59 results about "Network sampling" patented technology

The invention provides a hierarchical protection control system facing a regional power grid. The hierarchical protection control system comprises an in-situ level protection device, a substation area protection control system and a wide area protection control system, wherein the in-situ level protection device is configured according to a traditional protection scheme and adopts a direct sampling direct trip mode or a direct sampling network trip mode; the substation area protection control system is configured according to substations or voltage classes, utilizes an information sharing advantage to improve the backup protection performance, realizes protection functions of backup protection, redundancy protection, failure protection, and the like, and control functions of spare power automatic switching, low-frequency and low-voltage load shedding, and the like, and adopts a network sampling network trip mode; and the wide area protection control system is configured according to areas, realizes functions of wide area backup protection, self-adaptive adjustment of protection fixed values, low-frequency and low-voltage load shedding, power grid state evaluation, safety pre-warning, and the like, and realizes the interaction and the sharing of information of intelligent transformer substations in the areas through a wide area communication net. The hierarchical protection control system has the advantages that the reliability and the selectivity of relay protection can be improved on the basis that protection configuration of the existing power grid is not changed, so high requirements of power grid interconnection, new energy access, and the like on relay protection are satisfied.

Disclosed is an informed sampling technique for biasing a sample data set toward network data of interest for a particular application. Network data received at a network node (for example at a rate which is greater than a sampling rate for which the network node is configured) is chosen to be included in a sample set based on one or more predetermined signatures which are chosen to bias the sample set toward network data of interest for a particular application. For example, the sample set may be biased to include data of interest for fraud detection, spam detection, and intrusion detection. The particular signature(s) may be predefined by a user, or may be automatically generated by another network application. The invention may be implemented at various levels and nodes of a network. For example, the informed sampling may be implemented at a traffic monitoring function of a network router, a flow collector which receives network flow data from the router, or both.

The invention discloses a method and a system for authenticating identities of a remote network on the basis of multi-biometric features, the system comprises a multi-biometric features network sampling and controlling terminal, the remote network and a central authenticating server, the method comprises building communication connections by the multi-biometric features network sampling and controlling terminal and the central authenticating server, obtaining biometric features data through a biometric features sampling module, encrypting the biometric features data through a communication module, then, transferring to the central authenticating server through the remote network according to a TCP/IP protocol, matching the biometric features data with the biometric features data which is stored in the server by the central authenticating server to authenticate identities of biometric features holders, and sending control orders to the multi-biometric features network sampling and controlling terminal according to authenticating results. The invention enables a computer to be connected with a plurality of sampling equipment which is provided with biometric features as fingerprints, human faces, irises and retinas through the remote network in a mode of the central authenticating server.

The invention discloses a water discharge pipeline water quality automatic sampling. The water discharge pipeline water quality automatic sampling comprises a water quality monitoring center and a plurality of water sampling nodes, wherein the water quality monitoring center comprises a total controller as well as a first data processing unit and a local server which are connected with the total controller; each water quality sampling node comprises a sampling control box, a background controller and a second data processing unit; the sampling control box comprises a multi-mode communication adaptor, a power supply, a GPS positioner and a single chip microcomputer; a detection unit is provided; and the detection unit comprises an eddy-current sensor, a water sample collecting module and awater quality monitoring module. The water quality sampling nodes and the sampling control boxes are distributed by dividing areas, and information is subjected to block acquisition and block processing, so that sampling can be conducted rapidly, the sampled information is not in disorder, accurate positioning can be conducted when a component is maintained and changed, the water samples can be collected by four modes of collecting for many times and mixing the samples, sampling at fixed time, performing trigger sampling and performing network sampling, the collection modes are diversified, and the practicability is higher.

The invention discloses an SV network sampling-based double-circuit line single-phase earth fault distance measurement method which is used for measuring the fault phase voltage mutational volume, the fault phase current mutational volume, the zero-sequence current and the other two normal phase voltages of an I-circuit line at the protection mounting position of a double-circuit line on the same pole, and measuring the zero-sequence current of a II-circuit line; calculating the fault phase voltage when the I-circuit line operates normally, sequentially calculating the positioning vector function of each point on the I-circuit line of the double-circuit line on the same pole until the full length of the I-circuit line of the double-circuit line on the same pole, and selecting the distance between the point with the smallest positioning vector function value on the I-circuit line of the double-circuit line on the same pole and the protection mounting position of the double-circuit line on the same pole as the fault distance. According to the method, the voltage of the circuit before the fault does not need to be remembered, and a voltage protection loop of the circuit before the fault does not need to be set; the influence of the zero-sequence mutual inductive impedance among the lines, the transition resistance and the load current on the precision of measurement of the fault distance is eliminated, and no fault distance measurement dead zone exists in protecting the single phase grounding fault of the outlet single-circuit line.

