Optimal operation of heat exchanger networks based on structural information

A procedure for optimal operation of heat exchanger networks is discussed. For each possible bypass location, a change in the bypass fraction will have a monotonously increasing or decreasing effect on the outlet temperatures or utility duties. The signs of these effects are stored in two sign matrices which contain sufficient information to operate several classes of heat exchanger networks optimally, that is, to maintain the target temperatures while utility consumption at the same time is minimized. The method is used to pair inputs (bypass fractions) and outputs (target temperatures) in a decentralized control scheme with variable structure. Results based on simulations show that the method performs better, in terms of reduced utility consumption, than ordinary decentralized feed-back control.