Control of steam generation processes

A general-purpose control system for central heating plants is discussed. The controller structure includes the recently proposed generalized predictive control (GPC) algorithm and modern, distributed computing hardware. The simulations contrast the performance of the closed-loop system under the proportional-integral (PI) controller and the GPC self-tuning regulator. In the first two plots, the setpoint is held constant and a large step in steam demand applied. The GPC is more effective at rejecting this disturbance. Another pair of plots demonstrates the robustness of the GPC to changing plant dynamics, since the large step in the header pressure setpoint moves the nonlinear model to a new operating regime. The PI controller cannot provide good tracking at the new setpoint without retuning, and for larger setpoint changes, it loses stability, while GPC is still capable of maintaining it.