Optimised allocation of sectionalising switches and protection devices in distribution networks by using a reactive tabu search algorithm

Abstract

Optimised allocation of sectionalising switches and protection devices in strategic points of the distribution circuits improves the quality of supplied power and the reliability indices of the system. Sectionalising switches and protection devices are allocated in the system during the planning stage. In some cases, due to consumers habits, vulnerability to faults in some regions of the system, type of consumers and load grow, it is necessary to re-allocate some switches and protection devices in order to improve the reliability and the operation network conditions. Switches are allocated in distribution networks to obtain an optimal or optimised operation strategy and make the power supply re-establishment in case of contingency. Allocation of protection devices in distribution systems is related with the reliability and continuity of the power supply, avoiding significant impacts that a fault could have in terms of consumers outage, reducing times for fault location and system re-establishment. Number and types of protection devices to be installed in a particular feeder will depend on the importance of the system, magnitude and type of load, circuit configuration and exposition to risks of different nature that can lead the system to permanent or temporary fault conditions. Alternatives that minimises the number of switches to be manoeuvred should be preferred since the greater the number of manoeuvres, the greater its execution time, and consequently, the greater the consumers interruption time. In general, impact of devices allocation on customers is related to the frequency and duration of an interruption. In this work, an integer mixed non-linear programming model (IMNLP) with real and binary variables for solving the sectionalising switches and protection devices allocation problem, is presented. Constraints considered for the problem reflect technical and economical limitations, such as series protection devices coordination, number of available equipments, importance of the analysed feeder, circuit topology, quality of power supply and physical limitations of the system, among others. For solving this problem it is proposed a dedicated Reactive Tabu Search Algorithm (RTS). Results and optimised strategies for allocating protection devices and switches considering a real-life network restoration, are presented.