All ABET-accredited engineering programs mandate a culminating major design experience based on knowledge and skills acquired in earlier course work and incorporating realistic appropriate engineering standards and multiple realistic constraints. Some chemical companies organize their Manufacturing Innovation Process into a sequence of stages which typically include Need Identification, Product Design, Basic and Detailed Chemistry, Process Design, Equipment Design, Plant Design, Detailed Engineering and Vendor Specifications, Component Acquisition, Plant Construction Planning and Execution, Operating Procedure Development, Plant Commissioning and Start-up, and Production Planning, Scheduling, and Operation. Each of these stages involve the solution of many "design" problems that could be the subject of the culminating undergraduate chemical engineering design experience... (ABSTRACT ABBREVIATED)

The overarching goal of process design (Figure 1) is to find technologically feasible, operable, economically attractive, safe and sustainable processing pathways and process configurations with specifications for the connectivity and design of unit operations that perform a set of tasks using selected functional materials (e.g., catalysts, solvents, sorbents, etc.) to convert a set of feed-stocks or raw materials into a set of products with desired quality at a scale that satisfies the demand. Process synthesis and integration can further screen, optimize and improve these pathways for given techno-econo-environmental targets or objectives. These objectives may include, but are not limited to, minimizing the overall investment and processing costs, minimizing the energy consumption, minimizing the emissions or wastes, maxim-zing the profit, and enhancing the safety, operability, controllability, flexibility, circularity, and sustainability, among others... (ABSTRACT ABBREVIATED)

In this paper, we study the design optimization of methanol production with the goal of minimizing methanol production cost. One challenge of methanol production via carbon dioxide (CO2) hydrogenation is the reduction of operating costs. The simulation of methanol production is implemented within the Aspen HYSYS simulator. The feeds are pure hydrogen and captured CO2. The process simulation involves a single reactor and incorporates recycling at a ratio of 0.995. The methanol production cost is determined using an economic analysis. The cost includes capital and operating costs, which are determined through the equations and data from the capital equipment-costing program. The decision variables are the pressure and temperature of the reactor contents. The optimization problem is solved using a derivative-free algorithm, pyBOUND, a Python-based black-box model optimization algorithm that uses random forests (RFs) and multivariate adaptive regression splines (MARS). The predicted minimu... [more]

Initial design stages are inherently complex and often lack comprehensive information, posing challenges in evaluating sustainability metrics. Machine Learning (ML) emerges as a valuable solution to address these challenges. ML algorithms, particularly effective in predicting environmental impacts of new chemicals with limited data, enable more informed decisions in sustainable design. This study focuses on employing ML for predicting the environmental impacts related to human health, ecosystem quality, climate change, and resource utilization to aid in early-stage environmental impact assessment of chemical processes. The effectiveness of the ML algorithm, eXtreme Gradient Boosting (XGBoost) tested using a dataset of 350 points, divided into training, testing, and validation sets. The study also includes a practical application of the model in a cradle-to-cradle LCA of N-Methylpyrrolidone (NMP), demonstrating its utility in sustainable chemical process design. This approach signifies... [more]

Biogas is often considered as a source of renewable energy, for heat and power production. However, biogas has greater promise as a source of concentrated CO2 in addition to methane, making it a rich supply of carbon and hydrogen for the generation of fuel and chemicals. In this work, we use the concept of attainable region in the enthalpy-Gibbs free energy space to identify opportunities for effective biogas valorization that maximizes the conversion of CO2. The AR concept allows us to study a chemical process without knowing the exact reaction mechanism that the species in the process use. Deriving Material Balance equations that relate a reactive process's output species to its input species is sufficient to identify process limits and explore opportunities to optimize its performance in terms of material, energy, and work. The conversion of biogas to valuable products is currently done in two steps; the high temperature and endothermic reformer step, followed by the low temperatur... [more]

Hydrogen, as a clean and versatile energy carrier, holds immense promise for addressing the world’s growing energy and environmental challenges. However, hydrogen-based energy systems face challenges related to efficient storage methods, energy-intensive production, refueling processes, and overall cost-effectiveness. To solve this problem, a superstructure was developed that integrates overall technologies related to hydrogen energy transportation. This study synthesizes process pathways for hydrogen energy transportation method including energy carrier production, storage, and refueling, based on the developed superstructure. The techno-economic analysis was conducted to evaluate the performance of each transportation pathway and compare it with conventional fossil fuel transportation system. Process performance criteria, including unit production cost (UPC), energy efficiency (EEF), and net CO2 equivalent emissions (NCE), serve as indicators for process performance. By comparing tec... [more]

