Interpretable deep learning for guided microstructure-property explorations in photovoltaics

2019 ◽  
Vol 5 (1) ◽  
Author(s):  
Balaji Sesha Sarath Pokuri ◽  
Sambuddha Ghosal ◽  
Apurva Kokate ◽  
Soumik Sarkar ◽  
Baskar Ganapathysubramanian
Keyword(s):  
Surrogate Model ◽  
Photovoltaic Performance ◽  
Semiconductor Film ◽  
Deep Convolutional Neural Networks ◽  
Design Problems ◽  
Manual Exploration ◽  
Data Driven Approach ◽  
And Performance ◽  
The Relationship ◽  
Microstructure And Performance

Abstract The microstructure determines the photovoltaic performance of a thin film organic semiconductor film. The relationship between microstructure and performance is usually highly non-linear and expensive to evaluate, thus making microstructure optimization challenging. Here, we show a data-driven approach for mapping the microstructure to photovoltaic performance using deep convolutional neural networks. We characterize this approach in terms of two critical metrics, its generalizability (has it learnt a reasonable map?), and its intepretability (can it produce meaningful microstructure characteristics that influence its prediction?). A surrogate model that exhibits these two features of generalizability and intepretability is particularly useful for subsequent design exploration. We illustrate this by using the surrogate model for both manual exploration (that verifies known domain insight) as well as automated microstructure optimization. We envision such approaches to be widely applicable to a wide variety of microstructure-sensitive design problems.

scholarly journals Implications of doping on microstructure, processing, and thermoelectric performance: The case of PbSe

2021 ◽  
Author(s):  
Jann A. Grovogui ◽  
Tyler J. Slade ◽  
Shiqiang Hao ◽  
Christopher Wolverton ◽  
Mercouri G. Kanatzidis ◽  
...  
Keyword(s):  
Parent Material ◽  
Model Material ◽  
Material Behavior ◽  
Material System ◽  
Thermoelectric Efficiency ◽  
Lower Solubility ◽  
And Performance ◽  
Microscopy Data ◽  
The Relationship ◽  
Microstructure And Performance

Abstract In this work, we highlight the often-overlooked effects of doping on the microstructure and performance of bulk thermoelectric materials to offer a broader perspective on how dopants interact with their parent material. Using PbSe doped with Na, Ag, and K as a model material system, we combine original computational, experimental, and microscopy data with established trends in material behavior, to provide an in-depth discussion of the relationship between dopants, processing, and microstructure, and their effects on thermoelectric efficiency and thermal stability. Notable observations include differences in the microstructure and mass loss of thermally treated samples of Na- and Ag-doped PbSe, as well as findings that Na and K cations exist predominantly as substitutional point defects while Ag also occupies interstitial sites and exhibits lower solubility. We discuss how these differences in point defect populations are known to affect a dopants’ ability to alter carrier concentration and how they may affect the mechanical properties of PbSe during processing. Graphic Abstract

scholarly journals Mechanical Performance of Concrete Exposed to Sewage—The Influence of Time and pH

Minerals ◽  
2021 ◽  
Vol 11 (5) ◽  
pp. 544
Author(s):  
Justyna Czajkowska ◽  
Maciej Malarski ◽  
Joanna Witkowska-Dobrev ◽  
Marek Dohojda ◽  
Piotr Nowak
Keyword(s):  
Compressive Strength ◽  
Mechanical Performance ◽  
Negative Influence ◽  
Concrete Compressive Strength ◽  
Active Sludge ◽  
Strength Tests ◽  
Compressive Strength Of Concrete ◽  
And Performance ◽  
The Relationship ◽  
Microstructure And Performance

Contact of concrete with aggressive factors, technological structures, reduces their durability through microstructural changes. This work presents the results of research on determining the influence of post grit chamber sewage and sewage from the active sludge chamber in three different environments, i.e., acidic, neutral, and alkaline, on the structure and compressive strength of concrete. Compressive strength tests were carried out after 11.5 months of concrete cubes being submerged in the solutions and compared. To complete the studies, the photos of the microstructure were done. This made it possible to accentuate the relationship between the microstructure and performance characteristics of concrete. The time of storing the cubes in both acidic environments (sewage from post grit chamber and active sludge chamber) has a negative influence on their compressive strength. The compressive strength of cubes decreases along with the time. Compressive strength of cubes increases with increasing pH of the environment.

