Classifications

• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
  • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
  • H02J3/007—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources
  • H02J3/0073—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources for providing alternative feeding paths between load and source when the main path fails, e.g. transformers, busbars
• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
  • H02H1/00—Details of emergency protective circuit arrangements
  • H02H1/04—Arrangements for preventing response to transient abnormal conditions, e.g. to lightning or to short duration over voltage or oscillations; Damping the influence of dc component by short circuits in ac networks
• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
  • H02H3/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection
  • H02H3/08—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to excess current
• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
  • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
  • H02J3/001—Methods to deal with contingencies, e.g. abnormalities, faults or failures
  • H02J3/0012—Contingency detection

Definitions

• Redundant power supply especially for data centers, as well as methods and computer programs for their operation
• the invention relates to a redundant power supply that is particularly suitable for data centers.
• the power supply is designed redundantly.
• Such a system already expanded to include a generator and a local uninterruptible power supply, is known from US 2014/0191579 A1.
• a critical component of a data center is powered by two independent power supplies.
• both power supplies In order to ensure the operation of the critical component even if one of the power supplies fails, both power supplies must be dimensioned in such a way that each of the power supplies alone can reliably supply at least the critical component with power.
• every power supply must be designed in such a way that it can carry double the load compared to the undisturbed case. This is very uneconomical, since the design of the power supply for this rare fault leads to a considerable increase in the cost of the individual power supplies.
• a so-called "Isolated-Parallel Rotary Diesel UPS System” is disclosed in EP 1890 371 A1.
• a unified power supply network supplies several essential and several critical electrical loads and several diesel-powered emergency power supplies are provided, which take over the supply of the critical loads if the unified power supply fails.
• flywheels are provided as intermediate kinetic energy stores, which supply electrical energy via a generator after a power failure until the internal combustion engines of the emergency generators have been started using the kinetic energy of the flywheels.
• the emergency power units are coupled by means of an electrical bus system, which is intended to prevent a combustion engine that does not start or does not start in time from leading to an interruption in the supply of the critical load assigned to the corresponding emergency power unit.
• This system is also very expensive, both mechanically and in terms of circuitry.
• each load is immediately switched off (apart from the usual safety precautions such as circuit breakers and etc.) is supplied by two power supplies and is also coupled to the power supplies of the other load(s) via coupling switches and a busbar.
• each power supply does not have to be designed for twice (200%) the load to be carried in normal operation, but only for 133%, since the power lost from three power supplies if one power supply fails can be delivered.
• this is advantageously done without any intervention, i.e. without operator intervention and without changing the switching position of one of the elements in the current path, i.e. without any time delay.
• the redundant power supply according to claim 1 ensures an even load on the Stromversor conditions, especially in periods in which a load consumes less electricity than the other load (or is switched off in the extreme case), and also without any action. This has advantages with regard to the cooling of the power supply, for example, since the power loss is (more) evenly distributed over all four power supplies (and thus spatially).
• the concept according to the invention can be expanded to include additional power supplies and loads, advantageously in the form of additional subsystems that can be connected to the busbar by means of coupling switches, each consisting of two power supplies and one load, which means that the remaining power supplies also have to be borne in the event of a power supply failure
• the load can be further reduced or failures of more than one power supply can be intercepted if the design of the power supplies remains the same.
• a controller can be provided to handle other (rare) faults than the failure of one of the power supplies.
• This control can be implemented centrally or distributed or locally, for example in one or more coupling switches.
• a possible error is a fault in the busbar, for example a short circuit or short circuit. If such a fault is detected, the tie switches are opened.
• Another possible fault is a fault, again just a short circuit or short circuit, for example, in a line coming from a power supply, for example the line between one of the power supplies and the coupling switch connected to the power supply, or in the event of a line fault between one of the Power supplies and the load connected to the power supply without coupling switch. It can, at least in part, be the same line or a branch from which these two lines open. This error case is handled by opening the tie switch connected to the relevant power supply.
• the section switches have means for determining the current flowing through the respective section switch and/or the sign of the current flowing through the respective section switch, as well as means for transmitting a current value and/or a sign value to the Steering.
• the control can also be arranged centrally for this purpose or distributed to the individual section switches.
• the power supplies have means for generating a signal which signals whether the respective power supply is active or inactive, and means for transmitting this signal to the controller.
