You are here:
A Novel Model Predictive Control Scheme for Sustainability: Application to Biomass/Coal Co-gasification System
Citation:
Li, S., G. Ruiz-Mercado, AND F. Lima. A Novel Model Predictive Control Scheme for Sustainability: Application to Biomass/Coal Co-gasification System. 2018 AIChE Annual Meeting, Pittsburgh, PA, October 28 - November 02, 2018.
Impact/Purpose:
This abstract and presentation cover research done in CSS for novel process control methods for avoiding the generation of waste and improvements of chemical release profiles to be used in exposure assessments.
Description:
Industrial, business and government communities have begun to shift from economic stand-alone focus to inclusion of sustainability in the decision-making process. This shift is due to the adverse environmental impact and unsustainable development caused by human activities, including chemical industry releases. As a result, process systems engineering (PSE) approaches have been developed, especially for incorporating of sustainability into chemical process design and optimization [1]. However, the development of sustainability-oriented control schemes is scarce, especially when compared to the number of steady-state design and optimization studies available [2]. Process control plays a critical role in realizing the efficient, sustainable, and safe operation of the chemical and energy systems defined by the process design and optimization studies. The main barriers that has been prevented the integration of sustainability into process control are: 1) complexity of integrated large-scale chemical processes; 2) existence of multiple objectives for optimization involving social, economic, and environmental issues. To fill this gap, a novel sustainable model predictive control (MPC) strategy is proposed to drive the system to a sustainable operating point that is defined using a multi-objective optimization algorithm. In particular, the sustainable MPC is formulated based on dimensionless sustainability performance indicators of US EPA GREENSCOPE [3]. These established indicators are associated with the process state variables, releases, resource consumption, and products and can capture the sustainability information of the current process condition, including economic, environmental, and social aspects. Such sustainability indicators are used as hard/soft constraints in the implemented controller in order to maintain the process performance within a pre-defined sustainable zone where the sustainability performance indicator values are higher than desired thresholds. To explicitly visualize the multidimensional dynamic sustainability indicators, multivariate plotting method is developed by using of time-dependent radar diagrams. The developed method is illustrated via a biomass/coal co-gasification process for syngas production with the end goal of methanol manufacturing. For this application, the whole process model is developed in Aspen Hysys based on existing literature information [4-5]. With the established models in Aspen and a link for data communication between Aspen and MATLAB, a multi-objective optimization problem is solved to maximize profit and optimize the process sustainability performance (e.g., environmental release and resource use minimization), by employing a genetic algorithm-based approach developed in MATLAB. In this presentation, the details on the application results of this novel framework for improving sustainability performance are discussed, focusing on the tradeoffs between using coal and biomass for the sustainable production of chemicals. The results show that the proposed sustainable MPC can effectively drive the process to the optimal operating point, while maintain the process within sustainable zones during transients.