Huawei Mei

Huawei Mei

Auto-generated profile, based on publicly available information

Huawei Mei

North China Electric Power University 2 Department of Computer Baoding, 071003

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
29 July 2023 in Engineering Applications of Artificial Intelligence
Verified
Journal Article
Closed access
WCWangbin Cao
Wangbin Cao
YLYanping Liu
Yanping Liu
HMHuawei Mei
Huawei Mei
HSHonglin Shang
Honglin Shang
YYYang Yu
Yang Yu

21 October 2023 in Energy and Buildings
Verified
Journal Article
Closed access
HMHuawei Mei
Huawei Mei
YLYanping Liu
Yanping Liu
WCWangbin Cao
Wangbin Cao
YYYang Yu
Yang Yu

13 September 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
HMHuawei Mei
Huawei Mei
ZWZhenheng Wang
Zhenheng Wang
WCWangbin Cao
Wangbin Cao
JZJinghong Zhang
Jinghong Zhang
JLJincheng Li
Jincheng Li
QZQingyuan Zhu
Qingyuan Zhu
Accurate photovoltaic (PV) power forecasting is essential for the reliable operation of power systems. To enhance forecasting accuracy, this study proposes a Weather-Modulated State Tributary Space Model (WM-STSM) with a trend-seasonal enhancement strategy in a dual-branch forecasting framework. Specifically, WM-STSM differentiates the feedback contributions of historical states using content relevance and a local decay prior, while recursive state aggregation enables multi-hop, multi-path propagation, thereby improving the modeling of historical dependencies across time steps. A weather condition representation further modulates state feedback and readout, allowing the effective contribution of historical information to adapt to changing weather conditions. For trend-seasonal enhancement, first- and second-order differences with temporal downsampling are applied to the trend component, while depthwise one-dimensional convolution is applied to the seasonal component while preserving the original temporal resolution. This strategy strengthens the variation information of the trend component and the local fluctuations of the seasonal component, thereby improving the representation of their complementary temporal information. Comparative experiments on three public PV datasets against eight representative models demonstrate the superior overall performance of WM-STSM, including Transformer, Mamba, and Sessa. The proposed method achieves R2 values of 0.9983, 0.9944, and 0.9670 on the three datasets, respectively, indicating accurate and consistent forecasting performance.

1 September 2026 in Journal of Renewable and Sustainable Energy
Verified
Journal Article
Closed access
YWYusen Wu
Yusen Wu
ZWZhenheng Wang
Zhenheng Wang
HMHuawei Mei
Huawei Mei
Photovoltaic (PV) power generation offers significant potential due to its clean and renewable nature. To address the challenges posed by the volatility of PV generation on grid integration and scheduling, this paper proposes a hybrid forecasting method for 24-h-ahead PV power forecasting. We first use the Spearman correlation coefficient and feature selection experiments to improve data input quality. Subsequently, we introduce the Dual-Wing Collaborative Model (DWCM), incorporating a stacking algorithm, GRU-CNN-SE-LR, in the left-wing module. This module leverages the strengths of the Gated Recurrent Unit, Convolutional Neural Network, and the Squeeze-and-Excitation attention mechanism for initial forecasts. In the right-wing module, we integrate an Extreme Gradient Boosting (XGB) model optimized through random grid search (RGS), referred to as RGS-XGB, specifically tailored for short-term predictions. This model enhances accuracy by optimizing left-wing forecasts. The left wing focuses on longer-term forecasting (24 h), while the right wing addresses shorter-term predictions (12 h). By partially substituting left-wing predictions with those from the right wing, the DWCM demonstrates exceptional day-ahead forecasting performance. Validation experiments using historical data of varying resolutions confirm the DWCM's superiority over five advanced models in terms of mean absolute error, mean squared error, and relative absolute error, validating the advantages of the foundational model and short-term optimization strategy.