Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
IMPROVEMENTS IN OR RELATING TO DIRECT CURRENT PROTECTION SCHEMES
Document Type and Number:
WIPO Patent Application WO/2016/180921
Kind Code:
A1
Abstract:
There is provided a DC protection scheme (10, 110, 210, 310, 410, 510) for protecting a DC power transmission medium (16, 116, 216) within a DC electrical power network (12). The DC protection scheme (10, 110, 210, 310, 410, 510) comprises: a protection device (14, 114, 214) coupled in use to a DC power transmission medium (16, 116, 216), the protection device being operable to protect the DC power transmission medium (16, 116, 216) from an electrical fault; a measurement apparatus to selectively measure at least one electrical property of the DC power transmission medium (16, 116, 216); and a controller (18, 118, 218) programmed to: (i) determine the direction of the electrical fault by comparing the forward and backward travelling waves derived from the or each measured electrical property; (ii) determine whether the electrical fault is an internal or external fault by comparing the first and second frequency components derived from the or each measured electrical property, wherein the frequency of the first frequency component is higher than the frequency of the second frequency component; and (iii) operate the protection device (14, 114, 214) to protect the DC power transmission medium (16, 116, 216) if the determination of the direction of the electrical fault and/or the determination of whether the electrical fault is an internal or external fault meet a predefined criterion.

Inventors:
HA HENGXU (GB)
SRI GOPALA KRISHNA MURTHI SANKARA SUBRAMANIAN (GB)
Application Number:
PCT/EP2016/060670
Publication Date:
November 17, 2016
Filing Date:
May 12, 2016
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
GENERAL ELECTRIC TECHNOLOGY GMBH (CH)
International Classes:
H02H7/26; G01R31/08
Domestic Patent References:
WO2015043644A12015-04-02
Foreign References:
US20130015878A12013-01-17
CN103986132A2014-08-13
US20140300370A12014-10-09
Attorney, Agent or Firm:
MARSHALL, Caroline (The Belgrave CentreTalbot Street, Nottingham NG1 5GG, GB)
Download PDF:
Claims:
CLAIMS

1. A DC protection scheme for protecting a DC power transmission medium within a DC electrical power network, the DC protection scheme comprising:

a protection device coupled in use to a DC power transmission medium, the protection device being operable to protect the DC power transmission medium from an electrical fault;

a measurement apparatus to selectively measure at least one electrical property of the DC power transmission medium; and

a controller programmed to:

(i) determine the direction of the electrical fault by comparing the forward and backward travelling waves derived from the or each measured electrical property;

(ii) determine whether the electrical fault is an internal or external fault by comparing the first and second frequency components derived from the or each measured electrical property, wherein the frequency of the first frequency component is higher than the frequency of the second frequency component; and

(iii) operate the protection device to protect the DC power transmission medium if the determination of the direction of the electrical fault and/or the determination of whether the electrical fault is an internal or external fault meet a predefined criterion.

2. A DC protection scheme according to Claim 1 wherein the controller is programmed to derive each travelling wave from the or each measured electrical property by combining the or each measured electrical property with the surge impedance or admittance of the DC power transmission medium.

3. A DC protection scheme according to any one of the preceding claims wherein the controller is programmed to derive each travelling wave from the or each measured electrical property by: converting the or each measured electrical property to positive and zero sequence components; and combining the positive and zero sequence components with the positive and zero sequence surge impedances or admittances of the DC power transmission medium respectively.

4. A DC protection scheme according to any one of the preceding claims wherein the controller is programmed to obtain travelling wave norm values of the forward and backward travelling waves derived from the or each measured electrical property, and the controller is programmed to perform a comparison between the obtained travelling wave norm values of the forward and backward travelling waves to determine the direction of the electrical fault.

5. A DC protection scheme according to Claim 4 wherein the comparison between the obtained travelling wave norm values of the forward and backward travelling waves is performed within a duration that is twice the time it takes for each travelling wave to propagate from one end of the DC power transmission medium to the other end of the DC power transmission medium. 6. A DC protection scheme according to Claim 4 or Claim 5 wherein the controller is programmed to modify each travelling wave to remove a pre-fault component, and the controller is programmed to obtain the travelling wave norm values from the modified forward and backward travelling waves. 7. A DC protection scheme according to any one of Claims 4 to 6 wherein the determination of the direction of the electrical fault meets the predefined criterion when:

• the ratio of the travelling wave norm value of the forward travelling wave to the travelling wave norm value of the backward travelling wave exceeds a predefined threshold that indicates the presence of a forward directional fault; and/or

· either of the travelling wave norm values exceeds a predefined threshold that indicates the presence of a fault.

