Computer-Aided Design of a Perfluorinated Sulfonic Acid Proton Exchange Membrane Using Stochastic Optimization and Molecular Dynamic Method

Abstract

In this paper, a computer-aided polymer design (CAPD) framework with a stochastic optimization model is proposed for the design of perfluorinated sulfonic acid proton exchange membrane (PFSA-PEM) with desired properties. First, the requirements and target characteristics are identified and converted to property constraints. Then, the polymer design model is formulated as a stochastic optimization problem in which the operation temperature is treated as a random variable. Subsequently, the formulated stochastic mixed integer nonlinear programming problem is solved by a two-stage strategy. In stage I, molecular dynamics is utilized to simulate the target properties for different structures and quantities of side chains, thus establishing the quantitative structure−property relationship. In stage II, the operation temperature is considered subject to a specified probability distribution. The optimization model is solved to obtain the optimal polymer structure over the operating temperature range. Finally, a case study of PFSA-PEM design is given to illustrate the application of the CAPD framework.

The Computer-Aided Design Algorithm of Dyeing and Weaving Graphics from the Perspective of Public Art

“…The aided design algorithm described in this study is also contrasted with the classic aided design algorithm. In Reference [ 25 ], the design similarity index is used to calculate the measurement index. Figure 5 depicts a comparison diagram of the two options.…”

“…Scene visual understanding is one of the most widely utilized computer technologies in the field of the art design, and it is one of the most widely integrated computer technologies in the field of art design [ 24 , 25 ]. Aiming to replace the human eye and brain with computers, scene vision understanding is the process of using computers to simulate the visual function of the human eye and brain to perceive, recognize, and understand three-dimensional scenes and objects in the objective world, as well as to analyze complex objects in scene images by integrating with natural language.…”

Computer‐Aided Design of Crosslinked Polymer Membrane Using Machine Learning and Molecular Dynamics

“…Liang et al [5] proposed a computer-aided polymer design (CAPD) framework for the design of rubber tires. Guo et al [6] developed a CAPD framework with stochastic optimization algorithm, which is used for the design of perfluoro sulfonic acid proton exchange membranes (PEM). However, only the effects of repeating unit structure, molecular weight and side chain structure distribution of polymer on the properties were considered in previous work, the effect of crosslinking was ignored.…”

Using Artificial Neural Networks to Predict Physical Properties of Membrane Polymers

“…Recent research has demonstrated that it is possible to target specific properties and design compound, which will produce a membrane with those properties using CAMD [11]. Artificial neural networks (ANNs), which define a branch of machine learning, are capable of discovering complex patterns from large datasets.…”

“…Artificial neural networks (ANNs), which define a branch of machine learning, are capable of discovering complex patterns from large datasets. Applying the predictive power of ANNs to CAMD, specifically to predicting the properties of membrane polymers, provides reasonable results at a fraction of the computational cost of rigorous simulations [11,12], and avoids the costly and unfocused guess-and-experiment approach. This makes it possible to consider new membrane polymers on a scale which is orders of magnitude larger than either rigorous simulations or guess-and-experiment approaches would allow.…”