The ethanol production has become an important part of the worldwide economy driven by its use as renewable energy and environmentally clean fuel. To obtain a high-level quality product, a large quantity of energy is used in the distillation stage. This work proposes to optimize the bioethanol production process applying a closed-loop dynamic real-time optimization (CL-DRTO) framework associated with advanced control strategies in the ethanol distillation facilities to improve production and minimize energy losses. A high-fidelity computational platform is developed using a software-in-the-loop (SIL) technique to demonstrate the feasible application in real cases. OLE (Object Linking and Embedding) Automation integrates Matlab and Aspen Hysys software to simulate practical scenarios currently applied in the industry. The optimization and control algorithms are developed in Matlab and the ethanol distillation process is modeled in the Aspen Hysys. Different advanced control strategies, such as IHMPC (Infinite Horizon Model Predictive Control), MIMO FSP (Filtered Smith-Predictor) and DTCGPC (Dead-Time Compensator Generalized Predictive Controller), are used to overcome the process dynamic issues, for instance, strong nonlinearity, dead-times, long time constants and coupled loops, on ethanol distillation process. A CL-DRTO layer is developed in the proposed control structure to consider economic and production aspects, guiding the process to optimal operating points. Performance indices and computational effort are used to evaluate the control behavior for disturbance rejection and reference tracking scenarios. Furthermore, economic performance is analyzed to ensure the advantages of the proposal. Results show that the proposed computational platform is able to reproduce industrial scenarios with high-fidelity. In addition, preliminary results demonstrate that advanced control structure can improve the production and profitability of the bioethanol distillery and present a powerful alternative to replace classical controllers currently used in this type of industry.

乙醇生产已成为可再生能源和环境清洁燃料，已成为世界经济的重要组成部分。为了获得高质量的产品，在蒸馏阶段要消耗大量的能量。这项工作建议使用闭环动态实时优化（CL-DRTO）框架与乙醇蒸馏设备中的高级控制策略相关联来优化生物乙醇生产工艺，以提高生产效率并最大程度地减少能量损失。利用软件在环（SIL）技术开发了一种高保真计算平台，以演示在实际案例中的可行性应用。OLE（对象链接和嵌入）自动化集成了Matlab和Aspen Hysys软件，以模拟当前在行业中应用的实际方案。在Matlab中开发了优化和控制算法，并在Aspen Hysys中对乙醇蒸馏过程进行了建模。为了克服过程动态问题（例如强非线性），使用了不同的高级控制策略，例如IHMPC（无限远景模型预测控制），MIMO FSP（Filtered Smith-Predictor）和DTCGPC（Dead-Time Compensator Generalized Predictive Controller）。乙醇蒸馏过程中的停滞时间，长时间常数和耦合回路。在建议的控制结构中开发了一个CL-DRTO层，以考虑经济和生产方面，将过程引导至最佳操作点。性能指标和计算量可用于评估干扰抑制和参考跟踪方案的控制行为。此外，分析经济绩效以确保该建议的优势。结果表明，所提出的计算平台能够高保真地再现工业场景。此外，初步结果表明，先进的控制结构可以提高生物乙醇蒸馏厂的生产能力和盈利能力，并提供有力的替代品来替代目前在此类行业中使用的经典控制器。