35results about How to "Obvious fault features" patented technology

The invention provides an online low-current earth fault positioning method and positioning device based on a BP neural network and by means of a multi-source data fusion algorithm. Due to the fact that terminals arranged on different positions on a circuit accurately catch zero-sequence current transient signals of zero-sequence currents which exceed a starting value before a previous period of a starting value and after four periods of the starting value, signal characteristics of a fault transient signals are extracted by means of a Prony algorithm, a wavelet packet algorithm, an HHT algorithm and a fractal algorithm, the characteristics extracted by the algorithms are analyzed by means of the BP neural network, and a zone where a fault occurs is determined. The positioning device comprises a terminal and a main station, wherein the terminal is arranged on an overhead line tower or in a cable ring main unit, the input end of the terminal receives a zero-sequence current signal synthetized by a phase current signal of a CT secondary side of a power distribution circuit including an overhead line and a cable, the terminal is connected with the main station through optical cable communication or mobile communication, and the main station is arranged in a transformer substation or a dispatch center, comprises an optical cable communication module and a mobile communication module and receives a signal sent by the terminal. The method and the device are mature in technology and high in reliability.

The invention discloses a locating method and device for a one-phase earth fault region of a power distribution network. The locating method comprises the following steps of: accurately acquiring zero sequence current transient state signals within two periods before and after a zero sequence current exceeds a starting value through terminals arranged in different positions on a line; carrying out wavelet transformation and reconstruction on the zero sequence current transient state signals by the terminals; and analyzing a fault point region according to an approximation coefficient integral value of a detail component after the reconstruction. The locating device disclosed by the invention consists of a terminal and a master station, wherein the terminal is arranged on an overhead line tower or in a cable ring network cabinet, the input end of the terminal is used for receiving a phase current signal at the secondary side of a distribution line CT, the phase current signal is synthesized to obtain a zero sequence current signal, and the terminal is connected with the master station through optical fiber communication or mobile communication; and the master station is arranged in a converting station or dispatching center, composed of an optical fiber communication module and a mobile communication module and used for receiving a signal sent by the terminal. The locating method and device are matured in technique and high in reliability.

Disclosed is a multi-message fusion section locating method based on an extreme learning machine. After a fault happens, terminals extract feature vectors from zero-sequence transient current measured in real time according to a transient energy method, a wavelet method and a first half-wave method, upload the feature vectors to a master station and input the feature vectors into the extreme learning machine network where weight parameters are obtained after being trained, and the main station starts the multi-message fusion locating algorithm and outputs a section locating result. As the terminals installed at the multiple positions of a circuit can detect the transient zero-sequence current signals in real time, so that the section locating method based on the transient messages can be applied to the actual engineering. The influence on the method by factors like the grounding position, the grounding time and the grounding transition resistance is small, the extreme learning machine is intelligently fused and comprehensively analyzes the zero-sequence current transient information after the fault, and various algorithms are fused to achieve the complementary effect. The inherent defects of a single locating method are overcome, the transient zero-sequence current characteristic can be fully utilized, the fault section can be determined accurately, and the smaller the space between the terminals is, the more accurate the locating is.

The invention provides a power distribution networksingle-phase grounding fault handling method based on a dual-mode function, and belongs to the technical field of powerdistribution automation. The technical scheme is mainly suitable for a delaylow-resistanceinputgrounding mode of a neutral point, and can be compatible with grounding modessuch as non-grounding of the neutral point, grounding of the neutral point through an arc suppression coil and grounding of the neutral point through low resistance. According to the scheme, a transientpower direction protection/grounding overcurrent protection dual-mode function is deployed for a first-stage section switch of a 10kV outgoing line of a transformer substation, thetransientpower direction protection/grounding overcurrent protection dual-mode function or a voltage-time type feeder automation function is deployed for a later-stage section switch, and when grounding faults occur, rapid handling of single-phase grounding faults is realizedby logic cooperation of all stages of switches.

The invention provides a multi-domain semi-supervised fault diagnosis method for an axial plunger pump bearing, and the method comprises the steps: obtaining a source domain signal and a target domain signal, and carrying out the conversion of the source domain signal and the target domain signal, and obtaining a source domain sample and a target domain sample; wherein the source domain signal is a vibration signal of known partial fault information under a certain working condition; the target domain signal is a vibration signal of fault information of an unknown part under another working condition; performing source domain semi-supervised fault diagnosis on the source domain samples after the time-frequency transformation processing by adopting the trained semi-supervised fault diagnosis model to obtain all marked fault samples of the source domain; importing all the marked fault samples of the source domain and the target domain samples subjected to time-frequency transformation processing into a trained multi-target domain transfer learning network at the same time for fault diagnosis of each domain to obtain diagnosis results of all the samples; wherein the diagnosis result comprises normality, an inner ring fault, an outer ring fault and a rolling body fault. By implementing the invention, the problem of cross-domain fault diagnosis of the existing axial plunger pump bearing can be solved.

