DE19529458C2 - Method for determining the actual phase shift between current and voltage signals of any phases in three-phase networks - Google Patents

Description

The present invention relates to a method for Er averaging the actual phase shift between Current and voltage signals of any phase in rotation power grids and a use of the method for Kompen faulty phase connection of a blind power compensation controller or the like

In many measurement and control processes, knowledge of the Phase shift between current and voltage signals of one and the same phase (management) necessary.  

Assuming that there is a three-phase network with a symmetrical ohmic load and only one voltage and one current signal is measured, the phase shift between the current signal of one phase and the voltage signal of the same phase against zero potential is 0 °. If the voltage signal is taken from another phase and the voltage between two phases is measured, a phase shift occurs between the current and voltage signals. This phase angle is referred to below as the "connection angle". In a three-phase network with a neutral conductor, twelve different connection angles (0 °, 30 °, 60 °, ..., 330 °) can occur. If the current signal is measured in any phase of the assumed three-phase network, the following voltage signal taps are conceivable:

• - L1-N; L2-N; L3-N; L1-L2; L2-L3; L3-L1;
• - with reverse polarity: N-L1; N-L2; N-L3; L2-L1; L3-L2; L1- L3.

Each measured voltage signal has a different phase position Current signal. The division into 30 ° steps is due to the geome given the three-phase network.

The connection angle is constant with the connection unchanged and is not influenced by load-dependent phase shifts. It describes the angle between a current signal and one Voltage signal of the three-phase network assumed above. Becomes just a voltage signal and a current signal from one turn power supply removed (usual connection with compensation system gen) indicates the connection angle, in which ratio the the two signals are related to each other.  

This leads e.g. B. in reactive power compensation, that a z. B. caused by network inductance Phasesver shift between a mains current and a mains voltage not correctly compensated and therefore the reactive power cannot be optimally minimized.

DE-OS 32 40 625 describes a method for determining the Electrical power quality parameters of a three-phase network known, where a symmetrical three-phase reference system from "Mitspannungen" is used. To the quality parameters of the Elek tro energy include: frequency deviation, voltage deviation, Frequency swing, voltage swing as well as asymmetry, inequality and non-sinusoidal factors of the voltages. The procedure ren is based on the determination of your symmetrical components, according to their amount, the size of the electrical energy quality parameters is judged. First, the amplitude and initial phase of the reference voltages, then a symmetrical three phase reference system of the positive voltages, whose An Initial phase and amplitude of the previously determined values of the start initial phase and amplitude of the reference voltages is the same and the frequency matches the network frequency, and then who which the reference voltages phase from the input voltages subtracted wisely and the symmetrical component after the Difference between the input and reference voltages certainly.

From DE-OS 22 65 223 is a rotary current separator for a Three-phase network known, in which by means of two current transformers the three concurrent currents of the three-phase network through one Art circuit two currents are derived, with three phases existing networks each create 90 ° angles (67.5 ° and 22.5 °). The rotary current separator speaks extremely quickly Deviations in three-phase systems, so that he z. B. in front use in part as a shutdown or protective device can be detected.

The invention is therefore based on the object, a Ver drive to provide with the actual Phase shift between current and voltage signals arbitrary phases in three-phase networks can be determined can.

According to the invention, this object is achieved by adding and / or or switching off at least one purely capacitive or a purely inductive or a purely ohmic or an in their phase shift defined mixed load and ver the difference measured when switching on or off vector as a reference coordinate in a vector diagram.

It can be provided that the method comprises the following steps:

• (a) determining the network frequency f;
• (b) Creation and storage of a number s of discrete values with intervals of the duration Δt for a cos function cos _k and a sin function sin _k with the frequency f, where Δt applies to the intervals
• (c) Acquisition and storage of s discrete measured values
  with the measuring interval Δt for a current I _n and a voltage U _{n of} any phase of the alternating current;
• (d) Calculating the terms
  
• (e) Calculation of the active current
  and the reactive current
  where factor I takes into account an amplification of the measurement inputs and the number s;
• (f) connecting the load;
• (g) performing steps (c) through (e);
• (h) Calculation of the difference _vector D between a _vector = (I _{active, after} , I _{blind, after} ) after connection and a _vector = (I _{active, before} , I _{blind, before} ) before connection in the pointer diagram;
• (i) Determination of the connection angle ϕ _An based on the angle between the difference vector and the expected current vector of the connected load in the vector diagram according to:
  where ϕ _expected corresponds to the expected angle of the expected current vector of the connected load to the I _real axis, with
  ΔI _b = I _{blind, after} - I _{blind, before}
  ΔI _w = I _{active, after} - I _{active, before}
• (j) Rotation of the vector in the vector diagram by ϕ _on and determination of the true current values in accordance with
  I _{act, true} = I _{act, before} .cos (ϕ _An ) - I _{blind, before} .sin (ϕ _An )
  and
  I _{blind, true} = I _{blind, before} .cos (ϕ _An ) + I _{active, before} .sin (ϕ _An )
• (k) Calculation of the actual phase shift between I _eff and I _blind according to
  

It can further be provided that it also comprises the steps:

• 1. x times repetition of steps (c) to (k), the values determined each time for the connection angle ϕ _{An being} stored; and
• 2. Forming an average werts _An using the stored values ϕ _An .

