TARGETING STRATEGIES FOR THE SYNTHESIS AND ENERGY INTEGRATION OF NONISOTHERMAL REACTOR NETWORKS

In this paper, a constructive approach for the synthesis of nonisothermal reactor networks is presented based on a targeting methodology. Here, we determine a bound or target on a performance index for a given nonisothermal reacting system. The targeting model is based on mixing between different reacting environments and is formulated as a dynamic optimization problem, where the temperature, the feed distribution function, and an exit flow distribution function are the control profiles. In addition, the solution procedure is based on successive generation of reactor extensions that improve the target. Each reactor extension corresponds to the solution of a small nonlinear program; this procedure is repeated until no extensions that improve the objective function are generated. This technique ensures that only the simplest required model is solved. Having developed a general reactor targeting scheme, we also propose a new framework for integrating this scheme with an energy targeting approach. By discretizing the temperature profiles and allowing for heating or cooling profiles within the reactor, a combined representation for reactor and energy targeting is obtained. This model tums out to be a nondifferentiable nonlinear program, and novel smoothing techniques are presented for solution. Results on a typical process flowsheet optimization problem indicate significant increases in profit can be achieved by considering the reaction and energy synthesis schemes simultaneously.