FRS-based decision table reduction for the operation optimization of large coal-fired power units

Abstract

Large coal-fired power unit is a complex nonlinear system with more uncertainties to describe, evaluate and optimize. It is essential and difficult to determine the optimal targets in operation optimization of power units, especially considering the boundary constraints, operation conditions and system features. Fuzzy rough set (FRS)-based decision table reduction was introduced to clean the historian operation data efficiently without information losses. The result shows that the derived energy consumption decision rules can be used to determine the optimal targets quickly and dynamically for different boundary and operation conditions. It makes significant reference and promising prospects in energy-consumption diagnosis and operation optimization of power units.