Metaheuristic methods for improving the efficiency of distribution networks

Abstract

Relevance. Distribution networks (RS), mainly having a radial (open) configuration, traditionally relied on centralized production and supply of electricity from power centers to consumers. At the same time, they face a number of challenges, including increased electricity losses, deterioration in electricity quality, in particular voltage deviations from nominal values, low power factor, and equipment overload. In this context, the integration of distributed generation (RG) elements, such as photovoltaic power plants (FES), as well as compensating devices (CU), such as capacitor banks (CB), is one of the most promising areas for increasing efficiency and improving the quality of electricity in the RS. FES, as an element of the RG, mainly generate active power in close proximity to the consumer, reducing the load on the head sections of the network, and CB, in turn, are a classic means of compensating for reactive power, reducing the reactive component of the current and improving the voltage profile. When properly connected to the power grid, they lead, for example, to a reduction in EE losses, an improvement in the voltage profile, etc. But if they are connected incorrectly, they have the opposite effect. Therefore, the task of choosing the installation location and capacity of the RG and KB is a very important and urgent task, affecting primarily the improvement of the energy efficiency of the RS.

Goal. Analysis and selection of metaheuristic methods for optimal placement of RG and CU elements in radial distribution networks in order to reduce losses in them.

Methods. 35 different metaheuristic methods are considered to select the installation locations and the capacity of the RG and CU. Based on the results of the analysis, methods are determined, using which the probability of obtaining an optimal solution is very high. The proposed methodology includes the following steps: selection of methods; selection of installation locations and capacity of elements based on the selected methods; improvement of indicators by additional configuration. Three installation options are being considered: with two, three and four FES and a similar number of KB for the IEEE 33 and 69 node test distribution network. The steady-state PC mode is calculated using a two-stage method with reverse and forward running. The objective function is aimed at minimizing active power losses, improving the voltage profile, as well as voltage sensitivity and power loss factors.

Results. An analysis of the results obtained shows the effectiveness of the proposed methodology for finding the optimal solution, as well as the significant potential for synergy between RG and KU to reduce losses, improve electricity quality, delay grid modernization and increase economic profitability