GB1603504A - Method and apparatus for the control of real power to fluctuating loads - Google Patents

Description

(54) METHOD AND APPARATUS FOR THE CONTROL OF REAL POWER TO FLUCTUATING LOADS (71) We, DONALD JOHN CHEE-HING, a Citizen of Trinidad and Tobago, of High Street, Siparia, Trinidad, and KENNETH STEPHEN JULIEN, a Citizen of Trinidad and Tobago, of Departmental Electrical Engineering, University of the West Indies, Trinidad, do hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed to be particularly described in and by the following statement: The invention relates to a method and apparatus for compensating randomly varying real current and real power of a fluctuating load. Such fluctuating load commonly causes light flicker and voltage fluctuations, transient instability and power system frequency deviations.

Fluctuating loads such as electric arc furnaces, metal rolling mills, welding machines, etc.

cause highly fluctuating real and reactive power components to be drawn from the power supply. It is well known that the rapidly fluctuating reactive power component causes voltage fluctuations and poor power factor in the network. To compensate for these reactive effects several new static VAR compensators with associated control circuitry have been developed. The real power fluctuations also causes voltage fluctuations and light flicker effects, and more importantly can cause system frequency changes to develop in the supply network. It is known that the power system frequency must be kept within strict limits, otherwise frequency sensitive relays and protection equipment will operate causing interruption of power service, and in more severe cases can cause transient instability to develop in the supply system.The prior art to compensating for the fluctuating real power component was for the utility company to increase the short circuit capacity of the load bus through improving generation, transmission and distribution facilities, including improving the response of the load-frequency control devices of the generators. When the utility company cannot improve the short circuit capacity to the level required it becomes the responsibility of the consumer to employ load compensating devices to ensure that his load does not affect the integrity of the supply system. One known solution is for the consumer to provide part of his power requirements through installation of local generators. In another known method the consumer employs rotating equipment to compensate for the real power changes of his load.This is accomplished by pumping energy in and out of a flywheel attached to a synchronous machine. The method is however, applicable only to slow and periodic power changes and cannot be applied to random and rapidly real power changes.

There is evidently a need for a method and apparatus capable of compensating for load changes as fast as they occur, and capable of balancing and compensating each phase of a polyphase system. This becomes more important for relatively weak supply systems in which the rating of the dynamic load becomes comparable to the short circuit capacity of the supply bus.

According to one aspect of the present invention there is provided apparatus for compensating the randomly varying real current and real power of a load when coupled to an A. C. power supply, the apparatus comprising a compensating resistance arranged to be coupled with said load to the supply, and means for controlling the power drawn by said resistance from the supply, said means including controllable switch means, and means for the control of said switch means, this means for control being arranged to measure the current through said load and to derive a control from the real component of the load current.

According to another aspect of the invention, there is provided a method for compensating the randomly varying real current and real power of a load when coupled to an AC power supply, comprising measuring the current through said load, deriving a control signal in accordance with the real component of the load current and controlling the power drawn from said power supply by a compensating resistance. also coupled to said power supply, by means of applying the control signal to controllable switch means.

Accordingly, the technique disclosed herein in accordance with the present invention utilizes no rotating equipment and is based on static control of resistive compensators.

The resistive compensators are made to take a real power complementary to the real power demand of the main load, such that an equivalent constant load is presented to the supply network at all times, and a constant real energy flow is developed in the supply system. A control scheme is presented using signals derived from the load current to control the firing angles of the thyristors for proper compensation. The resistance of the compensator could be actual physical resistances, or an electrolytic cell, or any load that can be represented by an equivalent resistance.

Embodiments of the invention will now be described by way of example with reference to the accompanying drawings, in which: Figure 1 is a schematic diagram of a real power compensator in accordance with the present invention and applied to a single-phase system; Figure 2 shows the general current waveform of a general time varying load; Figure 3 illustrates the principle of establishing a constant supply real power; Figure 4 illustrates the current transducer reduced current signal; Figure 5 shows the output waveforms from the vector generator Figure 6 shows the schematic circuit of the slope measurement and pulse gating circuit in greater detail; Figure 7 illustrates the action of the circuit of Figure 6 through detailed waveforms; Figure 8 shows the circuit for on-off switching control of a number of resistive compensators;; Figure 9 shows the method of on-off switching of resistive compensators; and Figure 10 is a schematic diagram of a real power compensator in accordance with the present invention and applied to a three-phase system.

Figure 1 illustrates a real power compensator to compensate for the time varying real power demand of a single-phase load 1. The single-phase load 1 is connected to the single-phase power supply vab, where Vab = V sin wt. (1) The compensator consists of the compensating resistance Rab as represented by numeral 2, connected by a pair of inverse parallel thyristors T1 and T3 as represented by numeral 3 to the voltage supply Vab. The inverse thyristor pair T1 and T2 could be replaced by a single triac of suitable rating. The load 1 can be considered to be a general stochastic load taking a time varying distorted current iL(t) which varies randomly between some maximum peak value and a minimum value, as shown in Figure 2. The load current iL(t) can be decomposed by Fourier analysis to give the time varying DC component, the fundamental frequency component and the sum of the various higher harmonic components,

The fundamental frequency current component can itself be decomposed into the real current component ILp(t) sin wt which is in phase with the supply voltage V sin wt, and the orthogonal reactive current component 1LR t) cos wt which is in quadrature with the supply voltage, and where ILl(t) sin (wt - pI(t)) = ILp(t) sin wt - ILR(t) cos wt. (3) The prior art employs compensating devices as collectively represented by numeral 4, of Figure 1 to compensate for the time varying reactive curent ILR(t) cos wt as well as removal of the time varying harmonic terms.Whether these prior art compensating devices are used or not, the time varying real current component still remains flowing through the supply network, i.e.

is(t) = ILp(t) sin wt. (4) For a particular load, the fundamental real current component will vary randomly between a maximum peak value ILP max and a minimum value ILP min which could be zero, and where ILP min S ILP(t) S ILP max (5) It is the large random variation of this real current component ILp(t) sin wt flowing through the supply network that results in additional voltage fluctuations, system frequency deviations and transient instability, as outlined. If the real current flow can be made constant, then these deficiencies are totally eliminated.It is the function of the compensator as represented by 2 and 3 to take a real current component complementary to the load real current component at all times such that the supply real current component ia(t) assumes a constant amplitude, i.e.

is(t) = ILP max sin wt. (6) To achieve the equivalent contant supply current ILP max while the load real current ILp(t) varies, the resistive compensator must be controlled to take the complementary current ICp(t) such that IcP(t) sin wt = (ILP max - ILp(t)) sin wt. (7) In terms of power components, the instantaneous real power developed by the load is

while the real power developed by the compensator is

such that the constant power output from the supply network is PT = PL(t) + Pc(t) (10) This concept is illustrated in Figure 3.

Equation 5 indicates that the load real current varies randomly between maximum and minimum boundary values. The compensator real current then varies complementarily between the same boundary values, i.e.

ILP max 3 ICP(t) 3 ILP min (11) In one control embodiment, phase angle control of the thyristors T1 and Tz is employed to achieve fast, smooth and variable response of the compensator current. The value of the compensating resistance to be used is given by

with the range of thyristor firing angle a0 in each half-cycle being variable, i.e.

0 6 ct 6 1800. (13) To achieve proper compensation it is necessary to vary the equivalent value of the compensating resistance Rab through firing of the thyristors with control signals derived from the time varying load current. The remainder of the circuit illustrates means of deriving the required control signals.