In a particular embodiment of the invention it can be provided that step (s) comprises the following steps:

• (m-1) Forming a provisional mean ϕ _{An, provisional} Using all stored values ϕ _{An, provisional} ;
• (m-2) identification and deletion of the individual value φ _An, φ with the greatest deviation from the preliminary mean _{An, fig provisional;}
• (m-3) Form a provisional mean ϕ _{An, provisionally} using the remaining values ϕ _An and repeat this step up to a number y of remaining measured values for the connection angle ϕ _An ; and
• (m-4) Determination of the last provisional mean ϕ _An , _{provisionally} as the final mean ϕ _{An, end} for the connection angle.

Finally, the invention relates to the use of the Ver to compensate for a faulty phase conclusion of a reactive power compensation controller or the like

The invention is based on the surprising finding that by connecting and / or disconnecting a load with a known phase a circuit-related connection winch determined and in determining the actual Phase shift between current and voltage signals of any phases is taken into account. Furthermore lets when using the inventive method z. B. a faulty one for reactive power compensation controllers Determine the phase connection and the type of connection and with this knowledge, correct reactive power compensation  make.

Further features of the invention emerge from the claims. Advantages will be apparent from the description below, in one embodiment of the method according to the above lying invention with reference to the drawing in detail he is explained.

The following version applies to a three-phase connection to a compensation system on a three-phase network. From any phase, a current signal is fed to a reactive power compensation controller using a current transformer. The voltage signal is tapped in any manner (phase / phase or phase / zero) and from any phases and fed to the reactive power compensation controller. The additional phase shift between current and voltage caused by the connection can be from zero to 330 °. Due to the existing fixed phase relationship between the voltage phases, the additional phase shift can only occur in 30 ° steps due to the optional connection. These phase shifts are referred to in the following connection angle ϕ _An and are characteristic of the respective connection type.

Are z. B. with an incorrect phase connection current and Voltage signals from any phase using a Reactive power compensation system recorded, so would a wrong phase shift from the conventional method determined between the current and the voltage and hence an incorrect compensation from the blind power compensation system can be made. this will  however according to the present inventive method avoided as follows:

When a measuring cycle is triggered at the reactive power compensation system, 256 digitized values I _n and U _n of current and voltage are read into a memory. The measuring distances Δt are equidistant and the measuring time extends over exactly three periods of the fundamental wave. The two determined curves are stored in the memory. A sin and a cos curve are already permanently stored in the memory. These curves also extend over three periods. The active current I _active and the reactive current I _{blind are then determined} as follows

With

The correction factor factor I includes the gain of the measurement inputs and the factor 256 not taken into account. However, the results for I _eff and I _blind are only correct if the voltage and current signals fed in were tapped from the same phase. If signals from two different phases are fed in, a correction must be made before the measurement results are released. The determination of the correction angle (connection angle ϕ _On ) was determined as follows when the operating voltage was first applied:

The first time the operating voltage is applied, the present reactive power compensation system an An closing type detection (automatic connection angle detection) through, d. H. it recognizes itself in which phase and with what connection angle current and voltage path indicated are closed. This requires the reactive power compensation tion system first of all the exact network frequency f der Fundamental wave. The reactive power compensation system measures that Mains frequency f before each current / voltage measurement and inte grows the results. So she receives a remedy value of the mains frequency f of the fundamental wave.  

The reactive power compensation system takes a reactive current and active current measurement as described above. The result is a vector in the vector diagram (see drawing). Then the reactive power compensation system switches on a capacitor stage with the expected phase angle 90 ° and carries out a new measurement of the reactive and active current. The result is the vector (see drawing). The difference between the two vectors describes the connection of a capacitor stage (ϕ _expected = - 90 °) and should be a pure reactive current draw. In the present example (see drawing), the difference results in a current change (vector) in the active and blind areas. Since the expected current vector of the connected load is identical to the I _blind axis, the angle between the difference vector D and the I _blind axis must be determined. In the present example, this is 30 °. Accordingly, all vectors in the vector diagram have to be rotated by 30 ° in order to obtain the actual phase shift between current and voltage. The connection angle ϕ _An corresponds to the angle between the difference vector and the I _blind axis and is generally as follows

In the following, the true active current I is _active and reactive current I is _blind by rotation of the current vector, which takes place mathematically according to

I _{act, true} = I _{act, before} .cos (ϕ _An ) - I _{blind, before} .sin (ϕ _An )

and

I _{blind, true} = I _{blind, before} .cos (ϕ _An ) + I _{active, before} .sin (ϕ _An )

certainly. The real active and reactive currents can now be used to determine the actual phase shift ϕ _true according to the following equation

 and depending on it, one or more suitable con connect capacitor stage (s).