Introducing an identical benzodithiophene donor unit for polymer donors and small-molecule acceptors to unveil the relationship between the molecular structure and photovoltaic performance of non-fullerene organic solar cells

2019 ◽  
Vol 7 (46) ◽  
pp. 26351-26357 ◽  
Author(s):  
Youdi Zhang ◽  
Laitao Shi ◽  
Tao Yang ◽  
Tao Liu ◽  
Yiqun Xiao ◽  
...  
Keyword(s):  
Molecular Structure ◽  
Solar Cells ◽  
Small Molecule ◽  
Organic Solar Cells ◽  
Thiophene Ring ◽  
Photovoltaic Performance ◽  
And Performance ◽  
The Relationship

We designed an acceptor ITIC-SF by fluorinating the thiophene ring in the benzodithiophene segment of ITIC-S and investigated its effect on the morphology and performance.

scholarly journals Lyapunov-based Optimization of Edge Resources for Energy-Efficient Adaptive Federated Learning

2021 ◽  
Author(s):  
Claudio Battiloro ◽  
Paolo Di Lorenzo ◽  
Mattia Merluzzi ◽  
Sergio Barbarossa
Keyword(s):  
Wireless Network ◽  
Energy Efficient ◽  
Performance Metrics ◽  
Learning Task ◽  
Learning Performance ◽  
Deep Convolutional Neural Networks ◽  
Local Data ◽  
Data Driven Approach ◽  
Gradient Based ◽  
And Performance

The aim of this paper is to propose a novel dynamic resource allocation strategy for energy-efficient adaptive federated learning at the wireless network edge, with latency and learning performance guarantees. We consider a set of devices collecting local data and uploading processed information to an edge server, which runs stochastic gradient-based algorithms to perform continuous learning and adaptation. Hinging on Lyapunov stochastic optimization tools, we dynamically optimize radio parameters (e.g., set of transmitting devices, transmit powers, bits, and rates) and computation resources (e.g., CPU cycles at devices and at server) in order to strike the best trade-off between power, latency, and performance of the federated learning task. The framework admits both a model-based implementation, where the learning performance metrics are available in closed-form, and a data-driven approach, which works with online estimates of the learning performance of interest. The method is then customized to the case of federated least mean squares (LMS) estimation, and federated training of deep convolutional neural networks. Numerical results illustrate the effectiveness of our strategy to perform energy-efficient, low-latency, adaptive federated learning at the wireless network edge.

scholarly journals Lyapunov-based Optimization of Edge Resources for Energy-Efficient Adaptive Federated Learning

2021 ◽  
Author(s):  
Claudio Battiloro ◽  
Paolo Di Lorenzo ◽  
Mattia Merluzzi ◽  
Sergio Barbarossa
Keyword(s):  
Wireless Network ◽  
Energy Efficient ◽  
Performance Metrics ◽  
Learning Task ◽  
Learning Performance ◽  
Deep Convolutional Neural Networks ◽  
Local Data ◽  
Data Driven Approach ◽  
Gradient Based ◽  
And Performance

The aim of this paper is to propose a novel dynamic resource allocation strategy for energy-efficient adaptive federated learning at the wireless network edge, with latency and learning performance guarantees. We consider a set of devices collecting local data and uploading processed information to an edge server, which runs stochastic gradient-based algorithms to perform continuous learning and adaptation. Hinging on Lyapunov stochastic optimization tools, we dynamically optimize radio parameters (e.g., set of transmitting devices, transmit powers, bits, and rates) and computation resources (e.g., CPU cycles at devices and at server) in order to strike the best trade-off between power, latency, and performance of the federated learning task. The framework admits both a model-based implementation, where the learning performance metrics are available in closed-form, and a data-driven approach, which works with online estimates of the learning performance of interest. The method is then customized to the case of federated least mean squares (LMS) estimation, and federated training of deep convolutional neural networks. Numerical results illustrate the effectiveness of our strategy to perform energy-efficient, low-latency, adaptive federated learning at the wireless network edge.

