• Energy Research
• Aurora Universities Network close

This paper presents an application of the Mean-Variance Mapping Optimization (MVMO) algorithm to the identification of the parameters of Gaussian Mixture Model (GMM) representing variability of power system loads. The advantage of this approach is that different types of load distributions can be fairly represented as a convex combination of several normal distributions with respective means and standard deviation. The problem of obtaining various mixture components (weight, mean, and standard deviation) is formulated as a problem of identification and MVMO is used to provide an efficient solution in this paper. The performance of the proposed approach is demonstrated using two tests. Results indicate the MVMO approach is efficient to represented load models.

The characterization of random load behavior has been largely attempted through statistics-based model fitting. Remarkably, the use of Gaussian mixture model (GMM) has proven to be adequate to tackle the heterogeneity and variability of the statistical distribution of loads. In this paper, an application of the Mean-Variance Mapping Optimization (MVMO) algorithm to the identification of the parameters of GMMs, is presented. The feasibility of the proposed identification approach is demonstrated using historical data records from the Venezuelan transmission system portion that covers the Paraguana Peninsula.