Technique for field data based identification of static polynomial load model

The aim of the research is to develop and test technique for obtaining load static characteristics of large industrial enterprise. The research is initiated because the available data about Russian power system load have become outdated due to the change of power system structure, load content and behavior. At the same time the existing methods do not properly solve the problem of ill-conditioned data which is essential for such stochastic system as electrical load. The solution of this problem can be provided only on the basis of multiple experiments accounting different operating conditions of power system. The results of the first stage of such experimental work scheduled for the next several years by Russian System Operator is presented in the paper. To collect the data required for developing and testing of new multistage technique for identification of static polynomial load model, an active experiment was conducted. The proposed technique combines the successive approximation method and least-square method for filtering ill-conditioned data and estimation of polynomial model parameters. The obtained results confirm the efficiency of the proposed technique.