Attention Mechanism Based Calculation of Pseudo-measurements of Key Nodes

In order to solve the problems of insufficient real-time measurements in the system after access to large-scale distributed re-sources, low system redundancy, poor efficiency of traditional model training, and limited accuracy of pseudo-measurements, a key node pseudo-measurement modelling and calculation method based on an attention mechanism is proposed. First, a node comprehensive importance index is constructed using the weighted sum of three indicators: node connectivity, node density, and node power intermediation. A key node screening method based on comprehensive importance is proposed to measure the importance of nodes in the network from three perspectives: structure, state, and frequency. Second, the key nodes are taken as the research object, and a CNN-GRU pseudo-measurement generation model based on the attention mechanism is established to compute and deduce the pseudo-measurement of the injected power at the key nodes. Finally, simulation tests are carried out on an IEEE 33-node standard distribution network. The results show that the proposed method can correctly identify key nodes in the network and significantly improve the accuracy of the pseudo-measurement data.