BYPASS SELECTION FOR CONTROL OF HEAT-EXCHANGER NETWORKS

Decisions made during steady-state process design may put severe limitations on the achievable control performance or controllability of chemical plants. We show how different controllability measures may be used to evaluate controllability and select bypasses and appropriate pairings of heat exchanger networks. Flow rate dependence of beat transfer coefficients is included in the model. This is found to have a significant effect on control. Most of our results confirm good engineering practice: Prefer designs and bypass selections where all critical targets are controlled by either utility streams or bypasses with a direct effect. Consider bypasses over several exchangers in series as such total bypasses may enhance the controllability or the flexibility without having to install more control loops. Avoid bypass selections with two or more downstream paths to one critical target and designs where both output streams of one heat exchanger are critical targets.