Node Net Load Forecasting of Distributed Photovoltaic Considering Different User Characteristics

The randomness of Distributed Generation (DG), such as photovoltaic and wind power, has a great influence on the net load forecasting of nodes, and the power consumption modes of different types of users are greatly different. Traditional load forecasting methods are difficult to meet the prediction accuracy requirements. Firstly, the K-means clustering classification method is used to draw typical load curves for residential, business and industrial users. Then a short-term power forecasting model considering weather type index is proposed for different user types including distributed photovoltaic net load forecasting. The prediction model combines Elman neural network and wavelet neural network, which makes up for the defect that the wavelet neural network can not train the starting node, thus ensuring the integrity of data and improving the learning ability. Finally, the validity and effectiveness of the proposed method are verified by an example of a real area with distributed photovoltaics.