Junfang 君 Wei

Ms. Junfang 君 Wei


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
16 October 2024 in Symmetry
Verified
Journal Article
Open access
PDF file
YLYan-Ling Li
Yan-Ling Li
QSQianxing Sun
Qianxing Sun
JWJunfang 君 Wei
Junfang 君 Wei
CHChunyan Huang
Chunyan Huang
Solving shallow water equations is crucial in science and engineering for understanding and predicting natural phenomena. To address the limitations of Physics-Informed Neural Network (PINN) in solving shallow water equations, we propose an improved PINN algorithm integrated with a deep learning framework. This algorithm introduces a regularization term as a penalty in the loss function, based on the PINN and Long Short-Term Memory (LSTM) models, and incorporates an attention mechanism to solve the original equation across the entire domain. Simulation experiments were conducted on one-dimensional and two-dimensional shallow water equations. The results indicate that, compared to the classical PINN algorithm, the improved algorithm shows significant advantages in handling discontinuities, such as sparse waves, in one-dimensional problems. It accurately captures sparse waves and avoids smoothing effects. In two-dimensional problems, the improved algorithm demonstrates good symmetry and effectively reduces non-physical oscillations. It also shows significant advantages in capturing details and handling complex phenomena, offering higher reliability and accuracy. The improved PINNs algorithm, which combines neural networks with physical mechanisms, can provide robust solutions and effectively avoid some of the shortcomings of classical PINNs methods. It also possesses high resolution and strong generalization capabilities, enabling accurate predictions at any given moment.

27 July 2024 in Water
Verified
Journal Article
Open access
PDF file
YLYan-Ling Li
Yan-Ling Li
JWJunfang 君 Wei
Junfang 君 Wei
QSQianxing Sun
Qianxing Sun
CHChunyan Huang
Chunyan Huang
Accurate runoff prediction is crucial for watershed water resource management, flood prevention, and hydropower station scheduling. Data-driven models have been increasingly applied to runoff prediction tasks and have achieved impressive results. However, existing data-driven methods may produce unreasonable predictions due to the lack of prior knowledge guidance. This study proposes a multivariate runoff prediction model that couples knowledge embedding with data-driven approaches, integrating information contained in runoff probability distributions as constraints into the data-driven model and optimizing the existing loss function with prior probability density functions (PDFs). Using the main stream in the Yellow River Basin with nine hydrological stations as an example, we selected runoff feature factors using the transfer entropy method, chose a temporal convolutional network (TCN) as the data-driven model, and optimized model parameters with the IPSO algorithm, studying univariate input models (TCN-UID), multivariable input models (TCN-MID), and the coupling model. The results indicate the following: (1) Among numerous influencing factors, precipitation, sunshine duration, and relative humidity are the key feature factors driving runoff occurrence; (2) the coupling model can effectively fit the extremes of runoff sequences, improving prediction accuracy in the training set by 6.9% and 4.7% compared to TCN-UID and TCN-MID, respectively, and by 5.7% and 2.8% in the test set. The coupling model established through knowledge embedding not only retains the advantages of data-driven models but also effectively addresses the poor prediction performance of data-driven models at extremes, thereby enhancing the accuracy of runoff predictions.