研究目的
Developing a new technique based on artificial neural networks (ANN) for monitoring of the remote grid connected photovoltaic (PV) plant.
研究成果
The proposed ANN-based monitoring technique successfully predicts the power generated by the PV plant and the power consumed by the load without the need for communication infrastructure. The method is validated through simulations and shows good accuracy in predicting PV plant performance and diagnosing system outages.
研究不足
The proposed method does not employ any communication infrastructure, which may limit its applicability in systems requiring real-time data transmission. Additionally, the study focuses on a specific case study, which may not be generalizable to all PV systems.
1:Experimental Design and Method Selection:
The methodology involves using artificial neural networks (ANN) to monitor the PV power generated and the power consumed by the load at the distribution side, utilizing existing impedance relays’ measurements.
2:Sample Selection and Data Sources:
The study uses a case study of a 10 MWp PV grid connected plant to one of Kuwait Oil Company’s (KOC) power grids.
3:List of Experimental Equipment and Materials:
The study utilizes impedance relays, Matlab Simulink environment for simulation, and a local weather station for PV estimation.
4:Experimental Procedures and Operational Workflow:
The ANN is trained with data obtained from load flow simulations for different values of load and PV power. The network is then tested with testing data.
5:Data Analysis Methods:
The performance of the ANN is evaluated using mean square error (MSE) and gradient descent with momentum (GDM) learning function.
独家科研数据包,助您复现前沿成果,加速创新突破
获取完整内容