Patent Grant
7010436
The present invention relates to an apparatus for predicting a zero-crossing of an alternating current after occurrence of a fault current in a current path for determining the suitable time for opening an electric switching device arranged in the current path for breaking the current in the current path as well as a method for such a prediction.
“Electric switching device” is to be given a broad sense and covers not only such ones having a mechanical movement between different parts for obtaining an opening through physical separation of two parts in the current path, but also semiconductor devices, such as IGBTs or the like, which open by going to blocking state and by that breaking the current therethrough. “Electric switching device” also comprises so called transfer switches through which then a current in a current path may be broken upon occurrence of a fault current in the current path for switching in another current path instead to a load or the like.
It has within the electricity field been a long felt need of apparatuses and methods of this type. When such a fault current occurs, it is important that the electric switching device on one hand opens the current path, i.e. breaks the current, as soon as possible for not damaging different types of equipment connected to the current path, but it is on the other absolutely necessary that the alternating current changes direction, i.e. has a zero-crossing, before it is broken. However, the alternating current receives upon occurrence of said fault usually a direct current component (dc-component), the magnitude of which depends upon the time for occurrence of the fault, and this dc-component is superposed on the alternating current, which in the worst case may result in a duration of several periods of the alternating current before any zero-crossing occurs. For this sake, it has until now after occurrence of a fault simply been waited so long that a breaking definitely may be made in connection with a zero-crossing of an alternating current, in which it is assumed that the fault may have occurred at the most unfavourable time with respect to the dc-component. This long waiting means of course an imminent risk of greater damage on said equipment than would the breaking have taken place at an earlier time. The breaking will for this procedure of breaking the alternating current of course in most cases take place after the occurrence of a plurality of zero-crossings, since there has to be a considerable safety margin for not breaking to early.
It would therefor be desired to break the alternating current considerably earlier exactly when this is possible, i.e. predict a zero-crossing of the alternating current in the individual case so as to be able to obtain a breaking at an optimum time. It is for that sake not sure that it is always desired to break the current when the first zero-crossing occurs, since the dc-component may still be that great that the energy of an arc generated on a contact location would be to high and the amount of material burned away would be to large, so that the breaker or switching device may be partially destroyed or fail.
Another reason for desires of predicting a zero-crossing is in a switching device with breaking through contact separation the existence of the mechanical delay time interval of the contact system of such a switching device, which necessitates a start of the mechanic movement a certain period of time before the zero-crossing so that the breaking may take place at the zero-crossing.
It is pointed out that the invention is applicable to opening of current paths provided with all types of electric switching devices, since it is interesting to obtain a well controlled arcing time in the breaking chamber for conventional breakers through a said prediction, but the invention is particularly directed to so called hybrid breakers of the type described in the Swedish patent application 9904164-2 still unpublished of the applicant. In such a hybrid breaker having two branches connected in parallel in the current path, one in the regular current path through the switching device with a commutator, and one with a part having ability to block current therethrough in at least one blocking direction and conduct current therethrough in at least one direction, and a breaking contact member connected in series with said part, it is of great interest to be able to control the contact opening of the commutator to the zero-crossing of the alternating current for avoiding an arc. Since said part has to block for enabling an opening of the contact member without any current, when using parts in the form of rectifying diodes it is a condition that the commutator is not opened until a zero-crossing of the alternating current may be obtained. The corresponding problems are applicable to the hybrid breaker described in the Swedish patent application 9904166-7 still unpublished and owned by the applicant. Thus, there may both be a desire to predict a zero-crossing for being sure that a breaking really may take place and for determining the optimum time for the breaking, for example synchronise the breaking with the predicted time for a zero-crossing.
The object of the present invention is to provide an apparatus and a method of the type defined in the introduction, which make it possible to predict an early zero-crossing of an alternating current with a good exactness after occurrence of a fault current in a current path.
