Abstract:

The continuous upgrading and transformation of the low-voltage distribution network system has increased the complexity of the data information in the power grid.In order to effectively process the acquired multi-source data in real time and improve the calculation speed and fusion efficiency,this paper reforms the traditional neural network and proposes a Hermite orthogonal basis forward neural network based on the theory of polynomial interpolation and approximation.An algorithm model based on Hermite orthogonal basis forward neural network is built,and finally this algorithm is parallelized under the MapReduce framework.Through simulation verification, it is found that the model has higher processing efficiency,faster calculation speed and smaller error for multi-source data of low-voltage distribution network. Therefore, the multi-source data processing and fusion model of low-voltage distribution network based on Hermite orthogonal basis forward neural network proposed in this paper can better meet the real-time processing requirements of multi-source data in low-voltage distribution network.

Key words: low voltage distribution network, multi-source data fusion, Hermite neural network, MapReduce