Lianda Duan

ORCID: 0000-0003-0783-6436
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Energy Load and Power Forecasting
  • Computational Physics and Python Applications
  • Machine Fault Diagnosis Techniques
  • Solar Radiation and Photovoltaics
  • Photovoltaic System Optimization Techniques
  • Grey System Theory Applications
  • Advanced Neural Network Applications
  • Image and Signal Denoising Methods

Energy Research Institute
2023

China Institute of Water Resources and Hydropower Research
2022

Abstract In recent years, wind energy, as a ubiquitous, easy-to-capture cost-effective clean is accounting for sharp increase in the installed capacity of China’s new energy grid. However, its stochastic volatility has always brought challenges to generation scheduling farms. To better optimize management level farms and improve stability power grid connection, this paper proposes design forecasting system based on deep learning. Our architecture mainly includes data pre-processing,...

10.1088/1742-6596/2562/1/012043 article EN Journal of Physics Conference Series 2023-08-01

With the rapid development of wind and photovoltaic power generation, hydro-turbine generator units have to operate in a challenging way, resulting obvious vibration problems. Because significant impact on safety economical operation, it is great significance study causal relationship between other variables. The complexity unit makes difficult analyze causality mechanism. This paper studied correlation based data-driven method, then transformed into In terms correlation, traditional...

10.3390/en15031207 article EN cc-by Energies 2022-02-07

In order to replenish the oil in tank time and reduce loss caused by insufficient volume, we use wind turbines as an example build a 1,300 picture dataset containing different levels. And YOLOv5 network model based on PyTorch framework trains relevant dataset, are detected through training identify volume of tank, effect can meet industrial applications. The experimental results show that established this paper is 96% average accuracy level recognition, which effectively solves problem...

10.1117/12.2685639 article EN 2023-08-15
Coming Soon ...