DD253900A1 - METHOD FOR ERROR-TOLERANT POWER SUPPLY AND FAULT-TOLERANT POWER SUPPLY SYSTEM, ESPECIALLY FOR AUTOMATION PLANTS WITH DISTRIBUTED FUNCTIONAL UNITS - Google Patents

Abstract

Method for fault-tolerant power supply and fault-tolerant power supply system, in particular for automation systems with distributed functional units. The invention is everywhere applicable where a high level of reliability of the supply of electrical energy to be achieved. In a designed as a double or multiple ring power supply system with a feed point in each network node in all connected connection lines, the current is measured in magnitude and direction, the measurement results are fed to an existing of channel logic circuits and a Verknuepfungsschaltung evaluation logic, which in the event of a short or ground fault detects a short-circuited connection line or the short-circuited network node and switches off via disconnector. Fig. 2

Description

"" η

A ₁ j 1 (R ₁₁ φ E _lk ) VU _lk J Λ B ^R li * ^R rk ^} VU _rk ]

V ^

IHTKU

where F denotes binary drive signals for the assigned disconnect switches (TE), η denotes the number of rings of the power supply system, i and k run variables, r and I indices for distinguishing the reference directions from the two adjacent network nodes, and INTK an internal signal.

5. fault-tolerant power supply system, in particular for automation systems with distributed functional units according to claim 2 to 4, characterized in that the evaluation logic (A) the inputs of the digital logic circuit (V) upstream of the binary signals 0 channel logic circuits (KL) that each channel logic circuit ( KL) consists of two AND gates (D1, D2), of which the first is connected directly and the second via a delay circuit (TV) to the input of the channel logic circuit (KL), wherein the outputs of the AND gate (D 1, D2) are linked via an OR gate (D4), that each channel logic circuit (KL) contains a further AND gate (D3) whose η - 1 inputs exactly at the outputs of the logic circuit (V) the same reference direction (r or I ) are connected, which drive the circuit breaker (TE) η Φϊ or η Φ k and that the output of the third AND gate (D3) to the second input of the second (D2) and negier t is connected to the second input of the first AND gate (D 1).

For this 3 pages drawings

Field of application of the invention

The invention relates to the fault-tolerant power supply, in particular for automation systems with distributed functional units. Further applications are everywhere where as low as possible a high level of reliability of the supply of any consumer with electrical energy to be achieved, such as in the office, manufacturing, marine automation, laboratory measurement and lighting technology.

Characteristic of the known state of the art

Are known annular power grids. The ring structure is the simplest redundant structure. It makes it possible to tolerate a line break at any point in the ring system, without additional effort in the form of switches or measuring devices. Such a simple and relatively reliable power supply is shown in DE-OS 2905757. A disadvantage of this arrangement is that when a short circuit occurs, the supply for the entire power supply system fails. Therefore, methods have been developed which make it possible to disconnect short-circuited power supplies in ring-shaped arrangements. Such a method and operating according to this method arrangement is described in DE-OS 2531707. In this case, the short-circuited part of the power supply network is determined by means of a made in each network node measurement of the time from the occurrence of the short-circuit current to shutdown by the distance relay and turned off by the adjacent network nodes. Only then do the distance relays close the circuit again. This solution has the disadvantage that, in the event of a short circuit, there is a larger dead time due to the time measurement, which can lead to undesirable disturbances, especially in automation systems, to data losses and also in other power supply systems with high reliability requirements. Furthermore, the method requires relatively large distances between the distance relays installed outside the network nodes and between the individual network nodes themselves. In particular, in automation systems with distributed functional units such distances are often not given.

Object of the invention

The aim of the invention is to provide a mutually independent of the distance of the connected consumers working method and operating according to this method power supply system, which improved fault tolerance and higher reliability of the power supply can be achieved and which due to these properties, especially for automation systems with distributed functional units suitable is.

-3- Z53 9UU

Explanation of the essence of the invention

The technical problem to be solved by the invention is to realize a method for fault-tolerant power supply in power supply systems designed as a double or multiple ring with a feed-in point and a fault-tolerant power supply system operating according to this method, in particular for automation systems with distributed functional units, thus without short interruptions of the ring network in the Connection to an occurring short circuit, the detection and shutdown of the short-circuited line or the short-circuited network node is possible.