Since the connection angle ϕ _An between current and voltage can only vary in 30 ° steps, as explained above, no intermediate values should be determined. Due to the dynamics of the current and voltage signal, there are still different values for the connection angle ϕ _An . For this Grun de measuring the connection angle φ _to be followed by a mean of education. For this purpose, a series of conditions are placed on the measurements. As a first condition, the current change determined must be greater than 25 mA on the secondary side of the current transformer. The direction of the vector is irrelevant. Smaller changes in current are interpreted as zero levels.

Then the newly determined connection angle ϕ _{An is} compared with the last measured angle value. The difference may be a maximum of 4.5 °. If there is a major discrepancy, further processing is terminated. Nevertheless, the angle just measured is stored as the "last measured angle value".

During the calibration process, at least eight values in the Memory must be stored before the reactive power compensation system continues with its measurement.

In the subsequent operation, the An closing angle and the capacitor step power determined. The table for the angle values is up to twenty Measured values listed. Then the table is given a time lock. After a valid entry, the table for blocked the following one and a half hours. Other valid ones Measured values are not saved during this time.

Since the table has only 24 storage spaces, the 24th value overwritten the oldest entry. To this The angle values are updated again and again.

In order to obtain a result from the large number of connection angle values, an averaging must be carried out. This is done in the following way: After determining a provisional mean value wert _{On, provisionally} the individual measured value with the greatest deviation is searched for and deleted. Then a provisional mean ϕ _An , _{provisionally} calculated and the described method is used again. The mean value ϕ _{An, preliminary} , which was calculated from the last thirteen values, is used as the final angle value ϕ _{An, end} . As already described above, the phase shift between current and voltage with the different connection types can only assume twelve different values (0 × 30 °, 1 × 30 °, 2 × 30 °, ..., 11 × 30 °). Therefore, the nearest 30 ° grid is also searched for the determined angle value ϕ _{An, end} , and it is only with this value that work is continued.

Such averaging is particularly useful for strongly fluctuating network conditions advantageous.

Those in the foregoing description, in the drawings as well as features of the invention disclosed in the claims can be used individually or in any combination for the realization of the invention in its various NEN embodiments may be essential.

Claims (5)

1. Method for determining the actual phase shift between current and voltage signals of any phases in three-phase networks, characterized by switching on and / or off at least one purely capacitive or purely inductive or purely ohmic or a mixed load defined in their phase shift and Use of the difference vector measured when switching on or off as a reference coordinate in a vector diagram. 2. The method according to claim 1, characterized in that it comprises the following steps:

• a) determining the network frequency f;
• b) Creation and storage of a number s of discrete values with intervals of the duration Δt for a cos function cos _k and a sin function sin _k with the frequency f, where Δt applies to the intervals
  
• c) Detection and storage of s discrete measured values with the measurement interval Δt for a current I _n and a voltage U _{n of} any phases of the alternating current;
• d) calculation of the terms
  
• e) Calculation of the active current
  and the reactive current
  

where factor I takes into account an amplification of the measurement inputs and the number s;

• a) connection of the load;
• b) performing steps (c) to (e);
• c) Calculation of the difference _vector between a _vector = (I _{active, after} , I _{blind, after} ) after connection and a _vector = (I _{active, before} , I _{blind, before} ) before connection in the pointer diagram;
• d) Determination of the connection angle ϕ _An based on the angle between the difference vector and the expected current vector in the vector diagram according to:
  

where ϕ _expected corresponds to the expected angle of the expected current vector of the connected load to the I _real axis, with
ΔI _b = I _{blind, after} - I _{blind, before}
ΔI _w = I _{active, after} - I _{active, before}

• a) Rotation of the vector in the vector diagram by ϕ _on and determination of the true current values in accordance with
  I _{act, true} = I _{act, before} .cos (ϕ _An ) - I _{blind, before} .sin (ϕ _An )
  and
  I _{blind, true} = I _{blind, before} .cos (ϕ _An ) + I _{active, before} .sin (ϕAn)
• b) Calculation of the actual phase shift between I _eff and I _blind according
  

3. The method according to claim 2, characterized in that it further comprises the steps:

• 1. x times repetition of steps (c) to (k), the values determined each time for the connection angle ϕ _{An being} stored; and
• 2. Forming an average werts _An using the stored values ϕ _An .

4. The method according to claim 3, characterized in that step (m) comprises the following steps:

• 1. (m-1) Form a provisional mean ϕ _An , _{provisionally} using all stored values ϕ _An ;
• 2 (m-2) identification and deletion of the individual value φ _on with the greatest deviation from the provisional average value φ _{An, fig provisional;}
• 3. (m-3) Form a provisional mean ϕ _An , _{provisionally} using the remaining values ϕ _An and repeat this step up to a number y of remaining measured values for the connection angle ϕ _An ; and
• 4. (m-4) Determination of the last provisional mean ϕ _{An, provisionally} as the final mean ϕ _{An, end} for the connection angle.

5. Use of the method according to one of the preceding Claims to compensate for a faulty phase conclusion of a reactive power compensation controller or the like