GB2524352A

GB2524352A - Reactive power compensation

Info

Publication number: GB2524352A
Application number: GB1419898.0A
Authority: GB
Inventors: Michael Stephen Hill
Original assignee: Individual
Current assignee: Individual
Priority date: 2014-11-07
Filing date: 2014-11-07
Publication date: 2015-09-23
Anticipated expiration: 2034-11-07
Also published as: GB201419898D0; GB2524352B

Abstract

A controller operates with a varying target power factor and/or compensation range. In order to determine when to adjust the reactive power compensation by switching in/out stages of a multistage capacitor bank, the controller is provided with the step size provided by each switchable stage. The controller can thus switch in or out successive stages as necessary to compensate for the reactive power of the load and maintain the power factor close to target. To minimise repeated switching, the controller may be provided with a selectable sensitivity. The product of step size and sensitivity together with variable target power factor effectively creates a variable compensation range of reactive power levels within which no correction will be added or removed. When the reactive power level is outside the compensation range for a variable time duration the controller is operable to switch in/out shunt capacitors as necessary to return the reactive power level to within the compensation range. Preferably, a different discrete target power factor or different discrete sensitivities are selected when the active power level lies within different ranges, for example three groups of values: group 1 associated with lower levels of active power; group 2 associated with medium levels of active power; and group 3 associated with higher levels of active power.

Description

REACTIVE POWER COMPENSATION

Technical Field of the Invention

The present invention relates to the automatic control of reactive power compensation and in particular to improvements in optimisation of reactive power compensation.

Background to the Invention

In AC power supply systems, it is common to apply reactive power compensation in order to adjust the flow of reactive power in the power supply system arid/or to improve the power factor of the connected load towards unity. Typically, this is achieved by the provision of shunt capacitor banks and/or shunt reactors. The capacitor banks often consist of a number of switchable stages; there may also be a number of switchable shunt reactors. Individual capacitor banks, individual stages of a multistage capacitor banks and individual reactors can be switched into or out of service to compensate for the inductive or capacitive effects of the load. This step switching is controlled by a micro processer based controller often referred to as a reactive power controller or VAr controller or power factor controller or power factor regulator or power factor relay. The controller is operable to determine the measured active power in watts (W) drawn from the system and the reactive power in volt ampere reactive (VAr) drawn from the system. The controller can then switch in or out shunt capacitors and shunt reactors as necessary to adjust the reactive power drawn from and flowing in the system and/or correct the power factor of the connected load.

Known controllers typically operate with a settable target power factor. In order to determine when to adjust the reactive power compensation by switching in/out shunt capacitors/reactors, the controller is provided with the step size (VAr) provided by each switchable capacitor bank, switchable stage of a multistage capacitor bank and/or reactor. The controller can thus switch in or out successive capacitors and/or reactors as necessary to compensate for the reactive power drawn from the system by the load and correct the power factor of the load close to target.

To minimise repeated switching, the controller may operate only if a condition indicative of a need to switch persists for more than a settable time duration. To further limit switching, the controller may be provided with a settable sensitivity. The product of step size and sensitivity together with the target power factor effectively creates a compensation range of reactive power levels within which no correction will be added or removed.

In such known controllers, whilst sensitivity and target power factor are settable, they apply equally across the full range of operation. As such, any reactive power compensation applied is typically not optimal. This can result in increased transmission losses or other undesirable characteristics.

It is therefore an object of the present invention to provide improved methods and apparatus for reactive power compensation which at least partially alleviate or overcome the above problems.

Summary of the Invention

According to a first aspect of the present invention there is provided a method of operating a controller for an AC power supply system, the method comprising the steps of monitoring the active power level; monitoring the reactive power level; determining a target reactive power level based on the active power level and a target power factor; determining a compensation range either side of the target reactive power level; determining whether the reactive power level falls within or without the compensation range; wherein if the reactive power falls without the compensation range one or more reactive power compensation means are switched into or out of the system and wherein the target power factor and/or the compensation range vary dependent upon the active power level.