According to the underlying object of the method is achieved in that in each network node, the current is measured in each connected to the network node connection line in magnitude and direction that by comparing the current directions with each other and by comparing the current amounts each with a fixed upper and with a established lower limit that in the event of short circuit for exceeding the specified current limit causal, connected to the network node connection line or the network node itself identified as the cause, and that the thus determined connection line from the network node or the network node thus determined is disconnected from all interconnections. The essence of the method according to the invention is explained in more detail below:

A short-circuit or ground fault in a connecting line results in exceeding the fixed upper limit of the current amounts measured in the two connected network nodes, in each case in the relevant connection point. The setting of the short-circuit or ground fault characteristic upper current limits depends on the internal resistance of the feeding source, the resistance of the connecting lines, the load from the connected consumers and the topology of the system.

Depending on the location of the short-circuit or ground fault within the system, exceedances of the upper current limit values occur in further network nodes and non-short-circuited connecting lines, so that this criterion alone is insufficient to unambiguously determine the short-circuited line from the point of view of the individual network node. The invention therefore makes use of the knowledge that differs in one of the two connected by a short-circuit connection line network node and only there the current direction in the short-circuited connection line from the current directions of all other connected to this network nodes connecting lines, so that by comparing the current directions among themselves as a second criterion from the point of view of this network node, the short-circuited connection line can be clearly determined and separated from the network node. As a result of unilateral separation of the short-circuited connection line, the current amounts and current directions in the power supply system change so that even for the second network node to which the short-circuited connection line is still connected, the two criteria are met in the specified manner, which also by this network node the short-circuit Connection line can be clearly identified and separated. Connecting lines, which lead due to line interruption, separation due to short-circuit or ground fault or Potentialgleicheit the connected network nodes no electricity are either not included in the current direction comparison or the special comparison with such a connection line is to be evaluated in the respective network node in each case as a current direction difference. In order to determine such lines from the point of view of the two adjacent adjacent network nodes, a comparison of the current amounts with a specified lower current limit value is required in each case.

For the network node over which the power is supplied, in addition, that a connection line is then switched off because of short or earth fault, if only in her an exceeding of the upper limit of the current is detected and at least one parallel connection line is still intact. In addition, the invention also allows an internal short or ground fault to be detected in a network node and to respond accordingly. An internal short-circuit or ground fault in a network node has the effect that the Stromflußrichtung is the same in all intact connection lines from the point of view of the network node and in at least one connecting line, the exceeding of the upper limit of the current is detected.

The fault-tolerant power supply system according to the invention is characterized in that a controllable disconnect switch and a current measuring device for current amount and current direction is arranged on all power supply inputs of each network node and that each network node has a Auswertelogik which is connected to all disconnectors and current measuring devices of the network node, wherein the evaluation logic consists of a circuit that realizes the control of the circuit breaker exclusively as a function of the measured current amounts and directions. It is appropriate, the. Processing of the results consistently binary provide. For this purpose, each current measuring device has three binary outputs for the binary signals 0 for the exceeding of a fixed upper, U for falling below a fixed lower current limit value and for the current direction signal R. The evaluation logic present in each network node contains a digital logic circuit which uses the Boolean equations

- ⁰ PiAA {[ ^(R ri

• k = 1

^u ik] Λ [CR _ri 4 ^ V

k = 1 .k + i

with INTK

( _Rk V p _rk )]

[ ⁰ Ik ^V ° rk]

where F denotes binary drive signals for the assigned disconnect switches, η denotes the number of rings of the power supply system, i and k run variables, r and I indices for distinguishing the reference directions to the respective two adjacent network nodes, and INTK an internal signal.

A further advantageous embodiment is achieved in that 0 channel logic circuits are connected upstream of the inputs of the digital logic circuit for the binary signals, wherein each channel logic circuit consists of two AND gates, of which the first directly and the second connected via a delay circuit to the input of the channel logic circuit is, wherein the outputs of the AND gates are linked via an OR gate, that each channel logic circuit includes a further AND gate whose η - 1 inputs are connected exactly to the outputs of the logic circuit same reference direction which the disconnectors η Φ i or η * k and that the output of the third AND gate is connected to the second input of the second and negated to the second input of the first AND gate. The mode of operation of the fault-tolerant power supply system corresponds to the method. In the error-free case, all connection lines between the network nodes are intact and all disconnectors are closed. When a line short circuit occurs between the network nodes j and j + 1 in the connection line i, a short-circuit current flows in the power supply system and divides accordingly in the line network. The resulting node potentials are dependent on the internal resistance of the feeding source, the resistance of the connecting lines, measuring and disconnecting devices, the current load from the connected consumers and the location of the short circuit in the multiple ring system. Short-circuit currents which are detected by exceeding the upper current limit and which are relevant for the evaluation in the network nodes j and j + 1 occur in all connection lines between the network nodes j and j-1 and j + 1 and j + 2 and in the line i between the network nodes j and j + 1. In the connection lines k = 1 ... n; k Φ i between the network nodes j and j + 1 flow between the network node j and j + 1 corresponding equalization currents whose direction depends on the network node potentials. If network node j is at a higher potential than j.sub.j-1, rectified equalizing currents flow from j to j + 1 via the connecting lines k.phi..sub.L. According to the switch-off conditions, the one in connection line i first obtains between network nodes j and j + 1 in network node j + 1 arranged disconnectors