8. A DC protection scheme according to any one of the preceding claims wherein the forward and backward travelling waves are travelling voltage waves or travelling current waves.

9. A DC protection scheme according to any one of the preceding claims wherein the controller is programmed to obtain frequency component norm values of the first and second frequency components derived from the or each measured electrical property, and the controller is programmed to perform a comparison between the obtained frequency component norm values to determine whether the electrical fault is an internal or external fault.

10. A DC protection scheme according to Claim 9 wherein the determination of whether the electrical fault is an internal or external fault meets the predefined criterion when the ratio of the frequency component norm value of the first frequency component to the frequency component norm value of the second frequency component exceeds a predefined threshold that indicates the presence of an internal fault.

11. A DC protection scheme according to any one of the preceding claims wherein each frequency component is:

• a frequency component of the or each measured electrical property;

• a frequency component of a sequence component converted from the or each measured property;

• a frequency component of a forward or backward travelling wave derived from the or each measured electrical property; or

• a frequency component of a positive or zero sequence forward or backward travelling wave derived from the or each measured electrical property.

12. A DC protection scheme according to any one of Claims 4 to 7, 9 and 10 wherein each norm value is a root mean square value or an absolute mean value.

13. A DC protection scheme according to any one of the preceding claims wherein the determination of the direction of the electrical fault and the determination of whether the electrical fault is an internal or external fault meet the predefined criterion when the · electrical fault is identified as a forward directional fault and/or an internal fault.

14. A DC protection scheme according to any one of the preceding claims wherein the controller is additionally programmed to operate the protection device to protect the DC power transmission medium upon receipt of an external signal indicating the presence of an internal fault.

15. A DC electrical power network comprising a plurality of DC protection schemes according to any one of the preceding claims.

Description:
IMPROVEMENTS IN OR RELATING TO DIRECT

CURRENT PROTECTION SCHEMES

This invention relates to a direct current (DC) protection scheme and a DC electrical power network including a plurality of such schemes.

DC electrical power networks, such as DC grids, typically include a plurality of terminals each of which may be operatively associated with a power converter to interconnect the DC power transmission network with a respective alternating current (AC) network.

Respective pairs of terminals are interconnected by a section of DC power transmission medium, such as a transmission line or cable. Protection devices, such as circuit breakers and relays, are operated in the event of a fault to protect the DC electrical power network.

According to a first aspect of the invention, there is provided a DC protection scheme for protecting a DC power transmission medium within a DC electrical power network, the DC protection scheme comprising:

a protection device coupled in use to a DC power transmission medium, the protection device being operable to protect the DC power transmission medium from an electrical fault;

a measurement apparatus to selectively measure at least one electrical property of the DC power transmission medium; and

a controller programmed to:

(i) determine the direction of the electrical fault by comparing the forward and backward travelling waves derived from the or each measured electrical property, e.g. voltage or current;

(ii) determine whether the electrical fault is an internal or external fault by comparing the first and second frequency components derived from the or each measured electrical property, wherein the frequency of the first frequency component is higher than the frequency of the second frequency component; and

(iii) operate the protection device to protect the DC power transmission medium if the determination of the direction of the electrical fault and/or the determination of whether the electrical fault is an internal or external fault meet a predefined criterion. The DC protection scheme of the invention permits the detection of a fault and subsequent operation of a protection device, i.e. by way of a comparison between suitable values to identify the direction and/or type of the electrical fault, in a manner which requires only a measured electrical property of a DC power transmission medium to be acquired.

Such measurement(s) can be obtained locally, i.e. immediately adjacent to the controller and protection device associated with the DC power transmission medium, and so the invention is able to protect the DC power transmission medium without the need to establish a communication link with one or more remote elements that would otherwise be required to provide an indication of a fault occurring a long distance away from the said controller and protection device.

The aforementioned functionality is especially desirable when the considerable length of a DC power transmission medium, e.g. several hundred kilometres, means that communication with one or more remote elements is unreliable, or the resulting delay in receiving information once communication is established is such that an associated protection device cannot be operated quickly enough to provide meaningful protection. Moreover, the reliance solely on local voltage and current measurements permits the protection scheme of the invention to operate the protection device extremely quickly, e.g. within 0.5 millisecond of a fault occurring.

