Zhengfang  Wang

Dr. Zhengfang Wang


Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

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

Representative Publications

View All Publications
13 January 2021 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
BLBin Liu
Bin Liu
YRYuxiao Ren
Yuxiao Ren
HLHanchi Liu
Hanchi Liu
HXHui Xu
Hui Xu
ZWZhengfang Wang
Zhengfang Wang
PJPeng Jiang
Peng Jiang
A DNN architecture referred to as GPRInvNet was proposed to tackle the challenges of mapping the ground-penetrating radar (GPR) B-Scan data to complex permittivity maps of subsurface structures. The GPRInvNet consisted of a trace-to-trace encoder and a decoder. It was specially designed to take into account the characteristics of GPR inversion when faced with complex GPR B-Scan data, as well as addressing the spatial alignment issues between time-series B-Scan data and spatial permittivity maps. It displayed the ability to fuse features from several adjacent traces on the B-Scan data to enhance each trace, and then further condense the features of each trace separately. As a result, the sensitive zones on the permittivity maps spatially aligned to the enhanced trace could be reconstructed accurately. The GPRInvNet has been utilized to reconstruct the permittivity map of tunnel linings. A diverse range of dielectric models of tunnel linings containing complex defects has been reconstructed using GPRInvNet. The results have demonstrated that the GPRInvNet is capable of effectively reconstructing complex tunnel lining defects with clear boundaries. Comparative results with existing baseline methods also demonstrated the superiority of the GPRInvNet. For the purpose of generalizing the GPRInvNet to real GPR data, some background noise patches recorded from practical model testing were integrated into the synthetic GPR data to retrain the GPRInvNet. The model testing has been conducted for validation, and experimental results revealed that the GPRInvNet had also achieved satisfactory results with regard to the real data.

11 January 2021 in IEEE Sensors Journal
Verified
Journal Article
Closed access
YJYintao Ji
Yintao Ji
FZFengkai Zhang
Fengkai Zhang
JWJing Wang
Jing Wang
ZWZhengfang Wang
Zhengfang Wang
PJPeng Jiang
Peng Jiang
QSQingmei Sui
Qingmei Sui
Ground penetrating radar methods have been widely used for geological surveys, nondestructive inspections, advanced detections, and other subsurface structural detection processes. In this study, in order to fulfill the inversion processes for the aforementioned applications and reconstruct the permittivity of geo-structures with different actual sizes, a Deep Neural Network based inversion network was proposed to invert the relative permittivity of geo-structures from GPR B-scans. A network referred to as the Permittivity Inversion Network (PINet) utilizes a time dimension compression operation to address attenuation induced issues. In addition, it employs a global feature encoder in order to extract the global information, as well as automatically learn the spatial alignments between the GPR data and the permittivity images. This study constructed a universal dataset containing models with different actual sizes and target abnormities with different shapes and permittivity, for the purpose of training and validating the PINet. The inversion network was first validated using simulated GPR data. The inversion results demonstrated that the PINet was capable of accurately reconstructing permittivity images from GPR data with different central frequencies. Moreover, the performance of the PINet was verified using real GPR data. The position, rough shape and permittivity of the targets can be reconstructed.

25 July 2019 in IEEE Access
Verified
Journal Article
Open access
YDYanan Dong
Yanan Dong
JWJing Wang
Jing Wang
ZWZhengfang Wang
Zhengfang Wang
XZXiao Zhang
Xiao Zhang
YGYuan Gao
Yuan Gao
PJPeng Jiang
Peng Jiang
Periodically inspecting the operational conditions of tunnels, as well as diagnosing any potential structural damages, are vital to ensure uninterrupted and safe tunnel operations. However, the detections of multiple defect in tunnel linings are challenging tasks due to the size imbalance between cracks, spalling, and backgrounds. The currently used deep-learning based methods for multiple defect are dependent on multiple-stage networks, which has limited their scalability and complex frame-working processes. In order to accurately recognize the multiple defect at a pixel-level using only one-stage networks, a new method was proposed in this study. The proposed method integrated the basic SegNet with a focal loss function, and was referred to as an FL-SegNet method. The focal loss function was utilized to address the issue of the size imbalance by down-weighing the losses assigned to the well-classified examples, and then the training was focused on the hard examples. Furthermore, comparative experiments were conducted in order to evaluate the performances of the different methods. The experimental results demonstrated that the proposed FL-SegNet method was capable of accurately predictting the profiles of small-sized cracks and overlapping damages even under various noise conditions, and had successfully outperformed the two-stream method and the basic SegNet method in this regard. The performance metrices (MPA) and (MIoU) of the FL-SegNet method were found to be significantly higher than the state-of-the-art multiple defect detection approaches in different scenarios. Therefore, the results of this study provided a promising and effective solution for the future detection of multiple defect in tunnel linings.

