Time Series Forecasting via a Higher Order Neural Network trained with the Extended Kalman Filter for Smart Grid Applications

Time Series Forecasting via a Higher Order Neural Network trained with the Extended Kalman Filter for Smart Grid Applications

Luis J. Ricalde (Universidad Autonoma de Yucatan, Mexico), Glendy A. Catzin (Universidad Autonoma de Yucatan, Mexico), Alma Y. Alanis (Universidad de Guadalajara, Mexico) and Edgar N. Sanchez (Universidad de Guadalajara, Mexico)
DOI: 10.4018/978-1-4666-2175-6.ch012
OnDemand PDF Download:
List Price: $37.50


This chapter presents the design of a neural network that combines higher order terms in its input layer and an Extended Kalman Filter (EKF)-based algorithm for its training. The neural network-based scheme is defined as a Higher Order Neural Network (HONN), and its applicability is illustrated by means of time series forecasting for three important variables present in smart grids: Electric Load Demand (ELD), Wind Speed (WS), and Wind Energy Generation (WEG). The proposed model is trained and tested using real data values taken from a microgrid system in the UADY School of Engineering. The length of the regression vector is determined via the Lipschitz quotients methodology.
Chapter Preview

1. Introduction

In recent years, there has been an increasing attention on renewable energy generation and management due to the environmental and economic concerns. The first steps on integrating renewable energy sources began with hybrid wind and solar systems as complementing sources and as solution for rural applications and weak grid interconnections. Further research have implemented hybrid systems including several small scale renewable energy sources as solar thermal, biomass, fuel cells and tidal power. Since the production costs for photovoltaic and wind turbine applications have considerably reduced, they have become the primary choice for hybrid energy generation systems. The future of energy production is headed towards the scheme of integration of renewable energy sources with existing conventional generation systems (coal, natural gas, oil). This integration is defined as a smart grid. This new scheme increases the power quality since the production becomes decentralized and is the main reason for which Institutions have increased the research on this concept (Meiqin, et al., 2008). Microgrids integrate small-scale energy generation systems mainly from renewable energy and implementing complex control technologies to improve the flexibility and reliability of the power system (Wu, et al., 2009). The smart grids can be connected to the public network to support it or can be operated autonomously. The design of these systems integrates a distributed power generation system and a management unit composed of a communication network, which monitors and controls the interconnection between energy sources, storage devices and electrical loads with the aid of sensors to enhance the overall functionality of the electric power delivery system (Gellings, 2009).

Among renewable energy sources, wind energy is the one with the lowest cost of electricity production (Welch, et al., 2009), for this reason the interest in the study of wind energy as a source of electricity production is increased. However, integration of the wind energy into the grid of supply of electric energy is a real challenge mainly due to that the generated energy depends on meteorological conditions (wind speed, particularly) present at certain moment and those cannot be modified by human intervention (Lange, 2003); therefore it is necessary to have an adequate intelligent control to accomplish that integration (Werbos, 2009). To utilize this renewable energy in a large scale without fierce impact to the grid, in addition to apply energy storage to regular wind power output, accurate wind power output forecasting is an effective mean (Shi, et al., 2011). The forecast of wind power is a good way to improve the energy management of a wind power farm (Wu, et al., 2009). Integration of wind resource into the electricity transmission systems has as direct consequence environmental benefits because carbon dioxide emissions derived from electricity generation by traditional media are decreased; therefore, our planet is preserved for future generations.

Wind power prediction is also an essential process for (El-Fouly, et al., 2008):

  • 1.

    Wind farms units maintenance.

  • 2.

    Optimal power flow between conventional units and wind farms.

  • 3.

    Electricity marketing bidding.

  • 4.

    Power system generators scheduling.

  • 5.

    Energy reserves and storages planning and scheduling.

Many factors affect wind power prediction, including the presence of an accurate forecasting model for wind speed and direction at the farm site, the presence of accurate technique for wind speed simulation all over the farm layout and the existence of sufficient information about the farm characteristic and layout. The required forecasting horizon for wind farms output power depends on the required application, which can be unit maintenance, control, small power systems operation, interconnected power systems operation and maintenance planning (El-Fouly, et al., 2008). As for wind speed prediction, previous work has shown temperature is the most important meteorological parameter. Humidity and current wind speed have also been identified as key indicators (Welch, et al., 2009).

Complete Chapter List

Search this Book: