Remnant Life Estimation of Power Transformer Based on IFT and Acidity Number of Transformer Oil Saleh Forouhari, A. Abu-Siada Department of Electrical and Computer Engineering Curtin University of Technology Perth, WA m.forouhari@postgrad.curtin.edu.au Abstract—Power transformer is a key asset in any electrical transmission or distribution network. As a significant number of the global power transformers have reached the end of their expected designed life, utilities have given more concern to transformer condition-based maintenance to extend transformer operational life span and to retain the highest viable efficiency of the asset. Transformer failure statistics indicates that most of the failures occur before power transformers reach their expected operational life. Statistics also show that the main cause of transformer failures is attributed to the deterioration of dielectric insulation due to transformer ageing which is highly dependent on accumulated effects of moisture, temperature, Oxygen and acids in transformer oil. When paper insulation exhibits severe deterioration, it loses its tensile strength and its capability to withstand electrical faults significantly decreases. Interfacial tension and acid number of insulating oil are correlated with the number of years in which a transformer has been in service and are used as a signal for transformer oil reclamation. This paper introduces a new, simple and effective fuzzy logic-based model to estimate the remnant life of a power transformer based on the values of interfacial tension and acid number of power transformer insulating oil. Keywords—power transformer; fuzzy logic; acid number; interfacial tension number I. INTRODUCTION The global increase in the population of power transformers operating close or even beyond their expected technical life along with the day-by-day increase in power demand have led into more power transformer failures. In order to fulfil economic and technical criteria, transformer life cycle management has recently gained substantial attention. Based on an IEEE survey, the oil-immersed power transformers failure rate is about 0.00625 per unit year which means that over a course of 16 years, 10 percent of power transformers will be defective [1]. To enhance the reliability, to avoid catastrophic failures and to extend the operational life of power transformers, it is vital to adopt reliable diagnostic monitoring and well-developed condition-based management techniques. The implementation of these strategies would allow power transformers to operate with their most possible efficiency. Transformer failure mainly originates from the degradation of its insulating system which is comprised of insulating oil and paper insulation. Oil and paper insulation deterioration is due to electrical, mechanical, chemical and thermal stresses which power transformer experiences during its operational lifetime. Since the degradation of power transformer oil has a great impact on the aging of the entire power transformer paper insulation, implementing appropriate monitoring methods which are capable of detecting the progress of oil decomposition products at an early stage is crucial [2]. Economic constraints and high capital costs of replacing power transformers have led to an increasing demand of prolonging power transformers life span. A lot of efforts have been made so far by IEEE [3] to produce a life estimation model so as to help utilities achieve this purpose. This paper presents a new fuzzy logic-based method to develop a model estimating the remaining age of a power transformer by using a commonly accepted correlation between acid number, interfacial tension number of power transformer insulating oil and transformer remnant life. II. TRANSFORMER CHEMICAL AGEING The ageing of a power transformer is highly dependent on its paper insulation condition. Having a comprehensive understanding of paper insulation degradation process can help utilities be capable of extending power transformer lifetime. Power transformer paper insulation is made of cellulose which is a polymer consisting of repeated glucose monomers. The three main reactions which can break this polymeric chain to smaller chains are oxidation, pyrolysis, and hydrolysis [4, 5]. Along with accumulative impact of Oxygen, temperature and water on the degradation of paper insulation of a power transformer, acids either originated from oxidation of insulation oil or generated by acid hydrolysis of paper insulation accelerate the rate of paper degradation as well [2, 6]. Oil oxidation produces low-molecular and high-molecular weight acids [6, 7]. Low-molecular-weight acids such as carboxylic acids are more effective on the degradation process of cellulose than high-molecular-weight acids such as stearic or naphtenic acids since low molecular acids are more absorbed by the paper insulation, while high molecular acids are more soluble in the transformer oil. Furthermore, through acid hydrolysis of paper insulation, some other types of acids such as levulinic or formic acids are produced which increase the deterioration rate of paper insulation [6]. Thus, paper insulation degradation process inside a power transformer is a multi-factorial and accumulative 978-1-4799-8903-4/15/$31.00 ©2015 IEEE ISBN: 978-1-4799-8903-4 2015 IEEE 11th International Conference on the Properties and Applications of Dielectric Materials (ICPADM) 552