The provision of the above controller in the DC protection scheme of the invention therefore results in a non-unit DC protection scheme with absolute selectivity and rapid response times.

Preferably the controller is programmed to derive each travelling wave from the or each measured electrical property by combining the or each measured electrical property with the surge impedance or admittance of the DC power transmission medium.

The above features of the controller permits derivation of the travelling waves without the need for communication with remote elements, which could result in an undesirable delay in operating the protection device.

More preferably the controller may be programmed to derive each travelling wave from the or each measured electrical property by: converting the or each measured electrical property to positive and zero sequence components; and combining the positive and zero sequence components with the positive and zero sequence surge impedances or admittances of the DC power transmission medium respectively. This enables decoupling of the mutual inductance and admittance of two conductors of a bi-polar DC power transmission medium.

The direction of the electrical fault, i.e. forward or reverse, may be determined as follows.

In embodiments of the invention, the controller may be programmed to obtain travelling wave norm values of the forward and backward travelling waves derived from the or each measured electrical property. The controller may be programmed to perform a comparison between the obtained travelling wave norm values of the forward and backward travelling waves to determine the direction of the electrical fault. The comparison between the obtained travelling wave norm values of the forward and backward travelling waves may be performed within a duration that is twice the time it takes for each travelling wave to propagate from one end of the DC power transmission medium to the other end of the DC power transmission medium.

The foregoing configuration of the controller provides a reliable means for using local measurements to identify the direction of the electrical fault.

In further embodiments of the invention, the controller may be programmed to modify each travelling wave to remove a pre-fault component. The controller may be programmed to obtain the travelling wave norm values from the modified forward and backward travelling waves. This improves the accuracy of the comparison between the obtained travelling wave norm values to identify the direction of the electrical fault.

The determination of the direction of the electrical fault may meet the predefined criterion when:

· the ratio of the travelling wave norm value of the forward travelling wave to the travelling wave norm value of the backward travelling wave exceeds a predefined threshold that indicates the presence of a forward directional fault; and/or

• either of the travelling wave norm values exceeds a predefined threshold that indicates the presence of a fault.

The controller programmed in this manner advantageously improves the ability of the DC protection scheme of the invention to carry out its protective function. The forward and backward travelling waves may be travelling voltage waves or travelling current waves. In embodiments of the invention, the controller may be programmed to obtain frequency component norm values of the first and second frequency components derived from the or each measured electrical property, and the controller may be programmed to perform a comparison between the obtained frequency component norm values to determine whether the electrical fault is an internal or external fault.

The foregoing configuration of the controller provides a reliable means for using local measurements to identify the type of electrical fault.

In further embodiments of the invention, the determination of whether the electrical fault is an internal or external fault may meet the predefined criterion when the ratio of the frequency component norm value of the first frequency component to the frequency component norm value of the second frequency component exceeds a predefined threshold that indicates the presence of an internal fault. The controller programmed in this manner advantageously improves the ability of the DC protection scheme of the invention to carry out its protective function.

Each frequency component may be, but is not limited to:

• a frequency component of the or each measured electrical property;

· a frequency component of a sequence component converted from the or each measured property;

• a frequency component of a forward or backward travelling wave derived from the or each measured electrical property; or

• a frequency component of a positive or zero sequence forward or backward travelling wave derived from the or each measured electrical property.

Each norm value may be a root mean square value, an absolute mean value or another order norm value. In a preferred embodiment of the invention, the determination of the direction of the electrical fault and the determination of whether the electrical fault is an internal or external fault meet the predefined criterion when the electrical fault is identified as a forward directional fault and/or an internal fault. Programming the controller in this manner beneficially enhances the absolute selectivity of the DC protection scheme of the invention. In still further embodiments of the invention, the controller may be additionally programmed to operate the protection device to protect the DC power transmission medium upon receipt of an external signal indicating the presence of an internal fault.

Such a configuration advantageously extends the functionality of the DC protection scheme of the invention in circumstances where modest communication between, e.g. remote terminals in a DC electrical power network, is possible.

According to a second aspect of the invention, there is provided a DC electrical power network comprising a plurality of DC protection schemes as described hereinabove.

Such a DC electrical power network shares the benefits associated with the DC protection schemes included therein.