According to a second aspect of the present invention there is provided a controller for an AC power supply system, the controller comprising: active power monitoring means for monitoring the active power level; reactive power monitoring means for monitoring the reactive power level; compensation means operable to determine a target reactive power level based on the active power level and a target power factor and a compensation range; control means operable to determine whether the reactive power level falls within or without the compensation range and if the reactive power falls without the compensation range operable to switch one or more reactive power compensation means into or out of the system, wherein the target power factor and/or the compensation range vary dependent upon the active power level.

By varying target power factor and/or the compensation range dependent upon the active power level, improved compensation can be applied across a wider range of power levels.

Target power factor may vary continuously with active power level.

Preferably, target power factor has a number of discrete values, each value associated with a particular range of active power levels. In one particular embodiment there may be discrete target power factor values for predefined low, medium and high power ranges. In a preferred embodiment, the active power threshold level for each range differs depending upon whether the active power level is rising or falling.

The compensation range may be symmetrical or asymmetrical about the target reactive power level. In a preferred implementation, the compensation range upper arid lower boundaries may be defined by the product of step size and upper and lower sensitivities. Where the compensation range is symmetrical about the target reactive power level, sensitivity may be equal for the upper and lower range boundaries.

Where the compensation range is asymmetric, there may be different sensitivity for the upper and lower range boundaries.

Each sensitivity may vary continuously with active power level, Preferably, IS each sensitivity has a number of discrete values, each value associated with a particular range of active power levels. In one particular embodiment there may be discrete sensitivities for predefined low, medium and high active power ranges. In a preferred embodiment, the active power threshold level for each range differs depending upon whether the active power level is rising or falling.

In order to initiate switching the one or more reactive power compensation means into or out of the system, the method may require that a reactive power level outside the compensation range persists for a compensation range threshold time duration, Compensation range threshold time duration may be preset or may be varied. In particular, compensation range threshold time duration may be varied in response to any one or more suitable criteria including but not limited to: time of day, day of week, holidays, compensation effect, active power level and user inputs.

Additionally or alternatively, there may be discrete compensation range threshold S time durations associated with the upper and lower boundaries of the compensation range.

The compensation range may incorporate one or two nudging ranges.

Nudging ranges preferably have an outer boundary coincident with the corresponding boundary of the compensation range and another, inner, boundary within the compensation range. in some implementations, there may be a nudging range associated with both the upper and lower boundaries of the compensation range. In other implementations, a nudging range may be associated with only the upper or only the lower boundary of the compensation range.

The inner boundary of the nudging range may be defined with reference to IS step size, In response to determination that the reactive power is within the nudging range, one or more reactive power compensation means may be switched into or out of the system. In order to initiate switching the one or more reactive power compensation means into or out of the system, the method may require that a reactive power level within the nudging range persists for a settable nudging range threshold time duration Nudging range threshold time duration may be preset or may be varied, In particular, the nudging range threshold time duration may be varied in response to any one or more suitable criteria including but not limited to: time of day, day of week, holidays, compensation effect, active power level, and user inputs, Additionally or alternatively, there may be discrete nudging range threshold time durations associated with the upper and lower boundaries of the compensation range aM/or the inner and outer boundaries of each nudging range.

Preferably, the inner boundary of the nudging range may be selected such that switching in or out one reactive power compensation means will not result in the reactive power level leaving the compensation range. In particular, the inner boundary may be defined by the product of step size and nudging sensitivity.

The reactive power compensation means may comprise one or more switchable capacitors and/or one or more switchable reactors.

In a power system offering multiple infeeds, each infeed may be provided with reactive power compensation means comprising one or more capacitors arid/or one or more reactors, In such cases each reactive compensation means may be provided with a dedicated controller according to the present invention.

The dedicated controllers may be operable to detect the present status of a system with multiple infeeds and automatically adjust controller values accordingly.

In particular, the dedicated controllers may be operable to automatically adjust step size, whenever a change with respect to the parallel operation of infeeds is detected.

This allows the controller to recognise and allow for the fact that the effect of adding or removing compensation when infeeds are being operated in parallel is seen by the dedicated controller that initiated switching as different to the actual physical step size, According to a third aspect of the present invention there is provided an AC power supply system operable according to the method of the first aspect of the present invention or comprising a controller according to the second aspect of the present invention.