- 5 - £ 36 SUU

from the evaluation logic of the network node j + 1, the binary drive signal F for switching off the connection line i, since only in the network node j + 1 in the connection line i to the network node j, the current flow direction opposite to all other connected to the network node j + 1 connecting lines occurs.

After disconnecting the short-circuited connection line i from the node j + 1, the current in all connection lines k Φ i between the network nodes j and j + 1 changes its direction, whereby the shutdown conditions for the corresponding circuit breaker in the line i of the node j met in an analogous manner are. The short-circuited connecting line is thus separated out. The time required for this depends essentially on the switching time of the circuit breaker.

In case of an internal short circuit in the network node j, current of the same direction flows into the network node via all intact connected lines. If short-circuit current flows in a line whose all parallel lines are defective and thus switched off or interrupted, the resulting binary signal 0 for exceeding the fixed upper current limit value in the adjacent network node is forwarded to the logic circuit via the delay circuit arranged in the channel logic circuit. The time constant of the delay circuit is set so that the evaluation logic in the other network nodes of the power system are settled and the switching times of the circuit breaker have expired before the delay circuit turns on. With this arrangement, it is possible to respond to multiple shorts within the power distribution system. If a short circuit occurs that did not take place in the only intact line between two network nodes, the network nodes adjacent to the actual location of this short circuit have enough time to switch off the defective connection line. If the mains current does not fall below the upper limit 0 in the single connecting line between two network nodes during this time, the binary signal 0 for exceeding the fixed upper current limit value is sent to the logic circuit in the two network nodes which are only connected by this one connecting line forwarded. If the short circuit actually occurred in this single connection line, this is thus switched off after the time specified by the time delay circuit, since the signal 0 is still present even after the time delay has elapsed.

embodiment

The invention will be explained in more detail below using an exemplary embodiment.

Show it: . '

Fig. 1: The topology of a trained as a double ring fault-tolerant power supply system with 8 network nodes Fig.2: The structure of a network node

Fig. 3: The structure of the evaluation logic for a power supply system with η ring lines

The annular power supply system shown in Fig. 1 consists of two independent supply rings, which are interrupted in each case in the network nodes K1 ... K8. Within the network nodes K1... K8, the connection lines are interconnected via the disconnectors TEI 1, TEI2, TEr 1, TEr2 and the current measuring devices ME11, MEI2, MEr 1, MEr2 shown in FIG. In the connection point of the four lines LH, Ll 2, Lr 1, Lr 2 connected to each node, the load, for example a functional unit of an automation system to be supplied, is symbolically represented here by the load resistor RL.

The structure of all network nodes is the same. In the network node KI, the energy is fed from the feed point H via a feed-in line.

In the fault-free case, the current flows from the feed point H via the feed-in line and the network node K1 in the double-ring network and is divided equally long and equal lines on the two rings. In between two network nodes located parallel connection lines therefore flows the same amount and phase current. The connection of the node points of the network is such that each line Ll 1 j (line Ll 1 of the node j) of the line Lr1 _{j +} i (line Lr 1 of the node j + 1) and Ll 2j (line Ll 2 of the node j) corresponds to the line Lr2 _{j + :} (line Lr2 of the node j + 1). To close the ring, the nodes K1 and K8 are connected by the lines Lr 1i = LH ₈ and Lr2 ₂ = Ll 2 ₈ . Due to the annular structure, the current flows from the network node K1 in the lines Ll 1i and Ll 2, or Lr 1! and Lr2i in the respective same directions, which are different between LI ₁ , Ll 2i and LrI ₁ , Lr2i. That is, 11 flows opposite to 12, with 11 and 12 being the summons of the currents in the lines LMi and Ll 2, and LrI ₁ and Lr2, respectively. The current flow directions are maintained up to the current main sink of the system, which may vary depending on the power of the switched-on loads during operation. The structure of the individual network nodes is shown in FIG. 2. Via the connection lines LH, Ll2, Lr 1 and Lr 2, the line currents flow into or out of the illustrated network node. The connecting lines LH and Lr 1 or Ll 2 and Lr2 belong to the same line ring. The current flows via the circuit breakers TEI 1, TEl2, TEr 1, TEr2 and the measuring devices MEH, MEI2, MEr1, MEr2 into the node or out of the node, to which the load RL is symbolically connected. The disconnect switches TEI 1, TEI2, TEr 1, TEr2 separated by the evaluation logic A in the event of a short circuit, the connected lines from the network node. The switching conditions that activate the circuit breakers will be discussed in more detail.

