GB2326780A - Fault location on switch protected power circuit - Google Patents

Description

ELECTRICAL POWER SUPPLY CIRCUITS This invention relates to electrical power supply circuits, and more particularly to a method and means for locating faults in such circuits.

Suppliers of electrical power to the public are obliged to maintain such supplies whenever practical, and the protection of such low through to high voltage circuits must be balanced between the needs of safety to the public and the need for substantially constant supply.

Such circuits commonly incorporate a plurality of protection circuits, the simplest of which are fuses, and, in the event of an electrical fault being detected by a protection circuit, a sequence of events is triggered in accordance with set algorithms to maintain the faulted section of line disconnected and to restore connection to those customers closer to the point of supply than the faulted section.

In some cases, the algorithm involves a sequence of events which attempts to determine whether the fault is intermittent ie. lasts for typically less than about 5 seconds - or permanent - ie. lasts for typically more than 5 seconds.

Such a sequence, hereinafter referred to as an autoreclose sequence, involves: a) disconnecting the circuit; b) re-energising the circuit after a predetermined delay of, typically, 5 seconds; c) if the fault has disappeared, maintaining the circuit energised; d) if the fault persists - ie. is effectively permanent rather than intermittent - disconnecting the circuit.

As an alternative to step d), the algorithm may be designed to try a further re-energising sequence which is more tolerant of faults to allow time for other equipment, such as fuses or circuit breakers, to operate and disconnect the faulted section while the remaining sections stay on supply. Such re-energising may occur typically 5 seconds later with the associated equipment in a delayed operating mode (laid-back mode).

The above detailed autoreclose sequence is primarily used on circuits with only a single source of supply, although similar arrangements can be provided on circuits with two or more sources of supply. In such cases, directional protection must be incorporated whereby the autoreclosing sequence is polarised to re-energise towards the faulted section.

In all instances, delays, typically of 5 seconds or more, are built into the autoreclosing sequence to increase the probability that intermittent faults, or permanent faults covered by other means of protective disconnection, will be cleared from the system during the autoreclosing sequence - in practice it is found that about 95% of faults are intermittent and can therefore be cleared in this manner.

Fundamental to the successful elimination of intermittent faults in power supply circuits is the incorporation of a delay in the autoreclose sequence to give the fault time to clear. It is to be appreciated that such an arrangement, which can be defined as a fault conditioning mode, is not concerned with, and does not determine, the location of the intermittent fault.

It is not unusual for so-called intermittent faults to reoccur at later dates, for example as a result of periodic events such as windy weather, wet weather or the like.

Thus, as well as being able to eliminate the faults at the time of their occurrence, it would also be desirable to be able to locate the faults so that, if desired or considered necessary, action could be taken to prevent their re-occurrence at later dates.

According to the present invention there is provided a method of locating a fault in an electrical power circuit comprising the steps of providing a plurality of electrical switches, each equipped with a sensor associated with an adjacent part of the circuit and each having a slow (laid back) setting, an intermediate (normal-fast) setting and a quick (super-fast) setting, the arrangement being such that, on development of a fault within the circuit and on the establishment of a fault current between a source of electrical supply and the fault, the switches upstream of the fault each open on the intermediate setting and move to their quick setting, those switches upstream of the fault other than the switch immediately adjacent the fault then, after a first predetermined short period, typically 50 milliseconds, during which sensing indicates upstream conditions to be normal, reclosing and, after a further predetermined short period during which sensing indicates downstream conditions to be normal, moving to their slow setting indicative of normal conditions in their associated part of the circuit, while the switch immediately upstream of the fault, after said first predetermined short period, recloses, detects the fault a second time and opens in its quick setting more rapidly than the other up-stream switches, thereby indicating that part of the circuit containing the fault.

It will thus be appreciated that the above-detailed sequence can then be applied to the circuit in the conventional manner as the switches (circuit breakers) other than the one adjacent the fault are in slow (laid back) mode, only the switch adjacent the fault which is in quick (super-fast) mode will open (very quickly).

Thus, in the method of the invention, autoreclosing sequences are used in a fault locating mode to identify the section of the circuit which is faulted prior to any further autoreclosing sequences which may be used in accordance with the above described fault conditioning mode.

The emphasis of this approach is to provide such high speed fault location through a sequence of initial reclosure (fault location mode) that most customers connected other than to the faulted section are disconnected and reconnected so quickly they are not aware of the fault, and then to apply fault conditioning only to the faulted section. The faulted section is identified in all but the briefest of intermittent faults before the fault has cleared the system.

The method of the invention has equal application to circuits with either single or multiple sources of supply without recourse to directional protection.

It is preferable to use high speed vacuum circuit breakers because high speed operation is not inhibited by the presence of high speed opening arc products as occurs in, for example, oil based breakers.

The voltage sensing and the current sensing are preferably arranged to enable digital analysis to be conducted effectively in real time.

Conveniently the analysis of the data is by digital signal processing, and the algorithms are also stored digitally.

By way of example only, the method of the invention will now be described in greater detail with reference to the accompanying drawings of which Figs. 1 to 11 show, schematically, various alternative circuits associated with the method of the invention.

