Yunfei Liu

Mr. Yunfei Liu

orcid

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

16 May 2026 in Coatings
Verified
Journal Article
Open access
PDF file
YLYunfei Liu
Yunfei Liu
GBGuangtao Bao
Guangtao Bao
RFRui Fu
Rui Fu
TLTze Liang Lau
Tze Liang Lau
Protective coatings and surface-protection systems improve structural durability, but the long-term performance of durability-sensitive infrastructure also depends on the cyclic stability of supporting soil–rock mixture (SRM) foundations. In this study, undrained multistage strain-controlled cyclic triaxial tests were conducted on reconstructed SRMs with rock block contents of 0%, 10%, 20%, 40%, and 60% under confining pressures of 100, 200, and 400 kPa. Hysteresis-loop morphology, secant shear modulus, normalized shear modulus ratio, damping ratio, normalized damping ratio, and fitting parameters were evaluated. The results show that hysteresis loops evolved from narrow and steep to wider and fuller forms as strain amplitude increased, indicating stiffness degradation and enhanced hysteretic dissipation. The secant shear modulus decreased from 35.835 to 158.871 MPa to 3.296–12.854 MPa, corresponding to an overall reduction of approximately 85%–94%, while the damping ratio increased from 0.036 to 0.063 to 0.195–0.268. Higher rock block content and stronger confinement increased absolute stiffness, but rock block content advanced normalized degradation and damping development, whereas confinement delayed these normalized responses. These findings provide experimental evidence for dynamic-parameter selection, deformation-compatibility evaluation, and cyclic stability assessment of complex SRM foundations.

16 May 2026 in Applied Sciences
Verified
Journal Article
Open access
PDF file
YLYunfei Liu
Yunfei Liu
BXBinxiangyu Xiao
Binxiangyu Xiao
CLChunpeng Liu
Chunpeng Liu
TLTze Liang Lau
Tze Liang Lau
Severe transients in nuclear power plants (NPPs) are strongly coupled and highly nonstationary, which makes reliable short-horizon multivariate forecasting difficult for conventional sequence models. To address this challenge, this study develops a hybrid LSTM-Transformer forecasting framework for severe nuclear accident time series and uses the Sparrow Search Algorithm (SSA) as a task-oriented joint hyperparameter optimization tool for nuclear accident forecasting. In this framework, the self-attention mechanism captures long-range temporal dependencies and cross-variable interactions, while the LSTM component strengthens the modeling of short-term dynamics and local temporal memory. SSA is employed as a task-oriented joint hyperparameter optimization tool to adapt key model settings, including the number of attention heads, encoder depth, model dimension, LSTM hidden units, and dropout rate, for severe nuclear accident forecasting. In addition, a regularized training strategy combining dropout and validation-based early stopping is adopted to alleviate overfitting and improve training stability. The main comparison results are reported as mean ± standard deviation over 20 independent runs with the same data split and different random seeds. Experiments on high-fidelity PCTran/APR1400 simulations covering LOCA, LACP, and SLBIC scenarios, together with a severity-shifted LOCA test, demonstrate strong and statistically stable predictive performance. Across the three representative accident scenarios, the proposed framework achieves mean R2 values of 0.943 ± 0.009, 0.951 ± 0.007, and 0.946 ± 0.010, while maintaining about 30% lower mean nRMSE and nMAE than the strongest LSTM-Transformer baseline. A 2 × 2 ablation study shows that regularization mainly improves training efficiency, reducing the required epochs by a range of about 36–41%, whereas SSA primarily improves predictive accuracy through better hyperparameter selection. Their combination provides the best overall generalization. Cross-severity LOCA evaluation further confirms the robustness of the proposed model, yielding mean R2 = 0.885 ± 0.017 and mean nRMSE = 0.100 ± 0.010. The model also achieves low inference latency (P50 = 7.6 ms per sample), indicating its computational potential for near-real-time multivariate forecasting in safety-critical transient monitoring.