The measuring devices MEH, MEI2, MErI, MEr2 measure the flowing line current and evaluate it with respect to its direction, the falling below a fixed lower limit and the exceeding of a likewise fixed upper limit, each with a binary output signal Rl ₁ , Rl ₂ , Rri, Rr ₂ ; Uli, Ul ₂ , Uri, Ur ₂ ; 0I ₁ , 0I ₂ , Or ₁ , Or ₂ which is respectively entered into the evaluation logic Ae. The signal U is set by a measuring device when the lower current limit is exceeded. The signal O is present at the output of a measuring device when it has been determined that the upper current limit value has been exceeded. The direction signal R is defined for the registered in a connecting line direction of current from the point of view of the network node, ie, for example, set with outflowing current or vice versa. The evaluation logic illustrated in FIG. 3 is subdivided into a number of channel logic circuits KLH, KLI2, KLr 1, KLr2 corresponding to the number of lines LIi, Ll ₂ , Lr-i, Lr ₂ and a common logic unit. In the channel logic circuits KLH, KLI2, KLr 1, KLr2, the signals 0I ₁ , 0I ₂₁ Or ₁ , Or ₂ coming from the current measuring devices MEH, MEI2, MEr 1, MEr2 become 0

preprocessed, as will be described in more detail, and appear at the output of the channel logic circuits as signals Oli, 0I ₂ , Or ₁ 'and Or ₂ . The signals Fl ₁ , Fl ₂ , Fr _n , Fr ₂ at the output of the logic circuit V are control signals for the associated circuit breaker. ·

As illustrated in detail here, the logic circuit V does not implement the following Boolean equations:

= OIJj Λ [[(Rl ₁ Φ Rl ₂ ) VUl ₂ YES [(Rl ₁ Φ Rr ₂ )

Λ V

^ Rl ₁ φ Rr ₁ ) VUr ₁ ] V

δι ^! ₂ A

= 0I ₂

Λ I [(Ri ₂ φ Ri ₁ ) VUi ₁ YES

A [(Rl ₂ φ Rr ₂ ) VUr ₂ JV 1 'Λ Qt' A Or ¹

V Ur

= Or ^ Λ j [(Rr ₁ Φ Rl ₂ ) V UlJ A [(Rr ₁ Φ Rr ₂ ) [Rr ₁ Φ Rl ₁ ) V Ul ₁ ] V [INTK j

V [oil 'Λ cr' ₂ A

or

Fr ^ =

Or ' ₂ A ί [(Rr ₂ Φ Rl ₁ ) V Ul ₁ YES Λ [(Rr ₂ Φ Rl ₂ ) V Ul ₂ JV

V Ur ₁ J

with INTK =

k = 1

AV

k = 1

INTK is an internally generated in the logic circuit V signal, which is then set when inside the

concerned network node, for example, directly to the consumer Rl a short or ground fault has occurred. On

node-internal short circuit is characterized in that on all intact lines ejn short-circuit current flows, the current direction is the same in all intact lines. To explain the operation of the arrangement in case of error

Assuming that the defect occurs at the defect location DS between nodes K4 and K5 in FIG.

Open circuit:

When a line break occurs at the defect point DS no current flows on the line LI2 ₄ (ie on line

LI 2 of the node 4) corresponding to the line Lr2 ₅ (ie, the line Lr2 of the node 5). The measuring devices MEI2 ₄ and MEr2 ₅ (measuring devices in line LI 2 of the node 4 and measuring devices in line Ll 2 of the node 5) determine the falling below the lower current limit and set the signals Ul 2 ₄ and Ur2 ₅ for the Äuswertelogik A. line short circuit: '"-

