No abstract

No abstract

We offer a neural network model for forecasting the next day's hourly electric load of a city. We use a few ambient temperature account methods in the research to see how each of them affects the forecasting accuracy. Optimal meta parameters are determined to tune the neural network to give best forecasts. Among such meta parameters are the data history depth, data seasonality radius and regularization parameter of neural network weights. A multilayer perceptron is used to make forecasts. It is shown that the electric load can be forecasted most accurately when an additional neural network forecasts hourly ambient temperatures using actual hourly temperatures of the previous day and the weather station's temperature predictions for the forecast day.

No abstract

The results are presented of the analysis of the daily average values graphs for the time series of active power, air temperature, natural illumination, cloudiness and precipitation for Moscow. The method of one-dimensional singular spectral analysis were obtained the results of the dependence of active power on air temperature and natural illumination and the dependence of active power on cloudiness, precipitation and illumination. Mathematical model was proposed of one-dimensional singular spectral analysis of time series of natural illumination, cloudiness and precipitation is proposed to improve the results of forecasting active power. The results are presented of the decomposition of active power and meteorological factors for 2019 year. To implement the mathematical model of one-dimensional singular spectral analysis of time series of active power and meteorological factors was used the developed algorithm in the R- software Studio in language R. Forecasts daily graphs of active power were made using predictive data of natural illumination obtained using a forecast model based on a neural network during February 2019.The proposed version of the one-dimensional singular spectral decomposition allows finding the relationship between the selected harmonic components of power consumption and natural illumination and power consumption and air temperature. For analyzing and transforming the time series of meteorological factors were used the data of the meteorological hardware-software complex PAC "Meteo".