In this work, we present a follow-up work of reinforcement learning (RL)-driven process design using the Institute for Design of Advanced Energy Systems Process Systems Engineering (IDAES-PSE) Framework. Herein, process designs are generated as stream inlet-outlet matrices and optimized using the IDAES platform, the objective function value of which is the reward to RL agent. Deep Q-Network is employed as the RL agent including a series of convolutional neural network layers and fully connected layers to compute the actions of adding or removing any stream connections, thus creating a new process design. The process design is then informed back to the RL agent to refine its learning. The iteration continues until the maximum number of steps is reached with feasible process designs generated. To further expedite the RL search of the design space which can comprise the selection of any candidate unit(s) with arbitrary stream connections, we investigate the role of RL reward function and... [more]

In this paper, a hybrid heuristic rule-based and deterministic optimization-driven process decision framework is presented for the analysis and optimization of process flowsheets for end-to-end optimal (E2E0) pharmaceutical manufacturing. The framework accommodates various operating modes, such as batch, semi-batch and continuous, for the different unit operations that implement each manufacturing step. To address the challenges associated with solving process synthesis problems using a simulation-optimization approach, heuristic-based process synthesis rules are employed to facilitate the reduction of the superstructure into smaller sub-structures that can be more readily optimized. The practical application of the framework is demonstrated through a case study involving the end-to-end continuous manufacturing of an anti-cancer drug, lomustine. Alternative flowsheet structures are evaluated in terms of the sustainability metric, E-factor while ensuring compliance with the required pro... [more]

Superstructure optimization for process synthesis is a challenging endeavour typically leading to large scale MINLP formulations. By the combination of phenomena-based building blocks, accurate thermodynamics, and structural screening we obtain a new framework for optimal process synthesis, which overcomes prior limitations regarding solution by deterministic MINLP solvers in combination with accurate thermodynamics. This is facilitated by MOSAICmodeling’s generic formulation of models in MathML / XML and subsequent decomposition and code export to GAMS and C++. A branch & bound algorithm is implemented to solve the overall MINLP problem, wherein the structural screening penalizes instances, which are deemed nonsensical and should not be further pursued. The general capabilities of this approach are shown for the distillation-based separation of a ternary system.

We present new modifications to superstructure optimization paradigms to i) enable their robust solution and ii) extend their applicability. Superstructure optimization of chemical process flowsheets on the basis of rigorous and detailed models of the various unit operations, such as in the state operator network (SON) paradigm, is prone to non-convergence. A key challenge in this optimization-based approach is that when process units are deselected from a superstructure flowsheet, the constraints that represent the deselected process unit can be numerically singular (e.g., divide by zero, logarithm of zero and rank-deficient Jacobian). In this paper, we build upon the recently-proposed modified state operator network (MSON) that systematically eliminates singularities due to unit deselection and is equally applicable to the context of both simulation-based and equation-oriented optimization. A key drawback of the MSON is that it is only applicable to the design of isobaric flowsheets... [more]

This work presents an overview of the path towards the use of renewable and nonconventional resources for a sustainable chemical and process industry. The aim is not only to lead the way to meet the sustainable development goals but also to maintain the style and quality of life achieved by the technologies and products developed within this sector. Alternative raw materials are to be used and processed differently while a new paradigm for utilities is to be established. The development of technologies and their deployment faces several barriers that we as process engineers can help overcome by providing insight into the alternatives, the thresholds to achieve to become competitive, and strategic analyses.

In designing low-carbon processes, the unintended emission of CO2 remains a significant concern due to its global environmental impact. This paper explores carbon management within chemical processes, specifically examining the correlation between the process material balance (PMB) and CO2 emissions to understand and identify the potential for reducing these emissions. We interrogate the foundational issue of carbon discharge by analyzing the interplay among mass, energy, and entropy balances, which collectively influence the PMB. We introduce the concept of the Target Material Balance (TMB), which represents the material balance of a process corresponding to minimum CO2 emissions within the given constraints. We could ask what decisions in the design and operation of processes result in higher CO2 emissions than the TMB. We will focus on the interaction between reactions and recycles and how the arrangement of recycles in processes can inadvertently change the PMB, thereby increasing... [more]

Following the discovery of the least squares method in 1805 by Legendre and later in 1809 by Gauss, surrogate modeling and machine learning have come a long way. From identifying patterns and trends in process data to predictive modeling, optimization, fault detection, reaction network discovery, and process operations, machine learning became an integral part of all aspects of process design and process systems engineering. This is enabled, at the same time necessitated, by the vast amounts of data that are readily available from processes, increased digitalization, automation, increasing computation power, and simulation software that can model complex phenomena that span over several temporal and spatial scales. Although this paper is not a comprehensive review, it gives an overview of the recent history of machine learning models that we use every day and how they shaped process design problems from the recent advances to the exploration of their prospects.

