A Statistical Method for Estimating Maximum Electrical Loads in Urban Residential and Public Sectors

Abstract

Relevance: the main consumers of urban power supply systems are residential and public buildings; determining their maximum (design) electrical load is decisive when designing, reconstructing and justifying energy-saving measures. The current normative methods find the load through averaged coefficients as a function of the number of consumers and take into account neither the statistical dispersion of loads of homogeneous consumers nor the intergroup correlation, so that for some groups the design load is determined with an error of up to 18,9 %, on average 6,2 %, which leads to unjustified oversizing of cable cross-sections and transformer capacity or, conversely, to their overload.

Aim: to develop a statistically grounded new method and calculation formula for determining the maximum (design) electrical load of homogeneous consumers in the residential and public sectors of cities and to verify its accuracy against experimental measurements.

Methods: probability theory and mathematical statistics, multivariate statistical modelling of electrical loads, comparative analysis of normative documents and scientific literature, and comparison of calculated results with field measurements.

Results: a new formula is proposed that expresses the coincidence factor of homogeneous consumers through their number and load dispersion. Experimental verification in a residential district of Tashkent showed that the new method reduces the mean absolute error from 6,2 % to 3,4 %. The results can be used when selecting the capacity of urban distribution networks and transformer substations.