Yuyu Liu

Dr. Yuyu 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
27 August 2026 in Mathematics
Verified
Journal Article
Open access
PDF file
WZWei Zhang
Wei Zhang
YLYuyu Liu
Yuyu Liu
HLHong Li
Hong Li
Very short-term wind power forecasting without future numerical weather prediction (NWP) must exploit historical observations while controlling operating-state heterogeneity, erroneous analog retrieval, and unstable expert corrections. This study proposes state-aware dynamic retrieval and persistence-anchored expert fusion (SADR-PAEF). A continuous operating-state representation supports training-only historical retrieval, while persistence, analog-trajectory, and neural residual experts are coordinated by a state-aware gate. Persistence-biased initialization, entropy-based contraction based only on gate-weight concentration, and validation-selected anchoring constrain excessive deviations from persistence; gating entropy is treated as a heuristic stabilization factor rather than a calibrated reliability measure. State-Adaptive Online Conformal Prediction (SAOCP) provides state-conditional online interval calibration. Across five runs on the 2024 main-site dataset, SADR-PAEF achieved an MSE of 44.0526 ± 0.2307 MW2, 3.22% lower than DLinear, with significant four-horizon-aggregated squared-error improvements over all seven baselines. Independent retraining on the 2019 site yielded consistent performance under another site distribution, which is interpreted as site-level replicability rather than cross-site transfer or zero-shot generalization. At 90% nominal coverage, SAOCP achieved the lowest overall Winkler score and the lowest Winkler score at all four horizons, although Rolling CP was marginally closer to nominal coverage at 15 min. Persistence retained a slight MAE advantage. The proposed framework therefore primarily reduces squared-error and large-deviation risk while providing adaptive interval calibration under NWP-unavailable conditions.

8 October 2026 in Energies
Verified
Journal Article
Open access
PDF file
YLYuyu Liu
Yuyu Liu
WZWei Zhang
Wei Zhang
GHGaining Han
Gaining Han
HLHong Li
Hong Li
Local power changes, weak-output states, and changing interval coverage complicate ultra-short-term wind power forecasting. This study develops power-increment enhancement and signed adaptive conformal inference (PIE-SACI) to address these challenges within a joint point and probabilistic forecasting framework. XGBoost predicts power increments relative to the latest observation and is combined with a TimeXer temporal forecast; a Persistence-based low-power fallback (LPF) adjusts forecasts under weak-output conditions. Conditional residual quantiles provide base intervals, which are calibrated using the conventional signed CQR score, horizon-specific miscoverage states, and score histories updated only when target labels become available. Across 15, 30, and 60 min horizons on the primary dataset, the complete model reduces RMSE by 3.35–5.46% relative to TimeXer. Matched controls show that TimeXer adds only a modest incremental benefit when LPF is held fixed. At 90% nominal coverage, Signed-ACI achieves coverage errors of 0.0952–0.3998 percentage points and the lowest mean Winkler score among the compared methods at each horizon. Ablation and coverage-level analyses support the combined use of adaptive feedback and signed corrections. Independently retrained Hackberry experiments at 1, 2, and 4 h horizons show similar performance trends under the stated offline data-repair procedure. The results support the empirical integration of increment prediction, low-power adjustment, and delayed-feedback calibration for multi-horizon wind forecasting.

1 November 2026 in Mathematics
Verified
Journal Article
Open access
PDF file
LXLiu Xiao
Liu Xiao
YLYuyu Liu
Yuyu Liu
HLHong Li
Hong Li
This study examines whether integrating structured priors with target-aware user histories improves movie rating prediction beyond supervised calibration. CDIAC combines a CatBoost prior, a target-aware Deep Interest Network (DIN) fusion branch, and ridge-based calibration (AC). A chronological MovieLens 32M protocol separates model selection, calibration, and testing and uses temporal out-of-fold priors; calibration coefficients remain fixed during inference. Across five runs on 148,738 warm-start test ratings, the model with a 20-epoch training cap achieves RMSE 0.762500 ± 0.000655 and MAE 0.570105 ± 0.000635. Compared with XGBoost + AC using identical calibration labels, the paired RMSE difference is −0.001500, with a user-cluster bootstrap 95% interval of [−0.002834, −0.000195]. Training and inference require approximately 35.09 and 1.49 times as long, respectively. Component ablations support predictive contributions within the tested pipeline, without establishing DIN-specific superiority. AC reduces aggregate RMSE but increases absolute mean signed error in all four rating groups and pooled RMSE for ratings ≤ 2. These findings indicate a modest accuracy gain at substantially higher training cost. The evidence is limited to the evaluated temporal split and warm-start population: 97.87% of test ratings involve users with at least 50 prior interactions. Feature importance and SHAP characterize only the CatBoost branch.