RU2573171C1 - Method for determination of load static characteristics against voltage with abnormal deviations protection - Google Patents

Abstract

FIELD: electricity.

SUBSTANCE: in the load unit sequential changes in voltage take place, on-load power and voltage are measured and translated into relative units. At that the measurement of the on-load power and voltage is performed before and upon each I^th voltage change in the load unit in the form of three-phase active power P_1(i) and P_2(i) and valid average phase voltage U_1(i) and U_2(i), where i=1, 2 is a serial number of change in a pair for I^th voltage change, index 1 corresponds to measurement before voltage change while index 2 - upon voltage change, and against them values of load controlling effect indices KP_i are determined for each pair of measurements. The pairs of measurements are excluded when values of the load controlling effect indices are outside the limits of the preset confidence interval. The received pairs of measurements U_1(i) and U_2(i), P_1(i) and P_2(i) are filtered against values of the load controlling effect indices KP_i, the first approximation of base power values is defined for each pair of measurements at I^th voltage change

- for the first pair of measurements,

- for the next pairs of measurements in regard to which the pairs of measurements U_1(i) and U_2(i), P_1(i) and P_2(i) are translated into relative units as per ratios

thereafter coefficients a₀, a₁, a₂ of the polynomial approximant of the second degree

and mean-square deviation for voltage and power values are obtained in relative units of the obtained polynomial $$P_{*}=a_0+a_1\cdot U_{*}+a_2\cdot U_{*}^2,$$

where N is the quantity of pairs of measurements, and against the preliminarily defined coefficients a₀, a₁, a₂ together with the values of U_*1(i) and U_*2(i), P_*1(i) and P_*2(i) basis power values are specified for each pair of measurement in compliance with the ratio

with their further translation into relative units and repetition of operations for the determination of coefficients a₀, a₁, a₂, mean-square deviation σ and further specification of base power values P_base(i) until the mean-square deviation σ is decreased either to the minimum preset value or a value of its increase beginning with each repetition.

EFFECT: higher accuracy.

2 dwg, 2 tbl

Description

второй степени с коэффициентами аппроксимации a₀, a₁, a₂, которые соответствуют минимальному среднеквадратическому отклонению.The invention relates to the field of measurements in electrical engineering and can be used to determine the static characteristics of the load voltage U, which are in the form of a polynomial $$P_{*}(U_{*})=a_0+a_{\mathrm{one}}\cdot U_{*}+a_2\cdot U_{*}^2$$

second degree with approximation coefficients a ₀ , a ₁ , a ₂ , which correspond to the minimum standard deviation.

A known method for determining the static characteristics of the load voltage [Gurevich Yu.E., Libova L.E. The use of mathematical models of electric load in the calculation of energy systems and the reliability of power supply to industrial consumers. - M .: publishing house ELEKS-KM, 2008. - S. 211-215; Experimental studies of power systems / L.M. Gorbunova, M.G. Tailor, R.S. Rabinovich and others; under the editorship of S.A. Sovalova, - M .: Energoatomizdat, 1985. - P. 45-48], according to which the load node sequentially changes the voltage and measures the voltage and power at the load, followed by the conversion of the measured values into relative units, and the resulting characteristic is used in as a static load characteristic.

This method is the closest in purpose and technical result when used relative to the proposed, therefore, is selected as a prototype.

The condition for using this method is the stationarity of the test load. If irregular fluctuations and power drift take place, then this method cannot be used, since it leads to low accuracy in determining the static load characteristic. In this case, a repeat experiment is required. Repeated experiment is often associated with a number of technical and organizational difficulties and is not always possible. In addition, there is no guarantee that during the repeated experiment the load will be stationary and the use of this method to determine the static characteristics of the load will be successful.

Thus, the known method for determining the static characteristics of the load has a relatively low accuracy.

The objective of the invention is to develop a method to improve the accuracy of determining the static characteristics of the voltage load in the presence of irregular fluctuations and power drift.

The required technical result is to increase the accuracy of determining the static characteristics of the load in the presence of irregular oscillations and power drift.

The problem is solved, and the required technical result is achieved by the fact that in the method based on the fact that successive voltage changes are made in the load node, the power and voltage values are measured at the load, and the relative values are converted into relative units, according to the invention the load is produced before and after each I-st voltage change in the load node in the form of a three-phase active power P _{1 (i)} and P _{2 (i)} and the current average phase voltage value U _{1 (i)} and U _{2 (i)} , where e i = l, 2 - serial number of measurements in pair for the I-th voltage change, index 1 corresponds to the measurement before the voltage change, and index 2 - after the voltage change, which determine the values of the indicators of the regulatory effects of the load KP _i for each measurement pair in according to the ratio

,

,

exclude pairs of measurements, the values of the indicators of the regulatory effect of which do not fall within the specified confidence interval, filter the resulting pairs of measurements U _{1 (i)} and U _{2 (i)} , P _{1 (i)} and P _{2 (i)} by the value of the regulatory effect of the load KP _i determine the first approximation of the basis power values for each pair of measurements with the I-th voltage change