A reduced load current signal i'L(t) proportional to the load current iL(t) is obtained at outputs 6 and 7 of a current transducer 5, as shown in Figure 4. This load current signal is then processed by a vector generator 8. This device which is available commercially separates the fundamental real and reactive current components from the load current signal, as indicated by Equation 3. Figure 5 illustrates the waveforms obtained. The reactive current signal I"LR(t) cos wt obtained from outputs 9 and 10 of the vector generator 8 is not processed further, as only real power compensation is being considered. The real current signal I"Lp(t) sin wt taken from outputs 11 and 12 is processed in the slope measuring and pulse generating circuit of 13, which is shown in greater detail in Figure 6.The voltage supply vab at inputs 14 and 15 is required to provide power and voltage synchronization to circuit 13. This circuit measures the initial rate of rise of the time varying real current component, or

at each zero current and voltage crossing points and produces pulse gating signals at outputs 16 and 17. These pulse gating signals are fed to the gate-cathode circuit of the respective thyristors T1 and T2 for independent half-cycle control. The delay angles of the gating signals depend on the values of the measured slopes, as illustrated in Figure 7. When the slope measured is small, a pulse output is produced with a small delay angle al. When the slope measurement is large, a pulse output is produced after a larger delay angle a2 as depicted in Figure 7.

The current response of the thyristor controlled resistive compensator depends on the firing angle a(t) and is given by

ic(t) = 0 0 6 wt 6 a(t) (15) where 0 6 a(t) 6 1800 (16) When a(t) = 00, iL(t) = V/Rab and the maximum normal sinusoidal resistive current flows.

When a (t) = 1800, then ic(t) = 0 and no current flows. For intermediate values of a(t), a nonsinusoidal current pattern is obtained as illustrated in Figure 7. This current can be decomposed by Fourier analysis to give the time varying DC component, the fundamental frequency component, and the sum of the various higher harmonic components, i.e.

ic(t) = ICo(t) + Ici(t) sin (wt - l(t))

The fundamental component can be further decomposed to give the fundamental real and reactive current components, similar to Equation (3), i.e.

Ia(t) sin (wt - P)l(t)) = Icp(t) sin wt - IcR(t) cos wt. (18) The magnitude of the real current component depends on the firing angle a (t) and is given by

If prior art reactive and harmonic compensating devices as collectively represented by 4 are employed, the reactive component and higher harmonics are compensated, leaving the fundamental real current component of Equation (18). From Equation (6) it is required for ideal compensation that the resultant total supply real current be constant, i.e.

is(t) = ILP max sin wt (20) where from Equation 7.

is(t) = ILp(t) sin wt + ICp(t) sin wt. (21) From Equations (19, (20 and (21)

For any value of load current ILp(t) sin wt, there is thus a unique complementary value compensator current which results in maintaining the constant supply current of Equation (20). It is the function of the slope measurement and pulse gating circuit of 13 to translate the value of the measured slopes to pulse outputs with proper delay angles for firing the thyristors such that Equation (22) is satisfied.

In Figure 6, the real current signal taken from outputs 11 and 12, as shown in Figure 7(a), is first rectified by the full wave bridge rectifier of 18, giving unidirectional DC waveforms at outputs 19 and 20 as shown in Figure 7(b). The unijunction transistor circuit of 21 measures the initial slope of the DC waveforms as described and illustrated in Figure 7(c), producing pulse signals with proper delay angles at outputs 22 and 23, as shown in Figure 7(d). These pulse signals are amplified by the pulse amplifier 24. The amplified output taken at 25 is transformed by the isolating transformer 26, and the outputs at 16 and 17 are conveyed to the thyristors. Figure 7(e) shows the current Ic(t). Figure 7(f) shows the current component ICp(t).

It is evident that greater or lesser compensation action can be obtained for any particular situation by altering the value of the compensating resistance.

In another control embodiment, on-off control of the thyristors are employed intead of the phase-angle control as previously described. In this method the thyristors are either fully conducting with az 00, or the thyristors are fully non-conducting i.e. there is no gating signal. In this case a sinusoidal current always flows when the thyristors are conducting, and no harmonic components are generated. Since the control action is essentially on-off, the thyristors could be replaced by other contactors, such as vacuum switches or relays. This method is useful when large real power changes of the load can be permitted to be passed on to the power system, and exact compensation is not required.The resistive compensator is switched on only when the load power, or equivalently the value of the load signal real peak current I"Lp(t) sin wt falls below some chosen value.

