Anticipatory regulation of complex power systems
Abstract
Electric generation control is performed in a distributed manner to supply power to geographically defined control areas. The goal of generation control is to keep the inadvertent flow of power across a control area's boundary as small as possible. If a difference exists between the power supplied and the power demanded in a control area, the load deficit or surplus would be either borrowed from or stored as the kinetic energy in rotating machines on the grid. This thesis addresses the challenge of matching the power demand of a local area grid with the power delivered by a coal-fired power plant. An anticipatory controller for a model power plant is presented to prescribe the power output into the grid. The control system forecasts what the future demand of the power customers in a control area is likely to be and modifies the fuel input to the power generation facility in order to match the predicted demand. A neural network was found to be an adaptable and robust prediction mechanism for the highly nonlinear data found in the power consumption patterns in a residential area of the Commonwealth Edison grid. The corresponding control schedule of the power plant was tuned to match the anticipated demand using an iterative neural network approach. The use of neural networks and an iterative scheme allows the controller design in this research to be applied to a broad range of control problems. The control methodology presented takes into account limits in the magnitude and rate of control actions. Simulations show that this implementation of anticipatory control of electric power demand is effective and especially well suited to dynamic systems that include a dead time or control limitations. The response of the anticipatory neural network control system was shown to be more energy efficient than feedback control for a typical thermal power regulation facility and to have a much smoother, reduced control effort.
Degree
Ph.D.
Advisors
Tsoukalas, Purdue University.
Subject Area
Nuclear physics|Mechanical engineering|Energy
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.