Zhencheng Ye

Zhencheng Ye


Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 May 2021 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
CZChongzhen Zhang
Chongzhen Zhang
YTYang Tang
Yang Tang
CZChaoqiang Zhao
Chaoqiang Zhao
QSQiyu Sun
Qiyu Sun
ZYZhencheng Ye
Zhencheng Ye
JKJurgen Kurths
Jurgen Kurths
Semantic segmentation and depth completion are two challenging tasks in scene understanding, and they are widely used in robotics and autonomous driving. Although several studies have been proposed to jointly train these two tasks using some small modifications, such as changing the last layer, the result of one task is not utilized to improve the performance of the other one despite that there are some similarities between these two tasks. In this article, we propose multitask generative adversarial networks (Multitask GANs), which are not only competent in semantic segmentation and depth completion but also improve the accuracy of depth completion through generated semantic images. In addition, we improve the details of generated semantic images based on CycleGAN by introducing multiscale spatial pooling blocks and the structural similarity reconstruction loss. Furthermore, considering the inner consistency between semantic and geometric structures, we develop a semantic-guided smoothness loss to improve depth completion results. Extensive experiments on the Cityscapes data set and the KITTI depth completion benchmark show that the Multitask GANs are capable of achieving competitive performance for both semantic segmentation and depth completion tasks.

30 March 2021 in Processes
Verified
Journal Article
Open access
PDF file
ZYZhencheng Ye
Zhencheng Ye
MXMo Xiaoyan
Mo Xiaoyan
LZLiang Zhao
Liang Zhao
Liquefied natural gas (LNG) is a clear and promising fossil fuel which emits less greenhouse gas (GHG) and has almost no environmentally damaging sulfur dioxide compared with other fossil fuels. An LNG import terminal is a facility that regasifies LNG into natural gas, which is supplied to industrial and residential users. Modeling and optimization of the LNG terminals may reduce energy consumption and GHG emission. A mixed-integer nonlinear programming model of the LNG terminal is developed to minimize the energy consumption, where the numbers of boil-off gas (BOG) compressors and low-pressure (LP) pumps are considered as integer variables. A case study from an actual LNG terminal is carried out to verify the practicality of the proposed method. Results show that the proposed approach can decrease the operating energy consumption from 9.15% to 26.1% for different seasons.

31 January 2026 in Journal of Colloid and Interface Science
Verified
Journal Article
Closed access
XLXinyi Lu
Xinyi Lu
HHHaicai Huang
Haicai Huang
YBYihui Bao
Yihui Bao
YXYanyan Xia
Yanyan Xia
ZYZhencheng Ye
Zhencheng Ye
HCHouyang Chen
Houyang Chen

16 March 2018 in International Journal of Chemical Reactor Engineering
Verified
Journal Article
Closed access
PDF file
GHGuihua Hu
Guihua Hu
ZYZhencheng Ye
Zhencheng Ye
Wenli DuWenli Du
Wenli Du
feng qianfeng qian
feng qian
Gas-solid coupled heat transfer in an industrial isothermal acetylene hydrogenation reactor was carried out using computational fluid dynamics (CFD). A two-temperature porous medium model was established by adding source terms to energy equations of the solid and gas phases. The combination of a genetic algorithm with CFD methods is applied to optimization of the kinetic and process parameters of the reaction. The model was validated by comparing the simulated results with those obtained from a one-temperature porous medium model, a two-temperature porous medium model, and industrial data. The optimal hydrogen-to-acetylene ratio and inlet temperature are 1.78 and 326K, respectively. The optimized ethylene yield increase and hydrogenation selectivity are 0.53 % and 0.18 % higher than the values before optimization, respectively. Finally, the effects of the hydrogen-to-acetylene ratio and inlet temperature on the increase in ethylene yield and hydrogenation selectivity are analyzed. Therefore, the hydrogen-to-acetylene ratio and inlet temperature should be reasonably controlled during production.

15 May 2024 in Chinese Journal of Chemical Engineering
Verified
Journal Article
Closed access
KJKai Ji
Kai Ji
ZYZhencheng Ye
Zhencheng Ye
feng qianfeng qian
feng qian

27 September 2025 in Advanced Science
Verified
Journal Article
Open access
YBYihui Bao
Yihui Bao
XLXinyi Lu
Xinyi Lu
YXYanyan Xia
Yanyan Xia
ZYZhencheng Ye
Zhencheng Ye
HCHouyang Chen
Houyang Chen
Machine learning has emerged as a powerful tool for analyzing scanning transmission electron microscopy (STEM) images, yet its widespread application remains constrained by the scarcity of annotated training data. While deep generative models offer a promising solution, they typically struggle to reproduce the complex high‐frequency components that define experimental STEM images. Here, STEMDiff, a conditional diffusion model that transforms simple binary labels derived from crystal structures into realistic STEM images through a physical information embedding strategy, is proposed. By developing a novel Discrete Wavelet Transform‐based skip‐connection architecture, the high‐frequency bias inherent in diffusion models are addressed, enabling the preservation of experimental noise characteristics. This approach generates images that are quantitatively nearly indistinguishable from experimental data (17 fold improvement over previous methods) while retaining ground truth structural information. Fully convolutional networks trained exclusively on these synthetic images achieve high‐precision atomic column detection in experimental STEM images of WSe 2 and graphene, despite the presence of substantial background noise and contamination. This approach effectively eliminates the need for laborious manual annotation, providing a scalable solution to the data bottleneck in STEM image analysis. The principles underlying STEMDiff can extend to other scientific imaging modalities, accelerating advancements in materials design for water treatment.