東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Quantitative property-structural rel...

Wu, Ke.

FindBook

Google Book

Amazon

博客來

Quantitative property-structural relation modeling on polymeric dielectric materials.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Quantitative property-structural relation modeling on polymeric dielectric materials./
作者:	Wu, Ke.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2016,
面頁冊數:	185 p.
附註:	Source: Dissertation Abstracts International, Volume: 77-10(E), Section: B.
Contained By:	Dissertation Abstracts International77-10B(E).
標題:	Polymer chemistry. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10123933
ISBN:	9781339820743

Quantitative property-structural relation modeling on polymeric dielectric materials.
Wu, Ke.

Quantitative property-structural relation modeling on polymeric dielectric materials. - Ann Arbor : ProQuest Dissertations & Theses, 2016 - 185 p.

Source: Dissertation Abstracts International, Volume: 77-10(E), Section: B.

Thesis (Ph.D.)--Rensselaer Polytechnic Institute, 2016.

Nowadays, polymeric materials have attracted more and more attention in dielectric applications. But searching for a material with desired properties is still largely based on trial and error. To facilitate the development of new polymeric materials, heuristic models built using the Quantitative Structure Property Relationships (QSPR) techniques can provide reliable "working solutions". In this thesis, the application of QSPR on polymeric materials is studied from two angles: descriptors and algorithms. A novel set of descriptors, called infinite chain descriptors (ICD), are developed to encode the chemical features of pure polymers. ICD is designed to eliminate the uncertainty of polymer conformations and inconsistency of molecular representation of polymers. Models for the dielectric constant, band gap, dielectric loss tangent and glass transition temperatures of organic polymers are built with high prediction accuracy. Two new algorithms, the physics-enlightened learning method (PELM) and multi-mechanism detection, are designed to deal with two typical challenges in material QSPR. PELM is a meta-algorithm that utilizes the classic physical theory as guidance to construct the candidate learning function. It shows better out-of-domain prediction accuracy compared to the classic machine learning algorithm (support vector machine). Multi-mechanism detection is built based on a cluster-weighted mixing model similar to a Gaussian mixture model. The idea is to separate the data into subsets where each subset can be modeled by a much simpler model. The case study on glass transition temperature shows that this method can provide better overall prediction accuracy even though less data is available for each subset model.

ISBN: 9781339820743Subjects--Topical Terms:

3173488
Polymer chemistry.

Quantitative property-structural relation modeling on polymeric dielectric materials.
LDR:04171nmm a2200313 4500 001 2120457
005 20170719065342.5
008 180830s2016 ||||||||||||||||| ||eng d
020 $a 9781339820743
035 $a (MiAaPQ)AAI10123933
035 $a AAI10123933
040 $a MiAaPQ $c MiAaPQ
100 1 $a Wu, Ke. $3 3282393
245 1 0 $a Quantitative property-structural relation modeling on polymeric dielectric materials.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2016
300 $a 185 p.
500 $a Source: Dissertation Abstracts International, Volume: 77-10(E), Section: B.
500 $a Adviser: Curtis M. Breneman.
502 $a Thesis (Ph.D.)--Rensselaer Polytechnic Institute, 2016.
520 $a Nowadays, polymeric materials have attracted more and more attention in dielectric applications. But searching for a material with desired properties is still largely based on trial and error. To facilitate the development of new polymeric materials, heuristic models built using the Quantitative Structure Property Relationships (QSPR) techniques can provide reliable "working solutions". In this thesis, the application of QSPR on polymeric materials is studied from two angles: descriptors and algorithms. A novel set of descriptors, called infinite chain descriptors (ICD), are developed to encode the chemical features of pure polymers. ICD is designed to eliminate the uncertainty of polymer conformations and inconsistency of molecular representation of polymers. Models for the dielectric constant, band gap, dielectric loss tangent and glass transition temperatures of organic polymers are built with high prediction accuracy. Two new algorithms, the physics-enlightened learning method (PELM) and multi-mechanism detection, are designed to deal with two typical challenges in material QSPR. PELM is a meta-algorithm that utilizes the classic physical theory as guidance to construct the candidate learning function. It shows better out-of-domain prediction accuracy compared to the classic machine learning algorithm (support vector machine). Multi-mechanism detection is built based on a cluster-weighted mixing model similar to a Gaussian mixture model. The idea is to separate the data into subsets where each subset can be modeled by a much simpler model. The case study on glass transition temperature shows that this method can provide better overall prediction accuracy even though less data is available for each subset model.
520 $a In addition, the techniques developed in this work are also applied to polymer nanocomposites (PNC). PNC are new materials with outstanding dielectric properties. As a key factor in determining the dispersion state of nanoparticles in the polymer matrix, the surface tension components of polymers are modeled using ICD. Compared to the 3D surface descriptors used in a previous study, the model with ICD has a much improved prediction accuracy and stability particularly for the polar component. In predicting the enhancement effect of grafting functional groups on the breakdown strength of PNC, a simple local charge transfer model is proposed where the electron affinity (EA) and ionization energy (IE) determines the main charge trap depth in the system. This physical model is supported by first principle computation. QSPR models for EA and IE are also built, decreasing the computation time of EA and IE for a single molecule from several hours to less than one second.
520 $a Furthermore, the designs of two web-based tools are introduced. The tools represent two commonly used applications for QSPR studies: data inquiry and prediction. Making models and data public available and easy to use is particularly crucial for QSPR research. The web tools described in this work should provide a good guidance and starting point for the further development of information tools enabling more efficient cooperation between computational and experimental communities.
590 $a School code: 0185.
650 4 $a Polymer chemistry. $3 3173488
650 4 $a Materials science. $3 543314
690 $a 0495
690 $a 0794
710 2 $a Rensselaer Polytechnic Institute. $b Chemistry. $3 2098707
773 0 $t Dissertation Abstracts International $g 77-10B(E).
790 $a 0185
791 $a Ph.D.
792 $a 2016
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10123933