• the (central or distributed) controller receives the current values and/or sign values from all tie switches and optionally the signals from all power supplies and detects a busbar fault if the sum of all current values supplied by the tie switches exceeds a threshold value and/or all tie switches supply sign values, which show a current flow from the respective power supply to the busbar.
• threshold values for the current values can be defined and/or threshold values above which a section switch outputs a certain sign, so that tolerable (transient or permanent) effects do not lead to the section switch being switched off.
• the controller can be designed to take the course of time into account and only recognize the error or initiate measures when the described state lasts for a defined and/or configurable period of time.
• the time period can be selected as a function of the strength of the current flowing in the direction of the busbar fault or the electrical power flowing there, whereby shorter time periods are advantageously chosen for high power flows.
• the method described can also be used to detect so-called low-level faults, i.e. faults that cause fault currents that cannot be detected by conventional local short-circuit or fault current detectors because they cannot be distinguished locally from a permissible load case.
• errors can also be detected which occur in the line system to be assigned to an individual power supply.
• the signals mentioned above are also evaluated and the fault in the line system that can be assigned to a power supply is determined when all power supplies signal that they are active and the current value at one of the section switches corresponds to the negative value of the sum of the current values of all other section switches and/ or if the sign value of a tie switch is inverse to the sign values of all other tie switches.
• threshold values for the current values can also be defined for this error case and/or threshold values above which a section switch outputs a specific sign, so that tolerable, in particular transient, effects do not lead to the section switch being switched off. These thresholds may differ from or be the same as the thresholds used for the busbar fault.
• the controller can be designed to take into account the passage of time and only recognize the error or initiate measures when the described state for a defined and/or configurable period of time.
• the period of time can be selected as a function of the strength of the current flowing in the direction of the affected line system or the electrical power flowing there, with shorter periods of time advantageously being selected in the case of high power flows.
• the time periods can also deviate from or correspond to the time periods used for the busbar fault.
• the present invention also relates to a method for operating a redundant power supply according to the invention and a computer program which causes a processor to execute this method.
• the single FIGURE shows a schematic representation of a redundant power supply 1 to which two loads LI and L2 can be connected, according to a preferred exemplary embodiment of the present invention.
• the redundant power supply 1 can be viewed as the interconnection of two fundamentally known subsystems by means of a busbar 100 and coupling switches 15, 25, 35, 45.
• the first subsystem on the left in the illustration, has a first power supply module 10 and a second power supply module 20 .
• the first power supply module 10 is connected to a first line system 12 by means of a source-side safety device 11 .
• the safety device 11 on the source side is a safety device that is customary in the art.
• the second power supply module 20 is connected to a second line system 22 by means of a source-side safety device 21 .
• a source-side safety device 21 is a standard safety device.
• a first load LI is connected to the first line system 12, specifically by means of a load-side safety device 13.
• the load-side safety device 13 is also a customary safety device.
• the first load LI is also connected to the second line system 22, specifically by means of a load-side safety device 23.
• the load-side safety device 23 is also a conventional safety device.
• the second subsystem on the right in the illustration, has a third power supply module 30 and a fourth power supply module 40, which are connected via respective source-side safety devices 31, 41 to a corresponding third and fourth line system 32, 42.
• a second load L2 is connected to corresponding load-side safety devices 33, 43 both with the third and with the fourth line system 32, 42.
• a busbar is generally understood here to mean a possibly multi-pole rigid or flexible line which, in particular, does not necessarily have to have the shape of a rail.
• a first section switch 15 connects the first line system 12 to the busbar 100
• a second section switch 25 connects the second line system 22 to the busbar 100
• a third section switch 35 connects the third line system 32 to the busbar 100
• a fourth section switch 45 connects this fourth line system 42 with the busbar 100.
• the first load LI is connected to the first power supply 10 without a section switch, via the line system 12, and the first power supply is also connected via the line system 12 to the busbar by means of the first section switch 15.
• This enables the first power supply 10 to supply power to the second load L2 via the first line system 12, the first tie switch 15 and the busbar 100 if required, namely by means of the third and/or the fourth tie switch 35, 45 and the third and /or the fourth line system 32, 42.
• the section switches 15, 25, 35, 45 are all closed during normal operation. It is not important for the present invention whether the power supply modules 10, 20, 30, 40 supply DC or AC voltage. It goes without saying, however, that the nominal voltages of the power supply modules in the two subsystems must be at least approximately the same and that, in the case of alternating voltage, there must be phase synchronicity in order to avoid undesirable effects. Alternatively, it is of course also possible to couple two asynchronous AC systems by means of a DC voltage busbar and the corresponding rectifiers and inverters, in which case systems with different operating voltages can then also be coupled. As already explained above, the present invention provides considerable advantages in the design of the power supply modules 10, 20, 30, 40.
