87 results about "Dissolved gas analysis" patented technology

The invention discloses an on-line transformer fault diagnosis method based on support vector machines (SVM) and dissolved gas analysis (DGA), which belongs to the field of transformer fault diagnosis. The method comprises the following steps of: acquiring the concentration of fault feature gases including H2, CH4, C2H2, C2H4 and C2H6 in transformer insulating oil by a gas chromatographic analysis method, normalizing the data through a preprocessing system, sending the data to a classification diagnosis system assembled by an integrative learning method and formed by six SVM classifiers according to a decision process, classifying the measurement data through calculation, judging the running state of the transformer, and at last outputting the diagnosis results. In the method, the support vector machines (SVM) of the artificial intelligence technology are adopted to analyze the gases in the oil, and the relation between gas components and the running state of the transformer can be objectively and essentially reflected, thereby the accuracy of fault diagnosis is effectively improved.

The invention relates to a transformer fault diagnosis method based on a Bayesian network. According to the method, gas dissolved in oil of a transformer is analyzed by adopting a three-ratio method; data about gas is obtained in a real operation environment; study of structures and parameters of the Bayesian network is accomplished by adopting a TAN (Tree Augmented Naive) algorithm; a fault diagnostic model is established, and an expert system is utilized for correcting the fault diagnostic model; and the fault diagnostic model is used for diagnosing real-time operation states of the transformer. The method has the benefits that the problem about fault diagnosis for the transformer under the condition of uncertainty and lacking given information is solved, and meanwhile, an importance analytical method based on the Bayesian network is introduced to play a certain assistant role in analysis of the fault mechanism. The method can quickly and accurately diagnose the fault of the transformer, provide support for establishment of a maintenance decision for the transformer, effectively improve the maintenance efficiency, and lower the operation cost of a power system.

The invention discloses a dynamic correction method for reliability assessment of a large-sized oil-immersed power transformer; a transformer oil-paper insulating system is used as an assessment object, and two processes are involved; in the first process, with hotspot temperature calculation as the core, an HST-based (hotspot temperature) transformer aging fault model is established in conjunction with Weibull distribution and Arrhenius reaction law, winding HST is acquired by calculating, and fault rate gamma of the transformer oil-paper insulating system is solved; in the second process, with gray target correction analysis via dissolved gas in oil as the core, a relationship between transformer health state and its predicted life is established via grey theory, equivalent HST value is obtained, dynamic correction is provided for the basic model in the first process, and it is ensured that the assessment value is capable of well tracing and reflecting actual reliability level of the transformer. By using the method, it is possible to effectively solve the problems that, for instance, a transformer reliability analysis model has low accuracy and is difficult to popularize and apply.

The invention discloses an oil-immersed transformer fault diagnosis method based on a fault probability distribution model, aiming to provide the probability of various faults of the oil-immersed transformer according to analysis of dissolved gas in oil. The oil-immersed transformer fault diagnosis method includes the steps of collecting massive analysis data of the dissolved gas in the oil of the transformer and normalizing the data, utilizing a Sigmoid function to realize probability output on the basis of a support vector machine, computing according to the collected massive analysis data of the dissolved gas in the oil to acquire optimal parameters of the support vector machine and the Sigmoid function, fusing the probability output of multiple support vector machines into probability output of multi-class problems, and obtaining a diagnosis result by referring to the maximum probability classification and standard deviation. The oil-immersed transformer fault diagnosis method based on the fault probability distribution model has high accuracy without error diagnosis, can be used for processing multiple fault conditions effectively and has high practicability.

The invention discloses a method for discharge fault diagnosis of a transformer. The method is characterized in that a local discharge parameter and a parameter of gas dissolved in oil which relate to online detection of the transformer are combined, the two parameters are organically combined through a neural network, and combined diagnosis is performed on the type of a discharge fault happening in the transformer through the detected local discharge parameter and the detected parameter of gas dissolved in oil. According to the method, the accuracy for the discharge fault diagnosis is improved, and limitation of analysis of a local discharge signal and limitation of analysis of the gas dissolved in oil in the fault diagnosis are overcome; when the artificial neural network is established, training on the neural network can be completed by doing a series of typical experiments, and the method is easy and convenient to achieve.

