39 results about "N+1 redundancy" patented technology

Parallel UPS system controlled by bus relates to power source equipment technology with the following scheme: UPS parallel control system is put in and connected with UPS, realizing N+1 redundant parallel operation and every UPS will supply electricity to load and shares the load to meet the requirement of parallel operation having very high real time data process capability to ensure the real time control and reliability of UPS parallel control system increased greatly.

A power supply system is adapted to a platform including a plurality of processing devices including baseboard management controllers (BMC) for processing basic information units, each of which is equipped with a plurality of power supply units (PSU) and at least one secondary power supply unit interconnected to a power cable. The controller calculates the maximum power consumption of the processing device based on its configuration, an adequate number of power supply units being turned on to meet the maximum power consumption, and a redundant power supply corresponding to one power supply unit times the adequate number minus the maximum power consumption. When the processing device suffers from a power shortage, the secondary power supply unit or the power supply unit of the other processing device is selectively turned on so as to recover the redundant power supply, thus normally achieving the N+1 redundancy of power supply.

The invention discloses a system and a method for achieving power redundancy. The system includes: an alternate current detection module detects input state of an alternate current input end of one path and outputs the detected state information to a monitoring module; the monitoring module controls switching on or off of a switch in a switch module according to the input state information; and the switch module controls connection or disconnection between alternate current power voltage distributed by a power distribution module and a power module through the switching on or off of the switch. The system and the method can achieve single path alternate current input and double path alternate current redundancy input simultaneously under N+1 redundancy configuration of the power supply, does not need to increase quantity of power supplies and simultaneously improves reliability of a power system and reduces cost under the premise of meeting requirements for alternate current redundancy input of a fault-tolerance server.

The invention provides a power supply method and a power supply device of a cabinet server. The device comprises: at least two power modules, wherein N+1 redundant power supply is performed by the power modules, that is, one of the power modules is in a standby state, the rest of the power modules are used for supplying power to the cabinet server, N 1 is a natural number greater than 1, and all of the power modules are connected to a bus bar; a conducting module, wherein the conducting module comprises a copper bar, the copper bar is used to connect the bus bar, and a current output by the power module is conducted to the copper bar via the bus bar; and a power taking module, used for conducting the current on the copper bar of the conducting module to more than one node of the cabinet server. The invention can increase power supplied to a cabinet server.

The invention discloses a storage system with an N+1 redundancy function and a method. The system comprises a camera, storage equipment and a store-and-forward server, wherein the store-and-forward server receives a video stream sent by the camera and stores the video stream into the storage equipment; the storage equipment comprises N storage equipment parts and redundant storage equipment; and the store-and-forward server comprises a forward cache module, a storage management module, a redundant switching module, a redundant recovery module and a data backup module. The storage system effectively solves the problem that a large number of videos are lost after the storage equipment is damaged; the automatic data backup during the redundancy period ensures the data integrity after redundancy recovery and clear subsequent redundancy switching; the plan protection during redundancy switching and recovery ensures normal operation of the system; the stream cache during forwarding ensures the videos to be stored during redundancy switching and recovery; and the nondestructive storage of the system is ensured.

Disclosed is an apparatus and a method for implementing efficient redundancy and an expanded service coverage in a Radio Access Station (RAS) system. In the RAS system, if the main processor unit generates switching control signals in response to the sensed failures on sensing a failure in any of the channel cards and the transceivers or a failure in any of the high-power amplifiers, all supporting M (i.e., the number of FAs equal to or more than three) and K (i.e., the number of sectors equal to or more than three), between the transceivers and the predetermined Time Division Duplex (TDD) switches connected to the antennas, the RF switch unit switches a path based on the generated switching control signals so as to substitute the failed module either by one additional redundancy transceiver per M and K or by one additional redundancy high-power amplifier per M and K. As a result, an efficient N+1 redundancy structure is embodied.

A switch for switching packets. The switch includes a plurality of fabrics which switch portions of packets. The switch includes a port card connected to the fabrics and the network for receiving packets from and sending packets to the network. The port card has a mechanism for tolerating whether any one of the plurality of fabrics has a failure and still sending correct packets to the network. A method for switching packets. The method includes the steps of receiving packets at a port card from a network of a switch. Then there is the step of sending to fabrics of the switch portions of the packets as stripes from the port card. Next there is the step of switching the portions of the packets with the fabrics. Then there is the step of sending back to the port card the portions of the packets as stripes from the fabrics. Next there is the step of sending correct packets with the port card to the network even though one of the fabrics has a failure.

