RU2653173C2 - Method of measuring resistance to direct current - Google Patents

Abstract

FIELD: electricity.

SUBSTANCE: through the winding and reference resistor, which are connected in series, direct stabilized current is supplied. The current is produced by the current controller, then the voltage drops are measured on the winding and the reference resistor and their ratio is calculated. On the basis of this ratio, the required winding resistance is obtained. During the rise of the current in the winding up to the set value, the measurements of the winding current and voltage on it is done and the parameters of the winding are estimated. In this case, after the current rises up to a value close to the present value, the transmission factor of the current regulator is set. Its value is calculated on the basis of the measurements done during the current rise.

EFFECT: reduction of measurement time and accuracy.

4 dwg, 2 ex

Description

The invention relates to measuring equipment, namely to the field of electrical measurements of DC resistance of two-terminal devices having a large inductance.

GOST 3484.1-88 recommended a method for measuring the direct current resistance of the windings of power transformers, which consists in passing a direct current from a constant voltage source through a winding and an ammeter, and characterized by a measuring transient process in which the current in the winding and the voltage on it change simultaneously according to a nonlinear law. Measurement of current and voltage to calculate the resistance in the winding is possible only after the establishment time tust of their values: tust≈4.6⋅τ when setting the current and voltage with an error of δ = 1% and tust≈9.2⋅τ at δ = 0 , 1%, where τ is the transformer winding time constant equal to τ = Lobm / Robm. Due to the long measurement time by this method, in most modern instruments of this purpose, it is not a voltage source that is used, but current sources - a compensating type DC stabilizer.

A known method of measuring the DC resistance of the windings of a three-phase transformer (ed. St. USSR No. 788032, class MKI G01R 27/00, 1980). This method consists in the fact that currents of the same magnitude are passed simultaneously through the windings of the three phases of the transformer, and the magnetizing forces of the windings have the same direction. This ensures the absence of magnetization of the core of the transformer and the rapid establishment of measuring currents. The disadvantage of this method is the impossibility of its application when connecting the transformer windings with a triangle, a star without neutral output, as well as for single-phase transformers.

As a prototype closest to the claimed one, the method implemented in the invention “Device for measuring the active resistance of the windings of electrical equipment” (RF patent No. 2480774, class G01R 27/08, published in Bull. No. 12, 2013) was selected.

The method consists in passing a constant constant current through a winding and a reference resistor, the value of which is calculated based on a preliminary measurement of the winding resistance, measuring the voltage drop across the winding and the reference resistor and calculating their ratio, based on which the desired winding resistance is obtained. In addition, at the first stage, while the current in the winding rises to the calculated current, the supply voltage of the current stabilizer is increased to reduce the duration of the stage due to an increase in the rate of change of current, and at the second stage, when the current in the winding is almost equal to the calculated current, but continues finally installed, reduce it to reduce the power dissipation on the regulating transistor of the current stabilizer. Graphs of changes in current and voltage in the winding are shown in FIG. one.

Preliminary knowledge of the estimated value of the measured resistance allows you to calculate and set for each measured resistance the maximum possible current strength, limited only by the power of the current stabilizer.

In FIG. Figure 1 shows the dynamic error of current control in steady state. This is due to transients occurring in the current stabilizer. Such changes in current make an additional contribution to the error in measuring the resistance, and also increase the time of the second stage.

The indicated dynamic error is due to the fact that the parameters of the current regulator regulator, providing the best quality of the transition process, reducing the duration of stage 2 (Fig. 1) to negligible values, depend on the resistance and inductance of the load (in this case, the transformer windings), and these values at measuring the resistance of the windings of transformers of various types can vary thousands of times or more.

The objective of the present invention is to reduce the duration of the second stage of establishing current and improve the accuracy of measurement.

The problem is solved by using an adaptive current regulator, which, at the stage of increasing the measuring current, estimates the winding parameters of the measurement object and sets the transfer coefficient of the current regulator so as to ensure high accuracy of the current setting with a minimum transient duration.

The measurement consists of two stages. The first stage is the evaluation of the parameters of the winding with increasing current. The second stage is the current stabilization stage.

At the first stage, the maximum output voltage of the source is applied to the winding and the current rises to a certain fraction, for example, to (95-99)% of the set current. In this case, at the time of increasing current, current and voltage measurements are made to determine resistance and inductance estimates.

Consider the dependence of the voltage on the transformer winding on the current passing through it:

Here R _rm is the resistance of the winding to direct current, L _rm is the inductance of the winding.

Integrating equation (2) in the interval from t _k-1 to t _k by the trapezoidal method, we obtain:

where T = t _k -t _k-1 is the time interval between samples.

In combination with the measurement results in the previous step, we obtain a system of equations in a vector-matrix form:

where Z = [R _rpm L _rpm ] ^T is the vector of winding parameters; I matrix of samples of currents with a dimension of 2 × 2, and the components of the matrix:

i _{11 (k)} = i _{11 (k-1)} + T⋅i _k ;

i _{12 (k)} -i _{12 (k-1)} + T⋅i _{11 (k)} ;

i _{21 (k)} = i _{11 (k-1)} ;

i _{22 (k)} = i _{12 (k-1)} ;

U is a vector of stress samples of dimension 2 × 1, and the components of the vector:

u _{1 (k)} = u _{1 (k-1)} + u ^* _(k) ⋅ T;

u _{2 (k)} = u _{1 (k-1)} ;

u ^* _(k) = u ^* _(k-1) + u _(k) ⋅ T.

Then the solution of equation (4):

With well-known estimates of the parameters of the system, it is possible to build a current regulator that provides high speed and accuracy of establishing the output current at the regulation stage (second stage), and at the same time avoid system instability.