This object is according to the invention obtained by providing an apparatus of said type with members adapted to detect the current in the current path, an arrangement adapted to calculate the dc-level of the current, i.e. the displacement of the symmetry line of the alternating current with respect to the zero level thereof, and the decay of the dc-level with the time on the basis of values of the alternating current detected by said members, and said arrangement is adapted to predict the time for a future zero-crossing of the alternating current on the basis of at least the current values obtained through said current detection, the calculated dc-level, the calculated dc-decay and information about the period time of the alternating current, as well as a method according to the appended independent method claim.
The apparatus according to the invention designed in that way enables a reliable prediction of a future zero-crossing, since a future zero-crossing is calculated on the basis of said current values detected while considering both the dc-level of the alternating current and how rapidly it falls. It gets by this possible to control a breaker so that the mechanical movement of a contact member is started a certain period of time before a future zero-crossing for obtaining breaking exactly at the zero-crossing, would there be a desire thereof. There is neither any risk of making any attempt to break before any zero-crossing has occurred, since this is first predicted.
It is pointed out that although the apparatus is there for predicting a zero-crossing upon occurrence of a fault, such as a short circuiting, in a current path and is arranged for this sake, it may of course also be used for optimising the breaking of the current in the current path at normal load current, since it is there in anyway.
According to a preferred embodiment of the invention said current detecting members are adapted to detect the time for a zero-crossing of the current, and the arrangement is adapted to consider the time for a detected zero-crossing when predicting a time for a future zero-crossing of the alternating current. By firstly detecting a zero-crossing in this way and starting from this time when calculating a future zero-crossing the prediction of a future zero-crossing will be reliable.
According to a preferred embodiment of the invention said members adapted to detect the alternating current after occurrence of said fault current during a period of time of at least one period of the alternating current, and the arrangement is adapted to use current values resulting through detection of the alternating current during this period of time for calculating said dc-decay. By detecting the alternating current during at least one period the possible influences of harmonics upon the appearance of the alternating current and by that the possible influence thereof upon the time for predicted zero-crossings may be eliminated. The harmonics occurring during a whole period will namely be the same as those occurring during the next whole period and they will by that not influence the times for the predicted zero-crossings, thus, the prediction will be nearly insensitive to harmonics.
According to another preferred embodiment of the invention the apparatus comprises means adapted to integrate the alternating current detected by said members over a first and a second period of time of the same length as the first one and being substantially a period of the alternating current, and said arrangement is adapted to form the quotient of these two current integration values and utilise this for calculating said dc-decay. This constitutes a reliable way to calculate the dc-decay. It is pointed out that the second period of time starts after the first one, but 35 that the two may very well partially overlap each other.
According to another preferred embodiment of the invention the apparatus comprises members adapted to calculate the differential coefficient of the alternating current of the zero-crossing detected through information received from said current detection members, and the arrangement is adapted to use this differential coefficient value when calculating a future zero-crossing of the alternating current. The differential coefficient is then preferably determined on the basis of values of the alternating current detected closely before and closely after said zero-crossing.
According to another preferred embodiment of the invention the current detecting members are adapted to deliver the value of the alternating current of two consecutive current peaks to said arrangement, and the arrangement is adapted to form an average of these two current values for use as said dc-level when calculating said future zero-crossing of the alternating current. The dc-level may in this way easily be determined with the accuracy aimed at.
According to another preferred embodiment of the invention the apparatus is designed for an alternating current in the form of a three-phase alternating current, the arrangement is adapted to calculate the dc-level for two phases by determining an average of two consecutive current peaks of the respective phase, and the arrangement is adapted to calculate the decay with time of the dc-level on the basis of the relation between these two dc-levels and then use it when predicting a future zero-crossing. The dc-decay may in this way at three-phase faults be very rapidly calculated and a condition for an early prediction of a future zero-crossing of the alternating current is by that fulfilled.
According to another preferred embodiment of the invention the apparatus comprises also members adapted to calculate the ac-decay of the alternating current, i.e. the reduction of the amplitude of the alternating current with the time, on the basis of current values delivered by said current detecting members, which further improves the accuracy of the prediction, but it may require a longer time for calculation of the time for a future zero-crossing.
