研究目的
To develop a power prediction system for photovoltaic power plants using frequency domain identification and FOPDT modeling to address the intermittency and variability of PV power generation.
研究成果
The developed power prediction system using FOPDT modeling and Python software provides accurate predictions of PV power output, as validated with real data. It offers a portable and effective solution for integrating into SCADA systems, supporting better control and utilization of solar energy.
研究不足
The approach relies on accurate identification of the FOPDT model, which may be affected by changes in weather conditions and system performance. The software's portability is emphasized, but it may require specific data inputs and parameter tuning for different PV plants.
1：Experimental Design and Method Selection:
A frequency domain identification method is used to identify a first-order plus dead time (FOPDT) model of the PV power plant from operational data. The model is discretized to create an iterative calculation formula for power prediction. A Python-based software tool is developed for implementation.
2：Sample Selection and Data Sources:
One year of data (from 01/01/2017 to 31/12/2017) from the Jiahuashan PV power plant in Maoming, Guangdong, China, is used. Data includes hourly average power generation, irradiance, temperature, and other meteorological measurements.
3：List of Experimental Equipment and Materials:
PV power plant with Cadmium Telluride thin film technology, meteorological station for weather data, computer systems for software development and testing.
4：Experimental Procedures and Operational Workflow:
Collect operational data, identify FOPDT model using frequency response and least squares methods, discretize the model, develop Python software, and validate with real data.
5：Data Analysis Methods:
Use of MATLAB for model identification and validation, Python for software implementation and prediction, with comparisons to actual power outputs using l2-norm error minimization.
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