The invention discloses a transformer fault diagnosis method for optimizing a random forest model based on a particle swarm algorithm. The method comprises the steps: dividing a training set and a test set by taking a non-coding ratio of analysis data of dissolved gas in transformer oil as characteristic vector input; constructing a random forest model, and optimizing the random forest model through a particle swarm optimization algorithm to obtain two optimal parameters; and rebuilding a random forest model according to the obtained optimal parameters to identify the fault type of the transformer. According to the method, the fault diagnosis accuracy of the transformer is effectively improved, and a reliable basis is provided for operation and maintenance personnel to correctly judge theoperation condition of the transformer.

A distribution transformer fault diagnosis device based on multi-index comprehensive evaluation includes a sound acquisition unit, an oil temperature acquisition unit, a unit for analyzing gas dissolved in oil of a transformer, a fault diagnosis unit and a battery power supply module; the fault diagnosis unit includes an FM receiver module (7) whose output end is connected with an AD converter (8) and an ARM single-chip microcomputer (9) in sequence; and the sound acquisition unit includes a fixed support (2) on which a strong magnet (1) and a pickup head (3) are fixed, the pickup head is connected with a pickup (4), an output end of the pickup is connected with a signal conditioning circuit (5) and an FM transmitter module (6) in sequence, and the FM transmitter module is connected with the FM receiver module through FM electromagnetic waves. The distribution transformer fault diagnosis device based on multi-index comprehensive evaluation can satisfy requirements for performing fault detection on a distribution transformer in operation on site and judging the fault type, does not need a power supply to be provided on site, and is convenient to carry, low in signal acquisition and transmission cost, fast in diagnosis speed and convenient to use.

The invention discloses a transformer fault diagnosis method based on Bi-LSTM and analysis of dissolved gas in oil. The method comprises: collecting fault DGA monitoring data of each substation, carrying out normalization, sequence expansion, noise superimposing and the like on the data, and extracting fault feature information based on a non-coding ratio method; carrying out length ranking on a DGA sequence, carrying out grouping and filling, and classifying groups into a training set and a verification set; constructing a deep learning frame based on Bi-LSTM, inputting data, and carrying outtraining; and then carrying out diagnosis and network updating by combining actual test data to obtain a fault diagnosis model with the high diagnosis accuracy and portability. According to the invention, the influence of the noise and error on the diagnosis during the DGA data monitoring process is reduced effectively; and the Bi-LSTM-based transformer fault diagnosis model is constructed by considering the complex correlation between different sequences. With introduction of links of sequence sorting, grouping, filling and the like, a problem of different sampling lengths of different transformers in the actual engineering is solved by using the batch training strategy.

The invention provides a novel power transformer fault combined diagnosis model building method and a fault diagnosis method; the diagnosis model building method comprises the following steps: using at least two of the following diagnosis models to primarily diagnose the oil dissolving gas analysis data: a Naive Bayes diagnosis model, a RVM diagnosis model and a matter-element diagnosis model; carrying out weighted average for the primary diagnosis result, using the secondary RVM diagnosis model to make the secondary diagnosis, thus obtaining the combined diagnosis model; using the combined diagnosis model to make fault diagnosis. The method can fully utilize advantages of various diagnosis models, thus effectively improving the diagnosis precision and validity.

A system and method for field fluid sampling and analyzing dissolved gases in electrical equipment fluids is described. The system includes a disposable vial which is sealed with a self-sealing stopper. Fluid to be analyzed is inserted into the vial by a hollow needle. Also described is apparatus for evacuating and sealing the vial and apparatus for introducing fluid from electrical equipment into the vial.

A multi-functional gas chromatograph gas path flow, the gas path flow uses two inlets, three chromatograph columns, one two-arm thermal conductivity detector, two hydrogen flame detector, and two inlets parallel connecting to the carrier gas inlet, behind the first inlet, parallel connecting two chromatograph columns, one chromatograph column linking the hydrogen flame detector, another chromatograph column linking the nickel catalyst furnace and the second hydrogen flame detector through the first arm of the thermal conductivity detector; the second inlet followed by the third chromatograph column and the second arm of the thermal conductivity detector. The invention using one thermal conductivity detector, two hydrogen flame detectors, to process oil dissolved gas analysis, a sampling can complete seven gases analysis, simple operation and short analysis time; when processing oil gas content analysis, the gas path need not any changes, two times sampling can complete nine gases analysis, and it can process not only oil dissolved gas analysis in the electricity transformer, but also oil gas content analysis.

