Patrick Picher

No-load losses in transformer under overexcitation/inrush-current conditions: tests and a new model

Tests were conducted on several transformers rated at 100 kVA or less and on a power transformer rated 370 MVA in an effort to characterize the no-load losses and magnetizing resistance for transformers subjected to overexcitation and inrush current. Analysis of the results revealed that the magnetizing resistance changes as a function of the peak magnetization flux or the amplitude of the magnetic field. A new model of the instantaneous magnetizing resistance (IMR) as a function of the instantaneous flux has been developed and its dynamic use in the Electromagnetic Transients Program (EMTP) allows the form of the hysteresis cycle and the mean losses in overexcitation to be reproduced with a high degree of accuracy. The same model also accounts for losses due to the harmonics superimposed on the fundamental. The results showed that the IMR calculated under inrush current conditions is higher than that in overexcitation conditions.

Mitigation of Ferroresonance Induced by Single-Phase Opening of a Three-Phase Transformer Feeder

This paper presents a practical mitigation technique for the ferroresonance induced by the single-phase primary opening of a three-phase 5-legged (wound core) station service transformer. This ferroresonance problem was causing erratic behaviour of the automatic transfer switch between the main and reserve station service transformers. A nonlinear three-phase transformer model, based on the magnetic-circuit theory, was used to reproduce and acquire a deeper understanding of the ferroresonance phenomenon. The model, validated by field measurements, allowed comparison of the effectiveness of resistive and inductive loads for reducing the voltage on the open phase. In previous literature, it has been generally recommended to use a resistive load to damp ferroresonance. This paper demonstrates that an inductive load is more effective and has the advantage of being a more energy-efficient solution. The inductive-load mitigation technique is now in service in Hydro-Quebecs Deschambault substation

Study of the apparent load loss unbalance in three-phase transformers

In a three-phase transformer, the load losses are generally similar between phases, with some small variations due to the proximity of the tank on lateral phases. However, when a three-phase test configuration is used, the measured active losses in each phase are significantly unbalanced and do not reflect the actual losses in each phase. To provide a physical explanation for this phenomenon, measurements were taken on small magnetically coupled reactors and the experimental results were used to validate a simulation model based on the magnetic-circuit theory. The simulation model successfully reproduced the apparent load loss unbalance in a three-phase 47-MVA transformer. It is concluded that, even if the apparent losses are unbalanced, the total load losses in the transformer are correctly obtained by the summation of losses in each phase. The authors point out that since the FRSL (frequency response of stray losses) method is based on the comparison of losses to diagnose winding displacements, the three-phase test configuration cannot be used for this purpose

Experience with frequency response analysis (FRA) for the mechanical condition assessment of transformer windings

Measurement of the frequency response is now commonly used in the industry to assess the mechanical condition of transformer windings. The analysis of the results, the so-called Frequency Response Analysis (FRA), is based on comparison with a reference measurement which is either a previous measurement on the same unit, a measurement on an identical transformer or a measurement on another phase of a three-phase transformer. To ensure a good quality of the frequency response measurement, it is essential to follow the recommended practices reported in the recently published CIGRÉ and IEEE guides, and in the IEC standard. This paper presents case studies on the application of the method to detect mechanical displacement and some of the influencing parameters that can affect the interpretation of the results.

Numerical thermofluid analysis of a power transformer disc-type winding

This contribution describes the early results of a long term R&D collaboration established between IREQ and EFACEC. This partnership focuses on disc-type windings, which are the most common geometry used in real core-type transformers. One of the main objectives so far has been the comparison of two different in-house thermal models (THNM1 and THNM2) with the most accurate numerical approach, namely a tridimensional CFD model. Both codes proved to be robust and able to predict the thermal behavior, particularly for ON conditions. From this comparison, it is currently observed that THNM1 predictions tend to be closer to CFD flow patterns and disc temperatures, seemingly due to its CFD calibrated friction and convection heat transfer coefficients expressions. Further analysis will be performed in the future to compare such in-house models and CFD with measurements obtained under a tightly controlled experimental environment.

On-line monitoring of transformer bushings using a new decentralized measurement system

A new bushing insulation monitoring system based on decentralized measurements of the bushing-tap current is presented. The development of this decentralized system is driven by the need to reduce the implementation cost of bushing monitoring based on the comparison of specimens connected in parallel on the same electrical phase. GPS is used for time synchronization of the decentralized measurements. The phase and amplitude measurement accuracy specifications are respectively 0.01° and 0.1%. These specifications are required to allow monitoring of the insulation degradation as well as variations of the bushing capacitance and power/dissipation factor introduced by the fluctuation of the internal insulation temperature and ambient conditions (e.g. rain). Simulations of the new decentralized measurement system were performed to determine the detailed specifications. The system will be validated by comparing it with on-line measurements performed on nine 735/230-kV transmission transformers using a standard galvanic measurement system. These transformers are located in a substation with HVDC converters, which represents a challenging noisy environment for such accurate measurements.

Closure on "No-load losses in transformer under overexcitation/inrush-current conditions: tests and a new model"

Tests were conducted on several transformers rated 100 kVA or less and on a power transformer rated 370 MVA in an aim to characterize the no-load losses and magnetizing resistance for transformers subjected to overexcitation and inrush current. Analysis of the results revealed that the magnetizing resistance changes as a function of the peak magnetization flux or the amplitude of the magnetic field. A new model of the instantaneous magnetizing resistance (IMR) as a function of the instantaneous flux has been developed and its dynamic use in EMTP allows the form of the hysteresis cycle and the mean losses in overexcitation to be reproduced with a high degree of accuracy. The same model also allows account to be taken of losses due to the harmonics superimposed on the fundamental. The results showed that the IMR calculated under inrush current conditions is higher than that in overexcitation conditions.