- для первой пары измерений,

- for the first pair of measurements,

- для последующих пар измерений,

- for subsequent pairs of measurements,

relative to which they carry out translation in relative units of the measurement pair U _{1 (i)} and U _{2 (i)} , P _{1 (i)} and P _{2 (i)} in accordance with

и среднеквадратическое отклонение значений напряжения и мощности в относительных единицах от полученного полиномаafter which the coefficients a ₀ , a ₁ , a _{2 of the} approximating polynomial of the second degree are determined $$P_{*}=a_0+a_{\mathrm{one}}\cdot U_{*}+a_2\cdot U_{*}^2$$

and standard deviation of voltage and power in relative units from the resulting polynomial

where N is the number of measurement pairs, and according to the previously determined values of the coefficients a ₀ , a ₁ , and ₂ together with the values U _{* 1 (i)} and U _{* 2 (i)} , P _{* 1 (i)} and P _{* 2 (i )} specify the basis power values for each pair of measurements in accordance with the ratio

with their subsequent conversion to relative units and repeating the operations of determining the coefficients a ₀ , a ₁ , a ₂ , standard deviation σ and then clarifying the values of the base power P _{BAZ (i)} until, with each subsequent repetition, the standard deviation σ decreases or to set minimum value or before the start of its increase.

The method for determining static voltage loads is implemented as follows.

The proposed method allows to determine the static characteristics of the load voltage even in the presence of irregular fluctuations and power drift due to the fact that the measurement of voltage and power is carried out before and after each voltage change, after which the resulting pairs of measurements are filtered by the values of the regulatory effect of the load, and when translating power values in relative units for each pair of measurements, the values of the basis powers are chosen such that the standard deviation of the cut measurement measurements in relative units of the obtained static load characteristics were minimal.

Irregular fluctuations and power drift can be taken into account as a change in the value of the base power, which is used to convert the measured power values to relative units. In the proposed method, the measurement of voltage and power values is carried out immediately before and after the voltage change, which minimizes the likelihood of a change in the base power between such measurements. In order to exclude pairs of measurements between which a base power change still occurs, the proposed method provides for filtering of measurement pairs by the values of the regulatory effect of the load. Since the values of the base power for each pair of measurements are not known in advance, in the proposed method they are selected based on the condition of minimizing the standard deviation of the measurement results from the obtained static load characteristics. This allows you to minimize the effect of irregular fluctuations and power drift on the resulting static load characteristics, which expands the scope of the proposed method in comparison with the prototype.

The drawings show:

in FIG. 1 - static characteristic of the load and the results of voltage and power measurements, obtained as a first approximation of the basis power values;

in FIG. 2 is a static characteristic of the load and the results of voltage and power measurements obtained from the base power values corresponding to the minimum standard deviation.

Table 1 shows the measured voltage values U _{1 (i)} and U _{2 (i)} and power P _{1 (i)} and P _{2 (i)} , as well as the corresponding values of the regulatory effect of the load KP _i .

Table 2 shows the first approximation of the basis power value R _BAZi for each pair of measurements.

As an example, a method for determining the static characteristics of the active load power of an enterprise of OJSC "Sibkabel" in Tomsk by voltage is given. The load of the Sibkabel OJSC enterprise is very variable due to the characteristics of production, which is accompanied by irregular fluctuations and power drift, therefore, it is impossible to determine the static characteristics of the voltage load by known methods.

The successive voltage change in the load node is produced using the voltage regulation device under the load of the supply transformer. In accordance with the control actions, before and after each voltage change, the values of the three-phase active power P _{1 (i)} and P _{2 (i)} and the current average phase voltage value U _{1 (i)} and U _{2 (i)} are measured, where i is the serial number pairs of measurements, index 1 means that the measurement was made before the voltage change, and index 2 means that the measurement was made after the voltage change. Further, the obtained pairs of measured values of U _{1 (i)} and U _{2 (i)} , P _{1 (i)} and P _{2 (i)} (table 1) are used to determine the values of the indicators of the regulatory effects of the load KP _i for each pair of measurements in accordance with the ratio

Check whether the values of the indicators of the regulatory effects go beyond the confidence interval, for example 0.75 ÷ 2, and exclude pairs of measurements, the values of the regulatory effect of which are beyond its limits. Excluded values are shown in bold italics in Table 1.

The values of the regulatory effects of the load KP _i (table 1) are used to linearly filter the obtained measurement pairs U _{1 (i)} and U _{2 (i)} , P _{1 (i)} and P _{2 (i) to the} value of the regulatory effect of the load KP _i . Filtered pairs of voltage values U _{1 (i)} and U _{2 (i)} and power P _{1 (i)} and P _{2 (i)} are used to determine the base power values first in the form of a first approximation of its value for each measurement pair

- для первой пары измерений,

- for the first pair of measurements,

- для последующих пар измерений.

- for subsequent pairs of measurements.

Further, the obtained values of the base power P _{BAZ (i)} (table 2) together with the values U _{1 (i)} and U _{2 (i)} , P _{1 (i)} and P _{2 (i) are} converted into relative units

In this example, use a constant value of the base voltage (in this case, U _BAZ = 6200 V).