The invention relates to a transformer fault diagnosis method based on support vector description and K periapse neighbor. The method comprises the following steps: acquiring content data of characteristic gas H2, CH4, C2H6, C2H4 and C2H2 dissolved in transformer oil by virtue of a technology for analyzing dissolved gas in the transformer oil, preprocessing and normalizing the data, forming a transformer fault diagnosis sample set, establishing a multi-classification suprasphere model based on a single-classification support vector description method, diagnosing a fault of a transformer, further classifying an aliasing domain sample in a support vector description multi-classification process by using a K periapse neighbor classification algorithm, so that the fault diagnosis accuracy of the transformer is increased. By combining the multi-classification support vector description method and the K periapse neighbor classification method, small samples and imbalance sample sets can be accurately classified, the fault diagnosis precision of the transformer can be remarkably improved, and powerful support can be provided for making a state maintenance decision of the transformer.

A device for analyzing gas dissolved in power transformer oil based on Raman technique is featured as connecting sample cell with its optical system to gas compressing device by piping connecting spectrograph to computer through communication line for data exchange, fixing sample cell with its optical system and laser source on an installation plate. The sample cell with its optical system and spectrograph are independent units as they can exchange information to each other by lights.

A system, comprising at least one source for irradiating electromagnetic radiation into a sample fluid and a reference fluid resulting in a change in a temperature of the sample fluid and a change in a temperature of the reference fluid, and a processing subsystem that monitors and determines a concentration of a gas of interest dissolved in the sample fluid based upon a difference between the change in the temperature of the sample fluid and the change in the temperature of the reference fluid, wherein the reference fluid does not contain the gas of interest, and the electromagnetic radiation has a wavelength range corresponding to a spectral absorption range of the gas of interest.

The invention discloses a gas-dissolved-in-vegetable-insulating-oil analyzer with an oil-gas separation function and a detecting method. The gas-dissolved-in-vegetable-insulating-oil analyzer with the oil-gas separation function and the detecting method are characterized in that helium serves as carrier gas, and the vacuum degassing and minitype thermal conductance chromatographic instrument combined technology is utilized for analyzing hydrogen, oxygen, nitrogen, carbon monoxide, carbon dioxide, methane, ethylene, ethane, dimethylmethane and acetylene in vegetable insulating oil. A four-chromatographic-column analysis system is adopted, a valve is utilized for column switching, gas pump circulating gas is adopted for sample introduction, all samples enter a chromatographic column A at a time, meanwhile, a chromatographic column B and a chromatographic column D are connected to the tail end of the chromatographic column A in series, and after hydrogen, oxygen, nitrogen, methane and carbon monoxide are exhausted, the original state of a ten-way valve is switched. Carbon dioxide, ethylene, acetylene, ethane and dimethylmethane firstly generate peaks on a channel A of a thermal conductivity detector, and hydrogen, oxygen, nitrogen, carbon monoxide and methane generate peaks on a channel B of the thermal conductivity detector. The instrument is high in sensitivity to the gas and suitable for being applied and popularized in smart power grids.

Embodiments of the disclosure relate to systems and methods for monitoring and diagnosing transformer health. In one embodiment, a system incorporating a diagnostic apparatus for monitoring a power transformer can be provided. Various electrical current sensing elements and a dissolved gas analysis (DGA) apparatus are coupled to the transformer and to the diagnostic apparatus. The diagnostic apparatus can be configured to detect a through-fault in the transformer by executing an electrical current flow analysis based at least in part on electrical current values received from the electrical current sensing elements. The electrical current flow analysis involves comparing a ratio of a differential electrical current value and a restraining electrical current value to a threshold value. The diagnostic apparatus can also use DGA data provided by the DGA apparatus to detect an abnormal gas-related condition in the transformer. A transformer health related signal and / or a control signal that is based on the transformer health data, can be transmitted by the diagnostic apparatus.

The invention discloses a detection method for analysis of gas dissolved in insulating oil in high voltage electric oil filling equipment. An instrument system capable of performing single column separation, automatic feeding, program control switching and analysis by a single detector is adopted. The detector is a specific semiconductor combustible gas detector. One six-way valve and a quantitative sampling ring as a feeding mechanism are arranged inside the analysis system and are used for achieving the functions of sampling and feeding. The method specifically comprises the following steps: in a base line state, connecting one end of the quantitative ring with sample gas to be measured through an electromagnetic valve, connecting the other end of the quantitative ring with a micro vacuum pump and the electromagnetic valve, and directly feeding carrier gas into the detector through a chromatographic column; in a feeding state, firstly, vacuuming the sampling quantitative ring, when the vacuum degree is 10KPa, closing the electromagnetic valve of the vacuum pump, and switching on the electromagnetic valve at the sample gas end; subsequently turning the six-way valve, accessing the quantitative sampling ring to a carrier gas loop and accessing to the front end of the chromatographic column; and flowing H2, CO, CH4, C2H4, C2H6 and C2H2 into the detector for detection after chromatographic column separation.