Let a proportional regulator be used as a current regulator. The parameters of the proportional current controller are the transmission coefficient K _p and the setpoint current. The set current is defined as the value of the output signal at which the mismatch between the adjustable and set values is zero. Under the influence of disturbing influences, a deviation of the controlled quantity occurs. With a correctly selected transmission coefficient, a proportional controller allows achieving a certain stabilization accuracy (with an overestimated coefficient, self-excitation will be observed in the system, and with an underestimated one, the stabilization time and error will increase). Choosing the transmission coefficient based on the results of the identification step, it is possible to implement an adaptive current regulator, which will allow for a given type of regulator to obtain a stable current value within a small error (static regulation error), significantly reducing the dynamic regulation error.

Find the transfer coefficient of the current regulator K _p and the static error of current regulation δ.

Transfer function of the current regulator control loop:

Where

A = K _p ⋅K _c / (R _rm + R _p ),

τ = L _rpm / (R _rm + R _p ),

R _p is the resistance of the conductors connecting the winding to the device, K _with is the transfer coefficient of the current measurement circuit (in the simplest case, it is equal to the resistance of the reference resistor in the feedback of the current source).

When using a microcontroller as a current regulator, to strictly describe the processes in the current control loop, one should use the mathematical apparatus of z-conversion. In this case, the discrete transfer function of a first-order link with a zero-order storage element has the form (see Rotach V.Ya. Theory of automatic control. - M .: MEI, 2008, p. 291):

where d = exp (-T / τ), T is the sampling period of the analog-to-digital converter (ADC) of the microcontroller.

If the cutoff frequency of the open system of the current regulator is significantly less than the sampling frequency, you can replace the discrete representation of the system with a continuous one without significant loss of accuracy. Then the transfer function of the current control loop will have the form (6).

Let the selected cutoff frequency of the system

Then the condition is satisfied:

| W (j2πf _c τ) | = 1,

and the required transfer coefficient of the current regulator, taking into account (6), is approximately equal

Static current control error

taking into account (6) and (9) approximately

or as a percentage

Current establishment time in the second stage with an accuracy of 0.1%

When condition (9) is satisfied, approximately

Let's look at two examples.

Example 1. The high voltage winding of the transformer TC-630000/500 has a resistance of R _obm = 0.42 Ohm and an inductance of L _obm = 70 _GN at a current of 10A. Let the resistance of the conductors connecting the winding with the device, R _p = 0.2 Ohms. The sampling period of the ADC of the microcontroller is T = 100 μs. The transfer coefficient of the current measurement circuit K _s = 0.2 V / A. We choose the cutoff frequency of the system based on condition (8) f _c = 200 Hz (50 times less than the sampling frequency; in this case, the sampling effect is not significant). Then from (9) it follows K _p = 440000, from (12) - δ _% = 0.0007%, and from (14) - T _lane = 5.5 ms. Thus, when the current regulator transmission coefficient is set according to the data of measurements and calculations at the first stage, a high accuracy of current stabilization and a short transient time at the second stage are ensured.

Example 2. The low voltage winding of the transformer TMG-630/10 has a resistance of R _obm = 0.002 Ohms and an inductance of L _obm = 0.018 H at a current of 10A. The remaining conditions are the same as in the first example. The calculated parameters of the current regulator are K _p = 113, δ _% = 0.89%, T _lane = 5.5 ms.

If, as in the prototype, the transfer coefficient of the current regulator does not change, then it should be chosen from the conditions for ensuring the stability of the regulator for any measurement object. In the examples considered, this is the second case, i.e. should choose K _p = 113. Moreover, the static control error for the first example, as follows from formula (10), taking into account (6), will be equal to 2.6%, i.e. even worse than in the second example, and for the winding from the first example, the current settling time in the second stage, as follows from formula (13), T _per = 20.8 s, which significantly increases the total measurement time.

A significant difference between the proposed method and the prototype is that in the prototype the results of the calculation of the winding parameters in the first stage are used to calculate the voltage change of the power supply of the current regulator in the second stage, and in the proposed method, these data are used to calculate the transfer coefficient of the current regulator.

The proposed method was tested in modeling the measurement of the resistance of the high voltage winding of the TDTN-31500/110 transformer and the low voltage winding of the TMG-630/10 transformer. The voltage of the current source is 20V, the sampling period is 100 μs. The transfer coefficient of the current regulator, calculated for the first case, K _p = 286000, the measuring current strength is 6A. A graph of the current is shown in FIG. 3. Comparing this graph with the graph presented in the description of the prototype (Fig. 2), we see that in our case the duration of the transition process of the second stage is insignificant, and for the prototype it is 10 s.

When measuring the resistance of the low voltage winding of the transformer TMG-630/10, the calculated transmission coefficient of the current regulator K _p = 113. Measuring current - 10A. A graph of the current is shown in FIG. 4. It can be seen that the current is established in approximately 0.055 s, and the duration of the transition process of the second stage is 6 ms.

Chelenergopribor LLC has developed and manufactured a Trom-1 milliometer designed for measuring the resistance of the windings of electric machines, including transformers (see http://www.limi.ru/led.php?id_group=31). In the algorithm of operation of the measuring current regulator of this device, the claimed method is applied. The device passed state tests and was entered in the State Register of Measuring Instruments of the Russian Federation under No. 67448-17.

Claims (1)

A method for measuring the direct current resistance of windings of electrical equipment, which consists in passing a constant constant current produced by a current regulator through a winding and a reference resistor, measuring the voltage drop across the winding and the reference resistor and calculating their ratio, based on which the desired winding resistance is obtained moreover, while the current in the winding rises to a predetermined value, measurements of the winding current and voltage on it are performed and ku winding parameters, characterized in that after the rise to the current value close to the target, set the current regulator transfer coefficient, whose value is calculated based on measurements made during the current rise time.

Images