Xing-Fa Zi

Xing-Fa Zi


Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
9 May 2025 in Energies
Verified
Journal Article
Open access
PDF file
XZXing-Fa Zi
Xing-Fa Zi
Feiyi LiuFeiyi Liu
Feiyi Liu
MLMing-Yang Liu
Ming-Yang Liu
YWYang Wang
Yang Wang
Deep learning has become a widely used approach in photovoltaic (PV) power generation forecasting due to its strong self-learning and parameter optimization capabilities. In this study, we apply a deep learning algorithm, known as the time-series dense encoder (TiDE), which is an MLP-based encoder–decoder model, to forecast PV power generation. TiDE compresses historical time series and covariates into latent representations via residual connections and reconstructs future values through a temporal decoder, capturing both long- and short-term dependencies. We trained the model using data from 2020 to 2022 from Australia’s Desert Knowledge Australia Solar Centre (DKASC), with 2023 data used for testing. Forecast accuracy was evaluated using the R2 coefficient of determination, mean absolute error (MAE), and root mean square error (RMSE). In the 5 min ahead forecasting test, TiDE demonstrated high short-term accuracy with an R2 of 0.952, MAE of 0.150, and RMSE of 0.349, though performance declines for longer horizons, such as the 1 h ahead forecast, compared to other algorithms. For one-day-ahead forecasts, it achieved an R2 of 0.712, an MAE of 0.507, and an RMSE of 0.856, effectively capturing medium-term weather trends but showing limited responsiveness to sudden weather changes. Further analysis indicated improved performance in cloudy and rainy weather, and seasonal analysis reveals higher accuracy in spring and autumn, with reduced accuracy in summer and winter due to extreme conditions. Additionally, we explore the TiDE model’s sensitivity to input environmental variables, algorithmic versatility, and the implications of forecasting errors on PV grid integration. These findings highlight TiDE’s superior forecasting accuracy and robust adaptability across weather conditions, while also revealing its limitations under abrupt changes.

11 October 2025 in Electronics
Verified
Journal Article
Open access
PDF file
XZXing-Fa Zi
Xing-Fa Zi
Feiyi LiuFeiyi Liu
Feiyi Liu
MLMing-Yang Liu
Ming-Yang Liu
YWYang Wang
Yang Wang
Accurate short-term photovoltaic (PV) power generation forecasting is critical for the stable integration of renewable energy into the grid. This study proposes a Transformer model enhanced with an adaptive sparse self-attention (ASSA) mechanism for PV power forecasting. The ASSA framework employs a dual-branch attention structure that combines sparse and dense attention paths with adaptive weighting to effectively filter noise while preserving essential spatiotemporal features. This design addresses the critical issues of computational redundancy and noise amplification in standard self-attention by adaptively filtering irrelevant interactions while maintaining global dependencies in Transformer-based PV forecasting. In addition, a deep feedforward network and a feature refinement feedforward network (FRFN) adapted from the ASSA–Transformer are incorporated to further improve feature extraction. The proposed algorithms are evaluated using time-series data from the Desert Knowledge Australia Solar Centre (DKASC), with input features including temperature, relative humidity, and other environmental variables. Comprehensive experiments demonstrate that the ASSA models’ accuracy in short-term PV power forecasting increases with longer forecast horizons. For 1 h ahead forecasts, it achieves an R2 of 0.9115, outperforming all other models. Under challenging rainfall conditions, the model maintains a high prediction accuracy, with an R2 of 0.7463, a mean absolute error of 0.4416, and a root mean square error of 0.6767, surpassing all compared models. The ASSA attention mechanism enhances the accuracy and stability in short-term PV power forecasting with minimal computational overhead, increasing the training time by only 1.2% compared to that for the standard Transformer.

29 April 2026 in Materials
Verified
Journal Article
Open access
PDF file
XZXing-Fa Zi
Xing-Fa Zi
HYHongbin Yang
Hongbin Yang
MWMin Wang
Min Wang
DZDeqing Zhang
Deqing Zhang
JZJun Zeng
Jun Zeng
Feiyi LiuFeiyi Liu
Feiyi Liu
This study aims to develop and systematically evaluate a data-driven inverse-design framework for determining silver nanoparticle (AgNP) synthesis conditions that achieve a prescribed particle size. To this end, a forward surrogate model is first developed to learn the nonlinear mapping from synthesis parameters to particle size, and it is then coupled with target-conditioned inverse models, including conditional generative adversarial network (cGAN), conditional Wasserstein GAN (cWGAN), conditional Wasserstein GAN with gradient penalty (cWGAN-GP), Wasserstein GAN with gradient penalty (WGAN-GP), and other comparable frameworks, to generate feasible synthesis conditions. Three AgNP datasets covering microfluidic and chemical synthesis routes are used for evaluation, and the models are assessed using both experimentally observed target sizes and constructed targets spanning the attainable output range. The results show that conditional adversarial models generally outperform the non-adversarial baselines. Among them, cWGAN shows the most consistent performance across the three datasets, while cGAN remains competitive in the more difficult inverse-design cases. The proposed framework also captures the one-to-many nature of inverse design by producing multiple candidate synthesis conditions for a single target size. In addition, prediction errors increase near the lower and upper boundaries of the feasible size interval. Inverse design is therefore more challenging near these boundaries, although the main comparative conclusions remain unchanged under stricter validation. These findings support the use of forward-constrained conditional generative modeling for target-oriented AgNP synthesis design in limited-data settings.