Referring to the drawings, Fig. 1 shows a generic radial power supply system in which a number of customers are supplied with electricity from a single source S, the circuit incorporating a plurality of high speed vacuum circuit breakers Bn to Bn+N and Bn+N+1 onwards, together with a source circuit breaker B. The illustrated circuit includes a fault F between circuit breakers Bn+N and Bn+N+l.

When a fault develops, and on sensing the passage of an abnormal fault current, all the circuit breakers sensing this current - ie. Bn to Bn+N together with the source circuit breaker B - open at the same time on a first, normal-fast setting, and move their setting to a second, super-fast setting. The circuit breakers in question are multi-speed devices, the characteristics or responsiveness of which can be varied. The circuit breakers beyond the fault - ie. Bn+N+l onwards - remain closed as they do not detect the fault current. In this respect, and in all the drawings, X denotes a closed circuit breaker and @ denotes an open circuit breaker.

As shown in Fig. 2, and given that the source S remains alive, the source circuit breaker B recloses in its super fast mode and monitors the supplied voltage for a short period of, typically, 50 milliseconds to establish that the voltage is effectively maintained.

Given that the fault is in a section of the circuit other than that associated with the source circuit breaker B, this circuit breaker B then moves from its super fast mode to a slower, laid back mode in which it is more tolerant of the passage of fault current.

When the supply to the next circuit breaker Bn in the circuit has been re-energised for a short period, typically 50 milliseconds and no abnormal conditions are detected, then that circuit breaker Bn closes in its super fast mode and the process of checking that voltage is maintained and moves to its laid back mode in which it is more tolerant of the passage of fault current is repeated for circuit breaker Bn as it was for circuit breaker B.

This procedure occurs for each of the circuit breakers Bn to Bn+N1 as shown in Fig. 3, each of which is thus closed in sequence and moved to its laid back mode.

When the circuit breaker Bn+N controlling the faulted section recloses, it is in its super fast mode while all the other circuit breakers are in their laid back mode whereby, on passage of the fault current, the circuit breaker Bn+N opens very quickly while the remaining circuit breakers remain closed as shown in Fig. 4.

Thus the location of the fault, or at least that part of the circuit containing the fault, has been determined, and action can be taken to inform the system operator of fault performance otherwise regarded as transient.

If the circuit incorporates multiple sources, for example a generic ring system with dual sources S, S1 each having an associated source breaker B, B1 as shown in Fig. 5, further attempts to close onto the faulted section F will follow from other directions. If one of these other attempts occurs after an intermittent fault has cleared, then all circuit breakers adjacent to the faulted section will sense supplies to both sides and will reclose whereby the system will return to normal.

On the other hand, if the fault has not cleared, each circuit breaker Bn+N and Bn+N+l adjacent the faulted section will either move into lock out mode or into fault conditioning mode, and arrangements can be instituted to report back to a control centre with the status of the reconfigured network, as indicated in Fig. 6.

A development of the above concept can be applied to an open ring network, for example a dual source network with a normally open point Bn+N.2, as follows.

Following disconnection of the faulted section, supplies are lost to other downstream circuit breakers and to one side of the normally open point of the network as shown in fig. 7.

As indicated in Fig. 8 and after a relatively long period of time - typically about one minute - all of the circuit breakers which have lost supplies - ie. in the illustrated arrangement circuit breaker Bn+N+l in Fig. 8 - open and move to their superfast mode.

After a longer period of time - typically about one and a half minutes - the normally open point in the network, circuit breaker Bn+N+2, will close in super-fast mode as indicated in Fig.

9. The system is now configured to run from the configuration shown in Fig. 2 onwards regarding the supply at the normally open point Bn+N+2 as a source - see Figs 10 and 11.

Claims (5)

1. A method of locating a fault in an electrical power circuit comprising the steps of providing a plurality of electrical switches, each equipped with a sensor associated with an adjacent part of the circuit and each having a slow (laid back) setting, an intermediate (normal-fast) setting and a quick (super-fast) setting, the arrangement being such that, on development of a fault within the circuit and on the establishment of a fault current between a source of electrical supply and the fault, the switches upstream of the fault each open on the intermediate setting and move to their quick setting, those switches upstream of the fault other than the switch immediately adjacent the fault then, after a first predetermined short period during which sensing indicates upstream conditions to be normal, reclosing and, after a further predetermined short period during which sensing indicates downstream conditions to be normal, moving to their slow setting indicative of normal conditions in their associated part of the circuit, while the switch immediately upstream of the fault, after said first predetermined short period, recloses, detects the fault a second time and opens in its quick setting more rapidly than the other up-stream switches, thereby indicating that part of the circuit containing the fault.

2. A method as claimed in claim 1 in which the switches each comprise a high speed vacuum circuit breaker.

3. A method as claimed in claim 1 or claim 2 in which the voltage sensing and the current sensing are arranged to enable digital analysis to be conducted in real time.

4. A method as claimed in claim 3 in which analysis of data is by digital signal processing.

5. A method substantially as described with reference to the accompanying drawings.