Elektrik sistemlerinde bölgesel yük tahmini

Abstract

ÖZET Bu tezde çağımızın en önemli kavramlarından biri olan enerji tüketiminin gelecek yıllarda ne boyutlara ulaşacağı araştırılmıştır. Gelecekteki enerji tüketim seviyesini belirlemek amacı ile yük kavramından yola çıkılmıştır. Enerji tahminleri yerine puant yük talep tahminleri üzerinde çalışılmış, yük tahmin yöntemleri Türkiye'nin en hızlı gelişen bölgesi olan İstanbul İli Avrupa Yakası üzerinde uygulanmıştır. 1996 yılında Türkiye çapında gerçekleşen 94800 GWh lık enerji tüketiminin 12253 GWh inin İstanbulda gerçekleşmiş olması, Türkiye'deki enerji tüketiminde İstanbul'un ne kadar büyük bir paya sahip olduğunu apaçık göstermektedir. Bu da İstanbul için enerji tüketim veya yük tahminlerinin yapılmasının gerekliliğini ortaya koymaktadır. Bu amaçla yola çıkarak İstanbul İli Avrupa Yakasının 1998-2010 yılları arası olası enerji ve puant yük taleplerini tahmin edilmiştir. Bunu yaparken de üç ayrı yaklaşımda bulunulmuştur. İlk yaklaşımda İstanbul Avrupa yakasının geçmiş yıllarda kaydedilen toplam puant yüklerinden yola çıkarak gelecekteki puant yükleri için tahmin yapılmıştır. İkinci yaklaşımda 154 kV/O.G. indirici transformatör merkezlerinin geçmiş yıllardaki puant yüklerinden faydalanarak gelecekteki yükleri tahmin edilmiş, buradan da bölge geneline ulaşılmaya çalışılmıştır. Üçüncü yaklaşımda ise sosyal etkenleri göz önüne alarak tahmin yapılmaya çalışılmıştır. Son olarak her üç yöntemle bulunan değerler ve geçmiş yıllarda yapılan yük tahminleri sonucunda elde edilen değerler karşılaştırılmıştır.

SUMMARY Power system expansion planning starts with a forecast of anticipated future load requirements. Estimates of both demand and energy requirements are crucial to effective system planning. Demand forecasts are used to determine the capacity of generation, transmission and distribution system additions, and energy forecasts determine the type of facilities required. Load forecasts are also used to establish procurement policies for construction capital. Further energy forecasts are needed to determine future fuel requirements. A good forecast reflecting current and future trends tempered with a good judgment, is the key all planning indeed to financial succes. The accuracy of a forecast is crucial to any electric utility since, it dictates timing and characterisrics of a major system additions. A forecast that is to low can easily result in lost revenue from sales to neighboring utilities or even in load curtailment.On the other hand, forecasts that are too high can result in severe financial problems due to excessive investment in an electric plant that is not fully utilized or, equiavently, is operated at low capacity factor. Capacity factor is defined as the ratio of avarage energy supplied to maximum energy capability. Unfortunately an accurate forecast depends on the judgment of the forecaster. Good judgment cannot be emphasized enough in forecasting future requirements. Loads may be classified broadly as residental, commercial and industrial, and other. Residental costumers use energy for domestic purposes, whereas commercial and industrial costumers obviously use energy for commercial and industrial purposes. Other customers are municipalities or divisions of goverment using energy for street lighting and etc.Of the three classes of loads, residental loads have most constant annual growth rate and the most seasonal fluctuations. The seasonal variations of the residental component in many cases are responsible for the seasonal variations in system peak. Commercial loads are also characterized by seasonal fluctuations. Industrial loads are considered base loads that contain little wheather -dependent variation. Other loads, may have seasonal fluctuations depending on specific cases. It is well documented that most system peak demands occur as a direct result of seasonal wheather extremes. In many utilities the winter peak has usually been the highest peak. In preparing a forecast, the system planner is immediately confronted with the fallowing basics questions: 1 -Should peak demand be forecasted using energy and load factors, or should it be forecast separetely ? 2-Should the total forecast be determined by combining forecasts of appropriate load components, or should the total load forecast be directly obtained from historical total load data ? 3-Should simple forecasting methods be used, or should more formal mathematical procedures be investigated ? The answers to these fimdemantal questions set the direction to be taken by the forecaster to determine future requirements. Probably every utility answers these questions differently, indicating that no one approach will be satisfactory in all cases. Depending on the intended use of a forecast the anwers to these questions will differ and rightly so. Energy and load forecasts are open to forecaster. The utililties have used the two basic approaches succesfully. The next question to be considered by a forecaster is whether to forecast the total load directly or to assemble the total load forecast by appropriately combining forecasts of certain componets. Components consist of types of consumers, xngeographic areas, etc. An advantage of total load approach is that it is easier to use and totals are much smoother and more indicative of overall growth trends. On the other hand, the advantage of component approach is that abnormal conditions in growth trends of a certain compenent can be detected, thus preventing misleading forecast conclusions. Choosing a forecasting technique to use in establishing future loads requirements is a notrivial task in itself. Depending on the nature of load variations, on a particular method may be superior to another. Before choosing a particular method, abasic understanding of how a load behaves is essential. Forecasting is simply a systematic procedure for quantitatively defining future loads. Depending on the time period of interest, a specific forecasting procedure may be classified as a short- term, intermediate, or long term technique. In forecasting we can use extrapolation tecniques involve fitting trend curves to basic historical data adjusted to reflect the growth trend itself. With a trend curve the forecast is obtained by evaluating the trend curve function at the desired future point. Although a very simple procedure it produce reasonable results in some instance.Standart analytical functions are used ti trend curve fitting including: Y=a+bx Y=a+bx+cx2 Y= a+bx+cx2+dx3 Y=A(l+p)exp(x) Y=A.e.exp(ax) xmThe most common curve fitting technique for finding coefficients of a function is the metod of least squares. Regression analysis the best estimate of the model describing the trend can be obtained and used to forecast the trend.this approach is advantageous in forecing the forecaster to understand clearly the interrelationship between load growth patterns and the other factors. No one forecasting metod, it must be emphasized, is effective in all situations. The use of simple curve fitting tecniques is adequate for some utilities, and completely worthless for others. In any case, forecasting techniques must be used as tool to aid the planner; good judgment and experience can never be completely replaced. To obtain a regional forecast trend tecniques are applied to The European Part of Istanbul. To perform a succesful forecast, rigth historical data must be used in forecast. Because of this reason, simultaneous peak demand values of 154 kV/M.V. Transformer Units have been used in forecast. Istanbul load growth was studied with three approaches. In first approach total regional load growth is obtain from the historical peak demand data and future loads have been determined. In second approach total regional load was obtained from the annual peak demand data of the 154 kV/M.V. transformer units. In third approach total region load was obtained from the relationship between demographic factors and load growth. The results are shown in table XIVTo obtain the future loads of Istanbul T.E.K. and TÜBİTAK has performed forecasts and the result of these shown in table. XV