According to another preferred embodiment of the invention the detecting members are adapted to sample the value of the alternating current with a sampling frequency during at least a whole current period and a memory member is adapted to store the values sampled, and the arrangement is adapted to calculate the dc-level at a given time by forming the average of the current values stored for the period of time of a current period backwardly from said time and then use this dc-level in said prediction. It may then advantageously be assumed that the decay of the dc-level is exponential and the arrangement may be adapted to calculate the time constant thereof by dividing the dc-level obtained through division by the time differential coefficient thereof. This forms the basis for a possibility to predict a future zero-crossing of the alternating current without first having to detect any zero-crossing. More exactly, according to another preferred embodiment of the invention the arrangement is adapted to predict the dc-level at a future time on the basis of the dc-level and the decay of dc-level with the time calculated for said given time, and the arrangement is adapted to predict the value of the alternating current by subtracting, from the value of the alternating current measured a current period before the time last mentioned, the difference between the calculated dc-level a current period before the future time and the predicted dc-level of the current at said future time. By means of the current predicted in this way future zero-crossings thereof may be searched in different ways, for example by utilising the method of halve an interval.
According to another preferred embodiment of the invention said detecting members are adapted to detect the time for a peak value of the alternating current, and the arrangement is adapted to use this time as a reference for predicting future zero-crossings of the alternating current. A future zero-crossing may by this be predicted very early, and more exactly this may in a further development of this embodiment take place by the fact that the arrangement of such an apparatus also is adapted to predict the time for the zero-crossing of the alternating current following next to said peak value by adding ¼ of a current period and a first correction factor to the peak value time, and it is adapted to form said correction factor by a product of a constant d and
in which d is the part of the dc-level that remains after half a current period, imax said peak value of the current and dimax the peak value of a standardised differential coefficient of the current during the half period directly before the time for the peak value of the current, in which a standardisation is so selected that imax and dimax get the same numerical values when the current is a pure sine function.
According to another preferred embodiment of the invention the apparatus is adapted to carry out a prediction of the zero-crossing of the alternating current in an electric switching device comprising two branches connected in parallel in the current path, in which the first of them comprises a first contact member having two contacts movable with respect to each other for opening and closing and the second comprises a part with ability to block current therethrough in at least a blocking direction and conduct current therethrough in at least one direction, in which a second contact member having two contacts movable with respect to each other for opening and closing is connected in series with said part, and in which the switching device also comprises a unit adapted to control opening of said current path on the basis of said prediction by controlling the first contact member to open for transferring the current to said part when this is in or going into a conducting state and then the second contact member to open when said part is in a state of blocking current therethrough for breaking the current through the switching device. The apparatus according to the invention is particularly advantageous in connection with such an electric switching device, since it allows a contact opening of the first contact member at the zero-crossing of the current for avoiding an arc, whereupon the second contact member then may be opened when said part is in a blocking state, which in the case of a rectifying diode is after the next zero-crossing. This is also valid for an apparatus according to the appended claim 51, which relates to prediction of the zero-crossing of the alternating current in an electric switching device of the type described in the Swedish patent application 9904166-7 of the applicant still not available to the public. It is pointed out that it is important to “predict” or in advance determine the direction of the current for the predicted zero-crossing. This may be done in different ways, such as by determining the differential coefficient of the current at a given moment, detect a current peak value and so on.
According to another preferred embodiment of the invention the apparatus is designed for predicting a zero-crossing of an alternating current in the form of a multiple phase alternating current, in which a separately controllable electric switching device is arranged in said current path for the respective phase. According to the invention the arrangement is in this case adapted to calculate said future zero-crossing of the alternating current individually for each phase of the alternating current for individually for each switching device determining a suitable time for opening of exactly that switching device. It gets by this possible to obtain a breaking of the alternating current for each individual phase exactly when this is most suitable for the phase in question, and it gets also possible to co-ordinate the breaking of the alternating currents of the different phases with each other should there be a desire thereof. This means a very great improvement with respect to the way to proceed used so far, in which all phases have been broken simultaneously or with a certain fixed phase shift, after a delay resulting in a possibility to state with certainty that zero-crossings occur for all phases.
The phases may through the invention instead b broken at different times depending upon the dc-components they contain. It gets also possible to determine the order of the breaking of the phases depending upon the current values delivered by the current detecting members.
