研究目的
To tackle the problem of model self-adaptation when new categories of equipment failures occur in photovoltaic equipment using a new incremental learning method based on the self-organizing map (SOM) and probabilistic neural network (PNN).
研究成果
The ISOMPNN model effectively utilizes SOM's adaptive learning ability to build a more compact PNN, facilitating incremental learning of different types of new data. Experimental results on actual PV equipment data sets verify the method's effectiveness.
研究不足
The study does not explicitly mention limitations, but potential areas for optimization could include the computational cost of SOM training and the generalization of the model to other types of photovoltaic equipment.
1：Experimental Design and Method Selection:
The study proposes ISOMPNN, a new incremental learning method combining SOM and PNN for adaptive learning of photovoltaic device data.
2：Sample Selection and Data Sources:
Uses data from a certain type of three-phase PV inverter equipment of Delta Company in 2015 and
3：List of Experimental Equipment and Materials:
20 Not explicitly mentioned.
4：Experimental Procedures and Operational Workflow:
Involves training SOM for each category of data, constructing PNN with prototype vectors, and incremental learning for new data.
5：Data Analysis Methods:
Compares the performance of PNN with and without SOM processing, and evaluates incremental learning effectiveness.
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