The invention discloses a clock synchronization method of network sampling intelligent substation, wherein double master clock redundancy backups are used; the master clock is composed of a master clock A and a master clock B; an extended clock is composed of clock 1 to clock n; all extended clocks uniformly select the master clock A as the default time source, and then are uniformly switched to the standby master clock B when the master clock A becomes invalid. The invention solves the problems of intelligent substation synchronization system, puts forward a comprehensive solution for the clock synchronization system of network sampling intelligent substation, puts forward a combination mode of clock sources of the intelligent substation time synchronization system, and puts forward a way to track the satellite time and a seamless switching strategy between a GPS time source and a Beidou time source, greatly optimizes an abnormal fault tree structure of the intelligent substation time synchronization and protection malfunction fault tree structure caused by abnormal time synchronization, and can effectively improve the stability and reliability of the time synchronization system of intelligent substation.

The method comprises;for a certain king of data messages, according to a preset separating strategy the network device separates the data message in a accord with the separating strategy from said data messages; making the collection for the separated data message; the collected data message is packaged and then sends to the collector. The invention only makes processing for a certain data stream so as to avoid the processing inability of the collector due to the over massive data.

A method for alleviation of congestion in a mobile communications network includes detecting congested cells in the mobile communications network, identifying subscribers with active data sessions in the congested cells; and optimizing band-width usage for at least one of the identified subscribers. A bandwidth optimization system includes a network sampling interface to receive at least subscriber, cell and data session identifiers from a network data packet sampler, where the sampler identifies the identifiers from internal data traffic within a mobile communications network, and a network awareness engine (NAE) to at least cross reference the identifiers with external data traffic output by the mobile communications network to at least detect congested cells and associated subscriber data sessions emanating from the mobile communications network.

The invention discloses a bus protection system and method based on voltage phase locking networked sampling. The system comprises a PT merging unit, a plurality of spaced merging units, an interchanger and a bus protection device, the bus protection device comprises a FPGA, a message processor and a protection processor, and the method comprises the following steps of recording the arrival time of each frame of a sampling value message in real time, analyzing the sampling value messages, tracking the sending frequency of sampling values of the merging units, measuring network transmission extra delay errors, synchronizing the interpolation of the sampling values and performing relay protection on a bus. The bus is protected based on voltage phase locking networked sampling, the sampling values are output by the interchanger, the sampling value messages are output in a unified mode, failure analysis of a power system is facilitated, the number of the spaced merging units is set according to the requirement of a protected bus, the number is set flexibly, the engineering implementation is flexible, the synchronization logic and bus protection logic are relatively independent, and the system and method are suitable for being popularized and used in multiple areas.

The invention discloses a CAN network sampling point detection method, device and system and a storage medium. The CAN network sampling point detection method comprises the following steps: S1, activating a node network communication function, operating a tested controller, and dividing all bits contained in a message sent by the tested controller into Q time slices with predetermined values; S2,selecting data information of any bit unit in the message data, selecting Q time slices with predetermined values from all time slices in the selected data information of the bit unit, and carrying out interference operation on the Q time slices; and S3, recording the number P of all time slices before an error frame position occurs in the Q time slices, calculating a sampling point by using the number P of the time slices, the sampling point being P/Q, and continuously testing for multiple times to obtain an average value.

The invention discloses an HSR/PRP network sampling synchronization method based on data forwarding time-delay measurement, and the method comprises the steps: giving a timestamp to an SV message at the input and output ports of a basic communication unit RedBox of an HSR/PRP network; carrying out the measurement and accumulation of transmission time delay of the SV message on the RedBox, and writing the accumulated time delay into a message time domain, thereby achieving the precise measurement of the accumulated transmission time of the SV message among HSR rings and between the HSR rings and a PRP network; enabling an SV message receiving device (such as a protection device, a measurement and control device) to take a local clock as the reference, and deducing the transmitting moment of an SV message of a merging unit according to a value in a time domain of the received message, thereby achieving the synchronization of sampling data. The method can effectively solve a problem that the sampling synchronization of the HSR/PRP network depends on a public clock signal.

The invention discloses a cross-interval digital measurement method. The cross-interval digital measurement method comprises the following steps: taking the actual arrival time of a voltage sampling data value in any one of reports in any one of sampling periods meeting a preset condition during normal synchronization as a benchmark, calculating the relative time of two current sampling data values in the current report separately and taking the relative time as compensation time; according to the compensation time, taking any report in any sampling period during abnormal synchronization as abenchmark and calculating the synchronous interpolation time corresponding to each current sampling data value in the subsequent several continuous reports; according to the actual arrival time of thevoltage sample data value in the report serving as the benchmark and the synchronous interpolation time corresponding to each current sampling data value in the subsequent several continuous reports,calculating two current sampling data values at the same time of the voltage sampling data value in the report as the benchmark during abnormal synchronization. Data synchronization under the out-of-step condition can be realized by increasing a time measuring system in an electric energy meter and by using a synchronous compensation algorithm.

The invention provides a smart substation station domain protection control system framework which acquires all the spaced electrical quantity and switching quantity information of the whole station through network switches and realizes intra-station protection redundancy, optimization, supplementing and safety automatic control functions by adopting a network-sampling-network-tripping mode. The smart substation station domain protection control system framework comprises merging units and smart terminals which are spaced in the station, and a background monitoring system. A substation station domain protection control system is connected with at least one group of network switches via optical fiber interfaces. The acquired SV analog quantity transmitted by the merging units in the network and GOOSE switching quantity transmitted by the smart terminals are transmitted to the control system by the network switches. After parsing processing and protection control logic calculation of the control system, the GOOSE switching quantity requiring tripping is transmitted to the corresponding smart terminals. An MMS communication protocol between the background monitoring system and the control system is connected by the network switches. Different voltage levels of protection and control functions in the station can be realized.