According to a preferred embodiment of the invention the apparatus comprises means adapted to cooperate with an electrically controlled driving member adapted to obtain said opening of the electric switching device, and it is particularly advantageous if this driving member is an electromagnetic machine in the form of an electric motor. By using such a driving member it gets possible to very accurately control the movement of a movable part of the electric switching device for achieving said breaking and for example ensure that a separation of two contacts takes place in a very particular phase position of the alternating current. It may by this be taken full advantage of the prediction of a zero-crossing of the alternating current according to the invention. By the fact that said means for cooperation comprises a control unit in the form of an electronic unit adapted to control said driving member it is also possible to influence a movement of the movable part of the electric switching device when this has already started for making adaptions to possibly new predicted values of the zero-crossing. A co-ordination of an opening of the switching device with such a prediction may by that take place at a high accuracy.
The invention also relates to a device, a computer program and a computer program product according to the corresponding appended claims. It is easily understood that the method according to the invention defined in the appended set of method claims is well suited to be carried out through program instructions from a processor that may be influenced by a computer program provided with the program steps in question. Although not explicitly explained in the claims, the invention comprises such devices, computer programs and computer program products combined with a method according to any of the appended method claims.
Further advantages as well as advantageous features of the invention appear from the following description and the other dependent claims.
With reference to the appended drawings, below follows a description of preferred embodiments of the invention cited as examples.
In the drawings:
An electric switching device for alternating current of the type to which the invention is particularly well applicable is schematically illustrated in
A semiconductor device in the form of a diode 11, 12 having the conducting direction from the outer to the adjacent contact is connected between the two outer contacts and the next adjacent inner contact. The diodes may just as well both be directed with the conducting direction towards the outer contact.
The switching device has also a driving arrangement adapted to drive the inner cylinder 3 to rotate for movement of the movable contact part 5 with respect to the other contacts 7–10. The driving arrangement is in this case constituted by an integrated electric motor 13 schematically indicated, which may be of many different types.
An apparatus 14 for predicting a zero-crossing of the alternating current in the current path 2 is connected to the switching device. This apparatus has members 15 schematically indicated adapted to detect the current in the current path by detecting the direction and the magnitude thereof and by that also detect the time for a zero-crossing of the current. The detecting members are adapted to send signals with information about the current furtheron to an analogues/digital converter 16 for converting the analogues signals to digital signals. Filters 17, 18 are arranged in the signal path before and after the converter for filtrating out noise signals, especially high frequency noise signals, from the signals from the detecting members 15. The current information is sent further to an arrangement 19 adapted to make a calculation of the time for one or more future zero-crossings of the alternating current on the basis thereof. Furthermore, means 20 adapted to integrate the alternating current detected by the detecting members 15 over a first and a second period of time being just as long as the first one and substantially a current period are connected to the arrangement and adapted to send this information further to the arrangement 19, which is adapted to form the quotient of these two current integration values and utilise this time for calculation of the dc-decay of the alternating current, i.e. the development of the dc-component of the alternating current over time.
The apparatus has also members 21 adapted to calculate the differential coefficient of the alternating current at a zero-crossing detected through information from the current detecting members 15 and send this information further to the arrangement 19, which is adapted to use this differential coefficient value when calculating the time for a future zero-crossing.
The arrangement 19 is also adapted to calculate the dc-level of the alternating current at a given time, such as at a zero-crossing detected, on the basis of the signals from the current detecting members 15, and the arrangement may preferably make this by forming an average of the alternating current for two consecutive current peaks and consider this constituting said dc-level.
When the arrangement has in this way predicted a future zero-crossing it will send control signals to a control unit 22 adapted to control the motor 13 and by that the movement of the movable contact part 5 for obtaining a breaking procedure adapted to the time for the predicted zero-crossing. A number of other conditions are also considered and a coordinating with other phases takes place before the motor 13 is started. The control unit 22 is here constituted by an electronic unit adapted to control an electrically controllable driving member 13 in the form of an electric motor and drive the movable part 5 to rotate around the axle 4. By using such an electrically controllable driving member in the form of an electric motor and an electronic unit for co-ordination therewith, the movement of the movable part 5 may be controlled very accurately and adjusted or interrupted as long as it continues.