Detailed Description of the Invention

In order that the invention may be more clearly understood embodiments thereof will now be described, by way of example only, with reference to the accompanying drawings, of which: Figure 1 is an illustrative example of the compensation range boundaries for a prior art controller with a constant target power factor and constant sensitivity over its frill range of operation Figure 2 is a further illustrative example of the compensation range boundaries for a prior art controller with a different constant target power factor over its full range of operation to those in figure 1; Figure 3 is an illustrative example of the compensation range boundaries and one associated nudging range boundaries for a controller according to I 5 the present invention wherein target power factor is varied according to the active power level; Figure 4 is another illustrative example of the compensation range boundaries and one associated nudging range boundaries for a controller according to the present invention wherein target power factor is varied according to the active power level; Figure 5 is a frirther illustrative example of the compensation range boundaries and one associated nudging range boundaries for a controller according to the present invention wherein target power factor is varied according to the active power level; Figure 6 is an illustrative example of the compensation range boundaries and one associated nudging range boundaries for a controller according to the present invention wherein the compensation range boundaries and the nudging range boundaries are varied according to the active power level i.e. compensation range sensitivities and nudging range sensitivities are varied according to the active power eve1; Figure 7 is another illustrative example of the compensation range boundaries and one associated nudging range boundaries for a controller according to the present invention wherein the compensation range boundaries and the nudging range boundaries are varied according to the active power level i.e. compensation range sensitivities and nudging range sensitivities are varied according to the active power level; Figure 8 is an illustrative example of the compensation range boundaries and one nudging range boundaries for a controller according to the present invention wherein target power factor and compensation range sensitivities and nudgillg range sensitivities are varied according to the active power level; Figure 9 is a schematic illustration of a power supply system with multiple infeeds at 11kv according to the present invention comprising multiple reactive power compensation means and a dedicated controller for each reactive power compensation means.

Turning now to figure 1, a controller operates with a settable target power factor, in this example 0.95 inductive. In order to determine when to adjust the reactive power compensation by switching in/out 2.OMVAr stages of a multistage capacitor bank, the controller is provided with the step size (VAr) provided by each switchable stage, which is 2,00M VAr in this example. Typically, there would be multiple switchable stages, for example 3 switchable stages of 2MVAr which would together amount to 6MVAr, The controller can thus switch in or out successive stages and or as necessary to compensate for the reactive power of the load and maintain the power factor close to target. To minimise repeated switching, the controller may be provided with a selectable sensitivity, in this case 0.70. The product of step size and sensitivity (+/-I.4IvIVAr) together with target power factor effectively creates a compensation range of reactive power levels within which no correction will be added or removed.

When the reactive power level is outside the compensation range for a settable time duration the controller is operable to switch in!out shunt capacitors as necessary to return the reactive power level to within the compensation range. To minimise switching due to transient effects, there may be a minimum time threshold before switching is initiated Whilst the constant target power factor and constant compensation range is simple to implement, it can lead to some rather undesirable situations being maintained, For example: (a) When the active power level is below 4,259MW there is a possibility that the known controller will maintain a prolonged leading power factor situation.

(b) When the active power level is 5.00MW there is a possibility that the known controller will maintain the reactive power at 3.O43MVAr with a resulting PF of 0,S54inductive and a resulting apparent power of 5.SS3MVA, whereas the multistage capacitor bank would be better utilised by switching in a further 2MVAr stage thus reducing the reactive power to LO43MVAr with a resulting PE of 0.979inductive and a resulting apparent power of 5. 1O8MVA.

(c) When the active power level is 0,00MW there is a possibility that the known controller will maintain the reactive power at 4.687MVAr with a resulting PF of 0.9O5inductive and a resulting apparent power of 11.OSMVA, whereas the multistage capacitor bank would be better utilised by switching in a further 2MVAr stage thus reducing the reactive power to 2.687MVAr with a resulting PE of 0,966inductive and a resulting apparent power of 10.35MYA.

(d) When the active power level is 15.00MW there is a possibility that the known controller will maintain the reactive power at 6,33MVAr with a resulting PF of IS 0,92inductive and a resulting apparent power of 16,3MVA, whereas the multistage capacitor bank would be better utilised by switching in a fbrther 2JVIVAr stage thus reducing the reactive power to 4,33MVAr with a resulting PF of 0,9óiinductive and a resulting apparent power 15.6] MVA.

