AN INTELLIGENT DECISION SUPPORT TOOL FOR MANAGEMENT OF TREE MAINTENANCE IN POWER SYSTEMS

Abstract

Tree-related power outages represent significant costs and failure rates in power supply. The existence of trees implies that the solution to this problem should focus on optimizing pruning cycles. This optimization is of great importance because of the costs associated to such maintenance. This paper is intended to propose a methodology for the maintenance of trees based on data analysis. Starting from the information captured in the field, we use techniques and models based on fuzzy logic and genetic algorithms, which keeps maintenance tasks on the optimal time and place. The strategy is aimed at two key aspects: first, the history of failures caused by trees on medium-voltage power supplies, this information is used to calculate quality indicators (energy not supplied [ENS] and number of users affected [NU]*) only regarding to tree-associated failure. We use a fuzzy logic based system to ponder and find the critical circuits on which forest maintenance should be performed. Second, we construct an objective function based on treenetwork distance and growth patterns of trees per species in order to simulate forest maintenance. To this end, a genetic algorithm is used to determine the optimum pruning cycle of a given power supply. We use real historical data of medium voltage circuits for the system setting as well as for the evaluation of performance.

Keywords

• Fuzzy systems
• genetic algorithms
• maintenance
• power distribution networks
• trees.