Short-term load forecasting of Mazandaran province using expert systems

Thesis

Master's course

Abstract:

Short-term load forecasting, in the form of one-hour to several days load forecasting, has a significant impact on the operation of power systems. Because many energy management measures, such as setting a cost-effective plan to use existing power plants, planning the purchase of fuel needed by power plants, entering and exiting units, developing transmission lines and trans-distribution networks, as well as the amount of blackouts in case of shortage, are carried out based on this forecast.

In this project, a variety of forecasting models including time series, regression, final consumption and neural network models have been reviewed. Considering that the usual load forecasting in Mazandaran distribution company has been using traditional methods and since this method was not able to accurately predict the load of future days and specific days, the fuzzy method has been investigated to achieve the desired model. In this regard, he first mentioned the description of the fuzzy logic and the method of implementing the program in the environment of the article, and then using the information of the past years and the consumption chart and taking into account the impact of environmental factors, the short-term load of Mazandaran province has been predicted. The use of fuzzy logic method has resulted in increasing the accuracy and speed of forecasting, solving the problem of forecasting the load of certain days, reducing the size of the database along with increasing the ability to influence various factors.

The importance of load forecasting in power systems

Electricity industry is one of the infrastructure industries of a country and is considered a very important pillar in the growth and development of today's societies. Considering that, on the one hand, electricity industry projects require large investments and long periods of time, and on the other hand, with the existing technology, it is still not possible to store this energy on a large scale. Therefore, production planning should be done in such a way that it meets the demand of electrical energy. Therefore, load forecasting is considered as an important factor in the development and operation of power systems, and in fact, it is a tool that can be used to improve decision-making. Estimating the resource allocation process is mandatory for the development of the power supply network.

In planning the future development of a load estimation power system, it is very important and forms the basis of planning studies. The amount of load forecasting errors is of particular importance.

Difficulties in decision-making in this case increase when with a limited budget and the goal of minimizing costs on the one hand, and the pressure of experts and engineers of the power sector to purchase advanced and expensive equipment on the other hand, as well as the excessive expansion in the use of electrical energy. be faced If the amount of predicted load is less than the actual load, the reliability factor and as a result the quality of service will decrease and this may even lead to forced shutdowns. This problem makes it difficult for the officials to analyze the reliability of the system, and on the other hand, if the future load is predicted before the required amount, a lot of investment will be wasted and lead to financial need.

The impossibility of storing electric energy on the one hand, as well as spending huge economic costs to create new power plants, on the other hand, has pushed the general policies of electric energy production to a new direction in relation to electric energy management. The fact is that electric energy consumption is not constant and always fluctuates as a non-linear function of various time, environmental, economic parameters. The changes in electric load consumption have required electricity producing companies to predict the required information in different schedules for better energy management in power systems [9] and [13].

Increasing reliability and efficiency in the electricity industry, reducing costs and transaction costs in the electricity industry, providing more options for consumers and so on. As economic, social, and political motivations, the management of the electricity industry inevitably underwent a transformation, which is called the restructuring of the electricity industry. In the traditional environment, the production, transmission and distribution systems of electric energy are exclusively under the control of the government or a monopoly institution that has the task of monitoring the monopoly market and even monopoly prices. In unstructured power systems, consumers were unique distribution companies that received energy in bulk and according to specific rules from exclusive producers.These consumers received their energy only under the prices predetermined by the system operator, and the producers produced the energy required by the power grid only under the prices determined by the system operator under the traditional quantitative tariffs. In this type of economy, network losses are very high, so a lot of costs were wasted.

With the start of the restructuring of the electricity industry and the entry of competitive electric energy markets, the production, transmission and distribution parts were mostly assigned to private parties.

Although at the transmission level, this assignment is rarely done. Markets have been removed from government monopoly and operate independently as well as in coordination with the independent market operator.

In the distribution sector, by providing grounds for the participation of the private sector and the construction of power plants by domestic and foreign investors, as well as handing over the existing installed capacities and providing a basis for competition with the increase of producers, they tried to replace the competitive environment instead of monopolized conditions, which has been successful in many countries. A parallel network was established side by side in order to compete for electricity transmission. It is still managed in a monopoly manner and remains under the name of the national network. In the distribution sector, all activities from the point of receiving electricity from the transmission network to delivery to the final consumer are exclusive. First, the lines sector is separated from the subscriber sector, then this sector of lines is separated from the subscriber sector, so this sector, which is similar to the transmission sector, either remains in the possession of the government, or in some countries, it has been handed over to the private sector with different implementation approaches. In the customer service department, all matters from requesting and collecting claims and. It is done by retail companies. Since this sector has different conditions from the transmission and distribution sectors, the activities of this sector do not have exclusive activities, and therefore, retail companies can compete with each other for customer satisfaction at the same time, which will lead to quality improvement and cost reduction. Therefore, entrusting these matters to the private sector and creating a suitable environment for making this sector competitive has brought significant results. In the electricity economy, electricity producing companies are obliged to provide their consumers with high reliability, high quality, and reasonable prices, taking into account limitations such as environmental protection, with other partners in interconnected systems, taking into account limitations such as the power and type of existing power plants, the amount of fuel storage required by thermal power plants, and the amount of water in reservoirs. for the use of hydro power plants, etc. To achieve these goals, on the one hand, the equipment for power plants and transmission and distribution networks should be optimally used and exploited (minimum long-term investment) and on the other hand, the primary energies available for electricity production (types of fuels and water, etc.) should be optimally consumed [12]. From a technical point of view, the need for short-term load forecasting from a technical point of view can be summarized as follows: 1- Preparing a plan for using existing power plants, so that as much as possible, steam power plants are used as base load, hydro power plants are used for frequency control, and gas power plants are used to eliminate the lack of production during peak hours. Shortage

3- Preparation of the plan for leaving the units, the transmission lines of the transformers of the distribution network

4- Observance of the determined amount of energy of the water units, in this regard, due to the different amount of load at different hours of the day, it is possible to obtain the maximum efficiency from these power plants with the lowest cost by knowing the amount of water stored behind the dams in different months. Economic[8][12]

One of the future tasks of the distribution companies after starting the retail market is to buy the electric energy needed in their area from the wholesale or regional electricity market. The electric energy required by each company must be purchased from the electricity market, and the buyer must pay the relevant costs for the purchase of electricity, which include the requested power, consumed energy, and the penalty for failing the consumption test.