20 September 2021 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
JWJing Wang
Jing Wang
HLHanchi Liu
Hanchi Liu
PJPeng Jiang
Peng Jiang
ZWZhengfang Wang
Zhengfang Wang
QSQingmei Sui
Qingmei Sui
FZFengkai Zhang
Fengkai Zhang
Ground penetrating radar (GPR) enables infrastructure inspection using consecutive and long survey lines. However, the existing GPR data processing methods may lead to distortions or dislocations in the reconstructed shapes of the detected objects or even inconsistencies of the inverted dielectric values when performing inversion or object identification directly using GPR data obtained from the consecutive and long survey lines. To overcome these issues, this study proposed a novel deep neural network (DNN) architecture named GPRI2Net to simultaneously reconstruct the permittivity maps and categorize the object class labels from the GPR data of consecutive and long survey lines. GPRI2Net combined a convolutional neural network (CNN) based on DenseUnet and a recurrent neural network (RNN) based on bidirectional convolutional long short-term memory (Bi-ConvLSTM) to exploit the contextual information in and between the B-Scan segments extracted from the GPR data of a consecutive and long survey line. In addition, GPRI2Net performed inversion and object identification simultaneously using one network, which highly shared features between the two tasks and greatly reduced the computational complexity. Validation experiments were performed at two levels: first, using synthetic data based on the tunnel liner defects model and then using a sandbox model test in a realistic scenario. The results demonstrated that GPRI2Net can reconstruct consecutive permittivity maps and categorize the object class labels from GPR data with different dominant frequencies and survey line lengths and achieved a superior performance using the synthetic data compared to several other methods. Moreover, GPRI2Net also achieved satisfactory results using real-world GPR data.

11 June 2021 in IEEE Access
Verified
Journal Article
Open access
JWJing Wang
Jing Wang
KCKefu Chen
Kefu Chen
HLHanchi Liu
Hanchi Liu
JZJiaqi Zhang
Jiaqi Zhang
WKWenqiang Kang
Wenqiang Kang
ZWZhengfang Wang
Zhengfang Wang
The clutters of rebar in the ground penetrating radar (GPR) images may mask the echoes of the inner defects under the rebars, which adversely affects the identification of the inner structural defects in the reinforced concrete (RC). In this study, a deep learning-based method for rebar clutters removal and defect echoes enhancement in GPR B-scan images was proposed. The residual-inception blocks and attention modules were designed in the network as per the characteristics of the task. Validation of the method has been performed at three levels: first using synthetic data, next by synthetic data superimposed with actual concrete background noise, and finally using a sandbox model test in a realistic scenario. The results indicate that the proposed method had the ability to effectively remove rebar clutters and reconstruct complete defect echoes, with performance superior to those of the traditional deep-learning based methods. A RC structural defect identification experiment was performed to verify the effectiveness of the proposed method. The results indicated that after the rebar signals were removed using the method proposed, the accuracy for identifying the defects under the rebars had been improved from 0.208 to 0.850. The proposed method is promising for applying to practical engineering due to its fast processing and better performance.

20 February 2025 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
SGShuhua Gao
Shuhua Gao
YXYizhuo Xu
Yizhuo Xu
ZZZhaoqian Zhang
Zhaoqian Zhang
ZWZhengfang Wang
Zhengfang Wang
XZXiaoyu Zhou
Xiaoyu Zhou
JWJing Wang
Jing Wang
Microgrids equipped with hybrid energy storage systems (ESSs) are increasingly critical for balancing the intermittency of renewable energy sources (RESs) and the fluctuations in demand. This paper introduces a novel Multi-Agent Imitation Learning (MAIL) framework for real-time energy management in microgrids, particularly under real-time pricing conditions. The approach leverages a problem decomposition strategy, which separates the energy management task into two phases: first, optimal actions for each ESS are estimated using individual agents, each employing a deep neural network (DNN) trained to emulate an ideal mixed-integer linear programming (MILP) solver; then, a one-step online optimization reacts to these decisions to optimize the remaining system components holistically. Rigorous proofs establish the equivalence of the decomposed subproblems to the original optimization problem, ensuring integrity and effectiveness. Comparative simulation studies with real-world data reveal that our MAIL framework offers significant cost advantages and superior training efficiency relative to both traditional single and emerging multi-agent reinforcement learning (RL) methods. Notably, the operational costs achieved through our approach are markedly lower than those of its counterparts, closely approximating the theoretical minimum. This underscores the method’s proficiency and its potential as an effective solution for microgrid energy management. Moreover, the MAIL strategy’s multi-agent design allows for straightforward scalability to more extensive systems involving multiple ESSs and shows promising potential for cooperative management across interconnected microgrids.