Abstract: To meet the requirements of online computing speed of state estimation for the large-scale power grid,this paper proposes a fast rotation strategy of Givens orthogonal transformation based on row sparsity isostructuralism.This strategy takes advantage of the distribution characteristics of the state estimation measurement of the large-scale power grid and the sparsity of the weighted Jacobian matrix,dynamically selects rotating row pairs according to the row isomorphism pattern,and selects the rotation pivot and ranks the rows based on the row sparsity,which reduces the non-zero fill-in amount and the number of Givens rotation,and greatly reduces the computation time.The calculation results of a large-scale power grid show that the proposed strategy has a higher computation efficiency without changing the calculation accuracy,and the efficiency of the Givens rotation and the number of intermediate non-zero fill-in elements are effectively improved,which fully meets the requirements of real-time state estimation for the large-scale power grid.