Optimal flowsheet sensitivity in a sensitivity oriented environment

Abstract The optimal solution of flowsheet optimization problems usually results from deterministic specifications. However, several process parameters may be uncertain or subject to variations. An additional analysis enables us to quantify the trends in changing the optimal solution. An efficient algorithm for the evaluation of the parametric sensitivities of optimal flowsheets is described. Its implementation in a simultaneous-modular, steady-state simulator, which can generate analytically the first order information needed, is presented. The use of a range and null space decomposition of the problem, and of the sensitivity features of the simulator reduces drastically the computational cost as well as increases the quality of the results. The method is applied to an example and the results are compared with those of a numerical approach, as well as with the method without decomposition, with and without use of the simulator's sensitivity generation capabilities.