AU4304799A

AU4304799A - Method and device for reducing the current in the neutral conductor

Info

Publication number: AU4304799A
Application number: AU43047/99A
Authority: AU
Inventors: Mats Karlsson
Original assignee: Emerson Energy Systems AB
Current assignee: Vertiv Sweden AB
Priority date: 1998-05-06
Filing date: 1999-05-04
Publication date: 1999-11-23
Also published as: TR200003209T2; RU2000130721A; WO1999057793A1; SE9801582L; CN1308785A; CA2331710A1; DE1092255T1; JP2002514036A; HUP0102324A2; SE9801582D0; EP1092255A1; ES2158836T1; SE513884C2; BR9910214A

Description

WO 99/57793 PCT/SE99/00743 METHOD AND DEVICE FOR REDUCING THE CURRENT IN THE NEUTRAL CONDUCTOR Technical Field The present invention relates to an electric supply system having at least two phases 5 and a neutral conductor and a controllable load for each phase, and to a method in such a system. The invention also relates to a single-phase electrical system comprising a first and a second load, said loads being serially connected, and two serially connected trans 10 formers connected in parallel with the loads, a point between the loads and a point between the transformers, being connected by a conductor, and to a method in such a system. Description of Related Art 15 Such electrical systems are common in the art, the controllable loads, being, for ex ample, rectifiers. Object of the Invention When the neutral conductor of a three-phase electric supply system is connected 20 with a three-phase load, any asymmetries in the phase voltages, or in the load, will result in a current in the neutral conductor, causing a power loss in the neutral con ductor. It is also possible to create a neutral point other than the neutral point of the neutral point of the three-phase supply system in the load in some way. In this case, a current in the neutral conductor will result in a power loss in the unit creating the 25 neutral point. It is an object of the present invention to improve the performance of a two- or three-phase electrical system.

WO 99/57793 PCT/SE99/00743 2 It is also an object of the present invention to improve the performance of a single phase electrical system comprising two serially connected loads. Summary of the Invention 5 The object stated above is achieved according to the invention by an electric supply system as initially defined, said supply system comprising means for reducing the current in the neutral conductor, and by a method as initially defined, said method including reducing the current in the neutral conductor. 10 By reducing the current in the neutral conductor, the following effects are achieved: The size of the unit needed to balance the voltages of the system in case of an arti ficial neutral point being present is minimized. The voltage drop that may occur in the unit creating the neutral point is reduced or prevented. 15 According to a preferred embodiment, said means for reducing the current in the neutral conductor comprises: - means for measuring the voltage or current of each phase, - means for measuring the current in the neutral conductor, 20 - multiplying means for multiplying a signal corresponding to the phase voltage or current of each phase and a signal corresponding to the current in the neutral con ductor - means for feeding back the product of the phase voltage and the current to the load in a negative feedback loop. 25 Said apparatus may also comprise - interface means for converting the measured phase voltages or currents into con trol signals, - interface means for converting the measured current in the neutral conductor into a 30 control signal, and WO 99/57793 PCT/SE99/00743 3 - means for normalizing the output of said multiplying means by low-pass filtering. According to a preferred embodiment, said method comprising the following steps: - measuring the instantaneous phase voltage or phase current of each phase, 5 - measuring the instantaneous current in the neutral conductor, and for each phase performing the following steps: - multiplying a signal corresponding to the phase voltage or current of each phase by a signal corresponding to the current in the neutral conductor, - feeding back the product of the phase voltage and the current to the load in a 10 negative feedback. Preferably, the apparatus also comprises filter means for normalizing the output by low-pass filtering. 15 Said object is also achieved in a single-phase electric system as initially defined, said system comprising means for reducing the current in said conductor, and by a method in such a system, said method including reducing the current in said conduc tor. 20 By reducing the current in said conductor, the voltage across the two loads is bal anced. According to a preferred embodiment, the system comprises - a first measuring means for measuring the current in the conductor 25 - a second and a third measuring means for measuring the voltages across, or the currents through, said first and second loads, - means for multiplying the output signals from the first and the second, measuring means, processing the resulting signal and feeding it to the first load, and - means for multiplying the output signals from the first and the third measuring 30 means, processing the resulting signal and feeding it back to the second load.

