Longze Li

ORCID: 0000-0003-2808-9652
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Nuclear Materials and Properties
  • Nuclear Engineering Thermal-Hydraulics
  • Risk and Safety Analysis
  • Nuclear reactor physics and engineering
  • Combustion and Detonation Processes
  • Graphite, nuclear technology, radiation studies
  • Nuclear and radioactivity studies
  • Machine Learning in Materials Science
  • Probabilistic and Robust Engineering Design
  • Model Reduction and Neural Networks
  • Computational Fluid Dynamics and Aerodynamics
  • Intermetallics and Advanced Alloy Properties
  • Simulation and Modeling Applications
  • Safety and Risk Management
  • Color Science and Applications
  • Face recognition and analysis
  • Cavitation Phenomena in Pumps
  • Aluminum Alloy Microstructure Properties
  • Hydraulic flow and structures
  • Aluminum Alloys Composites Properties
  • Structural Integrity and Reliability Analysis
  • Face and Expression Recognition
  • Geological Modeling and Analysis
  • Anomaly Detection Techniques and Applications
  • Image Enhancement Techniques

University of Idaho
2018-2025

Fuzhou University
2023

Wuhan Ship Development & Design Institute
2018-2021

Xi'an Jiaotong University
2013-2018

Guilin University of Electronic Technology
2017

With the increased use of data-driven approaches and machine learning-based methods in material science, importance reliable uncertainty quantification (UQ) predicted variables for informed decision-making cannot be overstated. UQ property prediction poses unique challenges, including multi-scale multi-physics nature materials, intricate interactions between numerous factors, limited availability large curated datasets, etc. In this work, we introduce a physics-informed Bayesian Neural...

10.1038/s41598-024-61189-x article EN cc-by Scientific Reports 2024-05-08

Mechanical testing with sub-sized specimens plays an important role in the nuclear industry, facilitating tests confined experimental spaces lower irradiation levels and accelerating qualification of new materials. The reduced size results different material behavior at microscale, mesoscale, macroscale, comparison to standard-sized specimens, which is referred as "specimen effect." Although analytical models have been proposed correlate properties these lack broad applicability across...

10.1038/s41597-024-04329-2 article EN cc-by-nc-nd Scientific Data 2025-01-11

With the increased use of data-driven approaches and machine learning-based methods in material science, importance reliable uncertainty quantification (UQ) predicted variables for informed decision-making cannot be overstated. UQ property prediction poses unique challenges, including multi-scale multi-physics nature advanced materials, intricate interactions between numerous factors, limited availability large curated datasets model training, etc. Recently, Bayesian Neural Networks (BNNs)...

10.48550/arxiv.2311.02495 preprint EN other-oa arXiv (Cornell University) 2023-01-01

To reduce the loss of feature information, this paper proposes a facial extraction method based on Shallow and Deep Fusion Convolutional Neural Networks (SDFCNN). Firstly, shallow features are extracted by employing parallel multiple convolutional layers different scales. Then, obtained fused with those deeper layer. Finally, results further another layer for Softmax classification. Experimental show that, SDFCNN provide better representation than that DeepID network. The accuracy can reach...

10.1109/fskd.2017.8393320 article EN 2017-07-01

Mechanical testing with sub-sized specimens plays an important role in the nuclear industry, facilitating tests confined experimental spaces lower irradiation levels and accelerating qualification of new materials. The reduced size results different material behavior at microscale, mesoscale, macroscale, comparison to standard-sized specimens, which is referred as specimen effect. Although analytical models have been proposed correlate properties these lack broad applicability across...

10.48550/arxiv.2409.08306 preprint EN arXiv (Cornell University) 2024-09-11

Abstract It is well know that the two-fluid single pressure model currently widely used to analyze reactor transient accidents such as LOCA. Current mainstream safety analysis codes RALAP5 and CATHARE are based on this model. However has been proved be ill posed in sense equation system non-hyperbolic which will lead numerical unphysical oscillation. Currently use interfacial virtual mass force solve ill-posed non-hyperbolicity issue. The development of China’s self-owned high fidelity code...

10.1115/icone28-64434 article EN 2021-08-04

The natural circulation small modular reactor (NCSMR) is a 330 MW which has no coolant pumps (RCP) and active safety injection systems at all. mainly comprised of the pressure vessel (RPV) with integral pressurize r steam generator. RPV enclosed by vacuumed containment (PCV) PCV submerged in underground pool. A MELCOR model corresponding input deck are developed for RPV, PCV, pool takes role ultimate heat sink (UHS) accident situations. may crack leak some critical accidents as earthquake,...

10.1115/icone26-82267 article EN 2018-07-22

The Modular Accident Analysis Program version 5 (MAAP5) is a computer code that can simulate the response of light water reactor power plants during severe accident sequences. present work aims to typical Chinese pressure (PWR) induced by he pressurizer safety valve stuck-open with MAAP5. Three cases corresponding 1, 2 and 3 valves stuck open accidents without high injection (HPI) auxiliary feed (AFW) steam generator (SG) were analyzed. results showed HPI AFW failure would lead accident. And...

10.1115/icone22-30535 article EN 2014-07-07

The severe accident of CPR1000 caused by station blackout with the SG safety valve failure is simulated and analyzed using MELCOR code in this work. power plant response process results three different assumptions, which are no seal leakage nor auxiliary feed water, water exist, exist but separately, analyzed. According to calculation results, without pressure vessel would fail at 9576 s. When was supplied, vessel’s time delay nearly 30000s. exists, about 50 meaningful significant for...

10.1115/icone21-15152 article EN 2013-07-29

Vehicular color recognition is a challenging problem since the change of light conditions, interference background and difference subjective perceptions in dynamic complex environments. This paper proposes method closing to cognitive model human vision systems for vehicle recognition. Firstly, Spatial-Color (SC) normalized designed extract main images. Secondly, novel Wide Residual Convolution Neural Network (WRCNN) proposed global features, output provided by fully connected layer. Finally,...

10.1109/icice.2017.8479234 article EN 2017 International Conference on Information, Communication and Engineering (ICICE) 2017-11-01

Compared with the land-based nuclear power plant, operating conditions of offshore plant (ONPP) are much more complicated. For example, barge-mounted platform malfunction, which is as important natural events and human events, should be considered in safety analysis,. As a result, two dimension condition coupled barge reactor status development relevant procedures. On other hand, beyond design basis hazards induced by combination unique unanticipated external ONPP may lead to blind area both...

10.1115/icone26-81159 article EN 2018-07-22

The nuclear safety is the basic demand for development of energy. severe accident which design extended conditions that can lead to core melt in power plant, significant influence NPP design. risk marine plant (MNPP) higher than onshore limitation condition and limited layout space. Thus, mitigation measures are much more important mature research method referenced MNPP work. reactor a 100 MWt PWR with 2 loops. special residual heat removal system, passive removal, square steel containment...

10.1299/jsmeicone.2019.27.1044 article EN The Proceedings of the International Conference on Nuclear Engineering (ICONE) 2019-01-01

The marine nuclear power plant is a floating which supply for the offshore oil drill platform. It designed based on standard and experience of traditional onshore plant. reactor 100 MWt PWR type small modular with 2 loops. Each loops contains main pump, check valve, steam generator. pressurizer set one engineering safety features in are somewhat different from those plants. special residual heat removal system, passive removal, square steel containment suppression system prevention...

10.1299/jsmeicone.2019.27.1043 article EN The Proceedings of the International Conference on Nuclear Engineering (ICONE) 2019-01-01
Coming Soon ...