By use of symbolic factorization based on elimination tree theory and improved LU numerical decomposition algorithm based on sparse vector method,the efficiency of power flow calculation by Newton-Raphson method is improved.

Firstly, sparse vector method is discussed by factor matrix and elimination tree; then, based on given sparse matrix data structure and symbolic factorization, an algorithm in which sparse vector method is adopted to decompose LDLT is proposed; finally, combining with residuals variance calculation, two kinds of methods which apply sparse techniques in calculation are given.

The sparse vector method enhances the efficiency of matrix solution algorithms by exploiting the vector sparsity.

Sparse Matrix-Vector Multiply,Ab=x,is one of the important kernels in Scientific Computation,Text Retrieval and Data Mining.

A method of the mixing matrix estimation in underdetermined source separation is proposed,which is based on the linear clustering of sparse component.