Abstract
DEVELOPMENT OF DEEP LEARNING AND MODEL PRODUCTIVE CONTROL TO ENHANCE AUTOMATIC GENERATION CONTROL (AGC) IN INTERCONNECTED POWER SYSTEMS
Agwu Emmanuel Anayo*, Rufus Ogbuka Chime and I. I. Eneh
ABSTRACT
Every Country needs efficient, effective and balanced power system for the smooth running of its economy. To maintain power balance in the system, generation and load need to balance in order to uphold system frequency at the nominal value. Oscillation in the power system resulting from AGC problems could cause serious damages to generation and transmission assets on the power system. Cascadedcollapse of the system from AGC problem could lead to a wide scale blackout. Such wide scale blackouts would impact on the country’s Gross Domestic Product (GDP) as it is well known that electric energy usage impacts on and is a reflection of a country’s GDP. Control is the key component in this regard. This is where Automatic Generation Control (AGC) comes into play. Automatic Generation Control in an electric power system, is the control system for regulating the power output of many generators at different power plants, in response to changes in the load. The balance can be determined by measuring the system frequency. The key objective of this work is development of deep Learning model predictive control MPC to enhance Automatic Generation Control (AGC) in interconnected power system and regulating tie-line power exchange for enhanced power balance in the controlled areas. To overcome the above limitations, this study proposes the use of software to enhance the effectiveness of RNN-MPC model in AGC operations.
[Full Text Article] [Download Certificate]