The current transducer 5, and vector generator 8 perform the same functions as described previously resulting in a real current signal I"Lp(t) sin wt to be obtained at outputs 11 and 12 of the vector generator. The slope measurement and pulse generating circuit of 13, then measures the initial rate of rise of the time varying real current component, or dI11LP(t) LP(t) dt at each current and voltage zero crossing point. When these measured values are greater than or equal to some specified chosen value, no pulse output is obtained. The thyristors are not fired and no compensation action is obtained.When the measured values are less than the specified chosen value, pulse gating signals are produced at outputs 16 and 17 with the fixed delay angle of O", or near 0 , such that the thyristors are fully conducting and the maximum compensating current flows. The voltage supply vab at inputs 14 and 15 is again required for power and voltage synchronization for circuit 13.

When this on-off discrete control principle is used, it may be desirable to have a number of these resistive compensators to effect compensation in steps as the load real power fluctuates. Such an arrangement is illustrated in Figure 8, where the individual resistive compensators comprising compensating resistances R1 to R4, each connected in series with a respective switch Sl to S4 respectively, across the supply vab, can be switched on and off in any combination to compensate for the load changes. The resultant real power compensation action is shown in Figure 9, the compensation changing at instants t1 to tS at which the combination of operative compensators changes.

The compensator as described above for the single-phase case is equally applicable to polyphase voltage systems, and in particular for compensation on commercial three-phase systems. It becomes necessary to arrange three resistive compensators in star or delta configuration, but the delta arrangement is preferred to obtain independent phase control, as shown in Figure 10. This is important since three-phase dynamic loads produce different time varying currents in each phase of the three-phases, and different degrees of compensation are required simultaneously. The method of analysis and control becomes similar to that as described for the single-phase case, which in fact represents the analysis of one of the phases of a three-phase system.

In Figure 10 the three-phase load 27 is connected to the three-phase voltage supply Vab, Vbc, vca. The three-phase compensator 28a, 28b, and 28c consist of the resistors Rab, Rbc, and RCa connected in series with the reverse thyristor pairs T1 and T2 and T3. A three-phase current signal proportional to the load current is obtained by the three-phase current transducer 29a, 29b, and 29c. This is applied to the three-phase vector generator 30, which produces a three-phase real current output at 31a, 31b, and 31c. The three-phase real current signal is then processed in the three-phase slope measuring and pulse gating circuit 32, giving gating pulses at outputs 33a, 33b, and 33c.The three-phase voltage supply at inputs 34a, 34b, and 34c provide power and voltage synchronization for circuit 32. The gating pulses taken from outputs 33a, 33b, and 33c are fed to the gate-cathode circuit of each of the respective thyristors of the thyristor modules T1, T2 and T3. These thyristors are fired at the proper delay angles such that the three-phase compensating resistances draw complementary currents and the three-phase supply real current components assume constant amplitude.

For less precise control of the real power flow in a three-phase system, the alternative control embodiment utilizing on-off switching of the resistances could be used, following the method as illustrated for the single-phase case. A number of compensators may be employed to obtain required compensation action, similar to the single-phase case.

The advantages of the apparatus described are many. One advantage is that voltage fluctuations and light flicker caused by real power variations are eliminated. This is accomplished without the use of rotating compensators. Another advantage is that system frequency deviations are eliminated, minimized, or controlled, since the real power flow is now controlled. Another advantage is that system transient instability caused by large and random real power variations is eliminated, since a constant power flow can be achieved.

Another advantage is that compensation can be arranged within the same half-cycle of detection. Another advantage is that the DC component is eliminated, since the resultant supply real current has constant amplitude. Another advantage is that the real current component in each of the three-phases of a three-phase system is balanced.

Claims (25)

WHAT WE CLAIM IS:

1. Apparatus for compensating the randomly varying real current and real power of a load when coupled to an AC power supply, the apparatus comprising a compensating resistance arranged to be coupled with said load to the supply, and means for controlling the power drawn by said resistance from the supply, said means including controllable switch means, and means for the control of said switch means, this means for control being arranged to measure the current through said load and to derive a control from the real component of the load current.

