Integrating the energy available in industrial processes in the form of heat and work is fundamental to achieve higher energy efficiencies as well as to reduce process costs and environmental impacts. To perform this integration, a new framework for the optimal synthesis of work and heat exchange networks (WHEN) aiming to reduce capital and operating costs is presented. The main contribution of this paper is the elaboration of a new WHEN superstructure and mixed-integer nonlinear programming (MINLP) derived model. Strategies of changing variables are applied to reduce the number of decision variables from the model. The MINLP problem with a reduced number of decision variables is solved with a two-level meta-heuristic optimization approach, using Simulated Annealing in the combinatorial problem and Particle Swarm Optimization in the nonlinear programming problem. For the sake of validation, this methodology is applied to three case studies comprising two, five, and six process streams. Economic savings achieved outperform results reported in the literature from 1.0 to 7.2%. Also, the solutions obtained present non-intuitive WHENs that shows the importance of using superstructure-based mathematical programming for such a difficult decision-making task.

以热和功的形式集成工业过程中的可用能量，对于实现更高的能源效率以及降低过程成本和环境影响至关重要。为了执行此集成，提出了一种用于工作和热交换网络（WHEN）最佳综合的新框架，旨在减少资本和运营成本。本文的主要贡献是阐述了一种新的WHEN上层建筑和混合整数非线性规划（MINLP）衍生模型。应用更改变量的策略来减少模型中决策变量的数量。通过两级元启发式优化方法解决了决策变量数量减少的MINLP问题，在组合问题中使用模拟退火，在非线性规划问题中使用粒子群优化。为了验证起见，此方法论应用于包括两个，五个和六个过程流的三个案例研究。经济节省的表现优于文献报道的1.0％至7.2％。此外，获得的解决方案还存在非直观的WHEN，这表明使用基于上层建筑的数学编程进行如此困难的决策任务非常重要。