CZChiyu Zhao

Chiyu Zhao

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Chiyu Zhao has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Degradation Modeling and RUL Prediction for UAV Bearings Based on a Two-Phase Wiener Process with Stochastic Jumps

    by Ziyi Yu, Xin Zhao, Bincheng Wen, Haizhen Zhu, Changjun Li, Chiyu Zhao

    Accurately predicting the remaining useful life (RUL) of UAV bearings is challenging due to maneuver-shock-induced stochastic jumps during their two-phase degradation, while existing numerical methods are computationally too costly for UAV onboard computing. To address this, an analytical RUL prediction method considering stochastic jumps is proposed. A two-phase Wiener process incorporating stochastic jumps is constructed to model degradation processes involving shocks. Subsequently, a combined Kalman Filter–Rauch–Tung–Striebel Smoothing–Expectation Maximization (KF-EM-RTS) framework is developed for simultaneous online updating of drift and diffusion coefficients. Furthermore, utilizing Stein’s Lemma, an analytical expression under a fixed-change-point assumption for the RUL probability density function (PDF) of the proposed model is derived, thereby reducing the reliance on repeated numerical integration. Under the experimental settings used in this study, the analytical implementation reduces the single-point PDF calculation time by approximately 90% compared with the corresponding numerical integration implementation, which is important for compute-limited UAV platforms. Moreover, RMSE is decreased by 48% and 76% versus models ignoring jumps. This approach offers a lightweight solution for real-time predictive maintenance of UAVs.


  • Open Access

    Source term inversion of nuclear accidents based on the Optuna–GRU model

    by Chiyu Zhao, Sufen Li, Xingfu Cai, Yonggang Huo, Fei Wang

    Rapid and accurate inversion of radioactive source terms following a nuclear accident is critical for risk assessment and emergency response. Static neural networks are widely used for inversion because of their fast inference, but they struggle to capture the temporal dynamics of evolving source terms. To overcome this, we propose a source term inversion model that integrates a Gated Recurrent Unit (GRU) with Optuna’s automated hyperparameter optimization framework (Optuna–GRU). The model leverages simulated environmental radiation monitoring data from three consecutive time steps to predict source strengths at the next two steps. Trained and validated on data generated by the enhanced Gaussian Plume Model (GPM+), the model achieved a MAPE as low as 11.95% and an R2 of 0.9705 at the first prediction step, with R2 remaining high (0.9344) at the second step. Training required only 510.70 seconds, and inference was completed in under 1 ms, enabling real-time application. Sensitivity analysis revealed that performance is most affected by wind speed perturbations and far-field monitoring concentrations, while showing robustness against precipitation and release height uncertainties. By integrating Optuna, the method improves both network structure optimization and training efficiency, achieving high accuracy and strong extensibility.