Adaptive fuzzy temperature control for hydronic heating systems

In a hydronic heating system, a boiler heats hot water that is pumped through pipes to radiators throughout the dwelling. This article describes an alternative to a traditional PI algorithm for hydronic heating system control. The control algorithm described is an adaptive fuzzy control (AFC) algorithm. It compactly combines a unique adaptation algorithm that automatically learns the steady-state radiator valve positions for different operating set points with a simple rule-based fuzzy controller. Laboratory test results comparing the performance of the AFC with a variable-gain PI algorithm demonstrate that the AFC combines better temperature control with dramatically reduced battery consumption.