Sustainability of the chemical and materials industry (CMI) requires it to achieve net-zero emis-sions of greenhouse gases and other resources while making decisions that have a net-positive impact on nature and society. Many corporations, nations, and universities have pledged to meet such goals but systematic models, methods, and tools to guide this transition are missing. We pre-sent a framework to meet this need. It involves developing a comprehensive, open access model of the global CMI. In addition to existing technologies, this model includes emerging alternatives for renewable energy, circularization, and carbon capture, utilization and storage. Systematic methods help identify innovation opportunities and develop roadmaps that account for long-term changes such as technology evolution and climate change. Meeting the goal of net-zero emis-sions requires inclusion of life cycle impacts. Nature-positive decisions need to encourage eco-logical protection and restoration. Thi... [more]

The development of a bio-based economy has its foundations in the development of efficient processes to optimize biomass potential. In this context there are a multitude of molecules that can be either synthetized or recovered from biomass, among those the so-called 12 building-blocks reported by the US Department of Energy. Even if their identification and importance is clearly defined, research efforts concerning the purification or separation of these platform molecules are limited. To fill this gap, different configurations for the purification of levulinic acid recovered from lignocellulosic biomass are examined and compared in this work. In particular, hybrid configurations obtained by the combination of liquid-liquid extraction and distillation have been considered. It was demonstrated how a deep understanding of the subspace including all extraction-assisted simple column distillation configurations represents a fundamental step in the synthesis of different process alternative... [more]

Process Intensification (PI) is a vast and growing area in Chemical Engineering, which deals with the enhancement of current technology to enable improved efficiency; energy, cost, and environmental impact reduction; small size; and better integration with the other equipment. Since process intensification results in novel, but complex, systems, it is necessary to rely on optimization and control techniques that can cope with such new processes. Therefore, this review presents some advancements in the field of process intensification that are worthy of exploring in detail in the coming years. At the end, several important open questions that can be taken into consideration in the coming years are listed.

Advice and best practices for teaching conceptual process flowsheeting, simulation, and design at the third year undergraduate level. Discusses setting course goals, integration with the rest of the curriculum, and delivery techniques. Practical strategies for tutorials, exams, lectures, and projects. Training TAs for experiential learning workshops. Best practices in teaching distillation design. This is the Award Lecture for AIChE's David Himmelblau Award for Innovations in Computer-Based Chemical Engineering Education. Live lecture given via APMonitor.com as a part of the AIChE's Computing and Systems Technology division webinar series.

Due to the expanding concern on cleaner production and sustainable development aspects, a technology shift is needed for the hydrogen production, which is commonly derived from natural gas. This work aims to synthesise a large-scale bio-hydrogen network in which its feedstock, i.e., bio-methane, is originated from landfill gas and palm oil mill effluent (POME). Landfill gas goes through a biogas upgrader where high-purity bio-methane is produced, while POME is converted to bio-methane using anaerobic digestor (AD). The generated bio-methane is then distributed to the corresponding hydrogen sink (e.g., oil refinery) through pipelines, and subsequently converted into hydrogen via steam methane reforming (SMR) process. In this work, P-graph framework is used to determine a supply network with minimum cost, while ensuring the hydrogen demands are satisfied. Two case studies in the West and East Coasts of Peninsular Malaysia are used to illustrate the feasibility of the proposed model. In C... [more]

The attainable region (AR) is a region in a performance space in which all physically realizable reactor network designs must exist. ARs have been used since the 1960s for solving reactor network synthesis problems. The benefits of these methods are that the feasibility of a performance target can be assessed prior to running a synthesis routine, the solutions they give are guaranteed to be physically realizable, and a design can be made robust to uncertainties in feed and performance targets by assessing whether a solution and the range of its possible values lie within the AR, just to name a few. In this article, the theory of attainable regions is extended to bring these benefits to the heat exchanger network (HEN) synthesis problem. Basic properties of the HEN-AR are proven and a synthesis method using the AR is presented with examples.

In this talk, I give an overview of the chemical engineering curriculum at McMaster University as it relates to the 1.5 year process design sequence. The courses outside the design sequence were recently restructured and redesigned to create an environment with more modelling and algorithmic thinking/algorithmic problem solving. This includes a statistics course and a big data / machine learning course. The end result is that the design sequence is able to focus on state of the art tools and methods for process design because students receive many fundamental principles before the design sequence begins.

Artificial neural networks (ANNs) are used in various applications for data-driven black-box modeling and subsequent optimization. Herein, we present an efficient method for deterministic global optimization of ANN embedded optimization problems. The proposed method is based on relaxations of algorithms using McCormick relaxations in a reduced-space [\textit{SIOPT}, 20 (2009), pp. 573-601] including the convex and concave envelopes of the nonlinear activation function of ANNs. The optimization problem is solved using our in-house global deterministic solver MAiNGO. The performance of the proposed method is shown in four optimization examples: an illustrative function, a fermentation process, a compressor plant and a chemical process optimization. The results show that computational solution time is favorable compared to the global general-purpose optimization solver BARON.