The obtained values of U _{* 1 (i)} and U _{* 2 (i)} , P _{* 1 (i)} and P _{* 2 (i)} in relative units are used to determine the coefficients of the approximating polynomial (Fig. 1)

which can be determined by the standard least squares method according to the dependencies

Then the obtained coefficients, in the case under consideration, a ₀ = 1.387, and ₁ = -1.794, and ₂ = 1.392, together with the values U _{* 1 (i)} and U _{* 2 (i)} , P _{* 1 (i)} and P _{* 2 (i)} are used to determine the error in the form of the standard deviation of the voltage and power values in relative units from the resulting polynomial

where N is the number of measurement pairs.

Then the previously determined values of the coefficients a ₀ , a ₁ , a ₂ together with the values U _{* 1 (i)} and U _{* 2 (i)} , P _{* 1 (i)} and P _{* 2 (i)} are used to refine the values of the base power, for example, a second approximation of the basis power value is determined for each measurement pair

с коэффициентами а₀=1,481, а₁=-2,277, а₂=1,796, соответствующими минимальному среднеквадратическому отклонению σ=0,003 (фиг. 2). Возможно и ограничение процесса некоторым минимальным значением, например, до σ=0,01.with its subsequent conversion to relative units and repeating the operations of determining the coefficients a ₀ , a ₁ , a ₂ , standard deviation σ and then clarifying the values of the base power P _{BAZ (i)} until, with each subsequent repetition, the standard deviation σ decreases or to set minimum value or before the start of its increase. That is, the translation of the basis power values into relative units, the determination of the coefficients a ₀ , a ₁ , a ₂ , the determination of the standard deviation σ and the determination of the next approximation of the values of the basic power P _{BAZ (i) are} repeated until the standard deviation σ decreases with each subsequent repetition. In this example, the standard deviation continued to decrease to a minimum value of σ = 0.003, after which it began to increase. Therefore, the polynomial $$P_{*}=a_0+a_{\mathrm{one}}\cdot U_{*}+a_2\cdot U_{*}^2$$

with coefficients a ₀ = 1.481, and ₁ = -2.277, and ₂ = 1.796, corresponding to the minimum standard deviation σ = 0.003 (Fig. 2). It is also possible to limit the process to a certain minimum value, for example, to σ = 0.01.

Thus, the example of determining the static characteristics of the active load power by voltage of the enterprise of JSC Sibkabel shows the operability of the proposed method even in the presence of irregular load fluctuations and power drift.

Thanks to the introduction of an additional arsenal of technical means (method operations described above), the required technical result is achieved, which consists in increasing the accuracy of determining the static characteristics of the load in the presence of irregular fluctuations and power drift and taking into account possible anomalous distortions, since measurement pairs exclude pairs of measurements, the values of the indicators of the regulatory effect of which do not fall within a given confidence interval.

Claims (1)

A method for determining the static characteristics of a voltage load with protection against anomalous distortions, based on the fact that successive voltage changes are made in the load node, the power and voltage values are measured on the load, and the unit is converted into relative units, characterized in that the power and voltage values are measured on the load is produced before and after each I-st voltage change in the load node in the form of a three-phase active power P _{1 (i)} and P _{2 (i)} and the current average phase voltage value U _{1 (i)} and U _{2 (i)} , where i = 1, 2 is the serial number of measurements in pairs for the I-th voltage change, index 1 corresponds to the measurement before the voltage change, and index 2 - after the voltage change, from which the values of the indicators of the regulatory effects of the load KP are determined _i for each pair of measurements in accordance with the ratio

exclude pairs of measurements, the values of the indicators of the regulatory effect of which do not fall within a given confidence interval, filter the resulting pairs of measurements U _{1 (i)} and U _{2 (i)} , P _{1 (i)} and P _{2 (i)} by the value of the regulatory effect of the load KP _i , determine the first approximation of the basis power values for each pair of measurements with the I-th voltage change

- for the first pair of measurements,

- for subsequent pairs of measurements, relative to which the conversion is carried out in relative units of the measurement pair U _{1 (i)} and U _{2 (i)} , P _{1 (i)} and P _{2 (i)} in accordance with the ratios

after which the coefficients a ₀ , a ₁ , a ₂ , of the approximating polynomial of the second degree are determined $$P_{*}=a_0+a_{\mathrm{one}}\cdot U_{*}+a_2\cdot U_{*}^2$$

and standard deviation of voltage and power in relative units from the resulting polynomial

where N is the number of measurement pairs, and according to the previously determined values of the coefficients a ₀ , a ₁ , a ₂ together with the values U _{* 1 (i)} and U _{* 2 (i)} , P _{* 1 (i)} and P _{* 2 (i )} specify the basis power values for each pair of measurements in accordance with the ratio

with their subsequent conversion to relative units and repeating the operations of determining the coefficients a ₀ , a ₁ , a ₂ , standard deviation σ and then clarifying the values of the base power P _{BAZ (i)} until, with each subsequent repetition, the standard deviation σ decreases or to set minimum value or before the start of its increase.

Images