Aiming at the problem of insufficient reactive power support capacity of new energy stations under the increasing proportion of DC new energy power generation resources such as photovoltaic and energy storage connected to the power grid
a distributed collaborative optimization method for transient reactive power of new energy stations based on graph convolutional network (GCN) and bidirectional long short-term memory network (BiLSTM) is proposed. Firstly
by constructing the equivalent model of the new energy station system with multiple reactive power resource access
the transient reactive power characteristics of the new energy station under the access of multiple reactive power resources are studied
and the transient reactive power model of the new energy station is established. Then
with the optimization objective of the operating cost of reactive power resources in new energy stations
the transient reactive power collaborative optimization model of new energy stations is established
and the optimization model is solved by combining BiLSTM-GCN. Finally
the near-field simulation system model of new energy station is built
and the effectiveness of the transient reactive power collaborative optimization method proposed in this paper is verified by simulation.
XING L M, LIU W J.A data fusion powered bi-directional long short term memory model for predicting multi-lane short term traffic flow[J]. IEEE transactions on intelligent transportation systems, 2022, 23(9): 16810-16819.
KINGMA D, BA J.Adam: a method for stochastic optimization[C]//3rd International Conference on Learning Representations (ICLR 2015). San Diego, California, USA, 2015.
LU J G, ZHANG L, YANG H, et al.Improved strategy of particle swarm optimisation algorithm for reactive power optimisation[J]. International journal of bio-inspired computation, 2010, 2(1): 27.