WO 99/57793 PCT/SE99/00743 4 According to a preferred embodiment, said method comprises the following steps: - connecting at least two transformer windings in parallel with the serially con nected loads and serially connecting the windings, said transformer windings being 5 wound on the same core, - connecting a point between the two loads and a point between the two transformer windings by means of a conductor; and reducing the current in the conductor. According to a preferred embodiment, said method comprises the steps of: 10 - measuring the current in the conductor; - measuring the voltage across, or the current through, each load for each load: - multiplying a signal corresponding to the voltage across the load by a signal corre sponding to the current in the conductor; 15 - feeding back the result of the multiplication to the respective load in a negative feed-back loop. The invention utilizes the fact that the load having a greater current than the others will result in a current in the neutral conductor that is in phase with the curernt in 20 this phase, or the voltage of the phase if cosp~1, cp being the phase angle. If the phase current or voltage is in phase with the current in the neutral conductor, the average value of the product of current and voltage will be at a maximum. If the neutral current and the phase current, or phase voltage, have phase angles in the range of 900 - 2700, the average value of the product will instead be negative, with a 25 minimum at a phase angle of 1800. The invention offers the following advantages: It is particularly useful in two- or three phase systems in which an artificial neutral point is created, to minimize the current through this artificial neutral point. 30 Minimizing the neutral current is in itself always advantageous.

WO 99/57793 PCT/SE99/00743 5 In a single-phase system, a stable load distribution is achieved. Brief Description of the Drawings In the following, the invention will be further described with reference to the ac 5 companying drawings, in which Figure 1 shows a three phase system in which the inventive method and apparatus may be applied; Figure 2 shows the principles of an embodiment of the inventive method applied to a three-phase system; 10 Figure 3 shows the signal processing units applied in the feedback loop according to one embodiment of the invention; Figure 4A shows a single-phase system in which the inventive method and appara tus may be applied; Figure 4B shows the implementation of the inventive apparatus in the system of 15 Figure 4A. Detailed Description of Embodiments Figure 1 shows an electrical system with three phases, R, S and T and a neutral conductor N. A first rectifier 1 is connected to phase R, a second rectifier 3 is con 20 nected to phase S, and a third rectifier 5 is connected to phase T. The secondary in put terminals of the rectifiers 1,3, 5 are interconnected in a connection point NL. To balance the voltages across the rectifiers, a balancing unit 7 is connected. Such a balancing unit 7 may be implemented in several different ways. The balancing unit 25 creates an artificial neutral point in the system. If the sum of the three currents dif fers from zero, a current will flow through this point. To minimize the size of the circuits needed to create a stable artificial neutral point, this current should be minimized.

WO 99/57793 PCT/SE99/00743 6 The system shown in Figure 1 is only one example of a system in which an appara tus according to the invention may be useful. It may also be used in a lighting sys tem with lamps connected in a star configuration or a heating system. In such a system it will, for example, function to balance the phase voltages if the load is un 5 evenly distributed over the phases. Figure 2 shows the apparatus according to the invention applied in a three-phase network like the one in Figure 1. A first 11, a second 12 and a third 13 component are connected on the R phase, the S phase and the T phase, respectively. The com 10 ponents 11, 12, 13 may be the rectifiers shown in Figure 1. Each rectifier has a first and a second input terminal, a first and a second output terminal and a first and a second control input terminal. Instead of rectifiers, the components 11, 12 and 13 may be any kind of component having a controllable input current. An impedance Z1l, Z12 Z13 between the first output terminal and earth represents the load on the 15 first 11, second 12 and third 13 rectifier, respectively. The second output terminal is connected to earth. If the apparatus is to be applied to the system shown in Figure 1, the impedances Z 11, Z 12 and Z 13 can be connected in parallel (not shown in Figure 2). 20 On the first input terminals of each component, the respective phase voltage relative to earth is applied. The second input terminals are connected to the artificial neutral point. For each component 11, 12, 13, there is a feedback loop comprising the following 25 components: - An interface unit 21, 22, 23 respectively, converting an input voltage or current to a control signal, which may be a digital signal or an analogue electric signal, - A multiplying unit 31, 32, 33, respectively, - A processing unit 41, 42, 43. 30 WO 99/57793 PCT/SE99/00743 7 The input to each interface unit 21, 22, 23 is the respective phase voltage or current relative to earth. The output is a control signal having the same frequency and phase angle as the input voltage. The output signal of each interface unit 21, 22, 23 is fed to a multiplying unit, 31, 32, 33. 5 The current in the neutral conductor is measured in a measuring unit 35 and applied to the input terminal of a fourth interface unit 37, converting the input current signal to an output control signal having the same frequency and phase angle as the input current. The output signal from the fourth interface unit 37 is fed to the multiplying 10 units 31, 32, 33. Thus, the output signal from each multiplying unit 31, 32, 33 is the product of the respective phase voltage or current and the current in the neutral con ductor. This product is fed back to the control input of the corresponding rectifier 11, 12, 13 through a processing unit 41, 42, 43, respectively. 15 The processing units 41, 42, 43 may comprise several different units performing dif ferent functions, as will be discussed below in connection with Figure 3. If the current in the neutral conductor is zero the output signals from the multipliers 41, 42, 43 will be zero. If the current is different from zero, the output signals will 20 be sine waves with twice the frequency of the mains and average values different from zero. Assuming that the first rectifier 11 generates a stronger current in the neutral con ductor than the other ones 12,13, the current in the neutral conductor will have sub 25 stantially the same phase as the phase voltage of the R phase, if cos~1, cP being the phase angle. Thus, the output signal from the multiplying unit 41 will be strong and positive. In a negative feedback, a strong and negative signal will be applied to the control input terminal of the first rectifier 11, to reduce the input current to this rec tifier. This may be done, for example, by controlling the output current or the output 30 power.