The provided N+1 redundancy backup method for multiprocessing boards with N activation boards and one backup board comprises: composing the activation boards into a logical ring by address mapping, and backuping some running data of former board in the next one on same ring. Wherein, it also provides a processing board that comprises a main memory accessed by processor on same board to store self running data and a backup memory accessed by processor on another board to store at least some running data on said board. This invention reduces cost.

The invention discloses a redundancy-synchronization-isolation control method for the parallel connection or the serial connection of power devices. The characteristics of channels of light signals in an adopted light-way divider are uniform, the length of optical fibers is equal to or close to one another, and the two conditions guarantee the uniformity of the light-transfer characteristics of channels; and the difference of a light receiver and a light coupler lies in the case that voltage isolation does not need to be taken into consideration, and therefore the uniformity and the sensitivity of the light receiver are extremely superior to the light coupler. The uniformity of the transfer channels and the uniformity of parameters of the light receiver guarantee the synchronization of time of a control signal reaching each power device. The method produces the light-way divider into the mode of multiple-input-multiple-output, and a redundancy signal source is additionally arranged in a control system so that N+1 redundancy control can be realized.

The invention relates to a method for realizing service N+1 redundancy backup and load sharing, wherein N>=1, comprising: a step 1, setting a load sharing policy schedule; a step 2, the N+1 block of service treatment line cards shares a IP; a step 3, a interface line card completes the service redundancy backup and load sharing between service treatment line cards. When blocks of service treatment line cards go out of order at the same time, the service treatment still keeps a continuity; and when the flow of traffic of subscriber is oversized, the flow between blocks of service treatment line cards is imbalance, by which the service treatment caused appears congesting.

The invention discloses an automatic current digital control method based on CAN bus communication. In a main-driven competition mechanism, each of the N+1 parallel power supply modules publishes self information through CAN communication; each power supply can compete to be the main machine or the driven machine; about the main and the driven machine based on the competition mechanism, the main machine gives up a current feedforward ring, and takes an output positive-negative parallel bus pressure as the reference, and adjusts self output load current through a PID controller; the driven machine is added to the current feedforward ring, and adjusts self output load current to make the self output current value the same with the load current value of the main machine; each power supply module controls its load current itself to realize automatic current; if some power supply module has a fault, the power supply module quits work automatically, and the rest machines compete for main-driven machine again, which has no influence to the current sharing control system and the power supply system at present, overcomes the technology defect of redundant N+1 backup to the parallel power supply system in power supply technology field.

The invention discloses an N+1 redundant power system and a method for achieving intelligent power consumption reduction thereof. When the load is reduced, the method for achieving the intelligent power consumption reduction for the N+1 redundant power system comprises the following steps: judging whether the number of power modules in the normal mode is larger than one; when the number of the power modules in the normal mode is larger than one, detecting the total output power of the power system, and judging whether the total output power of the power system is smaller than or equal to a first preset value, wherein the total output power of the power system is the sum of power of the power modules in the normal mode, and the first preset value is related to the number of the power modules in the normal mode; when the total output power of the power system is smaller than or equal to the first preset value, switching one power module from the normal mode to the power consumption reduction mode. By means of the technical scheme of the system and the method, the complete machine power supply changeover efficiency of the power system can be raised.

The invention discloses a harmonic current counteracting module which comprises a direct-current voltage-stabilizing and current-tracking control unit, a drive unit, a three-phase full-bridge conversion circuit, a direct-current voltage-stabilizing capacitor bank, a three-phase single-harmonic filter circuit, a current sensor and a voltage sensor. The invention also discloses a harmonic current counteracting module group consisting of the harmonic current counteracting modules, and the harmonic current counteracting modules are connected in parallel and then connected with a low-voltage powerdistribution busbar through a circuit breaker. The harmonic current counteracting module is convenient for standardized protection by adopting a module structure, is convenient for dilatation, can improve system operation reliability due to N+1 redundancy, and has a harmonic current counteracting effect due to counteracting aiming at single power harmonic and reasonable design of the filter circuit.

Disclosed is a computer cabinet including: at least one computer server, and at least one module for the liquid cooling of the server. The cabinet includes at least three liquid cooling modules that communicate with each other using a collaborative protocol without master/slave technology, in such a way as to operate under N+1 redundancy where N is higher than or equal to 2, so as to be able to carry out a standard replacement of any one of the cooling modules without stopping the cooling of the computer cabinet and without stopping the operation of the server located in the computer cabinet, each of the liquid cooling modules including their own system for regulating the cooling and detecting failure.

