Search

The advent of Industry 4.0 has precipitated the incorporation of Artificial Intelligence (AI) methods within industrial contexts, aiming to realize intelligent manufacturing, operation as well as maintenance, also known as industrial intelligence. However, intricate industrial milieus, particularly those relating to energy exploration and production, frequently encompass data characterized by long-tailed class distribution, sample imbalance, and domain shift. These attributes pose noteworthy challenges to data-centric Deep Learning (DL) techniques, crucial for the realization of industrial intelligence. The present study centers on the intricate and distinctive industrial scenarios of Nuclear Power Generation (NPG), meticulously scrutinizing the application of DL techniques under the constraints of finite data samples. Initially, the paper expounds on potential employment scenarios for AI across the full life-cycle of NPG. Subsequently, we delve into an evaluative exposition of DL's advancement, grounded in the finite sample perspective. This encompasses aspects such as small-sample learning, few-shot learning, zero-shot learning, and open-set recognition, also referring to the unique data characteristics of NPG. The paper then proceeds to present two specific case studies. The first revolves around the automatic recognition of zirconium alloy metallography, while the second pertains to open-set recognition for signal diagnosis of machinery sensors. These cases, spanning the entirety of NPG's life-cycle, are accompanied by constructive outcomes and insightful deliberations. By exploring and applying DL methodologies within the constraints of finite sample availability, this paper not only furnishes a robust technical foundation but also introduces a fresh perspective toward the secure and efficient advancement and exploitation of this advanced energy source.

Deep Image Manipulation Localization (IML) models suffer from training data insufficiency and thus heavily rely on pre-training. We argue that contrastive learning is more suitable to tackle the data insufficiency problem for IML. Crafting mutually exclusive positives and negatives is the prerequisite for contrastive learning. However, when adopting contrastive learning in IML, we encounter three categories of image patches: tampered, authentic, and contour patches. Tampered and authentic patches are naturally mutually exclusive, but contour patches containing both tampered and authentic pixels are non-mutually exclusive to them. Simply abnegating these contour patches results in a drastic performance loss since contour patches are decisive to the learning outcomes. Hence, we propose the Non-mutually exclusive Contrastive Learning (NCL) framework to rescue conventional contrastive learning from the above dilemma. In NCL, to cope with the non-mutually exclusivity, we first establish a pivot structure with dual branches to constantly switch the role of contour patches between positives and negatives while training. Then, we devise a pivot-consistent loss to avoid spatial corruption caused by the role-switching process. In this manner, NCL both inherits the self-supervised merits to address the data insufficiency and retains a high manipulation localization accuracy. Extensive experiments verify that our NCL achieves state-of-the-art performance on all five benchmarks without any pre-training and is more robust on unseen real-life samples. The code is available at: https://github.com/Knightzjz/NCL-IML., Comment: Tech report. ICCV2023 paper

Neural Architecture Search (NAS) has emerged as one of the effective methods to design the optimal neural network architecture automatically. Although neural architectures have achieved human-level performances in several tasks, few of them are obtained from the NAS method. The main reason is the huge search space of neural architectures, making NAS algorithms inefficient. This work presents a novel architecture search algorithm, called GPT-NAS, that optimizes neural architectures by Generative Pre-Trained (GPT) model with an evolutionary algorithm (EA) as the search strategy. In GPT-NAS, we assume that a generative model pre-trained on a large-scale corpus could learn the fundamental law of building neural architectures. Therefore, GPT-NAS leverages the GPT model to propose reasonable architecture components given the basic one and then utilizes EAs to search for the optimal solution. Such an approach can largely reduce the search space by introducing prior knowledge in the search process. Extensive experimental results show that our GPT-NAS method significantly outperforms seven manually designed neural architectures and thirteen architectures provided by competing NAS methods. In addition, our experiments also indicate that the proposed algorithm improves the performance of finely tuned neural architectures by up to about 12% compared to those without GPT, further demonstrating its effectiveness in searching neural architectures.

Many problems in science and engineering can be represented by a set of partial differential equations (PDEs) through mathematical modeling. Mechanism-based computation following PDEs has long been an essential paradigm for studying topics such as computational fluid dynamics, multiphysics simulation, molecular dynamics, or even dynamical systems. It is a vibrant multi-disciplinary field of increasing importance and with extraordinary potential. At the same time, solving PDEs efficiently has been a long-standing challenge. Generally, except for a few differential equations for which analytical solutions are directly available, many more equations must rely on numerical approaches such as the finite difference method, finite element method, finite volume method, and boundary element method to be solved approximately. These numerical methods usually divide a continuous problem domain into discrete points and then concentrate on solving the system at each of those points. Though the effectiveness of these traditional numerical methods, the vast number of iterative operations accompanying each step forward significantly reduces the efficiency. Recently, another equally important paradigm, data-based computation represented by deep learning, has emerged as an effective means of solving PDEs. Surprisingly, a comprehensive review for this interesting subfield is still lacking. This survey aims to categorize and review the current progress on Deep Neural Networks (DNNs) for PDEs. We discuss the literature published in this subfield over the past decades and present them in a common taxonomy, followed by an overview and classification of applications of these related methods in scientific research and engineering scenarios. The origin, developing history, character, sort, as well as the future trends in each potential direction of this subfield are also introduced.

Exploring the mechanism of interfacial thermal transport and reducing the interfacial thermal resistance is of great importance for thermal management and modulation. Herein, the interfacial thermal resistance between overlapped graphene nanoribbons is largely reduced by adding bonded carbon chains by performing molecular dynamics simulations. And the analytical model (cross-interface model, CIM) is utilized to analyze and explain the two-dimensional thermal transport mechanism at cross-interface. An order of magnitude reduction in interfacial thermal resistance is found as the graphene nanoribbons are bonded by just one carbon chain. Interestingly, the decreasing rate of interfacial thermal resistance slows down gradually with the increasing of the number of carbon chains, which can be explained by the proposed theoretical relationship based on CIM. Moreover, by the comparison of CIM and traditional simplified model, the accuracy of CIM is verified and demonstrated in overlapped graphene nanoribbons. This work provides a new way to improve the interfacial thermal transport and reveal the essential mechanism for low-dimensional materials applied in thermal management.

The application of low-dimensional materials for heat dissipation requires a comprehensive understanding of the thermal transport at the cross interface, which widely exists in various composite materials and electronic devices. In this work, we proposed an analytical model, named as cross interface model (CIM), to accurately reveal the essential mechanism of the two-dimensional thermal transport at the cross interface. The applicability of CIM is validated through the comparison of the analytical results with molecular dynamics simulations for a typical cross interface of two overlapped boron nitride nanoribbons. Besides, it is figured out that the factor ({\eta}) has important influence on the thermal transport besides the thermal resistance inside and between the materials, which is found to be determined by two dimensionless parameters from its expression. Our investigations deepen the understanding of the thermal transport at the cross interface and also facilitate to guide the applications of low-dimensional materials in thermal management.

There has been an increasing demand for materials with special thermal properties, whereas experimental discovery is high-cost and time-consuming. The emerging discipline `Materials Informatics' is an effective approach that can accelerate materials development by combining material science and big data technique. Recently materials informatics has been applied to the design of novel materials such as thermal interface materials for heat-dissipation, and thermoelectric materials for power generation. This mini-review summarized the research progress on the applications of materials informatics for the thermal transport properties prediction and discovery of materials with special thermal properties, including optimal thermal conductivity, interfacial thermal conductance and thermoelectricity efficiency. In addition, some perspectives are given for the outlook of materials informatics in the field of thermal transport.