研究目的
To predict the performance ratio of photovoltaic (PV) pumping system based on grey clustering and second curvelet neural network to provide guidance for regulating the measurements of reducing losses of every part.
研究成果
The second curvelet neural network has the highest prediction precision and efficiency among the tested methods, correctly predicting the performance ratio of PV pumping system. Further research should focus on amending the main affecting factors, optimizing the structure of the second curvelet neural network, and improving the optimization algorithm.
研究不足
The research does not address the amendment of main affecting factors of performance level of PV pumping system, optimization of the structure of second curvelet neural network, and improvement of the optimization algorithm of second neural network to reduce calculation burden and improve prediction efficiency.
1：Experimental Design and Method Selection:
The research proposed a prediction model combining grey clustering and second curvelet neural network. The second curvelet neural network is constructed by combining the second curvelet transform and feed forward neural network.
2：Sample Selection and Data Sources:
Index values affecting the performance level of PV pumping system were collected from 2015/1 to 2017/12, with 30 data sets collected from the field.
3：List of Experimental Equipment and Materials:
The PV pumping system includes an array module with 12 PV units, a centrifugal pump, and a brushless DC motor.
4：Experimental Procedures and Operational Workflow:
The improved grey clustering algorithm processes the collected samples, and the second curvelet neural network is optimized by the firefly algorithm for prediction.
5：Data Analysis Methods:
The performance ratio prediction results are compared using traditional BP neural network and wavelet neural network to verify the effectiveness of the proposed method.
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
