Resumo:
Serious failures in power transformers are usually driven by problems in the
insulation structures, whether internal to the tank or in the bushings, and constitute one
of the most critical occurrences of the power system. Thus, the evaluation of the
insulation condition of a transformer is of great importance for the operational efficiency
of the transmission and distribution of electrical energy. However, as insulation systems
are complex, many challenges are seen in terms of measurement systems and modes
of analysis of critical parameters such as partial discharges (PDs), which are recognized
as a phenomenon that carries relevant information regarding the health of a high voltage
equipment. In this context, this dissertation addresses the development of an online
monitoring system for partial discharges for power transformers. The system applies the
electrical method to detect the discharge pulses through capacitive coupling in
condenser bushings and is integrated by a proprietary software for data processing and
communication with supervisory systems through the DNP 3 protocol. The robustness
of the developed system and the guarantee of integrity of the monitored equipment, as
well as the ability to correlate data to track the condition of the asset, were verified by
laboratory tests and real applications in substations. In the latter case, the results were
explored to show how information from a PD can be interpreted in practical situations to
analyze the conditions of the asset and assist in defining diagnoses.