A preferred embodiment of the invention will now be described, by way of a non-limiting example, with reference to the accompanying drawings in which:

Figure 1 shows a schematic view of a DC electrical power network including a plurality of DC protection schemes according to respective embodiments of the invention;

Figure 2 shows a schematic representation of a controller forming a part of each of the DC protection schemes shown in Figure 1 ;

Figure 3 shows a schematic representation of a fault direction detection sub-unit 26 that forms a part of the controller shown in Figure 2; and

Figure 4 shows a schematic representation of an internal fault detection sub-unit that forms a part of the controller shown in Figure 2.

A DC protection scheme according to a first embodiment of the invention is designated generally by reference numeral 10 and forms part of a DC electrical power network 12 in the form of a DC grid, as shown schematically in Figure 1. The first DC protection scheme 10 includes a first protection device 14 which, in the embodiment shown is a circuit breaker, although other types of protection devices such as a relay are also possible. The first protection device 14, i.e. circuit breaker, is electrically coupled in use to a DC power transmission medium 16, i.e. arranged in-line with the DC power transmission medium 16, and is operable to protect the DC power transmission medium 16 from an electrical fault, i.e. tripping the circuit breaker to clear the electrical fault inside the DC power transmission medium 6.

The DC protection scheme 10 also includes a first measurement apparatus (not shown) which is able to selectively measure the voltages and currents of the DC power transmission medium 16. In addition, the DC protection scheme 10 includes a first controller 18, as shown in Figure 2, that is programmed to:

(i) determine the direction of an electrical fault by comparing forward and backward travelling wavesderived from the measured voltages and currents;

(ii) determine whether the electrical fault is an internal or external fault by comparing first and second frequency components derived from the measured voltages and currents wherein the frequency of the first frequency component is higher than the frequency of the second frequency component; and

(iii) operate the first protection device 14 to protect the DC power transmission medium 16 if the determination of the direction of the electrical fault and the determination of whether the electrical fault is an internal or external fault meet a predefined criterion.

In the embodiment shown, the first controller 18 is programmed such that the determination of the direction of the electrical fault and the determination of whether the electrical fault is an internal or external fault meet a predefined criterion when the electrical fault is identified as a forward directional fault and an internal fault.

Each of the first protection device 14, the first measurement apparatus and the first controller 18 is operatively associated with a first terminal 20 of the DC electrical power network 12.

In the embodiment shown, the DC power transmission medium 16 is defined by a first transmission line, and more particularly by a first bi-polar transmission line, i.e. a transmission line incorporating two conductors (not shown) operating at different polarities. In other embodiments (not shown) the DC power transmission medium 16 may instead be defined by a first transmission cable which may or may not be bi-polar and could be mono-polar, i.e. could incorporate a single conductor operating at a single polarity. The DC power transmission medium 16 could also be defined by a mono-polar transmission line. When the DC power transmission medium 16 is defined by a monopolar transmission cable or line, the first measurement apparatus is configured to selectively measure the voltage and current of the DC power transmission medium 16.

Since the DC power transmission medium 16 is defined by a bi-polar transmission line, measured samples of positive voltage u mp and positive current i mp of the positive polar conductor and measured samples of negative voltage u mN and negative current i mN of the negative polar conductor can be obtained. Moreover, the sampling period Ts may, for example, be given by:

Ts = 1/96000 seconds

To decouple the mutual inductance and admittance of the two conductors of the DC power transmission medium 16, the measured voltages u mp , u mN are then converted to positive and zero sequence voltage components u m1 , u m0 , while the measured currents i mp , i mN are converted to positive and zero sequence current components i m1 , i m0 , as follows:

As indicated by reference numeral 22 in Figure 2, the first controller 18 is programmed to derive forward and backward travelling voltage and current waves f m1 , f m0 , b m1 , b m0 by combining the positive and zero sequence components u m1 , u m0 , i m1 , i m0 of the measured voltages and currents u mp , u mN , i mp , i mN with the positive and zero sequence surge impedances z c1 , z c0 and admittances y c1 , y c1 of the DC power transmission medium 16 respectively, as follows:

(positive sequence travelling voltage waves)

(zero sequence travelling voltage waves)

(positive sequence travelling current waves) (zero sequence travelling current waves) where

f m1 is the positive sequence forward traveling voltage or current wave;

b m1 is the positive sequence backward traveling voltage or current wave;

f m0 is the zero sequence forward traveling voltage or current wave;

b m0 is the zero sequence backward traveling voltage or current wave;

is the positive-sequence surge impedance in time

domain;

is the zero-sequence surge impedance in time

domain;

is the positive-sequence surge admittance in time

domain; is the zero-sequence surge admittance in time domain,

in which

R 1 and R 1 are respectively the positive-sequence and zero-sequence resistances per length of DC power transmission medium 16;

L 1 and L 0 are respectively the positive-sequence and zero-sequence inductances per length of DC power transmission medium 16;

C 1 and C 0 are respectively the positive-sequence and zero-sequence capacitances per length of DC power transmission medium 16.