The scenarios in (b), (c) and (d) above could be improved by increasing target power factor to 0.99inductive as is shown in figure 2. Nevertheless, this creates the unacceptable situation that when the active power level is below 9.825MW there is a possibility that the known controller will maintain a prolonged leading power factor situation.

In order to overcome these issues, the present invention envisages varying target power factor and/or compensation range in response to the active power level.

In particular, the controller of the present invention may be operable to select a different discrete target power factor or different discrete sensitivities when the active power level lies within different ranges. In the examples of figures 3 to 5, there are three groups of values: group 1 is associated with lower levels of active power group 2 is associated with medium levels of active power; and group 3 is associated with higher levels of active power.

In order to avoid repeated switching at threshold boundaries of active power, the threshold for switching from a higher active power range to a lower active power range is lower than the threshold for switching from a lower active power range to a higher active power range. In the present examples, respective thresholds are: (i) Threshold (inc-GltoG2) is 7.50MW. Change from group 1 values to group 2 values when the active power level increases above 7.50MW.

(ii) Threshold (inc-G2toG3) is 12.50MW. Change from group 2 values to group 3 values when the active power level increases above 12.50MW.

(iii) Threshold (dec-G3to02) is 10.00MW. Change from group 3 values to group 2 values when the active power level decreases below 10.00MW.

(iv) Threshold (dec-G2toGI) is 5,00MW, Change from group 2 values to group I values when the active power level decreases below 5.00MW.

Turning now to figure 3, the group values are as follows: Group-i values * Step size is 2.OOM VAr * Target N is 0.95inductive * Lower sensitivity is 0.70 * Upper sensitivity is 0.70 Group-2 values * Step size is 2.OOMVAr * Target PF is 0.97inductive * Lower sensitivity is 0.70 * Upper sensitivity is 0.70 Group-3 values * Step size is 2.OOM VAr * Target PF is 0.99inductive * Lower sensitivity is 0.70 * Upper sensitivity is 0.70 Particular advantages of the improved controller of present invention provided by the capability to accommodate the changing from one group of values to a different group of values as outlined above are explained further below: (a) When the active power level is below 4,25MW there is a possibility that the improved controller will maintain a prolonged leading power factor situation.

(b) When the active power level is 10.00MW there is a possibility that the improved controller will maintain the drawn reactive power at 3.9MVAr with a resulting PF of 0.932inductive and a resulting drawn apparent power of 10.73MVA.

This is a much improved situation compared with the performance of the known controller with a single group of values as described in relation to figure 1 above (i.e. 4.687M VAr with a resulting PF of O.9O5inductive and a resulting drawn apparent power of 11.O5MVA).

(c) When the active power level is 15.00MW there is a possibility that the improved controller will maintain the drawn reactive power at 3.537MVAr with a resulting PF of 0.9l3inductive and a resulting drawn apparent power of 15.41 IMVA.

This is a much improved situation compared with the performance of the known controller with a single group of values as described in relation to figure 1 above (i.e. a drawn reactive power of 6.33MVAr with a resulting if of 0.92inductive and a resulting drawn apparent power of 16.3OMVA).

In addition to the above, in some implementations of the improved controller, compensation range threshold time duration may be fixed or varied. The variation may be based on any one or more suitable criteria such as time of day, day of week, holidays, compensation effect, active power level and in response to user inputs. This means that in some implementations a different compensation range threshold time durations may be implemented at expected high and low demand periods. In particular, a shorter compensation range threshold time duration might be appropriate at lower active power levels where switching in/out of reactive power compensation means might have a significant effect. Mditionally or alternatively, there may be discrete compensation range threshold time durations associated with the upper and lower boundaries of the compensation range.

Further examples of operation of the improved controller with differing target power factors between groups are shown in figures 4 and 5. In these particular examples, target power factor in group I is 0.8oinductive in figure 4; and 0.9oinductive in figure 5. This allows the low power performance to be adjusted relative to that of the example of figure 3 without impacting on the medium and high power performance.