The Influence of Cognitive and Affective Reactions to Feedback on Subsequent Goals

2010 ◽  
Vol 15 (2) ◽  
pp. 121-131 ◽  
Author(s):  
Remus Ilies ◽  
Timothy A. Judge ◽  
David T. Wagner
Keyword(s):  
Performance Feedback ◽  
Business School ◽  
Structural Equation ◽  
Emotional Reactions ◽  
Equation Model ◽  
Proposed Model ◽  
And Performance ◽  
Dispositional Measure ◽  
The Relationship ◽  
Reactions To Feedback

This paper focuses on explaining how individuals set goals on multiple performance episodes, in the context of performance feedback comparing their performance on each episode with their respective goal. The proposed model was tested through a longitudinal study of 493 university students’ actual goals and performance on business school exams. Results of a structural equation model supported the proposed conceptual model in which self-efficacy and emotional reactions to feedback mediate the relationship between feedback and subsequent goals. In addition, as expected, participants’ standing on a dispositional measure of behavioral inhibition influenced the strength of their emotional reactions to negative feedback.

Cognitive Factors Mediate the Relation Between Age and Flight Path Maintenance in General Aviation

2016 ◽  
Vol 6 (2) ◽  
pp. 81-90 ◽  
Author(s):  
Kathleen Van Benthem ◽  
Chris M. Herdman
Keyword(s):  
Cognitive Functioning ◽  
Flight Path ◽  
General Aviation ◽  
Cognitive Factors ◽  
Pilot Performance ◽  
Pilot Error ◽  
Age Related ◽  
And Performance ◽  
Relationship Of ◽  
The Relationship

Abstract. Identifying pilot attributes associated with risk is important, especially in general aviation where pilot error is implicated in most accidents. This research examined the relationship of pilot age, expertise, and cognitive functioning to deviations from an ideal circuit trajectory. In all, 54 pilots, of varying age, flew a Cessna 172 simulator. Cognitive measures were obtained using the CogScreen-AE ( Kay, 1995 ). Older age and lower levels of expertise and cognitive functioning were associated with significantly greater flight path deviations. The relationship between age and performance was fully mediated by a cluster of cognitive factors: speed and working memory, visual attention, and cognitive flexibility. These findings add to the literature showing that age-related changes in cognition may impact pilot performance.

Work Engagement and Research Output Among Female and Male Scientists

2016 ◽  
Vol 15 (2) ◽  
pp. 55-65 ◽  
Author(s):  
Lonneke Dubbelt ◽  
Sonja Rispens ◽  
Evangelia Demerouti
Keyword(s):  
Work Engagement ◽  
Research Output ◽  
Research Performance ◽  
Daily Work ◽  
Time Control ◽  
Science Technology ◽  
And Performance ◽  
And Mathematics ◽  
Number Of Publications ◽  
The Relationship

Abstract. Women have a minority position within science, technology, engineering, and mathematics and, consequently, are likely to face more adversities at work. This diary study takes a look at a facilitating factor for women’s research performance within academia: daily work engagement. We examined the moderating effect of gender on the relationship between two behaviors (i.e., daily networking and time control) and daily work engagement, as well as its effect on the relationship between daily work engagement and performance measures (i.e., number of publications). Results suggest that daily networking and time control cultivate men’s work engagement, but daily work engagement is beneficial for the number of publications of women. The findings highlight the importance of work engagement in facilitating the performance of women in minority positions.

Sign in / Sign up

Export Citation Format

Share Document