A hybrid Genetic Algorithm and Monte Carlo simulation approach to predict hourly energy consumption and generation by a cluster of Net Zero Energy Buildings

Macro Theory & Simulations

Penlidis

et al. 2018

Self Cite

Traditional computational methods simulate the microstructure of polymer chains from input reaction conditions, but a need exists for predicting optimum reaction conditions in a computationally demanding multivariable space leading to the synthesis of predesigned microstructures and architectures. Herein, the intelligent Monte Carlo (IMC) approach, able to predict optimum reaction conditions for synthesizing copolymers with predefined, complex microstructures as input is introduced. This is rendered possible by a combination of kinetic Monte Carlo (KMC) simulation with artificial intelligence concepts, which enables a reasonably enhanced convergence to optimum reactions conditions. Chain shuttling polymerization is chosen as a first test case due to its complexity and the intricate multiblock microstructures that are formed; whose tailoring requires multiple parameters. The IMC approach locates optimum reaction conditions for the synthesis of olefinic multiblock copolymers with specific microstructures. This approach provides a new platform for identifying complex reaction conditions to “produce” and “tailor‐make” materials with precisely predefined microstructures and facilitates the development of meaningful structure‐property relationships.

Section: Model Developmentmentioning

confidence: 99%

Intelligent Monte Carlo: A New Paradigm for Inverse Polymerization Engineering

Mohammadi¹,

Macro Theory & Simulations

Penlidis

et al. 2018

Self Cite

“…Principally, in solving a multiobjective problem, difficulties may arise from the execution of the searching and decision‐making phases of optimization. This requires development of sophisticated computer codes and hybridization of computational algorithms in order to find the best optimal conditions and subsequently warrant multivariable optimization of two or more targets …”

Section: Model Developmentmentioning

confidence: 99%

“…Among different genetic algorithms, NSGA‐II is a unique multiobjective version of the family established primarily based on the domination concept. Undoubtedly, it can be considered as one of the most applied optimization techniques in different fields of science and technology …”

Section: Model Developmentmentioning

confidence: 99%

“…Among all “computationally intelligent” optimization methods and also classical deterministic and stochastic optimization techniques, the Non‐dominated Sorting Genetic Algorithm (NSGA‐II) is one of the most powerful methods to handle various multiobjective problems concurrently. Although NSGA‐II has been successfully applied to manage multiobjective optimization problems in different fields of study, it has never been employed in the manipulation of copolymer microstructure …”

Section: Introductionmentioning

confidence: 99%

“…Although NSGA-II has been successfully applied to manage multiobjective optimization problems in different fields of study, it has never been employed in the manipulation of copolymer microstructure. [23][24][25][26] In this work, a computer code was developed based on NSGA-II to optimize concurrently the MWD and CCD of ethylene/α-olefin copolymers in terms of polymerization temperature and concentrations of ethylene, 1 butene, cocatalyst, and hydrogen. In contrast to the ANN modeling approach, NSGA-II is an expert optimizer and has the capability of decision-making in case of complex multiobjective problems, hence it enables identification of all local optima and the global optimum.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Heuristic Search Strategy for Transforming Microstructural Patterns to Optimal Copolymerization Recipes

Mohammadi¹,

Macro Theory & Simulations

Penlidis

2018

Self Cite

Manipulation and optimization of copolymer microstructure for tailoring final properties is of great importance in macromolecular science and engineering. Uncovering the complexities of the interrelationships between copolymerization recipe and copolymer microstructure (a challenging field of study in its own right) is a multiobjective optimization problem, which has attracted a lot of attention in the last 10–15 years. In the present study, a powerful optimizer is developed based on the Non‐dominated Sorting Genetic Algorithm (NSGA‐II) for transforming desired microstructural copolymerization profiles, including molecular weight distribution and chemical composition distribution, back to optimal copolymerization recipes and operating conditions. The optimizer developed has the beneficial features of robust machine learning and multiobjective optimization based upon heuristic search strategies. The metallocene‐catalyzed ethylene/α‐olefin copolymerization is selected as a sufficiently complex system to challenge the proposed optimization tool. The developed computer code is used to explore copolymerization recipes (polymerization temperature and concentrations of ethylene, 1butene, cocatalyst, and hydrogen) needed to synthesize copolymers having desired microstructural features. Based on the results obtained, it is now possible to produce various grades or tailor‐make the copolymer structure by suggesting the “best” copolymerization recipe/conditions as reliably as possible.

Modeling and closed‐loop control of particle size and initial burst of PLGA biodegradable nanoparticles for targeted drug delivery

Baghaei

J of Applied Polymer Sci

Jafari

et al. 2017

Self Cite

An in-house computer code based on artificial intelligence has been developed and applied in modeling and closed-loop optimization of release behavior of Poly(lactic-co-glycolic acid) (PLGA) biodegradable particles. A series of micro-and nanoparticles were prepared via water-in-oil-in-water double emulsion to be loaded with albumin-fluorescein isothiocyanate conjugate as a typical drug. The interrelationship between input variables (molecular weight of polymer and stabilizer, polymer concentration, and sonication rate) and outputs (PLGA particle size and percentage of initial burst) was uncovered with the aid of artificial neural network modeling. The regression analysis confirmed acceptable correlation coefficients for the aforementioned responses, where the PLGA molecular weight played the most important role among the studied variables. Input variables needed to minimize PLGA size and PLGA initial burst were then obtained via multiobjective optimization performed by a genetic algorithm. PLGA nanoparticles were checked for particle size and particle size distribution using scanning electron micrographs.