Application of Time/Temperature Fire Curves for the Estimation of Fire Resistance of Transformer within Protective Structures

Abstract

At each power facility, a number of key elements can be identified thanks to which the power industry functions. One of these elements is transformers, which contain oil in their constructions. Being influenced by such factors as high oil temperature during operation of the device and actual flammability of this dielectric they become hazardous. Fires involving transformers are considered crucial due to the large amount of oil in contact with high-voltage elements. Therefore, it is advisable to prevent fires involving transformers. This condition can be met by maintaining the power equipment of transformer stations in good condition and without overheating. At the same time, transformers are currently a priority target for remote bombing due to Russian aggression against Ukraine. The high failure rate and the corresponding consequences for transformer reliability and safety require in-depth assessment. The paper briefly describes possible preventive measures that are currently being implemented in Ukraine to avoid incidents related to fires involving transformers. The application of standardized time/temperature fire curves is considered when assessing the fire resistance classes of protective structures in which transformers are located in today’s conditions. Based on the results of the scientific research, the current state of the art in determining the temperature effect of fire when assessing the fire resistance class of enclosing protective structures within which transformers are located has been analyzed. The purpose of further research, scientific tasks, object and subject of the research and further areas of work are formulated. Scientific research was conducted using analytical methods, in particular analysis and generalization of previously performed works. As the research tools, statistical data on fires, current requirements of regulatory documents, scientific achievements on the researched issue derived by other scientists, and statistical method for processing research results were used.