The fuzzy regression approach to peak load estimation in powerdistribution systems

摘要

This paper presents a new scheme based on the fuzzy regressionnanalysis for the estimation of peak load in distribution systems. Inndistribution systems, bus load estimation is complicated because thensystem load is usually monitored at only a few points. As a rule,nreceiving nodes are not equipped with stationary measuring instrumentsnso measurements of loads are performed sporadically. In general, thenonly information commonly available regarding loads, other than majorndistribution substations and equipment installations, is billing cyclencustomer kWh consumption. In order to model system uncertainty,ninexactness and random nature of customers' demand, a fuzzy systemnapproach is proposed. This paper presents possibilities of applicationnof the fuzzy set theory to power distribution system calculations.nUnreliable and inaccurate input data have been modeled by means of fuzzynnumbers. Trapezoidal and triangular forms of fuzzy numbers were used forndescription of input data. A regression model, expressing thencorrelation between a substation peak load and a set of customernfeatures (explanatory variables), existing in the substation population,nis determined. Simulation studies have been performed to demonstrate thenefficiency of the proposed scheme on the basis of actual data obtainednat two distribution system substations. The same data have been used fornbuilding standard linear regression models. Comparison of thenperformance of both methods has been done