WO 99/57793 PCT/SE99/00743 8 If the phase voltage or current has the opposite phase of the current in the neutral conductor, the product of the phase voltage or current and the neutral current will be negative. Thus, the feedback signal in this phase will be positive and the input cur 5 rent to the rectifier will be increased. Similarly, for a rectifier that contributes very little to the current in the neutral con ductor, the control signal applied to this rectifier will be correspondingly weak. 10 The processing of the output signals from the multipliers in the feedback loop in cludes the change of sign. In this way, a positive average output signal from a mul tiplier 41, 42, 43 will reduce the load on the corresponding phase. According to a preferred embodiment, the processing of the output signals includes the following steps: 15 - multiplication of the signal by a negative number - filtering of the sine wave signal to produce a signal without any AC component. Figure 3 shows an embodiment of a processing means, corresponding to one of the processing units 41, 42, 43 shown in Figure 2. An input signal from the multiplier, 20 which is a sine wave of twice the mains frequency, is first inverted in an inverting means 101, which may also be an amplifying means. The signal may also be sub jected to other processing, for example, integration in an integrating means 103. The conversion to a direct voltage is carried out in a filtering unit, which may be imple mented in any way known in the art. Here it is shown as an RC filter comprising a 25 resistance 107 and a capacitance 109. The output from the RC filter forms the con trol signal which is fed to the control input of the appropriate rectifier. The invention has been discussed in the above with reference to a three-phase sys tem only. As will be obvious to the person skilled in the art, it can easily be adapted WO 99/57793 PCT/SE99/00743 9 to a single-phase or two-phase system, without changing the implementation apart from the removal of one phase. Figure 4A shows a single-phase system in which the inventive idea may be utilized. 5 A voltage is applied across two loads 201, 203 that are serially connected. Two transformer windings 205, 207 on the same core are connected in parallel with the loads 201, 203. The two branches are interconnected by a connection 209 between a point 210 between the loads 201, 203 and a point 211- between the transformer windings 205, 207. The windings 205, 207 ensure an equal distribution of the volt 10 age across the two loads 201, 203. If the currents through the loads 201, 203 are not equal, there will be a current in the connection 209 between the two branches. In Figure 4B, an apparatus according to the invention has been connected to reduce the current in the connection 209 between the branches. The loads 201, 203, the 15 windings 205, 207 and the connection 209 are the same as in Figure 4A. In this em bodiment, the current in the connection 209 is measured and, if desired, converted to a control signal of the same frequency and phase angle, in a measuring device 212. The voltage across, or the current through, each of the two components 201, 203 is measured in a second and a third measuring device 215, 217, respectively. 20 The output signals from the first 212 and second 215 measuring devices are input to a first multiplying device 219. The output from the first multiplying device 219 is inverted and processed in a first processing unit 223 as explained above and fed to the control input of the first load 201. The output signals from the first 212 and third 25 217 converting devices are input to a second multiplying device 221. The output from the second multiplying device 221 is inverted and processed in a second proc essing unit 225 as explained above and fed to the control input of the second load 203.

WO 99/57793 PCT/SE99/00743 10 In analogy with the discussion above, this will lead to a positive signal being fed to a component that contributes negatively to the current in the conductor 209 and a negative signal being fed to a component that contributes positively to the current in the conductor 209, thus serving to reduce said current by balancing out the voltages 5 across the loads 201, 203. As will be obvious to the skilled person, two transformers comprising primary and secondary windings may be used instead of the transformers with duplicate wind ings on the same core. In case of two separate transformers, the outputs from the 10 transformers should be connected in parallel, to force the primary voltages to be equal. The same principle can be used for three-phase systems.