It will be understood that the symbol * in the above equations is intended to mean a convolution operation.

In other embodiments of the invention in which the DC power transmission medium 16 is mono-polar instead of bi-polar, it would not be necessary to use the positive and zero sequence components u m1 , u m0 , i m1 , i m0 of the measured voltages and currents u mp , u mN , i mp , i mN to derive the travelling voltage and current waves f m1 , f m0 , b m1 , b m0 . In such embodiments the travelling voltage and current waves f m1 , f m0 , b m1 , b m0 can instead be derived by combining the measured voltage and currents u mp , u mN , i mp , i mN with the surge impedances z c1 , z c0 and admittances y c1 , y c1 of the DC power transmission medium 16 respectively. It will be understood that it is not essential to derive both travelling voltage and current waves f m1 , f m0 , b m1 , b m0 , and the controller may instead be programmed to derive either the travelling voltage waves or the travelling current waves. A fault direction detection unit 24 of the first controller 18 includes two fault direction detection sub-units 26, each of which is configured to identify the direction of the electrical fault. Optionally the fault direction detection sub-units 26 may be replaced by a single fault direction detection sub-unit 26, particularly if the DC power transmission medium 16 is mono-polar instead of bi-polar.

A first of the fault direction detection sub-units 26 receives the positive sequence forward and backward travelling waves f m1 , b m1 , while a second of the fault direction detection sub-units 26 receives the zero sequence forward and backward travelling waves f m0 , b m0 . As shown by reference numeral 28 in Figure 3, each fault direction detection sub-unit 26 modifies each travelling wave (indicated as f m and b m for brevity) to remove a pre-fault component from the travelling wave f m , b m by, for example, applying the high pass filter or delta technique or wavelet transform or wavelet filter banks or differential operation on the forward and backward travelling waves.

As an example, the delta technique is employed here for removing the pre-fault component from each travelling wave f m , b m :

where

x is the input signal which is the travelling wave ;

y is output signal which is the modified travelling wave

W being the time window length, e.g. if the time window is 0.5 millisecond and the sampling period Ts is, as set out above, 1/96000, then the time window length is given by

As shown by reference numeral 30 in Figure 3, each fault direction detection sub-unit 26 then obtains travelling wave norm values m from the modified travelling waves Δf m , Δb m . Each travelling wave norm value ΔF m , ΔB m be a root mean square (RMS) value, an absolute mean value or another order norm value.

As an example, the RMS value ΔF m , ΔΒ™ can be obtained from a given modified travelling wave Δf m , Δb m , as follows:

where x is the input signal which is the given modified travelling wave Δf m , Δb m ;

y is the output signal which is the obtained RMS value ΔF m , ΔB m ;

W is again the time window length.

Subsequently each fault direction detection sub-unit 26 performs a comparison between the obtained travelling wave norm values ΔF m , ΔB m to determine whether two conditions are satisfied.

In the first condition, if the ratio R fb of the travelling wave norm value ΔFm of the forward travelling wave f m to the travelling wave norm value ΔBm of the backward travelling wave bm exceeds a predefined threshold Rsetl , this indicates the presence of a forward directional fault. Rsetl has a range [0,1] and may be set at, for example, 0.1.

In the second condition, if either of the obtained travelling wave norm values ΔF m , ΔB m exceeds a predefined threshold (which is Uset for the travelling voltage wave norm values and Iset for the travelling current wave norm values), this indicates the presence of a fault. Uset may be set at, for example, 0.1 per sub-unit, and Iset may be set at, for example, 0.05 per sub-unit.