Turning now to figure 6, this illustrates the possibility of varying upper and lower sensitivities to thereby vary the upper and lower boundaries of the compensation range. In this example the group values are as follows: Group-i values * Step size is 2.OOMVAr * Target PF is O.99inductive * Lower sensitivity is 1.40 * Upper sensitivity is 0.00 IS Group-2 values * Step size is 2.OOMVAr * Target PF is 0.99inductive * Lower sensitivity is 1.05 * Upper sensitivity is 0.35 Group-3 values * Step Size is 2,00MVAr * Target PF is 0.99inductive * Lower sensitivity is 070 * Upper sensitivity is 0.70 Particular advantages of this arrangement are explained below: (a) When the active power level is below 450MW there is no possibility that the improved controller will maintain a pr5longed leading power factor situation. This is an improved situation compared with the performance of the known controller (see figure 1) where there is a possibility that the known controller will maintain a prolonged leading power factor situation.

(b) When the active power level is 1000MW there is a possibility that the improved controller will maintain the reactive power drawn from the supply at 3.525MVAr with a resulting PF of O.943inductive and a resulting drawn apparent power of O.6OMVA. This is a much improved situation compared with the performance of the known controller (see figure 1) i.e. a drawn reactive power of 4,687MVAr with a resulting PF of O.9O5inductive and a resulting drawn apparent power of 11,O5MVA, (c) When the active power level is 15,OOv[W there is a possibility that the improved controller will maintain the reactive power drawn from the supply at 3.537MVAr with a resulting PF of O.973inductive and a resulting drawn apparent power of 15.4 IIMVA. This is much improved situation compared with the performance of the known controller (see figure 1) i.e. a drawn reactive power of 6.33MVAr with a resulting PF of O.92inductive and a resulting drawn apparent power of 16,30MVA, Further examples of operation of the improved controller with differing target power factor values between groups are shown in figures 7 and 8. In the example of figure 7, target power factor is 100 but upper and lower sensitivities are the same as in figure 6. In the example of figure 8, there are only two groups of values with a threshold (inc-G]toG2) of 5.00MW and a threshold (dec-G2toG]) of 7.50MW. The group values are as foflows: Group-I values * Step size is 2.OOMVAr * Target PF is 1.00 * Lower sensitivity is 1.40 * Upper sensitivity is 0.00 Group-2 values * Step size is 2.OOM VAr * Target N is 0.99capacative IS * Lower sensitivity is 1.75 * Upper sensitivity is -0.30 The values of the example of figure 8 are beneficial where it is desired to maintain a leading power factor at higher power levels but not at lower power levels.

In some implementations, the improved controller may be operable to switch in/out reactive compensation when the reactive power level is within a nudging range in order to potentially reach a more favourable operating condition, The inner boundary of an example nudging range is illustrated by a doffed line on figures 3 to 8.

The outer boundary of the nudging range is coincident with the lower boundary of the compensation range. The skilled man will of course appreciate that in other implementations, the nudging range may be biased toward the upper boundary of the compensation range. In some implementations, nudging range threshold time duration may be fixed or varied. The variation may be based on any one or more suitable criteria such as time of day, day of week, holidays, compensation effect, active power level and in response to user inputs. This means that in some implementations, the nudging operation can be permitted at any time the reactive power level is in the nudging range or nudging operation may be permitted only during particular time periods, whether reactive power compensation means are being switched into or out of the system, or in response to user inputs. In particular, it allows for different nudging range threshold time duration to be implemented at expected high and low active power periods. Additionally or alternatively, there may be discrete nudging range threshold time durations associated with the inner and outer boundaries of each nudging range.

The nudging range is by way of nudging sensitivities and is defined by the product of step size and nudging sensitivities. The nudging sensitivities are selected such that a nudging operation switch inlout of capacitors/reactors will not result in a condition outside of the compensation range. Typically, the inner nudging sensitivity would be equal to (1.00 -upper sensitivity).

