ENGLISH
|
JAPANESE
|
CONNECT WITH US:
Home
About
Contact
Log in
*
Home
Press release
Nov 27, 2018 20:30 JST
AI Drastically Cuts Number of Trials for Polymer Design
- Trials in polymer design processes can be cut to 1/40 by utilizing AI.
TOKYO, Nov 27, 2018 - (JCN Newswire) - Showa Denko (SDK), National Institute of Advanced Industrial Science and Technology (AIST), and Research Association of High-Throughput Design and Development for Advanced Functional Materials (ADMAT) have found that the number of trials for the design of polymers with desired properties can be cut to about one-fortieth by utilizing artificial intelligence (AI).
Figure 1: Concept of evaluation of number of trials for polymer design/verification
Figure 2: Demonstration of high efficiency of AI-based polymer search
Figure 3: Building AI
Researchers from SDK, AIST, and ADMAT conducted AI-based searches for polymers with desired properties, aiming to demonstrate the effectiveness of AI technology in the process of polymer design. As a model case, they focused on glass transition temperature, an index of heat resistance. Using 417 different types of structural data on polymers with known structures and glass transition temperatures, they conducted an AI-based search for a polymer with the highest glass transition temperature to see whether it is possible to shorten the development cycle.
First, randomly selected 10 sets of data were supplied as training data for AI. The Extended Connectivity Circular Fingerprints (ECFP) method was applied to the training data, digitizing structural features of polymers. Then, using Bayesian optimization*, the researchers made repeated efforts to predict and verify a polymer with the highest glass transition temperature out of the remaining 407 sets of data. Thus, they checked the number of trials required until the discovery of a target polymer. (Figure 1). To prevent the influence of the choice of data on the results, 500 examinations were conducted with different sets of initial training data, and the average number of trials was evaluated.
As a result, the researchers succeeded in discovering a target polymer with the highest glass transition temperature with an extremely small number of trials, namely, 4.6 times of trials on the average (Figure 2). This figure is about one-fortieth of the number of trials required under random selection of polymers, confirming the effectiveness of AI-based polymer design.
To build AI, it is necessary to convert the features of polymers into numerical values. By applying ECFP (a method for representing monomer structures) to this development work, the researchers found repeating units of molecules (such as functional groups) extracted automatically and structural features expressed appropriately in numerical vectors (Figure 3). Using AI built on these data, the researchers realized high-precision property predictions at the speed of 0.25 second per polymer. Thus, it became possible to make comprehensive property predictions for a large number of candidate polymers within a limited amount of time.
Furthermore, by using Bayesian optimization in prediction, the researchers discovered a polymer with the highest glass transition temperature out of about 400 types of candidate polymers, based on 13.6 sets of training data.
Prior to this study, it was believed that AI-based predictions with a small number of training data would be less accurate, and that a large amount of training data would be required. However, this development work suggests that AI would be effective in solving problems even in the area of developing most-advanced functional materials, where only a limited amount of training data seems to be available.
*[Bayesian optimization: A method for selecting the next candidate material to be examined by considering estimated prediction error, in addition to predicted values.]
We will continue working to improve the technology in order to utilize it in actual development of advanced functional materials. A detailed presentation regarding this development work will be made at 2018 MRS Fall Meeting scheduled for November 27 (U.S. Eastern time).
This development work has been subcontracted by NEDO (New Energy and Industrial Technology Development Organization) Ultra High-Throughput Design and Prototyping Technology for Ultra Advanced Materials Development Project (Ultra-Ultra PJ: P16010). By fully utilizing multiscale simulation and AI, Ultra-Ultra PJ aims to reduce the length of time required for development to one-twentieth compared with conventional ways of material development based on knowledge and experience.
Contact:
SDK
Public Relations Office
Phone: +81-3-5470-3235
Fax: +81-3-3431-6215
AIST
Media Relations Office
Phone: +81-29-862-6216
E-mail:
press-ml@aist.go.jp
ADMAT
S. Hatakeyama
Phone: +81-29-856-3580
E-mail:
s-hatakeyama@admat.or.jp
Sectors: Chemicals, Spec.Chem, Science & Nanotech, Engineering, Automation [IoT]
Copyright ©2024 JCN Newswire. All rights reserved. A division of Japan Corporate News Network.
Latest Release
Toyota Exhibiting at Beijing Motor Show 2024
Apr 25, 2024 16:25 JST
Honda Reaches Basic Agreement with Asahi Kasei on Collaboration for Production of Battery Separators for Automotive Batteries in Canada
Apr 25, 2024 11:10 JST
UNIQLO Sponsors KAWS + Warhol Exhibition Tour, Starting in Pittsburgh
Apr 25, 2024 09:00 JST
Mitsubishi Power Begins Commercial Operation of Seventh M701JAC Gas Turbine in Thailand GTCC Project; Achieves 75,000 AOH To-Date
Apr 24, 2024 17:19 JST
MC and Denka Sign J/V Agreement in Fullerene Business
Apr 24, 2024 17:02 JST
Mitsubishi Motors Posts Record Sales in the Philippines in FY2023
Apr 24, 2024 13:56 JST
NEC Develops High-speed Generative AI Large Language Models (LLM) with World-class Performance
Apr 24, 2024 13:25 JST
Fujitsu SX Survey reveals key success factors for sustainability
Apr 23, 2024 10:25 JST
Fujitsu and METRON collaborate to drive ESG success: slashing energy costs, boosting productivity with new manufacturing industry solutions
Apr 22, 2024 16:09 JST
NEC Strengthens Commitment to Space Industry with Investment in Seraphim Space Venture Fund II
Apr 22, 2024 15:09 JST
Soft Space Launches the First and Only JCB Payment Gateway in Malaysia
Apr 22, 2024 15:00 JST
TOYOTA GAZOO Racing takes a one-two in Croatian thriller
Apr 22, 2024 10:47 JST
First-ever Mazda CX-80 Crossover SUV Unveiled in Europe
Apr 19, 2024 13:50 JST
Fujitsu develops technology to convert corporate digital identity credentials, enabling participation of non-European companies in European data spaces
Apr 19, 2024 10:17 JST
Mitsubishi Heavy Industries and NGK to Jointly Develop Hydrogen Purification System from Ammonia Cracking Gas
Apr 18, 2024 17:01 JST
Toyota Launches All-New Land Cruiser "250" Series in Japan
Apr 18, 2024 13:39 JST
Fujitsu and Oracle collaborate to deliver sovereign cloud and AI capabilities in Japan
Apr 18, 2024 11:14 JST
Eisai: Research on Treatments for Alzheimer's Disease Based on Its Pathological Mechanisms Recieves Award for Science and Technology (Research Category)
Apr 18, 2024 10:53 JST
All-New Triton Confirmed as First Double-Cab Pickup Truck to Achieve 2024 Five-Star ANCAP Safety Rating
Apr 18, 2024 09:22 JST
Eisai's Antiepileptic Drug Fycompa Injection Formulation Launched In Japan
Apr 17, 2024 16:17 JST
More Latest Release >>