The comparison between the obtained travelling wave norm values ΔF m , ΔB m is performed within a duration that is twice the time T p it takes for each travelling wave f m , bm to propagate from one end of the DC power transmission medium 16 to the other end of the DC power transmission medium 16. The duration T p is calculated by multiplying the length of the DC power transmission medium 16 by the velocity of a given travelling wave fm , bm. This comparison is reset after the duration T p has lapsed. If both the first and second conditions are satisfied, the corresponding fault direction detection sub-unit 26 provides a first output indication DDB_FWD that indicates the presence of a forward directional fault. Each fault direction detection sub-unit 26 has a reset delay of Tdsetl that is normally set at 1 millisecond. If at least one of the first and second conditions are not satisfied, the corresponding fault direction detection sub-unit 26 identifies the electrical fault as a reverse directional fault.

The fault direction detection unit 24 identifies the electrical fault as a forward directional fault if either or both of the fault direction detection sub-units 26 provides the first output indication DDB_FWD, and thereby provides the first output indication DDB_FWD to a trip logic block 32 of the first controller 18.

In this manner the first controller 18 is programmed to obtain travelling wave norm values

ΔFm, ΔBm of forward and backward travelling waves f m , b m derived from the measured voltages and currents u mp , u mN , i mp , i mN , and then perform a comparison between the obtained travelling wave norm values ΔF m , ΔB m to identify the direction of the electrical fault.

An internal fault detection unit 34 of the first controller 16 includes two internal fault detection sub-units 36, each of which is configured to identify the type of electrical fault. Optionally the internal fault detection sub-units 36 may be replaced by a single internal fault detection sub-unit 36, particularly if the DC power transmission medium 16 is monopolar instead of bi-polar. As shown in Figure 4, each internal fault detection sub-unit 36 obtains frequency component norm values Δb mH , ΔB mL of first and second frequency components, i.e. high and low frequency components Δb mH , Δb mL , derived from the measured voltages and currents UP, U N , ip, I N - More specifically, the high and low frequency components Δb mH , ΔbmL may be frequency components of an input signal in the form of:

• a given measured voltage or current UP, U N , ip, I N ;

• a positive or zero sequence component u m1 , u m0 , i m1 , i m0 of a given measured voltage or current UP, U N , ip, I N ;

· a given forward or backward travelling wave f m , b m ; or

• a given positive and zero sequence forward or backward travelling wave f m1 , f m0 , bm1 , bmO- More particularly, a first of the internal fault detection sub-units 36 is configured to receive an input signal in the form of a positive sequence forward or backward travelling wave f m1 , m1 , the measured voltage up or current ip of the positive polar conductor or the positive sequence component u m1 , i m1 of the measured voltage UP or current ip, and a second of the internal fault detection sub-units 36 is configured to receive an input signal in the form of a zero sequence forward or backward travelling wave, the measured voltage U N or current I N of the negative polar conductor or the zero sequence component U m0 , i m0 of the measured voltage U N or current I N .

For each internal fault detection sub-unit 36, the high and low frequency components

Δb mH , Δb mH are obtained by passing the input signal through high and low frequency band pass filters 38,40 respectively. Preferably the backward travelling wave b m is selected as the input signal, with the reference forward direction being from the DC bus to the DC power transmission medium 16.

In the embodiment shown in Figure 4, the high frequency band pass filter 38 has a frequency band of 2-24 kHz, and the low frequency band pass filter 40 has a frequency band of 1.5-3 kHz. It will be appreciated that these frequency band values are merely selected to illustrate the working of the invention, and other frequency band values may be used as long as the frequency band of the high frequency band pass filter 38 is higher than the frequency band of the low frequency band pass filter 40. Each band pass filter 38,40 can either be implemented by wavelet transform, as a finite impulse response (FIR) band pass filter banks, or as an infinite impulse response (MR) band pass filter.

As an example, each band pass filter 38,40 can be designed as a zero-pole-based band pass filter in which the expression of frequency response is shown in the following equation:

where is the pole of lower frequency

;

σΐ = 500 is the decaying factor for f1 ; The algorithm for the two band pass filters 38,40 are shown as:

As indicated by reference numeral 43 in Figure 4, each internal fault detection sub-unit 36 then obtains frequency component norm values from the high and low frequency components Again, each frequency component norm value Δb mH ,

ΔBmL may be a root mean square (RMS) value, an absolute mean value or another order norm value.

As an example, the RMS value Δb mH , ΔBmL can be obtained from a given frequency component, as follows:

where x is the input signal which is the given frequency component Δb mH , Δb mL ,

y is the output signal which is the obtained RMS value Δb mH , ΔB mL

W is again the time window length.