• a first subsystem is usually already present here, with the usual design of the power supply modules such that each of the power supply modules can supply the load of the first subsystem alone. If a second subsystem is to be installed, the first subsystem is simply expanded by adding two branches to the line systems and the tie switch and then connected to the new busbar - no further changes are required and the (now actually oversized) power supply modules can continue be used or, if necessary, replaced by more efficient modules.
• the redundant power supply 1 also has a controller 110, the function of which is described in detail further below.
• the control can be implemented centrally or decentrally, ie distributed to other components, for example the section switches 15, 25, 35, 45.
• the presentation of the signal and/or command lines that may be required between the power supply modules and the controller and/or between the section switches and the control or the section switches among themselves were omitted for the sake of a better overview.
• the direction of the currents i1, i2, i3, i4 is defined for the present consideration without loss of generality such that there is a positive value when the current flows from the line system 12, 22, 32, 43 to the busbar 100 and vice versa a negative value Value is present when the current from the bus bar 100 to the respective line system 12, 22, 32, 42 flows out.
• the second operating scenario differs from the first operating scenario in that the power consumption of the first load LI is (significantly) lower than the power consumption of the second load L2, for example the first load LI consumes only 70% of its nominal power, while the second load L2 consumes 100% of its rated power.
• the first error scenario has already been discussed above and relates to the failure or maintenance of a power supply module. Without restricting the generality, the case should be considered that the first power supply module 10 is not available or the source-side safety device 11 is defective or has been triggered due to an error and as a result no energy is fed into the first line system 12 from the first power supply module 10 can.
• the failure of the first power supply module 10 is completely transparent, the electrical power continues to be routed via the redundant power supply paths of the first load LI and can be called up equally by the load LI from both power supply paths, as in the fault-free case will.
• the second fault scenario relates to a fault in the busbar.
• this fault situation more current or power flows into the busbar than is drawn from elsewhere, i.e. il + i2 + i3 + i4 > 0 or, for many practical applications, il + i2 + i3 + i4 > iS, where iS is a defi nable threshold value, so that normal operation is recognized as long as this threshold value is not exceeded.
• a self-test signal is generated by each power supply module and output to controller 110 .
• a positive self-test signal indicates that the respective controller is active and working properly.
• the instantaneous value of the current is measured by each coupling switch, ie il..i4, and a signal representing this value is transmitted to the controller 110.
• the controller 110 evaluates the received signals and detects a fault if the criteria already briefly mentioned above are met, i.e. if positive self-test signals are received from all power supply modules and the sum of the current values exceeds a threshold value.
• the criterion that the sum of the current values exceeds a threshold value is used to differentiate this second error scenario from the two operating scenarios and the first error scenario, because both in the two operating scenarios and in the first error scenario it applies that the sum of the instantaneous values of the currents is at least approximately Resulting in zero, i.e. below the threshold value iS.
• This error detection also works when the redundant power supply 1 is in the second operating scenario, i.e. the two loads LI and L2 have different power consumptions, and it also works when the second error scenario occurs in addition to the first error scenario.
• simpler section switches are used which, instead of determining the instantaneous current value, only determine a direction of the current, or, to put it another way, the sign of the current flowing through the respective section switch. Then the second error scenario can be recognized by evaluating the sign. An error occurs when all signs supplied by the section switches are the same. However, the detection of the second error scenario with these simpler coupling switches is only possible if the two loads LI and L2 have at least approximately the same power consumption, which is still advantageous compared to the case in which the error is not detected at all.
• the section switches only output a sign value if the magnitude of the current flowing exceeds a specific threshold value implemented in the section switch using hardware or software.
• the controller 110 preferably causes all coupling switches to open, as a result of which the faulty busbar 100 is isolated. In addition, an alarm can be given to an operator.
• the third error scenario is that a fault occurs in one of the line systems 12, 22, 32, 42 that can be assigned to the individual power supply modules, while all four power supply modules 10, 20, 30, 40 are working without faults, for example if in the line section between the first power supply 10 and the first coupling switch 15 and/or a fault occurs in the wiring harness between the first power supply 10 and the first load LI.
• this negative signal is present or the positive signal is not present, so that controller 110 can distinguish between the first and third fault scenarios and can initiate the measure already described in principle above, namely isolating first line system 12 by opening it of the first dome switch 15.
• a corresponding alarm can be issued to an operator.