The function of a switching device of the type illustrated appears more in detail from the Swedish patent application mentioned above but it will here be briefly summarised: when a desire of breaking a current in the current path 2 is born, for example by the fact that the detecting members 15 detect a very high current in the current path 2, which may be caused by a short circuiting therealong, it will then be possible for obtaining the quickest possible breaking to detect the direction of the alternating current and make the rotation direction of the cylinder 3 and by that the movable contact part 5 depending thereupon, but a very high accuracy at the very breaking is given priority with respect to the highest possible speed in the present invention. In the closed position according to
When the voltage over the switching device changes direction no current will flow therethrough, but a voltage will be built up across the diode 11 then reverse biased and the rotation movement of the movable contact part 5 is now continued in the same direction as before, so that the galvanic connection between the contact 8 and the contact 10 is broken, in which this breaking may take place without any arcing, since no current flows through the contact place at the breaking instant. The entirely open position in
The general construction of an electric switching device according the Swedish patent application 9904166-7 mentioned above is schematically illustrated in
An apparatus 14 according the invention for controlling or operating the electric switching device is connected thereto and the construction thereof is the same as described above for the embodiment according to
The function of this electric switching device is as follows: when there is a desire of breaking the current in the current path 2, for example by the fact that the detecting member 15 detects a very high current in the current path, which may be caused by a short circuiting therealong, it is determined in the way described above through the result of the detection when it is most suitable to break the current through the respective electric switching device. Once it has been determined that a given electric switching device shall be opened, the control unit 22 takes first a decision of which two contact members, here the contact members 37 and 38 (se
The apparatus according to the invention has the object to predict a future zero-crossing or several future zero-crossings of the alternating current for obtaining the breaking procedure according to above being an optimum with respect to the location thereof on the time scale. How this is intended to take place in the practise will now be explained with reference to
It is illustrated in
During a period of time t2 of a good whole period of the alternating current, i.e. somewhat more than 20 ms, after the short circuiting detected the value of the alternating current is detected and registered, in which two zero-crossings 23, 24 are detected. A first prediction of future zero-crossings 25–27 is then made at the time t3. The predictions of the zero-crossings 25 and 27 are made on the basis of the zero-crossing 23 measured and the prediction of the zero-crossing 26 on the basis of the measured zero-crossing 24. By basing the prediction on whole periods (instead of half periods) the prediction gets rather independent of both even and odd harmonics.
It is illustrated in
The differential coefficient of the alternating current at a zero-crossing detected is further calculated by measuring the current at two times close to the zero-crossing at t4 and divide the difference in current level between these with the time, as shown through the points 30 and 31. The reading of the current then always takes place on the side of current zero on which the long half wave of the alternating current is located, i.e. on the side with a positive dc-addition. For determining the dc-decay of the alternating current the alternating current is integrated over a first and a consecutive (possibly with a certain overlap) time period being just as long as the first one, which each is substantially a period of the alternating current, and the quotient of these two current integration values is then formed for utilising 0them when calculating the dc-decay.
The following formula is preferably used for predicting a future zero-crossing:
tpred=tm+T+dc×(1−d2)/s
In which these stand for the following:
According to another preferred embodiment of the invention a whole period of the current is stored in a buffer memory. The dc-level and the decay thereof are continuously calculated through integration of the buffer memory. A period of the current may at each time be predicted through assuming that the current gets the same as it was a period backwardly in the time minus the current dc-decay.
The prediction according to the invention gets a high accuracy, and it is particularly well suited for a multiple phase alternating current with a separately controllable switching device arranged in the current path for the respective phase, since a breaking of the different phases may take place at times suitable for each phase.
A method for predicting a future zero-crossing according to another preferred embodiment of the invention will now be explained with reference to
Idc*(t)=idc*(t−1)+(imesu(t)−imesu(t−T))/T
T is the number of samples of a current period.
