TEMPERATURE DEPENDENCY OF ELECTRICAL NETWORK LOAD

Abstract

The dependency on outdoor temperature is one of the most important characteristics of load. First of all, this dependency is expressed there, where electrical heating or different air conditioning devices are used. For example, in Lapland (Finland), where electrical heating is of great importance and deviations in outdoor temperature are large, the load increase caused by outdoor temperature may be up to 100% compared to the load at normal temperature. Usually, however, the temperature dependency of load is smaller, especially in the case of industrial load. The dependency of electrical network load on temperature and other weather factors is widely been covered in different papers. In [1] a regression-based adaptive weather-sensitive short-term load forecasting approach has been presented. Proposed algorithm is robust and not very sensitive to weather forecast accuracy considering only the data having a great influence on consumption, e.g. only minimal, maximal and average daily temperatures and average daily wind speed are used. Also the extent and inertia of the temperature influence are considered. Weather impact on load diversity is covered in [2]. To relate critical points in the load curve to weather factors, a multiple regression was applied. It is proposed that daily maximum temperatures are the dominant weather factors, followed by humidity at times of maximum temperature, daily minimum temperature, wind speed and hours of sunshine. When considering the temperature, the effect of the previous day temperature is important [3]. It is indicated that