2. Apparatus according to claim 1 wherein said means for the control of said switch means are adapted for phase control of said switch means.

3. Apparatus according to claim 2 wherein said switch means include switching elements comprising one or more silicon controlled rectifiers or triacs.

4. Apparatus according to claim 2 or claim 3 including means for deriving control variables for the compensator from the load by steps comprising: a) obtaining a current signal proportional to the load current, b) deriving the fundamental real and reactive current components from the current signal, and c) measurement of the real current component to generate properly delayed pulse gating signals which are to be applied to the switching elements.

5. Apparatus according to claim 1 wherein said means for the control of said switch means are adapted for the on-off control of said switch means.

6. Apparatus according to claim 5 wherein said switch means includes switching elements including one or more silicon controlled rectifiers, triacs, vacuum switches or relays.

7. Apparatus according to claim 5 or claim 6 including a plurality of said compensating resistances.

8. Apparatus according to any of claims 5 to 7 including means for deriving control signals for the compensator from the load by steps comprising: a) obtaining a current signal proportional to the load current b) deriving the fundamental real and reactive current components from the current signal, and c) measurement of the real current component and comparing it with a chosen reference value to generate gating signals at a fixed delay angle of 0 , or near 0 , or suppressing the generation of gating signals entirely.

9. Apparatus according to any preceding claim wherein the or each compensating resistance is arranged to be connected by said switch means in parallel with the load across the power supply.

10. Apparatus according to any of claims 1 to 9 wherein said switch means comprises a pair of inverse parallel controllable semi-conductor switches arranged to connect the said compensating resistance controllably to the power supply.

11. Apparatus according to any of claims 1 to 10 wherein said controllable switches are thyristors.

12. Apparatus according to any preceding claim and adapted for connection to a poly-phase power supply.

13. Apparatus according to claim 12 and adapted for connection to a three-phase power supply.

14. Apparatus according to claim 12 or claim 13 wherein the load is a poly-phase load, to be supplied from the poly-phase power supply, and wherein the apparatus includes, in respect of each phase component of said load, a said compensating resistance and a said switch means for the appropriate control of the power drawn by that resistance.

15. Apparatus according to any preceding claim wherein said means for controlling are operative to control the power drawn by said resistance to be substantially complementary to the fluctuating real power drawn by the load.

16. A method for compensating the randomly varying real current and real power of a load when coupled to an AC power supply, comprising measuring the current through said load, deriving a control signal in accordance with the real component of the load current and controlling the power drawn from said power supply by a compensating resistance, also coupled to said power supply, by means of applying the control signal to controllable switch means.

17. A method according to claim 16 wherein said controllable switch means are phase controlled.

18. A method according to claim 17 in which control variables for the control of the switch means are derived from the load by steps comprising: a) obtaining a current signal proportional to the load current; b) deriving the fundamental real and reactive current components from the current signal; and c) measurement of the real current component to generate properly delayed pulse gating signals which are to be applied to the switching elements.

19. A method according to claim 16 wherein said controllable switch means are on-off controlled.

20. A method according to claim 19 in which control signals for the control of the switch means are derived from the load by the steps comprising: a) obtaining a current signal porportional to the load current; b) deriving the fundamental real and reaction current components from the current signal, and c) measurement of the real current component and comparing it with a chosen reference value to generate gating signals at a fixed delay angle of 0 , or near 0", or suppressing the generation of gating signals entirely.

21. A method according to any of claims 16 to 20 wherein the supply frequency is the commercial frequency such as 50Hz or 60Hz.

22. A method according to any of claims 16 to 21 wherein the power supply is the commercial three phase system.

23. A method according to any of claims 16 to 21 wherein the power supply is a poly-phase system other than the commercial three-phase system.

24. Apparatus for compensating the randomly varying real current and real power of a load when coupled to an AC power supply, substantially as hereinbefore described with reference to the accompanying drawings.

25. A method for compensating the randomly varying real current and real power of a load when coupled to an AC power supply substantially as hereinbefore described with reference to the accompanying drawings.