Subsequently each internal fault detection sub-unit 36 performs a comparison between the obtained frequency component norm values Δb mH , ΔB mL to determine whether the ratio of the frequency component norm value Δb mH of the high frequency component

Δb mH to the frequency component norm value ΔB mL . of the low frequency component

ΔbmL exceeds a predefined threshold Rset2, i.e. whether the following discriminative condition is satisfied:

If the ratio of the frequency component norm value ΔB MH of the high frequency component bto the frequency component norm value ΔE the low frequency component

ΔbmL exceeds the predefined threshold Rset2, this indicates the presence of an internal fault. Rset2 has a range [0,1] and may be set at, for example, 0.2.

If the above discriminative condition is satisfied, the corresponding internal fault detection sub-unit 36 provides a second output indication DDBJNTNL that indicates the presence of an internal fault. Each internal fault detection sub-unit 36 has a reset delay of Tdset2 that is normally set at 1 millisecond. If the above discriminative condition is not satisfied, the corresponding internal fault detection sub-unit 36 identifies the electrical fault as an external fault.

The internal fault detection unit 34 identifies the electrical fault as an internal fault if either or both of the internal fault detection sub-units 36 provides the second output indication DDBJNTNL, and thereby provides the second output indication DDBJNTNL to the trip logic block 32 of the first controller 18.

In this manner the first controller 18 is programmed to obtain frequency component norm values Δb mH , Δb ML of high and low frequency components Δb MH , Δb ML derived from the measured voltages and currents U P , U N , i P , I N , and then perform a comparison between the obtained frequency component norm values Δb mH , Δb ML to identify the type of electrical fault.

As mentioned earlier, the first controller 18 is programmed such that the determination of the direction of the electrical fault and the determination of whether the electrical fault is an internal or external fault meet the predefined criterion when the electrical fault is identified as a forward directional fault and an internal fault. Thus, when the trip logic block 32 receives both first and second output indications DDB_FWD, DDBJNTNL, the first controller 18 operates the first protective device 14 to protect the DC power transmission medium 16 from the electrical fault.

In addition, the first controller 18 will operate the first protection device 14 upon receipt of an external signal 42 indicating the presence of an internal fault, e.g. from a second controller 118 associated with the second terminal 44 at the other end of the DC power transmission medium 16 and configured to operate a second protection device 1 14 in the form of a further circuit breaker. In this manner the second controller 1 18 and second protection device 1 14, along with a second measurement apparatus (not shown), together define a second DC protection scheme 110 which provides overlapping, backup protection for the DC power transmission medium 16.

The DC electrical power network shown in Figure 1 also additionally includes another DC power transmission medium 116 which extends between the second terminal 44 and the third terminal 46.

Each terminal 20, 44, 46 electrically interconnects a corresponding first, second or third power converter 48, 50, 52 and associated first, second or third AC network 54, 56, 58 via the aforementioned DC electrical network 12, i.e. the aforementioned DC grid.

The second terminal 44 also has a third DC protection scheme 210 operatively associated therewith, with the third DC protection scheme 210 including a third controller 218, a third protection device 214 (in the form of a still further circuit breaker), and a third measurement apparatus (not shown).

In a similar manner, the third terminal 46 has a fourth DC protection scheme 310 operatively associated therewith, with the fourth DC protection scheme 310 again similarly including a fourth controller 318, a fourth protection device 314 (in the form of a yet further circuit breaker), and a fourth measurement apparatus (not shown).

The third and fourth DC protection schemes 210, 310 provide overlapping protection for the other DC power transmission medium 116. The DC electrical power network 12 shown in Figure 1 also further includes a further DC power transmission medium 216 which extends between the third terminal 46 and the first terminal 20.

The third terminal 46 additionally has a fifth DC protection scheme 410 operatively associated therewith (which includes a commensurate fifth controller 418, fifth protection device 414 (again a circuit breaker), and fifth measurement apparatus (not shown)), while the first terminal 20 has a sixth DC protection scheme 510 operatively associated therewith (similarly with a commensurate sixth controller 518, sixth protection device 514 (again a circuit breaker), and sixth measurement apparatus (again not shown)).

The fifth and sixth DC protection schemes 410, 510 provide overlapping protection for the further DC power transmission medium 216. In the embodiment of the DC electrical power network 12, i.e. DC grid 14, shown in Figure 1 , each of the second, third, fourth, fifth and sixth DC protection schemes 110, 210, 310, 410, 510 is identical to the first DC protection scheme 10 described hereinabove.