The specific effects of nudging operation are illustrated with reference to figure 5 where the group values are as follows Group-I values * Step size is 2.OOM VAr * Target N is 0.9oinductive * Lower sensitivity is 0.70 * Upper sensitivity is 0.70 * Inner nudging sensitivity is 0.30 * Outer nudging sensitivity is 0.70 Group-2 values * Step size is 2.OOM VAr * Target N is 0.97inductive * Lower sensitivity is 0,70 * Upper sensitivity is 0.70 * Inner nudging sensitivity is 0.30 * Outer nudging sensitivity is 0.70 Group-3 values * Step size is 2.OOMVAr * Target PF is 0.99inductive * Lower sensitivity is 0.70 * Upper sensitivity is 0.70 * Inner nudging sensitivity is 0.30 * Outer nudging sensitivity is 0.70 Particular examples of advantages of nudging operation are explained below taking reference to figure 5: (a) Without nudging and when the active power level is 5.00MW the controller will not initiate switching even though the maintained reactive power drawn from the supply is close to 3,822M VAr resulting in a PF of 0,793inductive and a drawn apparent power of 6.293'v1VA. However with nudging the controfler will switch in a further 2MVAr stage thus reducing the drawn reactive power to 1.822MVAr with a resulting PF of 0.94oinductive and a resulting drawn apparent power of 5.319MVA.

(b) Without nudging and when the active power level is 10.00MW the controller will not initiate switching even though the maintained reactive power drawn from the supply is close to 3,9O6MVAr resulting in a PF of 0,93 linductive and a drawn apparent power of 0.74MVA. However with nudging the controller will switch in a further 2MVAr stage thus reducing the drawn reactive power to L906M VAr with a resulting PF of 0.982inductive and a resulting drawn apparent power of 10.1 8OMVA.

IS (c) Without nudging and when the active power level is 13,00MW the controller will not initiate switching even though the maintained reactive power drawn from the supply is close to 3,537MVAr resulting in PF of 0.973inductive and an apparent power drawn from the supply of 15.4IIMVA. However with nudging the controller will switch in a further 2MVAr stage thus reducing the reactive power drawn from the supply to I.537MVAr with a resulting PF of 0.995inductive and a resulting drawn apparent power of 3.O79MVA, In some implementations with multiple infeeds, there are provided multiple reactive compensation means, As an example, figure 9 shows a 4 x 33/11kv transformer substation with four ÔMVAr, 11kV multistage capacitor banks, each multistage capacitor bank having three stages of 2MVAr capacitors. In such implementations where each multistage capacitor bank is associated with an individual infeed, each multistage capacitor bank may be provided with a dedicated controller that uses measurements taken from the associated infeed, The controllers may be operable together because each controller is able to determine the present operating scenario of the infeeds and hence to vary step size used in each group of values accordingly.

When all bus-section circuit breakers are open no infeeds from the transformers are being operated in parallel. Each controller will measure a compensation effect due its initiated switching equal to the physical step size of 2MYAr, In this case each controller will automatically adjust step size within all groups of values to 2MVAr, When a bus-section circuit breaker is closed so that the infeeds from IS transformer-] and transformer-2 are operated in parallel the controller-] measures the active power and the reactive power supplied by transformer-i. This is 50% of the total supplied by transformer-i and transformer-2, Similarly, the controller-2 measures the active power and the reactive power supplied by transformer-2. This is 50% of the total supplied by transformer-i and transformer-2. Accordingly, it is required for optimum control of reactive power compensation that the controller-i values and the controller-2 values are adjusted to consider a step size of MVAr, The controller-t will measure a compensation effect due to its initiated switching equal to 50% of the physical step size of 2MVAr, In this case the controller-i will automatically adjust step size within all groups of values to EVIVAr.

Likewise, the controller-2 will measure a compensation effect due to its initiated switching equal to 50% of the physical step size of 2MVAr, In this case the controller-2 will automatically adjust step size within all groups of values to IMVAr.

When a second bus-section circuit breaker is then closed so that the infeeds from transformer-I and transformer-2 and transformer-3 are then operated in parallel.

In such a case: -The controller-i measures the active power and the reactive power supplied by transformer-I This is 33,33% of the total supplied by transformer-i and transformer-2 and transformer-3; -The controller-2 measures the active power and the reactive power supplied by transformer-2. This is 33.33% of the total supplied by transformer-i and transformer-2 and transformer-3 and -The controller-3 measures the active power and the reactive power supplied by transformer-3, This is 33.33% of the total supplied by transformer-i and transformer-2 and transformer-3.

Accordingly, it is required for optimum control of reactive power compensation that the controller-] values and the controller-2 values and the controller-3 values are adjusted to consider a step size of 0,67MVAr.