The invention discloses a series arc detection method for a low-voltage power distribution network, and belongs to the field of electrical engineering measurement. The method is characterized by comprising the following steps: step a, setting monitoring nodes in a to-be-monitored line; step b, synchronously viewing the to-be-detected line and acquiring the voltage at the two ends; step c, performing filtering processing on the Uo; step d, obtaining a Us-Uo waveform; step e, executing the next step if the series arc fault occurs, and returning to the step b if the series arc fault does not occur; step f, counting the number of faults; and if a preset threshold value is exceeded, cutting off the circuit or sending an alarm signal; if not, returning to the step f. According to the series arc detection method for the low-voltage power distribution network, the series fault arc is detected by measuring the voltage difference between the head end and the tail end of the line, the method for judging the existence of the series fault arc by using the voltage information has the advantages that the influence of the load type and the load power is small, and the fault characteristics are more obvious, so that the judgment accuracy is higher.

The invention discloses a series arc fault on-line detection method and device, and the method employs loop current and power voltage signals of a three-phase motor, carries out the grouping of collected signals to determine the number of phases where an arc fault happens, and facilitates the quick troubleshooting of the arc fault. Then power supply harmonic waves and load noise in the current andvoltage signals can be separated by using a kernel principal component analysis method, a fifth principal component and a sixth principal component with obvious fault characteristics are selected tocarry out characteristic analysis, and the kurtosis and skewness of the fifth principal component and the kurtosis and skewness of the sixth principal component are taken as four-dimensional fault characteristic vectors; a firefly algorithm is adopted to optimize penalty coefficients and kernel function parameters of a support vector machine, a feature vector set is used for training the optimizedsupport vector machine to obtain a series fault recognition model, loop current and power supply voltage signals are collected in real time and input into the series fault recognition model to achieve real-time detection of series faults. The method has a fault positioning function and is high in fault identification accuracy.

The invention relates to a gearbox fault feature extraction method based on a scale space and improved sparse representation. The method comprises the following steps: dividing an original signal into a plurality of frequency band component signals by using a scale space according to a frequency band, wherein the frequency band component signals take the meshing frequency of each order of a gear as a center; constructing a noiseless DCT dictionary for the frequency band component signals; constructing an observation dictionary used for performing sparse decomposition on the frequency band component signals, including adding noise signals into atoms of a noiseless DCT dictionary in an orthogonal matching pursuit algorithm iteration process based on a singular value decomposition algorithm to obtain a noisy DCT dictionary; selecting an optimal atom in the observation dictionary by taking a fusion index of the minimum variance and the inner product as a criterion; based on an orthogonal matching pursuit algorithm, using the observation dictionary to carry out sparse reconstruction on the frequency band component signals; and analyzing the demodulation spectrum and then extracting the fault characteristic frequency of the gearbox. The method is used for extracting the fault features of the gearbox, and has higher time-frequency resolution and reconstruction precision compared with other methods.

The invention provides a gear box fault diagnosis method based on a PSOOBP-CS algorithm, which is characterized in that gear box fault feature data is extracted based on ICEEMDSN and MPE algorithms, existence of intermittent components is judged firstly, and a corresponding decomposition algorithm is adopted accordingly, so the operation efficiency is improved, and the residue of white noise added in the signal decomposition process by only adopting the ICEEMDAN algorithm in the traditional feature extraction method is reduced. According to the invention, test data is collected through a gear case working condition data collection device and transmitted to a fault diagnosis unit in a computer, an Elman-TCL neural network stored in the fault diagnosis unit is optimized by using a PSOOBP-CS algorithm, the search capability of PSO and the global search capability of CS are superposed, the particle speed and position in the PSO algorithm are optimized, and the particle global optimal position of the PSO algorithm is optimized, so the defects of premature convergence and easy falling into local optimum are overcome, and the fault diagnosis effect is good.

The invention discloses a protection method for a turn-to-turn short-circuit fault of a controllable reactor. The protection method comprises the following steps of 1, acquiring the electrical quantity of the controllable reactor, namely the grid-side three-phase voltage, three-phase current and the control-side three-phase current; 2, calculating the control side zero-sequence current, the network side negative-sequence voltage and the negative-sequence current according to the electrical quantity; 3, taking the control side zero-sequence current as a protection starting threshold to carry out protection starting judgment; 4, calculating a negative sequence power direction angle by using the grid-side negative sequence voltage and the negative sequence current, and taking the negative sequence power direction angle as a judgment basis of a fault position; and 5, when a small turn-to-turn fault occurs, processing the negative sequence power direction angle by using a dynamic mean value method, and then judging the fault occurrence position in the system. According to the protection method, the control side zero sequence current and the grid side negative sequence power direction principle are combined, so that the sensitivity of the control side turn-to-turn short circuit protection of the controllable reactor can be improved; the problem of negative sequence power direction angle fluctuation during small turn-to-turn short circuit is solved by adopting a dynamic mean value method, and the protection reliability is improved.