The invention discloses a data source end synchronization method for improving the reliability of an intelligent transformer substation. In an SV networking mode, the synchronization method is improved from a data source, the relying of the whole intelligent transformer substation system on a time synchronization method is effectively reduced, the problems of whole station main protection exiting caused under the conditions of second loss, second hopping and the like during satellite signal receiving abnormality such as signal loss or low signal intensity and low signal-to-noise ratio are solved, and the operation reliability of a network sampling system is greatly improved. The improvement method has the advantages that the additional addition of the time synchronization system is not needed, the existing system structure is not changed, a time synchronization device and an MU need the improved design, and no requirement is required for a protection device, so that the applicability range is wider, and the large-area popularization condition is realized. The method is not applicable to the cross-transformer-substation application, such as wide area protection and optical fiber differential protection.

The embodiment of the invention discloses an automatic searching method and device for precision and decomposition rank of a recurrent neural network, and the method comprises the steps: initializing a super network, and carrying out the following operation: sampling a sub-network from the super network, and carrying out the performance evaluation of the sampled sub-network; automatically searching a target hyper-parameter combination according to a performance evaluation result, and updating hyper-parameters of the hyper-network according to the target hyper-parameter combination; aiming at the super network after the hyper-parameter adjustment, returning to execute the sub-network sampling operation, and performing the performance evaluation operation on the sampled sub-network; and outputting the current decomposition rank of the super network and the quantized value of each layer of network in response to the condition that the performance evaluation result meets a preset condition. In the embodiment of the invention, according to the performance evaluation result, the decomposition rank of the whole super network and the quantized value corresponding to each layer of network are not searched and updated, so that the purpose of quickly searching the optimal decomposition rank required by model compression and the optimal quantized value of each layer of network is achieved.

The invention discloses a low-speed port scanning detection method for a high-speed network sampling data collection scene, which comprises the following steps of: firstly, carrying out system sampling on a public data set, and then extracting traffic characteristics by utilizing a sketch structure comprising four counters and two hash tables; firstly, adopting a K-means algorithm to cluster traffic characteristics, then verifying and labelling the traffic in a cluster where a known scanning stream is located based on a rule, and then constructing a training set with a complete label; and finally, training a classification model for port scanning detection by using a supervised machine learning algorithm. The classification model detects low-speed port scanning activities of TCP and UDP in a high-speed network under a sampling data collection scene, and the method is still effective for low-speed scanning attacks with the duration time exceeding 50 days. According to the invention, detection of port scanning events in massive high-speed traffic is realized by using a limited memory, and the method is used by a network manager for safety event monitoring in a high-speed network.

The invention discloses a wave length mutual multicasting type process level optical network structure. The wave length mutual multicasting type process level optical network structure is characterized in that a plurality of groups of process level network devices are interconnected with a plurality of groups of bay level multicasting devices in a group-to-group correspondence manner through intermediate optical nodes, and all the process level network devices are connected with bay level semi-broadcasting or full-broadcasting devices; the intermediate optical nodes include optical access nodes and optical distribution nodes, wherein the optical access nodes are optical nodes on the process level side, and the optical distribution nodes are optical nodes on the bay level side; each group formed through the group-to-group correspondence interconnection includes at least two of the process level network devices and at least two of the bay level multicasting devices. The wave length mutual multicasting type process level optical network structure provided by the invention has the advantages that the required number of optical fibers is greatly reduced, the network structure is more orderly and simpler, the implementation is simple, and the maintenance is easy; through the wave length mutual multicasting of the intermediate optical nodes, the optical layer multicasting function and the sharing of network information on an optical layer are achieved, and the ''network sampling and network hopping'' method can be used; automatic network restructuring can be further achieved when optical cross connectors are adopted in the optical distribution nodes.

The invention provides a method and a device for determining a link prediction algorithm type. The method for determining the link prediction algorithm type comprises the following steps: acquiring asampling network obtained by performing network sampling on an initial network; calculating a multivariate closure probability corresponding to a multivariate closure node structure composed of a plurality of nodes in the sampling network; calculating a structured heterogeneous index of the sampling network according to the multivariate closure probability; and determining a link prediction algorithm type corresponding to the structured heterogeneous index in a preset link prediction algorithm type set, and taking the link prediction algorithm type as a target link prediction algorithm type ofa link prediction algorithm adopted for link prediction of nodes in the initial network. According to the link prediction algorithm type determination method provided by the invention, the target link prediction algorithm type of the link prediction algorithm for performing link prediction on the nodes in the initial network is determined by calculating the structured heterogeneous index of the sampling network, so that the selection range of the link prediction algorithm is reduced, and the efficiency of selecting the link prediction algorithm is improved.