Zhonglin Zhang

Zhonglin Zhang

Auto-generated profile, based on publicly available information

Zhonglin Zhang

School of Building Services Science and Engineering, Xi’an University of Architecture and Technology, Xi’an, People’s Republic of China

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
1 January 2026 in Engineering Research Express
Verified
Journal Article
Closed access
ZZZhonglin Zhang
Zhonglin Zhang
DCDengfeng Chen
Dengfeng Chen
JYJingyao Yang
Jingyao Yang
JLJiahao Li
Jiahao Li
HXHaiyan Xiao
Haiyan Xiao
QLQianqian Liu
Qianqian Liu
Photovoltaic panels are susceptible to environmental conditions and manufacturing defects during operation. Timely automated detection is crucial for maintaining power generation efficiency and ensuring equipment safety. This paper presents a lightweight enhanced YOLOv11n model for detecting photovoltaic panel defects through visible–infrared image fusion. The proposed model integrates a cross-modulation fusion network that extracts structural texture features from visible images and thermal anomaly features from infrared images, enabling effective cross-modal fusion without explicit geometric registration. Moreover, to enhance both detection accuracy and adaptability for edge deployment, the model incorporates a Semantic-Aware Downsampling Module (NAMDown), a Lightweight Feature Extraction Module (LFEM), and a Location Quality Estimation Head (LQE-Head). Furthermore, the InterpIoU loss function, tailored for small targets, is employed to enhance regression performance. Experiments are conducted on a self-constructed dual-modal dataset for photovoltaic panel defects. The results indicate that the proposed model achieves 81.33% accuracy and surpasses existing lightweight detection methods in both accuracy and recall. Finally, the model is deployed on the Jetson AGX Orin edge platform. With TensorRT optimization, the average inference delay is reduced to 29.4 ms, demonstrating a favorable balance between accuracy and real-time performance.

1 June 2026 in Engineering Research Express
Verified
Journal Article
Closed access
DCDengfeng Chen
Dengfeng Chen
MLMuqing Li
Muqing Li
ZZZhonglin Zhang
Zhonglin Zhang
THTuohang He
Tuohang He
YLYingjun Liu
Yingjun Liu
ZDZhenping Dong
Zhenping Dong
Accurate detection of hidden road defects using ground penetrating radar (GPR) is crucial for pavement maintenance but remains challenging due to complex subsurface noise and blurred defect boundaries. To address these issues and automate GPR data interpretation for infrastructure health monitoring, this study proposes bidirectional feature infusion detection transformer (BDFI-DETR), an enhanced real-time detection model based on the real-time DETR framework. Specifically, a bidirectional feature pyramid network is introduced to improve cross-scale feature interaction. A content-aware reassembly of features operator replaces traditional upsampling to adaptively reconstruct spatial structures and preserve boundary details. Furthermore, a Semantic-Detail Infusion mechanism is embedded to collaboratively integrate high-level semantic and low-level detailed features, significantly enhancing the recognition of small and irregularly shaped defects. Experimental results on a real-world GPR dataset demonstrate that BDFI-DETR achieves a mean average precision of 93.2%, outperforming the baseline by 3.7%, while maintaining a real-time inference speed of 104.1 frames per second. Comprehensive comparisons with mainstream models and robust noise-degradation evaluations confirm that the proposed method provides a highly accurate, robust, and efficient automated solution for large-scale subsurface anomaly detection, facilitating intelligent pavement management.

6 November 2026 in Energies
Verified
Journal Article
Open access
PDF file
HLHaoming Li
Haoming Li
KLKunyang Li
Kunyang Li
ZZZhonglin Zhang
Zhonglin Zhang
DCDengfeng Chen
Dengfeng Chen
Extreme weather stresses the power, thermal and computing infrastructure of data centers, yet energy and computing resources are still scheduled separately. This paper presents a risk-aware energy–compute co-scheduling framework for multi-microgrid integrated energy systems hosting heterogeneous data centers. Learning and optimization exchange a single quantity: a risk-conditional exchange-capability credit. The empirical conditional value-at-risk (CVaR) tail of the net-load distribution, widened by a dispersion- and hazard-calibrated buffer, sets the import headroom of a single-step rolling mixed-integer linear program (MILP) that executes one of five operating postures chosen by offline-trained implicit Q-learning agents, subject to device, storage, queue, migration and unserved-energy constraints. Scenario-level tests over eight held-out scenarios per weather type and three replications produce positive energy-not-served (ENS) margins against rule-based and constant-posture baselines in all six comparisons, and a ten-scenario library confirms a 10/10 sign agreement. In a recorded-event replay, the ENS ratio falls from 9.751% to 2.292%. Sensitivity and scalability tests further confirm that the framework remains effective under parameter variation and at larger system sizes.

9 March 2026 in Engineering Research Express
Verified
Journal Article
Closed access
LZLiang Zhao
Liang Zhao
JLJiahao Li
Jiahao Li
ZSZhan Ping Song
Zhan Ping Song
DCDengfeng Chen
Dengfeng Chen
ZZZhonglin Zhang
Zhonglin Zhang
XLXu Li
Xu Li
As a critical defect affecting the safety and durability of tunnel structures, detection of joint fissures plays a vital role in ensuring tunnel construction safety.This study proposes a joint detection model that integrates attention and super-resolution mechanisms into YOLOv8n, specifically designed for complex tunnel environments. Based on YOLOv8n, a lightweight multi-scale residual attention (MSRA) module is designed to enhance the representational ability of target features and strengthen the semantic expression of feature information. It also achieves a balance between lightweight design and multiscale feature extraction capability. Meanwhile, to improve model diversity and discrimination while reducing computational overhead, a lightweight multi-scale pooling module (DSAP), which combines multi-scale pooling, dual-channel attention, and feature recalibration, is introduced. In addition, to ensure efficient utilization of feature information, a combination of an efficient attention mechanism, a progressive upsampling strategy, and a residual learning module is designed. This component, termed the progressive super-resolution (APSR) module, is designed for image restoration tasks. Finally, experiments were conducted on a self-built dataset. The results demonstrate that the proposed model outperforms other mainstream models in terms of precision and recall, achieving a balance between reducing computational overhead and improving detection accuracy. The mean average precision (mAP) of the proposed model in the generalization experiment reaches 75.6%, exceeding that of other models, thereby verifying its superior robustness.