserve services are included in this paper to illustrate the application of the proposed method: (1) dispatch considering simple bids; (2) dispatch considering complex bids.

Electric power industry is moving towards an increasing competition, with a variety of services previously provided by electric utilities being progressively unbundled [1], [2], [3], [4], [5], [6] and [7]. Initially the attentions have mainly focused on the energy market but ancillary services (AS) are gaining more and more importance. Ancillary services have an essential role in the operation of electricity markets [8], [9], [10], [11] and [12]. AS are generally defined as those services that are important to achieve targeted goals on power system security, reliable operation, frequency stability, as well as on voltage level and stability [13], [14] and [15], in order to cover generation and transmission contingencies. The definition of ancillary services prices and how those prices should change regarding the operating decisions is becoming a major issue due to the changes in the market structure [9] and [16]. The creation of a market for ancillary services aims at achieving mutual benefits for buyers and sellers. The Independent System Operator (ISO) aims at obtaining the necessary resources in order to maintain the system security at minimum cost, while providers look for the maximization of their profits from the sale of ancillary services. Ancillary services are necessary, according to the Federal Energy Regulatory Commission (FERC), to support the power transmission from sellers to buyers, whereas the control areas and transmission utilities have the obligation to maintain a reliable operation of interconnected transmission system [17]. Regulation services – Regulation Up (RU) and Regulation Down (RD) – are the first level acting reserves which are updated around every 4 s depending on the system operator. The regulation keeps the continuous matching between load and generation and is controlled by Automatic Generation Control (AGC) systems [18] and [19]. The Spin Reserve (SR) is one of the most important AS required for maintaining power system reliability following a major contingency. In the same way, Non-Spin Reserve (NSR) has an important role in keeping the system security, since this reserve is used when SR cannot satisfy the reserve needs. Short-term prediction of day-ahead reserve requirement helps the ISO making effective and timely decisions. Moreover, based on these forecasts, market participants can make decisions concerning their bidding strategies for day-ahead SR market [18] and [19]. Two alternative methods are available for energy and ancillary service provision: simultaneous procurement and separate procurement [20], with advantages and drawbacks of each of these solutions. Because most ancillary services are provided by the same equipments that produce energy, the energy and AS markets are linked so any problem in one of these markets may cause a problem in the other [21]. In recent years, ancillary service markets design has become an important work area. Designing options of bid selection protocols and settlement rules for ancillary service markets are presented in [22]. An ancillary framework for managing power system security, based on a market for ancillary services, is presented in [23]. A detailed formulation of simultaneous energy and ancillary service auctions for integrated market systems by an optimal power flow is presented in [24]. A Security Constrained Unit Commitment model for energy and ancillary services auction, which can be used by an ISO to optimize reserve requirements in electricity markets, can be found in [25]. In [13], Vale et al. present an approach using linear programming and genetic algorithm approaches that has been used in MASCEM, a multi-agent electricity market simulator, to deal with ancillary services market simulation. An Artificial Neural Network approach for day-ahead RD, RU and NS forecasting which has been included in MASCEM [26] is described in [19]. [27] presents two mathematical models, the Security Constrained Unit Commitment (SCUC) and the Security Constrained Economic Dispatch (SCED) that are used to perform Unit Commitment and Economic Dispatch respectively in CAISO (California ISO) day-ahead and Real Time Markets. It is well known that the economic selling bids of energy that sellers present to the market operator can be simple or integrate complex conditions. The simple bids are economic offers of energy sale that sellers present for each time period and for each generation unit of which they own with a price and a quantity of energy. In most of the cases the bids are not simple and integrate complex conditions of sale in order to achieve the producers (sellers) objectives and the needs of the ISO. The complex bids are those that comply with the requirements for the simple bids and also integrate any of the allowed conditions, such as: minimum remuneration for each bid, minimum working hours, and minimum consecutive working hours. Thus, in order to accommodate the objectives of the producers (sellers) and the needs of the ISO this paper proposes a methodology for AS market management and simulation. This methodology includes two distinct phases: the AS requirements forecasting and the AS dispatch. The AS requirements forecast based on Artificial Neural Networks (ANN) has been previously proposed by some of the authors of this paper [19]. The ANN is trained using AS historic data and are able to determine the amount of power to be guaranteed for each AS. The AS dispatch considers the bids submitted to the market and includes a market clearing mechanism based on linear programming and other considering complex bids based on mixed integer non-linear programming. After this introduction section, this paper is organized as follows: Section 2 deals with ancillary services in electricity markets. Section 3 presents the methodology for the problem under study. Section 4 presents two case studies and the discussion of the obtained results. The presented case studies use real AS data from California Independent System Operator (CAISO) to forecast AS requirements. Finally, Section 5 presents the most relevant conclusions of the presented work.

Providing adequate decision support for electricity market players requires the use of decision support tools which can be effectively based on simulation studies. In order to make decision support effective, all business opportunities must be considered. Ancillary services are an important component of electricity markets as they represent a business opportunity for market players. Therefore they must be included in electricity market simulation, so that it is possible to support market participants’ decisions. This paper proposes a deterministic approach to solve a bid-based ancillary services dispatch in an electricity market and use an Artificial Neural Network approach for day-ahead prediction of Regulation Down, regulation-up, Spin Reserve and Non-Spin Reserve requirements. A case study, with real data from the California Independent System Operator, allows comparing the results obtained with the method without complex bids which is based on linear programming with those obtained with the method considering complex bids based on mixed integer non-linear programming. Electricity market operation shows the importance of using adequate methodologies for ancillary services dispatch in a market environment. The inclusion of ancillary service markets in an electricity market simulator, like MASCEM, allows electricity market players to make simulation studies, considering a combined participation in several markets, including energy and ancillary services markets, allowing them to define adequate bidding strategies.