If a short or ground fault occurs at the assumed defect location DS in FIG. 1, a short-circuit current flows in the entire system. The current is divided equally between the two parallel rings. The upper limit in the current measuring devices can certainly be set so that the exceeding of the upper current limit is detected in each of the two parallel intact line sections. In each channel logic circuit of each network node, the signal OHj and Orij are then set. The processing of the above-mentioned Boolean equations shows that only the line is switched off, flows in the short-circuit current and their Stromflußrichtung is opposite to that of all other connection lines of a network node. But this is true only for the line Ll 2 ₄ = Lr 2 ₅ , so that the circuit breaker TEI2 ₄ and TEr2 ₅ open. The sequence of the shutdown depends on the current potential of the network nodes K4 and K5. A line short can be detected autonomously in each network node if and only if the line parallel to the relevant line is intact

 Internal short circuit:

A short circuit within a network node K1 to K8 is characterized in that a short circuit current flows in all intact lines connected to the network node, the currents in all intact lines having the same direction (eg all currents flow in).

In the logic logic V, the above-mentioned Boolean equation for INTK is realized, so that an internal short can be detected in each network node, if at least two intact lines are connected to the respective network node. If an internal short circuit is detected, the INTK signal is set, which only switches off all connected and still intact lines. This switching operation breaks the supply of the relevant network node static and transferred the connected lines in a de-energized state. This can be done for example by an electromagnetic switch with mechanically blocked end position. The mechanical blockage can be released again after repair by the maintenance personnel, so that the circuit breaker can turn on again. If, between two adjacent network nodes Kj and Kj _{+ Ί} , both connection lines are still intact, the channel logic circuits of these lines KLI Ij, KLI 2j, KLr 1 j + _u KLr I ₁ + and the gates D111 _jf D4I1 _{j (} D1 \ 2 - _V D4I2j; D1r1j ₊ · ,, D4r1 j + · ,, D1 r2j + τ, D4r2j + _{λ are switched} through and the signals Ol V ₁ , OI2j, Or1 _{i + 1} , Or2 _{J +} ί corresponding to the signals Ol 1 \, Ol 2 |, OrIj _{+ 1} , Or2j _{+ 1} .

If only one intact connection line exists between two adjacent network nodes Kj and Kj +1, for example connection line 1 between network nodes K4 and K5, D1l1 ₄ and D1r15 are disabled and in the channel logic circuits KLH ₄ and KLrI ₆ the delay circuits TV become active. The signals OH ₄ and OrI ₅ are now forwarded time-delayed with the occurrence of corresponding amounts of current to the logic circuits V of the network nodes K4 and K5. Thus, it is possible to tolerate and treat multiple shorts in the system. If a further short-circuit occurs after switching off connecting lines, it is not possible to decide in the nodes between which only one connecting line is intact whether the short-circuit has taken place in this one line or not. The time-delayed setting of the signal causes the short-circuited line within the time delay can be switched out of the system, if it is not this one line between two network nodes. However, if the short-circuited line is the one which connects only two post-nodes, it will be disconnected successively from the two connecting network nodes after the set delay time, as already described above.

Claims (4)

- ι - £ .Ό £. Claims: 1. A method for fault-tolerant power supply, in particular for automation systems with distributed functional units, in which the power supply system is formed as consisting of point-to-point connections between the functional units double or multiple ring with a feed point, characterized in that in each network node ( K) the current in each connected to the network node connection line (L) is measured in magnitude and direction that by comparing the current direction with each other and by comparing the current amounts each with a fixed upper and a fixed lower limit in the short circuit for exceeding the fixed current limit, connected to the network node connection line or the network node itself detected as the cause, and that the connection line thus determined by the network node or the network node thus determined is disconnected from all interconnections. 2. fault-tolerant power supply system, in particular for distributed equipment automation systems, structured as a double or multiple ring with a feed point consisting of point-to-point connections between the network nodes, the interconnections including current measuring devices and disconnect switches controlled by trip circuits, characterized in that a controllable disconnector (TE) and a current measuring device (ME) for current amount and current direction is arranged at all power supply inputs of each network node (K) and that each network node (K) has a Auswertelogik (A) with all circuit breakers (TE) and Strommeßeinrichtungen (ME) of the network node is connected, wherein the Auswertelogik (A) consists of a circuit arrangement that realizes the control of the circuit breaker (TE) exclusively as a function of the measured current amounts and directions. 3. Fault-tolerant power supply system, in particular for automation systems with distributed functional units according to claim 2, characterized in that the current measuring devices (ME) each have three binary outputs for the binary signals 0 for exceeding a fixed upper, U for falling below a specified lower current limit and for Current direction signal R have. 4. fault-tolerant power supply system, in particular for automation systems with distributed functional units according to claim 2 and 3, characterized in that the evaluation logic (A) includes a digital logic circuit (V), the Boolean equations