Duofa Ji

Dr. Duofa Ji

School of Civil Engineering, Harbin Institute of Technology, Harbin 150090, China

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
5 June 2024 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
Duofa JiDuofa Ji
Duofa Ji
CLChenxi Li
Chenxi Li
Changhai ZhaiChanghai Zhai
Changhai Zhai
ZCZelin Cao
Zelin Cao
Solving partial differential equations is a fundamental task for computational electromagnetic and mechanical wave modeling, which hold utmost significance in remote sensing and geophysics. The importance of efficient partial differential equations solving methods lies in their ability to provide rapid simulations and real-time predictions. However, as the scale and dimensionality of the problems increase, the efficiency of current numerical methods becomes a concern. Therefore, this study leverages the inherent connection between the temporal-spatial stepping processes and recurrent neural networks as well as convolutional layers to propose a general-purpose recurrent convolutional neural network (RCNN) platform for solving partial differential equations. The RCNN platform rigorously executes the temporal-spatial iterations involved in discretized forms of partial differential equations. Furthermore, benefiting from the latest advancements in deep learning, the platform enhances the efficiency of convolutional computations. By applying the RCNN platform to electromagnetic, acoustic wave, and seismic wave modeling, among others, its reliability and high efficiency in solving various high-dimensional partial differential equations are demonstrated.

22 May 2018 in Earthquake Engineering & Structural Dynamics
Verified
Journal Article
Closed access
PDF file
Weiping WenWeiping Wen
Weiping Wen
Changhai ZhaiChanghai Zhai
Changhai Zhai
Duofa JiDuofa Ji
Duofa Ji

15 November 2019 in Entropy
Verified
Journal Article
Open access
PDF file
YZYonggang Zhao
Yonggang Zhao
Duofa JiDuofa Ji
Duofa Ji
XYXiaodong Yang
Xiaodong Yang
LFLiguo Fei
Liguo Fei
Changhai ZhaiChanghai Zhai
Changhai Zhai
It is still an open issue to measure uncertainty of the basic probability assignment function under Dempster-Shafer theory framework, which is the foundation and preliminary work for conflict degree measurement and combination of evidences. This paper proposes an improved belief entropy to measure uncertainty of the basic probability assignment based on Deng entropy and the belief interval, which takes the belief function and the plausibility function as the lower bound and the upper bound, respectively. Specifically, the center and the span of the belief interval are employed to define the total uncertainty degree. It can be proved that the improved belief entropy will be degenerated to Shannon entropy when the the basic probability assignment is Bayesian. The results of numerical examples and a case study show that its efficiency and flexibility are better compared with previous uncertainty measures.

6 October 2021 in Journal of Earthquake Engineering
Verified
Journal Article
Closed access
Duofa JiDuofa Ji
Duofa Ji
JLJin Liu
Jin Liu
Weiping WenWeiping Wen
Weiping Wen
Changhai ZhaiChanghai Zhai
Changhai Zhai
WWWei Wang
Wei Wang
Evangelos I. KatsanosEvangelos I. Katsanos
Evangelos I. Katsanos
This study aims to develop a reliable ground motion model (GMM) for CAV by using ground motion (GM) recordings from the PEER NGA-West2 database. A total of 17,684 GM recordings are chosen and randomly separated into the training, validation, and testing datasets. The DNN is advanced by incorporating the refined second-order (RSO) neuron. The effect of seismological and site-specific parameters on the predicted CAV is investigated. The comparative assessment of four existing models with the RSO-DNN model of this study highlights the superior prediction skill of the latter one since the RSO-DNN model is found to be associated with considerably less error.

30 April 2020 in Journal of Earthquake Engineering
Verified
Journal Article
Closed access
Weiping WenWeiping Wen
Weiping Wen
Duofa JiDuofa Ji
Duofa Ji
Changhai ZhaiChanghai Zhai
Changhai Zhai
This manuscript quantifies the effects of aftershocks with pulse characteristics on the cumulative damage of structures. The revised randomization method is developed to generate the artificial mainshock-aftershock (MSAS) ground motions showing pulse characteristics. The efficiency of period normalization is discussed from the point of view of dispersion, and the effects of the aftershocks are quantitatively studied. It is found that strong aftershock can contaminate the efficiency of period normalization, and the effects of strong aftershocks in the near-fault region on cumulative damage can exceed 20% and reach 40%. The equations are proposed for the prediction of damage induced by pulse-like MSAS sequences.

26 January 2024 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
YCYouming Chen
Youming Chen
Duofa JiDuofa Ji
Duofa Ji
QMQiang Ma
Qiang Ma
Changhai ZhaiChanghai Zhai
Changhai Zhai
YMYuhong Ma
Yuhong Ma
Recorded seismic signals are of fundamental importance in earthquake engineering, serving as vital data for analyzing and comprehending various ground motion parameters that are crucial for seismic design. However, ground motions usually suffer from noise contamination, resulting in baseline drift. The drift can introduce significant biases in ground motion parameters, especially those associated with displacement time histories. To address the aforementioned issues, this paper proposes a seismic signal denoising generative adversarial network (SDGAN), which is a conditional Wasserstein GAN composed of a generator and a discriminator. SDGAN is trained on a database constructed from the STanford EArthquake Dataset (STEAD) and the INSTANCE dataset. We compared the trained model with the Butterworth filter, a spectral filtering-based method, and DeepDenoiser, a deep learning-based denoising method. The results demonstrate the SDGAN’s efficacy in the simultaneous removal of various types of noise and baseline drift with minimal information loss, even in scenarios characterized by significantly low signal-to-noise ratio (SNR). Moreover, we substantiate the exceptional generalization performance of SDGAN using the real noisy seismic signals recorded in Kyoshin Network (K-Net).