Distribution system reconfiguration using a modified Tabu Search algorithm

Abstract

This article presents an efficient meta-heuristic method for reconfiguration of distribution systems. A modified Tabu Search (MTS) algorithm is used to reconfigure distribution systems so that active power losses are globally minimized with turning on/off sectionalizing switches. TS algorithm is introduced with some modifications such as using a tabu list with variable size according to the system size. Also, a random multiplicative move is used in the search process to diversify the search toward unexplored regions. The Kirchhoff algebraic method is adopted to check the radial topology of the system. A salient feature of the MTS method is that it can quickly provide a global optimal or near-optimal solution to the network reconfiguration problem. To verify the effectiveness of the proposed approach, the effect of load variation is taken into consideration and comparative studies are conducted on three test systems with rather encouraging results. The obtained results, using the proposed MTS approach, are compared with that obtained using other approaches in the previous work.

Introduction

The subject of minimizing distribution systems losses has gained a great deal of attention due to the high cost of electrical energy and therefore, much of current research on distribution automation has focused on the minimum-loss configuration problem. There are many alternatives available for reducing losses at the distribution level: reconfiguration, capacitor installation, load balancing, and introduction of higher voltage levels. This research focuses on the reconfiguration alternative.

Network reconfiguration is the process of changing the topology of distribution systems by altering the open/closed status of switches. Because there are many candidate-switching combinations in the distribution system, network reconfiguration is a complicated combinatorial, non-differentiable constrained optimization problem. Two types of switches are used in primary distribution systems. There are normally closed switches (sectionalizing switches) and normally open switches (tie switches). Those two types of switches are designed for both protection and configuration management. The change in network configuration is achieved by opening or closing of these two types of switches in such a way that the ‘radiality’ of the network is maintained.

The reconfiguration algorithms can be classified by the solution methods that they employ: those based upon a blend of heuristics and optimization methods, those making use of heuristics alone, and those using some from of artificial intelligence (AI). Numerous researchers advocate the use of a blend of heuristics and optimization techniques. The blend of the two types of technique permits the problem to retain a certain degree of accuracy, while assuring convergence and an acceptable solution time.

In Ref. [1], a branch exchange method that considered the on–off conditions of the sectionalizing switches in discrete numbers was developed [1]. Since the method is based on heuristics, it is not easy to take a systematic way to evaluate an optimal solution.

Two different methods with varying degree of accuracy to approximate power flow in systems were proposed in Ref. [2]. The search method has an acceptable convergence characteristic. However, it can get stuck in local minimum. The method is very time consuming due to the complicated combinations in large-scale systems.

An expert system for feeder reconfiguration, based upon extensions of the rules of Ref. [1] was presented in Ref. [3], with the potential of handling realistic operating constrains. The approach taken is set up a decision tree to represent the various switching operations available. This strategy is efficient for trees that are not too large. However, as a search tree becomes larger, a great amount of time can be spent searching for the optimal solution. To guarantee an optimal solution an exhaustive tree search should be used.

A linear programming method using transportation techniques and a new heuristic search method for comparison with previously developed heuristic techniques which are based on an optimal load flow analysis were presented in Ref. [4]. This study indicates that linear programming, in the form of transportation algorithms, is not suitable for application to feeder reconfiguration since the power loss function is not linear whilst heuristic approaches, although not optimal, can provide substantial saving if properly formulated.

Based on partitioning the distribution network into groups of load buses, the line section losses between the groups of nodes are minimized [5]. By dividing the distribution network into groups of busses, the combinatorial nature of the reconfiguration problem is overcome, while simultaneously minimizing losses.

In recent years, meta-heuristic methods have been studied for solving combinatorial optimization problems to obtain an optimal solution of global minimum. Typical meta-heuristic methods include Simulated Annealing (SA), Genetic Algorithm (GA), and Tabu Search (TS).

A two-stage solution methodology based on a modified simulated annealing technique for solving the reconfiguration problem of distribution systems was proposed in Ref. [6]. In Ref. [7], a modified SA technique for network reconfiguration for loss reduction in distribution systems was presented. An efficient perturbation scheme and an initialization procedure determining a better starting temperature for the simulated annealing approach were proposed. This method can get a solution better than that obtained using the method presented in Ref. [5]. This solution algorithm gives a near-optimal solution but this method does not work so well in the case of load variation.

A GA based method for feeder reconfiguration was proposed in Ref. [8]. Strings which represent switch status, a fitness function consisting of total system losses, and penalty values of voltage drop limit and current capacity limit were formed. Sample results demonstrate that, although the minimal loss solutions were obtained, solution time was prohibitive.

An artificial neural network based method for feeder reconfiguration was presented in Ref. [9]. However, such technique can encounter difficulties, such as getting trapped in local minima, increased computational complexity, and not being applicable to certain objective functions. This led to the need of developing a new class of solution methods that can overcome these shortcomings.

A parallel Tabu Search (PTS) based method for feeder reconfiguration has been proposed in Ref. [10]. PTS introduces two parallel schemes. One is the decomposition of the neighborhood with parallel processors to reduce computational efforts. The other is the multiplicity of the tabu length to improve the solution accuracy. PTS algorithm gives results better than results obtained by SA, parallel Simulated Annealing (PSA), GA, and parallel Genetic Algorithm (PGA). In Ref. [11], a TS algorithm for solving the problem of network reconfiguration in distribution systems in order to reduce the resistive line losses under normal operating conditions was presented. A method for checking system radiality based on an upward-node expression, which has been developed in solving the problem of restorative planning of power system was proposed. In Ref. [12], an efficient hybrid algorithm of SA and TS method for feeder reconfiguration to improve the computation time and convergence property was proposed. In Ref. [13], a modified Tabu Search (MTS) based algorithm for reconfiguration of distribution systems has been proposed. The TS algorithm was introduced with some modifications such as using a tabu list with variable size to prevent cycling and to escape from local minimum. Also, a constrained multiplicative move was used in the search process to diversify the search process toward unexplored regions.

Zhang et al. [14] presented an Improved Tabu Search (ITS) algorithm for loss-minimization reconfiguration in large-scale distribution systems. In ITS algorithm, mutation operation, a main operator used in genetic algorithm, is introduced to weaken the dependence of global search ability on tabu length. In addition, the candidate neighborhood, which only contains several optimal switch exchanges in each tie switch associated loop network, is designed to improve local search efficiency and to save a large amount of computing time. The ITS algorithm in Ref. [14] was applied to the 119-node system and gave an optimal solution.

In this article, an enlarged version of Ref. [13] is introduced to solve the reconfiguration problem. The proposed method is applied to large-scale networks to show the effectiveness of the modified Tabu Search algorithm. In comparison with Ref. [14] in which the mutation operation of GA is used to weaken the dependence of global search ability on tabu length, on the other side, we use a dynamic tabu list with variable size according to the system size and a multiplicative move is applied to diversify the search process and improve the local search efficiency of Tabu Search to reach the global solution. Also, the effect of variation of load is taken into consideration to show the capability of the proposed algorithm (MTS) to work at different load levels.

To verify the effectiveness of the proposed method, comparative studies are conducted on three test systems with rather encouraging results. The proposed method is applied to a 16-node system, a 69-node system, and a 119-node system. The results, obtained using the proposed MTS approach, are compared with results obtained using other modern techniques to examine the performance of the proposed approach.