Analysis of power consumption presents a very important issue for power distribution system operators. Some power system processes such as planning, demand forecasting, development, etc.., require a complete understanding of behaviour of power consumption for observed area , which requires appropriate techniques for analysis of available data. In this paper, two different time-frequency techniques are applied for analysis of hourly values of active an d reactive power consumption from one real power distribution transformer substation in urban part of Sarajevo city . Using the continuous wavelet transform (CWT) with wavelet power spectrum and global wavelet spectrum some properties of analysed time series are determined. Then, empirical mode decomposition (EMD) and Hilbert -Huang Transform (HHT) are applied for the analyses of the same time series and the results show ed that both applied approaches can provide very useful information about the behaviour of power consumption for observed time interval and different period (frequency) bands. Also it can be noticed that the results obtained by global wavelet spectrum and marginal Hilbert spectrum are very similar, thus confirming that both approaches could be used for identification of main properties of active and reactive power consumption time series .

Electricity demand forecasting is one of the most important components in the power system analysis. Furthermore, it is difficult and complicated process to forecast energy consumption. This study deals with modeling of the electrical energy consumption in Bosnia and Herzegovina in order to forecast future consumption of electrical loads based on temperature variables using machine learning methods. We used three different  machine learning methods for analyzing short term forecasting. The methods were trained using historical load data, collected from JP Elektroprivreda electrical power utility in BiH, and also considering weather data which is known to have a big impact on the use of electric power. Comparing the results it was seen that prediction for 500 hours is pretty good in range from 92,92% for reactive power till 98.84% for active power. Four different parameters were analyzed mean absolute error, root mean squared error, relative absolute error and root relative square error. The best results for apparent power were gotten with linear regression and are presented as for mean absolute error 9.84, root mean squared error 13.62, relative absolute error 14.06%, root relative squared error 14.39%. It is also seen from the results that,  the short term power consumption can be predicted which is important for maintaining of the voltage at the consumer side.