Name
Zueva Victoria Nicolaevna
Scholastic degree
•
Academic rank
associated professor
Honorary rank
—
Organization, job position
Kuban State Technological University
Web site url
—
Articles count: 2
In this work, we consider the design and development of
neural network software module for prediction of
electricity consumption in the system of support of
decision-making power control. Two prediction models
support the software module: regression model and
neural network model, based on multilayer perceptron.
Software development to predict power consumption in
the system of decision-making today is one of the priority
directions in the Russian power industry. Therefore, the
work associated with the development of methods and
algorithms of forecasting of power consumption in the
power sector, is surely relevant
The article discusses the use of regression methods of
forecasting the deterministic time series on the example
of the load curve. Forecasts of the load curve of
electrical equipment are the demands of consumers and
their security in EPS. All predictive tasks are based on
prediction models. Electricity consumption is happening
on an electronic level; storing electricity on an industrial
scale is impossible, the consumption depends on many
random factors. Therefore, generally, we use a
combination of mathematical and heuristic models. This
is the daily task of power systems and many technical,
economic and commercial decisions on the management
regimes depend on its solutions. Development of
methods of forecasting of the energy consumption in the
system of decision-making today is one of the priority
directions in the Russian power industry. Therefore, the
work associated with the development of methods and
algorithms of forecasting of power consumption in the
power sector is still relevant