Interactive Evolutionary Design Method of Tile Shape by GAN Reflecting Adjectives

2021 ◽  
Author(s):  
Mao Takegishi ◽  
Kaoru Arakawa
Keyword(s):  
Design Method ◽  
Evolutionary Design

Profiles evolutionary design and evaluation of mixing performance of screw elements for intermeshing counter-rotating twin-screw kneader

2016 ◽  
Vol 230 (17) ◽  
pp. 3076-3091 ◽  
Author(s):  
Jing Wei ◽  
Dongming Zhou ◽  
Aiqiang Zhang ◽  
Yuliang Yang ◽  
Dabing Chen
Keyword(s):  
Mathematical Model ◽  
Cross Section ◽  
Design Method ◽  
Design Parameters ◽  
Evolutionary Design ◽  
Mixing Performance ◽  
Distribution Density Function ◽  
Different Types ◽  
Twin Screw ◽  
Dispersive Mixing

The profile designs of screw elements are very crucial to improving the mixing performance for the intermeshing counter-rotating twin-screw kneader. In order to find the inherent law among different profiles of screw elements and to derive a mathematical model which evolves different types of end cross-section profiles of screw elements, a universal mathematical model of end cross-section profiles of screw elements is presented in this work. Different types of profiles of screw elements, including those traditionally used can be obtained by evolutionary design method after changing the tooth number of female and male elements and other design parameters. Several typical screw elements with different types of end cross-section profiles are selected to analyze the mixing performance. Spatial flow patterns in flow channel of the screw elements are presented with particle tracking analysis by using mesh superposition technology. The mixing performance of different screw elements including distributive and dispersive mixing, contrastive analysis for some parameters, such as residence time distribution density function, segregation scale, rate of stretching, rate of shear, etc. are proposed. The results in this paper provide a data basis for the profiles selection of screw element for intermeshing counter-rotating twin-screw kneader.

A generative evolutionary design method

2006 ◽  
Vol 17 (1) ◽  
pp. 49-63 ◽  
Author(s):  
Patrick Janssen
Keyword(s):  
Design Method ◽  
Evolutionary Design

Product Pedigree Based Conceptual Structure Evolutionary Design

2012 ◽  
Vol 215-216 ◽  
pp. 612-615
Author(s):  
Xing Zu Wang ◽  
Bo Yang ◽  
Kong Yao ◽  
Chang Qing Gao
Keyword(s):  
Conceptual Design ◽  
Design Process ◽  
Design Method ◽  
Conceptual Structure ◽  
Structure Design ◽  
Classification Model ◽  
Evolutionary Design ◽  
Variant Design ◽  
Interaction Field ◽  
The Relationship

By applying biological pedigree concept into product conceptual structure design, an conceptual structural design approach is set forth to make full use of creative and innovative intelligence of living systems for design automation. According to the relationship between different product structures with same function, the classification model of product pedigree is established, which can be used as unified functional isomorphic model during the intelligent conceptual design process. On the other hand, according to the evolutional logic from quality change to quantities change, pedigree based product evolutionary design strategy is given. Product pedigrees which are homology dependent are used as basic evolutionary models, based on which, the parameter based, interaction field based and interaction mater based evolutional conceptual design method are proposed, respectively. Based on the relation network of the pedigree units, multiple variant design schemes can be obtained, which is a strong backbone to the innovative design process. Finally, design examples are used to demonstrate the successful application of the proposed pedigree based evolutionary method in the product variant design.

scholarly journals Evolutionary Design Method for Multipurpose Batch Plants on the Basis of Cyclic Production.

1994 ◽  
Vol 27 (1) ◽  
pp. 57-64 ◽  
Author(s):  
Tetsuo Fuchino ◽  
Kazuo Yamaguchi ◽  
Masaaki Muraki ◽  
Toyohiko Hayakawa
Keyword(s):  
Design Method ◽  
Evolutionary Design ◽  
Batch Plants ◽  
Cyclic Production

scholarly journals Evolutionary design of molecules based on deep learning and a genetic algorithm

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Youngchun Kwon ◽  
Seokho Kang ◽  
Youn-Suk Choi ◽  
Inkoo Kim
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Deep Learning ◽  
Design Method ◽  
Network Models ◽  
Molecular Structures ◽  
Evolutionary Design ◽  
Neural Network Models ◽  
Practical Challenge ◽  
Significant Attention

AbstractEvolutionary design has gained significant attention as a useful tool to accelerate the design process by automatically modifying molecular structures to obtain molecules with the target properties. However, its methodology presents a practical challenge—devising a way in which to rapidly evolve molecules while maintaining their chemical validity. In this study, we address this limitation by developing an evolutionary design method. The method employs deep learning models to extract the inherent knowledge from a database of materials and is used to effectively guide the evolutionary design. In the proposed method, the Morgan fingerprint vectors of seed molecules are evolved using the techniques of mutation and crossover within the genetic algorithm. Then, a recurrent neural network is used to reconstruct the final fingerprints into actual molecular structures while maintaining their chemical validity. The use of deep neural network models to predict the properties of these molecules enabled more versatile and efficient molecular evaluations to be conducted by using the proposed method repeatedly. Four design tasks were performed to modify the light-absorbing wavelengths of organic molecules from the PubChem library.

Sign in / Sign up

Export Citation Format

Share Document