Improving the life cycle management of power transformers transforming data to life

Power transformers are essential in the transmission of alternating current and are the most costly equipment in the substation. Although their failure rate is relatively low (less than two percent), the in-service failure of a power transformer can be very catastrophic. The restoration cost of in-service power transformer failures increases by up to seventy-five percent and loss of revenue by sixty percent. It is therefore prudent for utility companies to optimize the life of power transformers and mitigate in-service failures. “Early Detection Saves Lives”, a slogan associated with the fight against cancer is also very relevant to the life cycle management of power transformers. The health of the power transformer is reflected through its insulation strength. There are several on-line and offline tests used to assess the health of power transformer. These give rise to an enormous amount of data to be analyzed. Many utilities manually analyze these results and only focus on the most recent test results. This method of analysis is time consuming and is subject to human error. It abets the back-log of testing and not correcting developing faults in a timely manner. There are benefits that can be gained from using a computer program to store, analyze and trend power transformer data, as well as predict and monitor developing power transformer faults. Software are available to test and monitor power transformer oil via online devices; while other maintenance software analyze single electrical field tests. There is the potential to gain greater benefits from having one computer program that incorporates oil and electrical field tests results, as well as the age, loading and inspection data of the power transformers to determine their health status or condition index. This paper outlines the proposed architecture and features of a custom built Power Transformer Maintenance Software (PTMS). PTMS facilitates a paradigm shift from reactive maintenance (such as practiced at the Jamaica Public Service Company Limited) to a predictive maintenance approach. This study proposes the use of a customized integrated data management system to improve the life cycle management of power transformers.