The controller-i will measure a compensation effect due to its initiated switching equal to 33.33% of the physical step size of 2MVAr. In this case the controller-I will automatically adjust step size within all groups of values to O.67MVAr. Likewise, the controller-2 will measure a compensation effect due to its initiated switching equal to 33.33% of the physical step size of 2MVAr. In this case the controller-2 will automatically adjust step size within all groups of values to O,67M VAr. Likewise, the controller-3 will measure a compensation effect due to its initiated switching equal to 33.33% of the physical step size of 2MVAr. In this case the controller-3 will automatically adjust step size in all groups of values to 0.67MVAr.

It is desirable that no simultaneous switching of stages occurs, In order to avoid this outcome, the controller of each multistage capacitor bank can operate on a different compensation range threshold time duration. For example: -controller-i compensation range threshold time is 600s, -controller-2 compensation range threshold time is 6 tOs, -controller-3 compensation range threshold time is 620s, -controller-4 threshold compensation range threshold time is 630s.

Each time a controller initiates the switching of a stage the compensation effect is calculated by the controller. This is the difference between the controller's measurement of reactive power at the instant just prior to and just after the switching.

The controller then checks that the compensation effect is consistent with its step size, If not consistent the controller will automatically select the appropriate step size. The appropriate step size may be selected from limited options provided to the controller and time of setup. For example -for the case being considered step size options would be 2MVAr, 1MVAr, 0.67MVAr and 0.5MVAr. Alternatively, the controller could automatically adjust step size according to the latest compensation effect calculated from measurements of reactive power at the instant just prior to and just afier the last switching event.

The above embodiments are described by way of example only. Many variations are possible without departing from the scope of the invention as defined in the appended claims.

Claims

CLAIMSL A method of operating a controller for an AC power supply system, the method comprising the steps of: monitoring the active power level; monitoring the reactive power level; determining a target reactive power level based on the active power level and a target power factor; determining a compensation range either side of the target reactive power level; determining whether the reactive power level falls within or without the compensation range; wherein if the reactive power falls without the compensation range one or more reactive power compensation means are switched into or out of the system and wherein target power factor and/or the compensation range vary dependent upon the active power level.
2. A method as claimed in claim I wherein the target power factor varies continuously with the active power level.
3, A method as claimed in claim I wherein the target power factor has a number of discrete values, each value associated with a particular range of active power levels.
4. A method as claimed in claim 3 wherein a active power threshold level for each range differs depending upon whether the active power level is rising or falling
5. A method as claimed in any preceding claim wherein the compensation range is defined by the product of the step size of the reactive power compensation means and sensitivities.
6. A method as claimed in claim 5 wherein the sensitivities are equal for the upper and lower range boundaries.
7. A method as claimed in claim S wherein there are different sensitivities for the upper and lower range boundaries.
8. A method as claimed in any one of claims 5 to 7 wherein the sensitivities vary continuously with active power level.

