MODELING A DETECTOR SYSTEM FOR ILLEGAL CONNECTION ON ELECTRICITY DISTRIBUTION

Abstract

Distribution of electricity involves significant Technical as well as Non-Technical Losses (NTL). Illegal consumption of electricity or electricity theft constitutes a major share of NTL. Electricity theft is a growing problem worldwide. A system for detecting illegal connections on electricity distribution was modeled to address this problem. Several methods implemented by illegal consumers for stealing electricity were studied, the impacts of the NTL were also discussed. Customers’ historical billing data for 36 months gotten from BEDC Electricity PLC, Akure Business Unit were analyzed. Twelve (12) selected customers were also inspected. The customers were grouped into four (4) categories according to the nature of their operations, for the analysis - Hotels, Commercial Customers, Industrial Customers and Banks. The results gotten from the analysis were used to categorize the customers into three (3) different consumption bands- Large customers, Medium customers and Small Customers. A Model for detecting illegal connections on electricity distribution was successfully developed using a combination of four different technologies. The usage procedure basically involved feeding the system with data generated through a simulated program. The rules developed compare the consumption of these customers giving due consideration to their consumption size (Large, Medium or Small Customers), the season of the year (Summer or Winter), the day of the week (Weekend or weekday), the time of the day (Mid night, Early Morning, Morning, Late morning, Afternoon, Evening, Night, Late night) and the Voltage level (Normal or abnormal). Twenty-four (24) customers, two (2) representing each class of customer, were fed into the model and their consumption were simulated for week. The system date was adjusted to accommodate both summer and winter seasons. The Model was able to identify 122 suspicious consumption.