Evaluating the effect of presence of distributed generation unit on retailer's performance in the electricity market

Dissertation

Master's degree

Field: Electrical Engineering-Power Orientation

Abstract:

In recent years, with the increasing expansion of electricity consumption, there is a need to increase energy production and use renewable sources in electric energy production for reasons Environmental and economic has increased, among which dispersed production units have been taken into consideration. In the vertical structure, the same policies are often considered for different consumers, while in the restructured system, it is possible for consumers to participate by participating in the electricity market.

In this thesis, the effect of the presence of a distributed thermal generation unit on the performance of the retailer in the electricity market in the short and medium term is studied. The retailer in question in this thesis determines the future contract and the proposed selling price to its consumers, taking into account the uncertainty in the share market prices and consumer demand. The innovation made in this thesis is that it is assumed that the retailer has a small-scale heat-dispersed production unit at his disposal, and in addition to participating in the shared and future market, he can provide a part of his required load through the small-scale unit in some hours. In order to model the uncertainty of current market price and consumer demand, sets of scenarios have been used. Due to the high volume of the problem, in order to reduce the volume of calculations with optimal accuracy, a scenario reduction method has been used. The presented two-stage stochastic programming model is a mixed integer programming problem (MILP) that is solved using GAMS software. The proposed model has been tested in two case studies with a short-term and a medium-term time frame, respectively. The results obtained for different values ??of the operating cost of the thermal unit and different values ??of the penetration coefficient of the thermal unit show that the use of distributed energy sources in the retail sector, in addition to increasing the retailer's profit, also reduces risk and reduces the selling price offered to consumers. Keywords: electricity market, future market, distributed generation, subscription market, retailer, uncertainty. style="direction: rtl;">1-1     Preface

Among the issues that are currently facing the decision makers and policy makers of the electricity industry in many countries of the world, the thought of changing the shape of the structure of this industry is in line with the process of increasing efficiency and competition in other industries. The need to move in this direction is undeniable for various reasons such as capital-intensiveness, the economic inefficiency of the traditional structure and its monopoly.

With the increase in electricity consumption, the need to increase energy production and the use of renewable sources in the production of electrical energy has increased due to environmental and economic reasons, among which scattered production units are being considered [1]. It is expected that in the near future, the use of scattered production units in the retail sector of the electricity market will increase significantly. Retail companies act as intermediaries between producers and consumers in the electricity market. The activities of retailers include buying energy from the wholesale market in order to sell it to consumers. For a medium-term period, these companies must decide on their position in the future market to overcome the risk that arises due to the variability of share market prices. The important issue is for retailers to gain accurate knowledge and awareness about shared market prices during the time period of future contracts so that they can make the right decisions about these contracts.

Today, with the creation of competition in the power systems and its restructuring, many issues of the past have changed and new issues and uncertainties have been introduced in the issues, which have created very strong motivations for the use of random programming in solving problems.Stochastic programming provides a suitable modeling framework in which decision-making problems are properly formulated under conditions of uncertainty [2] and [3]. Stochastic programming relies on having information about the distribution functions of uncertain parameters such as the share market price. When non-deterministic parameters are modeled using continuous or discrete distribution functions, it will be possible to formulate a mathematical programming problem that considers the uncertainty in these parameters. Each non-deterministic parameter is modeled by sets of outcomes or scenarios, so that each scenario represents a possible realization of the non-deterministic parameters with a corresponding probability of occurrence. Usually, the number of scenarios required to represent an uncertain parameter is very large. For this purpose, scenario reduction methods are used to reduce the number of scenarios, while preserving the random characteristics of uncertain parameters. Therefore, it makes sense to model the risk associated with decisions made by market agents. This work can be done directly by entering the risk values ??in the stochastic programming problem [4].

The current thesis, while presenting one of the examples of restructuring in the electricity industry (the role of the small-scale dispersed production unit on the performance of the retailer in the electricity market), will examine the results of its presence in the electricity market.  In this chapter, the general goals of the plan and the contents of the future chapters will be introduced.

The retailer considered in this thesis to determine the future contract and the selling price offered to its consumers must take into account the uncertainty in the share market prices and consumer demand, and must also consider the possibility that if the selling price offered to consumers is not competitive enough, the consumer may choose another retailer. After deciding on the future market and choosing the selling price, the retailer must determine his purchase and sale in the shared market.

The innovation of this thesis is that it is assumed that the retailer has a small-scale thermal dispersion production unit at his disposal and can, in addition to participating in the shared and future market, provide a part of his required load through the small-scale unit in some hours and thereby achieve maximum profit. Considering that the market price and consumer demand in the proposed model have uncertainty, the presented model is random. The presented two-stage stochastic medium-term programming model is a mixed integer programming problem (MILP) modeled using GAMS software. These concepts include the electricity market, various representatives participating in the electricity market, shared market, future market and distributed generation sources. Then, random programming is introduced and the scenario generation method used in this thesis is presented. This method is based on time series models and a scenario reduction method. Chapter 2 ends with the introduction of risk measures that are used in stochastic planning and an overview of the conducted research. In Chapter 3, the proposed method to solve the medium-term planning problem of the presence of a small-scale dispersed production unit is fully investigated. The retailer in question seeks to determine future market issues and determine the selling price offered to consumers. In this season, consumers' electricity demand is provided by the retailer and through purchases from the future market, shared market and thermal production unit. The shared market price and consumer demand are modeled as a stochastic process using sets of appropriate scenarios. Consumers' response to the price offered by the retailer is modeled by the price rationing curve. Future contract curves are used to model the retailer's market power in the future market. Risk modeling is done by CVaR. Finally, the formulation of the problem is presented.