RU2354982C1 - Method of diagnosing transformer core - Google Patents

Abstract

FIELD: physics; measurement. ^ SUBSTANCE: present invention relates to inspection technology. The method involves phase no-load loss measurement successive short-circuiting of the low-voltage winding of one its phases and generating two other phases. The high voltage phase winding, the low voltage winding of which is short-circuited, is excited using direct magnetising current. Graphs of open circuit power loss versus magnetising current are plotted. Comparing these curves with previous measurements, the state of the transformer coil is diagnosed. ^ EFFECT: unambiguity of obtained results independent of remnant magnetisation of the transformer core during no-load loss test at low voltage.

Description

The invention relates to the field of electrical engineering, in particular to electromagnetic energy converters.

A known method for diagnosing a magnetic system by measuring the loss of idle at low voltage. Losses of idling at low voltage for transformers are measured in order to verify the absence of inter-sheet short circuits in the transformer core, to detect an unequal number of turns in parallel connected coils (branches), erroneous connections of windings or switching devices and other similar defects.

In the process of acceptance tests, when the transformer has passed all stages of technological processing and its active part is in the tank with oil, measurements of losses at low voltage, in addition to detecting possible defects not detected during the assembly process (which is extremely rare), give loss values for comparison with the data obtained in the future when testing transformers before commissioning or after repair [1].

Losses at low voltage are measured at the factory at a frequency of 50 Hz and a voltage of not more than 10% of the nominal voltage, usually at a voltage of 220 or 380 V. Usually, the losses of three-phase transformers are measured with single-phase excitation in order to know the losses of each phase separately, and therefore, be able to compare them and make sure that the transformer has the correct loss ratio and has no defects. In phase-by-phase measurement of losses, three experiments are necessary with the reduction of a three-phase to a single-phase one by alternately short-circuiting the winding of one of its phases and exciting two other phases. Shorting one of its windings is done so that there is no magnetic flux in this phase, and therefore no loss.

Three experiments are performed in the following sequence when the transformer is powered from the low voltage side:

first experiment - short-circuit the winding of phase a, feed the windings of phases b and c of the transformer and measure the loss P '_0bc;

second experiment - short-circuit the winding of phase b, feed the windings of phases a and c of the transformer and measure the loss P '_0ac;

third experiment - they short-circuit the phase c winding, feed the windings of phases a and b of the transformer and measure the loss P ' _0ab .

The windings of this or that phase are closed at the corresponding inputs of any of the transformer windings (high and low voltage), while being guided by the connection scheme of its windings.

Measured in phase loss compared with each other. In the absence of defects in the transformer, the losses P ' _0bc and P' _0ab should be practically (5%) equal to each other, and the losses P ' _0ac , depending on the design, are approximately 25-50% greater than the losses P' _0bc and P ' _0ab .

The disadvantage of this method is the strong dependence on the magnitude of the residual magnetization of the cores of the transformer, which is the result of a sudden shutdown of the transformer and a current break. The magnitude of the magnetization is a random variable and depends only on the phase of the current at the time of switching off the transformer.

Therefore, to obtain reliable results of losses in steel, it is necessary to demagnetize the transformer. The demagnetization procedure, which consists in alternately supplying direct current to one of the windings of each rod, also does not allow us to unequivocally state that the core is demagnetized to zero residual induction, since it is not possible to control the residual magnetization of the core. Thus, by measuring the losses P ' _0ab , P' _0ac , P ' _0bc , it is impossible to unambiguously determine whether the spread of values is the result of residual magnetization or a developing defect. Moreover, it is impossible to predict the nature of this defect.

The aim of the present invention is to increase the uniqueness of the results, regardless of the residual magnetization of the core of the transformer, thereby increasing the information content of the method.

The proposed method for the diagnosis of the magnetic system of the transformer consists in the artificial preliminary magnetization of the core to a certain known value with subsequent measurement of losses.

The essence of the method is as follows.

For magnetization in accordance with the method of measuring losses at low voltage from the low voltage side, we have:

first experiment - they short-circuit the phase a winding, feed the windings of phases b and c of the transformer, supply direct current to the high-voltage winding of phase A and measure the loss P '_0bc;

the second experiment - they short-circuit the phase b winding, feed the windings of phases a and c of the transformer, supply direct current to the high-voltage winding of phase B and measure the loss P '_0ac;

the third experiment - they short-circuit the winding of phase c, feed the windings of phases a and b of the transformer, apply direct current to the high-voltage winding of phase C and measure the loss P ' _0ab .

Despite the fact that high voltage can be transformed on the high voltage side of the two phases of the transformer, applying a constant voltage to the high voltage phase with a shorted low voltage winding of the same phase is absolutely safe, since the voltage does not transform on this phase.

Thus, the dependences of the power loss and the no-load current on the magnitude of the magnetization current can be removed. Comparing these characteristics with previous measurements, it is possible to diagnose the state of the magnetic system of the transformer.

Measurements in the magnetization mode make it possible to identify defects caused by a change in the geometry of magnetic structures under the influence of a magnetic field, which appear only in the operating mode and disappear with stress relief. For example, a structural element, attracted to the core, forms a bridge for the eddy current circulation, forming local heating, after removing the field, the bridge breaks, and when conducting an idling experiment at low voltage, normal results will be obtained.

Bibliographic list

1. Ashryatov A.K. Measurement of idle losses of power transformers. - “Electric stations”, 1948, No. 5, p. 34-36.

Claims (1)

A method for diagnosing a transformer’s magnetic system by phase-by-phase measurement of no-load losses when alternately short-circuited the low-voltage winding of one of its phases and excitation of the other two phases, characterized in that the high-voltage winding of the phase, the low-voltage winding of which is shorted, is excited by a direct magnetizing current, and the dependences are removed power of idling losses from the magnitude of the magnetization current and, comparing these characteristics with previous measurements, they diagnose the state of the magnetic system we are a transformer.