Wai-Kin Wong

Mr. Wai-Kin Wong

orcid

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

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

Representative Publications

View All Publications
25 February 2017 in Atmosphere
Verified
Journal Article
Open access
PDF file
WWWang-chun Woo
Wang-chun Woo
WWWai-Kin Wong
Wai-Kin Wong
Hong Kong Observatory has been operating an in-house developed rainfall nowcasting system called “Short-range Warning of Intense Rainstorms in Localized Systems (SWIRLS)” to support rainstorm warning and rainfall nowcasting services. A crucial step in rainfall nowcasting is the tracking of radar echoes to generate motion fields for extrapolation of rainfall areas in the following few hours. SWIRLS adopted a correlation-based method in its first operational version in 1999, which was subsequently replaced by optical flow algorithm in 2010 and further enhanced in 2013. The latest optical flow algorithm employs a transformation function to enhance a selected range of reflectivity for feature tracking. It also adopts variational optical flow computation that takes advantage of the Horn–Schunck approach and the Lucas–Kanade method. This paper details the three radar echo tracking algorithms, examines their performances in several significant rainstorm cases and summaries verification results of multi-year performances. The limitations of the current approach are discussed. Developments underway along with future research areas are also presented.

19 September 2015 in arXiv
Verified
Preprint
Open access
PDF file
XSXingjian Shi
Xingjian Shi
ZCZhourong Chen
Zhourong Chen
HWHao Wang
Hao Wang
DYDit-Yan Yeung
Dit-Yan Yeung
WWWai-Kin Wong
Wai-Kin Wong
WWWang-chun Woo
Wang-chun Woo
The goal of precipitation nowcasting is to predict the future rainfall intensity in a local region over a relatively short period of time. Very few previous studies have examined this crucial and challenging weather forecasting problem from the machine learning perspective. In this paper, we formulate precipitation nowcasting as a spatiotemporal sequence forecasting problem in which both the input and the prediction target are spatiotemporal sequences. By extending the fully connected LSTM (FC-LSTM) to have convolutional structures in both the input-to-state and state-to-state transitions, we propose the convolutional LSTM (ConvLSTM) and use it to build an end-to-end trainable model for the precipitation nowcasting problem. Experiments show that our ConvLSTM network captures spatiotemporal correlations better and consistently outperforms FC-LSTM and the state-of-the-art operational ROVER algorithm for precipitation nowcasting.

30 June 2020 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
SCSongjian Chai
Songjian Chai
ZXZhao Xu
Zhao Xu
YJYouwei Jia
Youwei Jia
WWWai-Kin Wong
Wai-Kin Wong
Deployment of PV generation has been recognized as one of the promising measures taken for mitigating the environmental issues worldwide. To seamlessly integrate PV and other renewables, accurate prediction is imperative to ensure the reliability and economy of the power system. Distinguished from most existing methods, this work presents a novel robust spatiotemporal deep learning framework that can generate the PV forecasts for multiple regions and horizons simultaneously considering corrupted samples. Within this framework, the Convolutional Long Short-Term Memory Neural Network is employed to exploit the temporal trends and spatial correlations of the PV measurements. Besides, given the collected PV measurements might be subject to various data contaminations, the correntropy criterion is integrated to give the unbiased parameter estimation and robust spatiotemporal forecasts. The performance of the proposed correntropy-based deep convolutional recurrent model is evaluated on the synthetic solar PV dataset recorded in 56 locations in US offered by NREL. The comparative study is conducted against benchmarks over different sample contamination types and levels. Experimental results show that the proposed model can achieve the highest robustness among the rivals.

5 October 2017 in arXiv
Verified
Preprint
Open access
PDF file
XSXingjian Shi
Xingjian Shi
ZGZhihan Gao
Zhihan Gao
LLLeonard Lausen
Leonard Lausen
HWHao Wang
Hao Wang
DYDit-Yan Yeung
Dit-Yan Yeung
WWWang-chun Woo
Wang-chun Woo
With the goal of making high-resolution forecasts of regional rainfall, precipitation nowcasting has become an important and fundamental technology underlying various public services ranging from rainstorm warnings to flight safety. Recently, the Convolutional LSTM (ConvLSTM) model has been shown to outperform traditional optical flow based methods for precipitation nowcasting, suggesting that deep learning models have a huge potential for solving the problem. However, the convolutional recurrence structure in ConvLSTM-based models is location-invariant while natural motion and transformation (e.g., rotation) are location-variant in general. Furthermore, since deep-learning-based precipitation nowcasting is a newly emerging area, clear evaluation protocols have not yet been established. To address these problems, we propose both a new model and a benchmark for precipitation nowcasting. Specifically, we go beyond ConvLSTM and propose the Trajectory GRU (TrajGRU) model that can actively learn the location-variant structure for recurrent connections. Besides, we provide a benchmark that includes a real-world large-scale dataset from the Hong Kong Observatory, a new training loss, and a comprehensive evaluation protocol to facilitate future research and gauge the state of the art.

1 March 2015 in Atmospheric Research
Verified
Journal Article
Closed access
GWGaili Wang
Gaili Wang
WWWai-Kin Wong
Wai-Kin Wong
YHYang Hong
Yang Hong
LLLiping Liu
Liping Liu
JDJili Dong
Jili Dong
MXMing Xue
Ming Xue

28 November 2018 in Meteorological Applications
Verified
Journal Article
Open access
PDF file
KCKelvin T. F. Chan
Kelvin T. F. Chan
JCJohnny C L Chan
Johnny C L Chan
WWWai-Kin Wong
Wai-Kin Wong
The rainfall distribution associated with landfalling tropical cyclones (TCs) along the South China coast are examined using radar data from Hong Kong Observatory (HKO). This preliminary study relates the landfall direction, and possible rotations/transitions of TC rainfall asymmetries before and after the landfall. Three types of landfalling TCs on the South China coast are categorized: from the east, normal to the coast and from the southwest. For those from the east, the rainfall maximum rotates anticyclonically from the southwest before later landfall to the northwest. TCs making landfall perpendicular to the coast have a rainfall maximum remaining in the southwest quadrant throughout the landfall process. For TCs making landfall from the southwest, the rainfall maximum rotates cyclonically from the southeast before later landfall to the northeast. The relative importance of the impact of land properties and vertical wind shear on these rainfall asymmetries is investigated by using a numerical model. Sensitivity tests show that the rotations/transitions of landfalling TC rainfall asymmetry appear to be insensitive to land properties (e.g. moisture availability, surface friction and topography), but result primarily from changes in the orientations of the vertical wind shear in the environment. However, the surface friction of the landmass appears to contribute to the enhancement of rainfall intensity.