Abstract:

The node admittance matrix is a sparse matrix, and the short circuit current calculation needs to query the admittance matrix data. In order to keep querying the element numerical value according to row and column and to further improve the efficiency of querying the line number according to the numerical value, which can facilitate the storage and subsequent matrix processing, we propose an improved orthogonal list method for constructing the highly balanced binary tree. Based on productive capacity table stores,  the data node pointer field is expanded so as to form a balanced binary tree. The tree’s whose height is maintained at (O(log2 n)),and its average search length is maintained at (O(log2 n)).It can reduce operation time complexity and improve the efficiency of numerical query. At the same time, in order to ensure the fairness of the test results, the time to construct the highly balanced binary tree is included in the total time for comparison. The corresponding examples verify the efficiency of the improved method.

Key words: sparse matrix, orthogonal list, highly balanced binary tree, query, efficiency