Abstract
Continuous improvement is vital for organisations to sustain their day-to-day operations. The electricity distribution companies are confronted with operational challenges such as frequent failure of aging equipment, high maintenance cost, dealing with uncertainty, changes in regulation by the National Energy Regulators of South Africa (NERSA), and expectations of continuous reliability improvements. Many utilities and organizations throughout the world have established strategies and procedures for improving the performance and reliability of their systems, to operate effectively and efficiently. This research seeks to investigate gaps in Reliability Centered Maintenance (RCM) of equipment in power distribution substations, and further develop a reliability maintenance algorithm using Artificial Neural Network to optimise the maintenance strategy. The major challenge in electricity distribution companies is associated with the aging infrastructure that frequently fail, due to poor equipment health prediction, which leads to poor reliability of electricity supply to customers. Utility companies depend on asset management plans to accomplish their maintenance activities in sustaining equipment condition. Studies have shown that frequent failures of distribution substation equipment are contributory factors to power outages that lead to poor performance of electrical distribution systems. A major requirement in power distribution companies is to supply customers with reliable electricity with minimal to zero electricity supply interruptions, excluding the loadshedding events. The optimised RCM algorithm in this research was developed using Artificial Neural Network (ANN). The algorithm has the potential of improving the reliability of distribution substations with aging infrastructure by reducing the failure rate, durations, and cost associated with aging power infrastructural maintenance. The optimal algorithm was achieved by predicting a precise health status of each component of equipment before an appropriate maintenance strategy was applied. The changes in reliability were measured based on the number of substation equipment failures and the number of unplanned outages. The Key Performance Indicators were utilised to measure the system performance and reliability. System Average Interruption Duration Index (SAIDI) has improved by lowering the average monthly customers power interruptions by 19,15 % lesser time spent without electricity. The System Average Interruption Frequency Index (SAIFI) was improved by lowering the average monthly power interruptions frequency by 20,63 % lesser outage frequency. The Distribution Supply Loss Index (DSLI) was
vi
improved by dropping the average monthly power losses by 7,5 % power losses saved. The Customers Average Interruption Duration Index (CAIDI) was improved by lowering the average customer interruption by 36.1% improvement from the average period before the RCM optimization. The Average System Availability Index (ASAI) was improved by shortening the average annual interruption durations by 1,36%. The ASUI was improved by increasing the average duration of electrical power availability by 1.28%. The total RCM costs were reduced by the average of R964,342.00 annually between 2019 and 2022. The The optimal RCM using ANN algorithm haoptimal RCM using ANN algorithm hass improved the reliability of the selected improved the reliability of the selected aged aged substationsubstation equipment by lowering the annual outage duration, equipment by lowering the annual outage duration, frequencies,frequencies, and power and power losseslosses.. SAIDI was improved by lowering the average monthly power interruptions durations. The SAIFI was improved by lowering the average monthly power interruptions frequency leading to lesser outage frequency. The DSLI was improved by dropping the average monthly power losses of power transformer annually. The developed algorithm enhanced the effectiveness of determining the health status of the equipment applicable to system reliability judgement, maintenance regulation, maintenance planning, prioritization of maintenance, and maintenance policy formulation. As a recommendation, the optimal algorithm will guide assets and maintenance managers towards effective decision-making, by balancing the reliability with associated maintenance cost.