• the controller 110 sends a disconnect command to the source-side safety device 11 and the load-side safety device 13 of the first line system 12 in order to completely isolate the fault location.
• the methods described above are used to detect so-called low-level errors, ie errors that cannot be detected, or at least cannot be reliably detected, by conventional trigger criteria that can be implemented locally.
• Conventional, locally recognizable triggering criteria mean in particular: the instantaneous current value exceeds a maximum current value (this situation is referred to below as a high-level error) or the current-time curve has a characteristic that is above a triggering criterion, for example the well-known evaluation of I 2 t (this situation is hereinafter referred to as moderate level error).
• high-level error detection moderate-level error detection and low-level error detection, for example by means of appropriately designed section switches.
• tie switches have a locally implemented high-level error detection, for example means for comparing the instantaneous current value with a configurable or fixed maximum value.
• high-level error detections are known to those skilled in the art and offer the advantage that arithmetic operations can be switched off very quickly and in particular without delay, and consequential damage can thus be avoided.
• a typical high level error is a short circuit.
• tie switches have a moderate level error detection, for example in the form of a tripping curve.
• a moderate level error detection for example in the form of a tripping curve.
• This can be implemented completely locally in the section switch, and many electromechanical or electrothermal implementations are known for this.
• the tripping curve can also be implemented electronically by the section switch and the central or decentralized control by evaluating the measured current values over time by the Control is made, which is particularly advantageous if, as in the illustrated embodiment, a central controller controls a number of coupling switches.
• the moderate level error detection can be combined with the low level error detection.
• a typical moderate level error is, for example, an inadmissibly long-lasting excessive current flow between the both subsystems, for example caused by a detected and treated error according to the third error scenario combined with excessive power consumption in the affected subsystem with very low power consumption in the other subsystem at the same time.
• a low-level error detection is preferably implemented as described in detail above, in which a central controller 110 or a controller distributed, for example, to the individual section switches preferably the measured current values of all section switches, alternatively the sign of the Current flows through all section switches, recorded and evaluated according to the procedures described. This can be combined with the moderate level error detection.
• electronic switches are used as coupling switches, so-called Solid State Circuit Breakers (abbreviation SSCB, sometimes also SCCB) being particularly preferred.
• SSCB Solid State Circuit Breakers
• the use of electronic switches has the advantage that the section switches can be opened and faults isolated so quickly that the usual protective devices 11, 13, 21, 23, 31, 33, 41, 43 cannot be triggered due to the fault (which would lead to functional restrictions) is avoided.
• the tripping characteristic of the tie switches is selected to be faster than the tripping characteristic of the source-side protective devices 11, 21, 31, 41 and/or than the tripping characteristic of the load-side protective devices 12, 23, 33, 43.
• the coupling switch(es) to be switched are switched in accordance with the above detailed description before one or more power supply module(s) are impermissibly loaded due to a fault, For example, too high a current is called up for too long and/or before a load is not supplied with the required electrical power due to a fault, for example the supply voltage drops below the minimum permissible value.
• the criteria mentioned above i.e. loading a power supply module with too high a current and/or supplying a load with too low a voltage, possibly also taking into account a time over which the respective critical state lasts, as an alternative or additional switching criteria can be used for switching the section switches.
• the present invention entails a certain amount of implementation effort, but this should be worthwhile in most cases, because only the present invention enables the use of significantly smaller power supply modules (as already explained, the power supply modules only have to be used for 133% of the normal load can be designed, compared to 200% of the normal load for isolated subsystems not equipped according to the invention).
• all relevant error scenarios can be detected in good time and the errors can be automatically isolated in such a way that the operation of the loads LI, L2 is not impaired.
• the respective error can be remedied by a maintenance team without the loads L1, L2 having to be switched off for this purpose.
• controller includes processors and processing units in the broadest sense, for example general purpose processors, graphics processors, digital signal processors, application-specific integrated circuits (ASICs), programmable logic circuits such as FPGAs, discrete analog or digital circuits and any combination thereof, including any other processing units known to those skilled in the art or hereafter developed.
• processors can consist of one or more devices. When a processor consists of multiple devices, they may be configured to process instructions in parallel or sequentially.

Landscapes

• Engineering & Computer Science (AREA)
• Power Engineering (AREA)
• Supply And Distribution Of Alternating Current (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=81850684

Family Applications (1)

Country Status (4)

Family Cites Families (7)

• 2022
  • 2022-04-27 CN CN202280003158.7A patent/CN115568290A/zh active Pending
  • 2022-04-27 WO PCT/EP2022/061183 patent/WO2022233673A1/de active Application Filing
  • 2022-04-27 EP EP22726026.2A patent/EP4104270A1/de active Pending
  • 2022-04-27 US US17/910,748 patent/US20240204518A1/en active Pending

Also Published As

Similar Documents

Legal Events