Design and implementation of I/O performance prediction scheme on HPC systems through large-scale log analysis

Article Engineering, Electrical & Electronic

A new ensemble deep graph reinforcement learning network for spatio-temporal traffic volume forecasting in a freeway network

Pan Shang et al.

Summary: A new ensemble deep graph reinforcement learning network is proposed in this study to construct a high-precision traffic volume forecasting model. The experimental results demonstrate that the model can effectively integrate two different graph neural networks and achieve better prediction results compared to traditional ensemble methods.

DIGITAL SIGNAL PROCESSING (2022)

添加到收藏夹

Article Engineering, Aerospace

A novel direct performance adaptive control of aero-engine using subspace-based improved model predictive control

Qian Chen et al.

Summary: The direct performance predictive control method improves the engine's operating performance with better control effect compared to traditional methods. This paper proposes a subspace-based improved model predictive control method and a high-precision real-time estimation method of nonlinear onboard performance parameters based on spherical unscented Kalman filter. The methods successfully achieve direct performance adaptive control and reduce computational complexity while maintaining accuracy. The simulation results demonstrate improved dynamic performance of 32% compared to traditional control.

AEROSPACE SCIENCE AND TECHNOLOGY (2022)

添加到收藏夹

Proceedings Paper Computer Science, Software Engineering

Simulation-Based Performance Prediction of HPC Applications: A Case Study of HPL

Gen Xu et al.

PROCEEDINGS OF 2020 IEEE/ACM INTERNATIONAL WORKSHOP ON HPC USER SUPPORT TOOLS (HUST) AND THE WORKSHOP ON PROGRAMMING AND PERFORMANCE VISUALIZATION TOOLS (PROTOOLS) (2020)

添加到收藏夹

Proceedings Paper Computer Science, Hardware & Architecture