Disclosed is an apparatus and a method for implementing efficient redundancy and an expanded service coverage in a Radio Access Station (RAS) system. In the RAS system, if the main processor unit generates switching control signals in response to the sensed failures on sensing a failure in any of the channel cards and the transceivers or a failure in any of the high- power amplifiers, all supporting M (i.e., the number of FAs equal to or more than three) and K (i.e., the number of sectors equal to or more than three), between the transceivers and the predetermined Time Division Duplex (TDD) switches connected to the antennas, the RF switch unit switches a path based on the generated switching control signals so as to substitute the failed module either by one additional redundancy transceiver per M and K or by one additional redundancy high-power amplifier per M and K. As a result, an efficient N+l redundancy structure is embodied.

The invention relates to an energy-saving intelligent alternating current voltage stabilizing power supply, comprising a main control circuit, a driving circuit of a voltage stabilizing power supply, an autotransformer, a protection circuit, an energy-saving control circuit and the like. A novel intelligent high frequency PWM (pulse width modulation) integrated chip is taken as a control circuit, an Infineon high-end MOS (metal oxide semiconductor) tube is taken as the driving circuit of the voltage stabilizing power supply, and a diode and a super fast recovery diode form a clamping circuit to be taken as an MOS tube protection circuit; a modular power supply forms a distributed power supply system and is designed into an N+1 redundant power system in a parallel manner; an inlet wire terminal of a power supply is provided with a primary power supply noise filter, and interference of a power grid is inhibited; a tantalum electrolytic capacitor is taken as a power supply system absorption circuit and is connected with the power supply system in parallel; and an LCD (liquid crystal display) device displays parameters, an RS232 standard upper computer interface is arranged, and the energy-saving intelligent alternating current voltage stabilizing power supply provided by the invention has the advantages of fast input voltage response, fast load change response, overvoltage, undervoltage and overcurrent protection, automatic temperature detection, stable output frequency and low power consumption.

Disclosed is a method of communication between a plurality of liquid cooling modules of a cooling system for one or more one computer servers, in which: the cooling modules communicate with each other in a manner that operates in N+1 redundancy where N is greater than or equal to 2, so as to enable a standard replacement of any one of these cooling modules without stopping the cooling and without stopping the operation of the server or servers, this communication being ensured by a collaborative protocol without master/slave, before switching from an active mode where it is cooling to a backup mode where it is no longer cooling, the redundant cooling module verifying beforehand that a data set is consistent across all these cooling modules and that this consistency is maintained for a predetermined duration.

There is described a circuit to supply electric power to N electrical panels. The circuit comprises N+1 blockbars, each one of the blockbars being for connection to an automatic transfer switch (ATS) of a power source, each one of the blockbars being connected to at least two other ones of the N+1 blockbars via a segregating switchgear. The circuit further comprises N+1 UPS units, each one of the UPS units being for connection to a dedicated one of the blockbars via a dedicated switchgear. The circuit further comprises 2N power outputs, each one of the N electrical panels being connected to 2 of the 2N power outputs, each one of the 2N power outputs being connected to 2 of the N+1 UPS units. This configuration allows for N+1 redundancy at the generator and UPS levels to feed the electrical distribution with 2N redundancy.

The invention provides a N+1 redundancy high-voltage frequency converter with double standby power units, which is characterized by consisting of a multiple secondary winding transformer and 3m+2 power units, wherein the 3m+2 power units comprise two standby power units, the rest 3m power units are evenly divided into three groups, and each group respectively forms one phase line of the frequency converter by connecting m power units in series, namely a first phase line, a second phase line and a third phase line respectively; one end of one standby power unit and one end of the other standby power unit are mutually connected together and connected with one end of the first phase line, while the other ends are respectively connected with one end of the second phase line and one end of the third phase line; and one end of the first phase line, one end of the second phase line and one end of the third phase line, which are not connected with the standby power units form a triphase output end of the frequency converter. The high-voltage frequency converter realizes N+1 unit redundancy function only by the 3m+2 power units, saves one power unit compared with 3m+3 power units of the prior art, and reduces the realization cost.

The invention relates to a medium-voltage CS-APF (Cascaded Shunt Active Power Filter) dual-hysteresis fault-tolerant control method based on a voltage space vector method, and aims to solve the problem of high failure probability of a cascaded circuit existing in a CS-APF and effectively reduce the cost. In a non-failure state, the rapidness of current tracking is mainly considered when a current tracking error is large, and accuracy is mainly considered when the error is small; in a failure state, an optimal non-failure voltage vector is selected to equivalently replace a failure voltage vector as far as possible by making full use of the characteristics of a great number of redundant voltage vectors of a cascaded inverter and frequent coincidence of positions, so that the output current of the CS-APF can be used for accurately tracking a reference value in the failure state. Compared with the conventional frequently-used N+1 redundant fault-tolerant way, the method has the advantages that additional hardware redundant resources do not need to be added, cost is reduced, and the working reliability of a medium-voltage active power filter is improved remarkably; the method can be popularized into a CS-APF of any order.