(479f) A Versatile Model for Design and Verification of Chemical Products

The focus of current chemical industries has been shifting from process design of business-to-business (B2B) chemicals to product and process design of business-to-consumer (B2C) products^[1]. Chemical product design determines the structure and composition of a system of single or multiple species that satisfies a set of desired properties and functions. Traditionally, chemical products are designed and developed through heuristic rule-based and/or trial-and-error experiment-based approaches. Although these approaches often lead to safe and reliable product designs, it is not practically feasible to evaluate all alternatives or to obtain the optimal solutions. During last three decades, efforts have been made to develop databases, design methods and associated software tools for chemical products. Nowadays, model-based design approaches have become popular because of the ability to rapidly design/screen candidates over a much larger design space. However, chemical product design problems are multiscale and multidisciplinary in nature, which need various models and associated software tools from different disciplines such as computational chemistry, thermodynamics, material science, chemical engineering, industrial engineering, electronic engineering, data science and artificial intelligence for different types of chemical products^[2]. As the chemical products are different, and chemical product design problems are also different, so is their corresponding representation by mathematical models and data. In addition, one factor not often considered in product design is the verification of the product performance during application. Therefore, to develop a computer-aided model and tool for chemical product design with or without associated performance verification, and, capable of solving a wide range of problems, a generic model from which the needed problem-specific model could be generated and used would be advantageous.

Based on the Grand Product Design model^[3], a versatile model is proposed in this work together with numerical analyses and solution strategies for a wide range of chemical product design and verification problems, which requires the additional model equations representing the verification process^[4]. The versatile model consists of a collection of sub-models such as the product sub-model, verification process sub-model, the product quality sub-model, the cost sub-model, the pricing sub-model, the economic sub-model, environmental impact sub-model. For any design and/or verification problem, the necessary sub-model classes are selected. Then, the models needed for the specific design problem within each selected sub-model class are retrieved from a model library together with the associated model parameters.

The resulting design and/or verification problem, represented by a collection of model equations, represent different types of simulation-optimization problems, such as NAE (non-linear equation solution), NLP (non-linear programming), MILP (mixed-integer linear programming) and/or MINLP (mixed-integer non-linear programming). An appropriate solution strategy is developed to solve the defined problem. The presentation will highlight the structure of the versatile model, the rules and steps needed to generate the problem-specific models, the associated solution strategies, and, the solution of the specific design-verification problems. The illustrative examples will highlight molecular design with desired properties, blend and/or formulation design, and molecular design integrated with verification of the product performance through a process model. In each case, the generation and solution of the problem specific models from the versatile model will be highlighted.

REFERENCES:

[1] Zhang L, Fung KY, Wibowo C, Gani R. Advances in chemical product design. Reviews in Chemical Engineering, 2018, 34, 319-340.

[2] Zhang L, Mao H, Liu Q, Gani R. Chemical product design – recent advances and perspectives. Current Opinion in Chemical Engineering, 2020, 27, 22-34.

[3] Fung KY, Ng KM, Zhang L, Gani R. A grand model for chemical product design. Computers & Chemical Engineering, 2016, 91, 15-27.

[4] Chai S, Liu Q, Liang X, Guo Y, Zhang S, Xu C, Du J, Yuan Z, Zhang L, Gani R. A grand product design model for crystallization solvent design. Computers & Chemical Engineering, 2020, 135, 106764.