There is internal instability in the practical application of rooftop photovoltaic.Therefore,a bidirectional interactive control model for rooftop photovoltaic energy of source network storage integration is studied based on source network and me-ta-learning to solve this problem.The load prediction model is constructed by the multivariate learning method,and the load prediction of multiple controllers such as source,network and storage in the equipment layer is realized by the model.The load prediction results are conveyed through the microgrid layer and used as the basis to control the main network and microgrid of the distribution network.The energy exchange layer receives the control results.After processing,the optimal control scheme is formulated and then transmitted to the photovoltaic energy of the microgrid layer and the equipment layer for interactive con-trol.The energy exchanger in the model uses the edge computing framework to collect and analyze the control data,and these analysis results are used as the basis for the control strategy of the energy exchanger under multiple working conditions to real-ize the bidirectional interactive control of photovoltaic energy.The test results show that the control model can accurately pre-dict the load variation and control the power variation under multiple working conditions,and has a relatively good control effect.
维普
万方数据