9, A method as claimed in any one of claims 5 to 8 wherein the sensitivities have a number of discrete values, each value associated with a particular range of active power levels.0, A method as claimed in any preceding claim wherein to initiate switching the one or more reactive power compensation means into or out of the system, the method may require that a reactive power level outside the compensation range persists for a threshold compensation time duration.11. A method as claimed in claim 10 wherein the compensation range threshold IS time duration is preset.12. A method as claimed in claim 10 wherein the compensation range threshold time duration is varied.13. A method as claimed in claim 11 or claim 12 wherein there are discrete compensation range threshold time durations associated with the upper and lower boundaries of the compensation range.14. A method as claimed in any preceding claim wherein the compensation range incorporates one or more nudging ranges.15. A method as claimed in claim 14 wherein the or each nudging range has an outer boundary coincident with the corresponding boundary of the compensation range and another, inner, boundary within the compensation range.16. A method as claimed in claim 15 wherein the inner and outer boundary of the or each nudging range is defined with reference to the step size.17. A method as claimed in any one of claims 14 to 16 wherein to initiate switching the one or more reactive power compensation means into or out of the system, the method requires that a reactive power level within the nudging range persists for a nudging range threshold time duration.8, A method as claimed in claim 17 wherein the nudging range threshold time duration is preset.
9. A method as claimed in claim 17 wherein the nudging range threshold time duration is varied.IS 20. A method as claimed in claim 18 or claim 19 wherein there are discrete nudging range threshold time durations associated with the inner and outer boundaries of the or each nudging range.21. A method as claimed in any one of claims 15 to 20 wherein the inner boundary and outer boundary are defined by the product of step size and nudging sensitivities.22. A method as claimed in any preceding claim wherein the reactive power compensation means comprises one or more switchable capacitors and/or one or more switchable reactors and provided with a dedicated controller, wherein the method further comprises detecting the present status of a system with multiple infeeds and automatically adjusting controller values accordingly.23 A method as claimed in claim 22 wherein the method includes automatically adjusting step size, whenever a change with respect to the parallel operation of infeeds is detected.24. A controller for an AC power supply system, the controller comprising: active power monitoring means for monitoring the active power level; reactive power monitoring means for monitoring the reactive power level; compensation means operable to determine a target reactive power level based on the active power level and a target power factor and a compensation range; control means operable to determine whether the reactive power level falls within or without the compensation range and if the reactive power falls without the compensation range operable to switch one or more reactive power IS compensation means into or out of the system, wherein the target power factor and/or the compensation range vary dependent upon the active power level.25. A controller as claimed in claim 24 wherein the target power factor varies continuously with active power level.26. A controller as claimed in claim 24 wherein the target power factor has a number of discrete values, each value associated with a particular range of active power levels.27. A controller as claimed in claim 26 wherein active power threshold levels for each range differ depending upon whether the power level is rising or falling 28. A controfler as claimed in any one of claims 24 to 27 wherein the compensation range is defined by the product of step size and sensitivities.29. A controller as claimed in claim 28 wherein the sensitivities are equal for the upper and lower range boundaries.30. A controller as claimed in claim 28 wherein the sensitivities differ for the upper and lower range boundaries.31. A controller as claimed in any one of claims 27 to 30 wherein the sensitivities vary continuously with the active power level.32. A controller as claimed in any one of claims 27 to 31 wherein the sensitivities have a number of discrete values, each value associated with a particular range of active power levels.33. A controller as claimed in any one of claims 24 to 32 wherein the controller is operable to initiate switching the one or more reactive power compensation means into or out of the system, if a reactive power level outside the compensation range persists for a compensation range threshold time duration.34, A controller as claimed in claim 33 wherein the compensation range threshold time duration is preset.35. A controller as claimed in claim 33 wherein the compensation range threshold time duration is varied.36. A controller as claimed in claim 34 or claim 35 wherein there are discrete compensation range threshold time durations associated with the upper and lower boundaries of the compensation range.37. A controfler as claimed in any one of claims 24 to 36 wherein the compensation range incorporates one or more nudging ranges.38. A controfler as claimed in claim 37 wherein the or each nudging range has an outer boundary coincident with the corresponding boundary of the compensation range and another, inner, boundary within the compensation range, 39. A controfler as claimed in claim 38 wherein the inner boundary and outer boundary of the or each nudging range is defined with reference to step size.40. A controfler as claimed in any one of claims 37 to 39 wherein the controller is operable to initiate switching the one or more reactive power compensation means into or out of the system, if a reactive power level within the or each nudging range persists for a threshold nudging time duration.4L A controller as claimed in claim 40 wherein the nudging range threshold time duration is preset.42. A controller as claimed in claim 40 wherein the nudging range threshold time duration is varied.43, A controller as claimed in claim 41 or claim 42 wherein there are discrete nudging range threshold time durations associated with the inner and outer boundaries of the or each nudging range.44. A controller as claimed in any one of claims 37 to 43 wherein the inner boundary and outer boundary are defined by the product of step size and nudging sensitivities.45. An AC power supply system comprising a controller operable according to the method of any one of claims I to 23 or comprising a controller according to any one of claims 24 to 44.46. A controller as claimed in claim 45 wherein the reactive power compensation means comprises one or more capacitors and/or one or more reactors and wherein each reactive power compensation means is provided with a dedicated controller operable according to the method of any one of claims I to 23 or provided with a dedicated controller according to any one of claims 24 to 44.47, A controller as claimed in claim 46 wherein the dedicated controllers are operable to detect the present status of a system with multiple infeeds and automatically adjust controller values accordingly.48. A controller as claimed in claim 47 wherein the dedicated controllers are operable to automatically adjust step size, whenever a change with respect to the parallel operation of infeeds is detected.