Data Reconciliation for Energy System Flowsheets

Abstract Data reconciliation is a key step of data extraction from existing plants. While there are many publications on data reconciliation generally, for Heat Integration Analysis (HIA) they are scarce. Reconciliation also determines the result. For HIA reconciling data for an individual heat exchanger is insufficient and incorrect. In this work complete heat exchanger networks are considered within the data reconciliation scope. Two methods are compared: i) iterative method using local NLP and ii) simultaneous method applying global NLP. The iterative method aims at reducing the complexity of a problem by decomposing the problem into two steps. In the first step the temperatures are reconciled and in the second the mass flow-rates are reconciled at fixed (reconciled) temperatures obtained during the first step. The simultaneous models reconcile the mass flow-rates and temperatures of the HEN within one step. The results indicated that the simultaneous approach achieves better accuracy and should be preferred when accuracy is the main focus but the iterative approach is simpler and still a feasible approach to be used when computational complexity is an issue.