Claims

1. An apparatus for use in an electric supply system having at least two phases (RS, T) and a neutral conductor (N) and a controllable load (11, 12, 13) for each phase, 5 said apparatus comprising means for reducing the current in the neutral conductor, characterized in that said means comprises: - voltage measuring means (21, 22, 23) for measuring the voltage of each phase (R, S, T) or phase current measuring means (21, 22, 23) for measuring the current of each phase (R, S, T), 10 - current measuring means (35) for measuring the current in the neutral conductor (N) - multiplying means (31, 32, 33) for multiplying a signal corresponding to the phase voltage of each phase and a signal corresponding to the current in the neutral con ductor 15 - feedback means for feeding back the product of the phase voltage, or phase cur rent, and the current to the load in a negative feedback loop.

2. An apparatus according to claim 1, characterized in that it comprises - interface means (21, 22, 23) for converting the measured phase voltages or phase 20 currents into control signals - interface means (37) for converting the measured current in the neutral conductor (N) into a control signal.

3. An apparatus according to any one of the claims 1 or 2, characterized in that it 25 comprises - filtering means (41, 42, 43) for normalizing the output of said multiplying means (31, 32, 33) by low-pass filtering.

4. A method of reducing the current in the neutral conductor in an electric system 30 having at least two phase conductors (R, S, T) and one neutral conductor (N), and a WO 99/57793 PCT/SE99/00743 12 controllable load (11, 12, 13) for each phase, said method being characterized by the steps of - measuring the phase voltage or phase current, of each phase (R, S, T), - measuring the current in the neutral conductor (N), and 5 for each phase (R, S, T), performing the following steps: - multiplying a signal corresponding to the phase voltage by a signal corresponding to the current in the neutral conductor, - feeding back the product of the phase voltage, or phase current, and the current to the load in a negative feedback. 10

5. A method according to claim 4, characterized by the step of - normalizing the product by low-pass filtering.

6. A single-phase electrical system comprising a first (201) and a second (203) load, 15 said loads (201, 203) being serially connected, and two serially connected windings (205, 207) connected in parallel with the loads (201, 203), a point (210) between the loads (201, 203) and a point (211) between the windings (205, 207), being con nected by a conductor (209), characterized in that it comprises 20 - means for reducing the current in said conductor (209).

7. A system according to claim 6, characterized in that it comprises: - a first measuring means (212) for measuring the current in the conductor (209) - a second and a third measuring means (215, 217) for measuring the voltages across 25 said first (201) and second (203) loads, - means (219) for multiplying the output signals from the first (212) and the second (215), measuring means, processing the resulting signal and feeding it to the first load (201), and WO 99/57793 PCT/SE99/00743 13 - means (221) for multiplying the output signals from the first (212) and the third (217) measuring means, processing the resulting signal and feeding it back to the second load (203). 5

8. A system according to claim 6, characterized in that it comprises: - a first measuring means (212) for measuring the current in the conductor (209) - a second and a third measuring means (215, 217) for measuring the currents through said first (201) and second (203) loads, - means (219) for multiplying the output signals from the first (212) and the second 10 (215), measuring means, processing the resulting signal and feeding it to the first load (201), and - means (221) for multiplying the output signals from the first (212) and the third (217) measuring means, processing the resulting signal and feeding it back to the second load (203). 15

9. A method of balancing the voltage across two serially connected loads in a sin gle-phase system, comprising the following steps: - connecting a transformer winding in parallel with each of the serially connected loads and serially connecting the windings, said transformer windings being wound 20 on the same core, - connecting a point (210) between the two loads (201, 203) and a point (211) be tween the two transformer windings (205, 207) by a conductor (209); characterized by the step of reducing the current in the conductor (209). 25

10. A method according to claim 9, characterized by the steps of: measuring the current in the conductor (209); measuring the voltage across each load (201, 203) for each load: WO 99/57793 PCT/SE99/00743 14 - multiplying a signal corresponding to the voltage across the load (201, 203) by a signal corresponding to the current in the conductor (209) - feeding back the result of the multiplication to the respective load (201, 203) in a negative feed-back loop. 5

11. A method according to claim 9, characterized by the steps of: - measuring the current in the conductor (209); - measuring the current through each load (201, 203); for each load: 10 - multiplying a signal corresponding to the current through the load by a signal cor responding to the current in the conductor (209) - feeding back the result of the multiplication to the respective load (201, 203) in a negative feed-back loop.