It is also assumed that the decay of the dc-level is exponential and the time constant τ is calculated by dividing the dc-level obtained through the division by the time differential coefficient thereof according to
τ=−idc*(t)/(didc*(t)/dt)
the dc-level of the current may by this be calculated at an arbitrary time, so that this for the sample t+t1 gets:
idc(t+t1)=idc(t)*exp(−t1/τ)
The current at the sample t+t1 may by means of this be predicted according to:
ipred(t+t1)=imesu(t+t1−T)−(idc(t+t1−T)−idc(t+t1))
Thus, a value of the alternating current in a future time is predicted by subtracting from the value of the current measured a current period before the time last mentioned the difference between the dc-level calculated a current period before the future time and the predicted dc-level of the current at said future time. By means of the predicted current future zero-crossings may be searched by means of for example the method of halve an interval.
It is schematically illustrated in
t1pred=t0+T/4+korr1
t2pred=t1pred+T/2+korr2
Korr1 and korr2 are calculated by means of the quotient of the maximum current (imax) and the maximum differential coefficient (dimax) during the last half period according to
korr1=Ad(1-imax/dimax)
korr2=Bkorr1d,
in which A, B are constants, and dimax the peak value of the standardised differential coefficient 44 the half period directly before, in which the standardisation is such as for a pure sine function imax=dimax. d is the part of the dc-level remaining after half a period.
It is finally schematically illustrated in
y1r−y2r−dcr(1/√{square root over (d)}−√{square root over (d)})=2 ymax
y1s−y2s−dcr(1/√{square root over (d)}−√{square root over (d)})=2 ymax
where d indicates how great part of the dc-level remains after ½ current period, and 2ymax is the distance between two consecutive current peaks in absence of dc-decay. d may be cancelled out from this equation system and by that the dc-decay be calculated.
The decay of the dc-level with time may be calculated in the corresponding way upon occurrence of a fault current in a one phase alternating current by determining the value of the alternating current of three consecutive current peaks through said current detection, and by then writing a corresponding equation system with a comparison of the first two current peaks in the first equation and the second and third current peak in the second equation.
By the fact that the method according to the invention for predicting a zero-crossing of the alternating current allows a large content of harmonics a very accurate prediction may be made also upon for example a one or two phase short circuiting of a generator or when the fault location contains an arc. The apparatus according to the invention is advantageously used for predicting a zero-crossing of the alternating current in a current path in a switch gear for electricity supply within industry or in distributions or transmission networks, and the prediction preferably takes place for an alternating current in a current path having a voltage on intermediate voltage level, i.e. between 1–52 kV. However, the invention is not restricted to alternating voltages on these levels.
Furthermore, the invention is particularly applicable to prediction of a zero-crossing of an alternating current in the current path through an electric switching device adapted to take an operation current of 1 kA, preferably at least 2 kA.
The invention is of course not in any way restricted to the preferred embodiments described above, but many possibilities to modifications thereof would be apparent to a person with ordinary skill in the art without departing from the basic idea of the invention as defined in the appended claims.
The invention is as already mentioned applicable to all types of electric switching devices.
| Number | Date | Country | Kind |
|---|---|---|---|
| 0002125 | Jun 2001 | SE | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/SE01/01263 | 6/7/2001 | WO | 00 | 6/4/2003 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO01/95354 | 12/13/2001 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4754360 | Nakada | Jun 1988 | A |
| 4959558 | Kay et al. | Sep 1990 | A |
| 5510591 | Tremblay et al. | Apr 1996 | A |
| 5563459 | Kurosawa et al. | Oct 1996 | A |
| 5793594 | Niemira et al. | Aug 1998 | A |
| 5804991 | Hu | Sep 1998 | A |
| 6597999 | Sinha et al. | Jul 2003 | B1 |
| 20030067723 | Suzui et al. | Apr 2003 | A1 |
| 20030202303 | Parker | Oct 2003 | A1 |
| 20040090720 | Larsson et al. | May 2004 | A1 |
| Number | Date | Country |
|---|---|---|
| 0137298 | May 2001 | WO |
| 0137300 | May 2001 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20040090719 A1 | May 2004 | US |