The invention discloses a degasser applied to analysis of a dissolved gas in oil by a photoacoustic spectroscopy principle so as to separate oil and gas of transformer insulation oil. The degasser is mainly composed of an oil cylinder, a gas cylinder, a gas chamber, oil chambers and a photoacoustic cavity. Characteristic gas separated by vacuum degassing can be transferred into the photoacoustic cavity to be used in photoacoustic spectrometry. The oil chambers, the gas chamber and the photoacoustic cavity of the degasser are special chamber components, wherein each oil chamber is of a double-hole bottom-cover structure, and oil can be returned safely in cooperation with a liquid level sensor. Two symmetrical oil chambers are arranged on the degasser to achieve the functions of dual-direction oil return and oil channel cleaning; meanwhile, single-point oil sample collection can be prevented, and the multi-point circulation collection of the oil sample can be realized.

The invention relates to an automatic analyzer for dissolved gas in insulating oil. The gas analyzer comprises a degassing mechanism, an autoinjection mechanism, a chromatogram mechanism and a circuit control part, wherein the autoinjection mechanism comprises a degassing disc, a degassing motor, a background light source and a camera, the degassing motor drives the degassing disc to rotate, N vertical syringe mounting through holes are uniformly arranged in the circumference of the upper surface of the degassing disc, and the background light source and the camera are used for detecting the volume of insulating oil in syringes; the autoinjection mechanism comprises a gas feeding motor, a belt, a gas feeding head support, a gas feeding head and a push rod motor, the gas feeding motor drives the gas feeding head support to move up and down via the belt, and the gas feeding head is mounted on the gas feeding head support; and when the degassing disc rotates to a set position, the upper end of a to-be-detected syringe on the degassing disc just corresponds to the gas feeding head, the lower end of the to-be-detected syringe just corresponds to the upper end part of a push rod of the push rod motor, and a hollow syringe needle in the gas feeding head is connected with the chromatogram mechanism. The automatic analyzer has a high automation degree and small errors.

A method for dissolved gas analysis is presented. The method includes the steps of irradiating a fluid with electromagnetic radiation; and determining a concentration of a gas as a function of a temperature change of the fluid in response to the irradiation. A device for such an analysis of dissolved gases in a fluid, and a system having such device are also described.

The invention relates to the technical field of power equipment state monitoring and fault diagnosis, and specifically relates to a clustering analysis method for the state monitoring data of a transformer. The invention takes the lead in proposing an analysis method for gas dissolved in oil which gives consideration to the state of the transformer under the same operation conditions (the same service age section, the same voltage level and the same load rate), overcomes a key difficulty in inspecting a latent fault of equipment, and improves the safe and stable level of a power grid. An inspection system which covers the data extraction of the dissolved gas in oil, the cleaning fusion and clustering analysis is successfully developed, effectively avoids a passive situation of the frequentfault alarm and alarm leakage of the state monitoring of the transformer, and achieves the integrated optimization of the risk, efficiency and cost with one action.

The invention discloses a transformer fault diagnosis method based on deep learning, and belongs to the field of transformer fault diagnosis. The method comprises the following steps: firstly, carrying out duplicate removal and abnormal value processing on concentration data of fault characteristic gases H2, CH4, C2H2, C2H4 and C2H6 collected by an analysis method of dissolved gas in oil, fillingmissing values by using a random forest method, and then carrying out normalization processing on the data to form a training set sample and a test set sample; establishing a three-layer stack type sparse noise reduction auto-encoder model, and rewriting a cross entropy loss function in a traditional classification model into a Focal loss function; according to the method, hyper-parameters are determined through class sample weights, white Gaussian noise is added into input, an auto-encoder is made to fully extract effective features, and therefore an effective feature extraction model is obtained, and a Softmax classifier is used for outputting a diagnosis result of the model. Compared with the existing methods such as a three-ratio method, an SVM and a BP neural network, the transformerfault diagnosis method provided by the invention has good diagnosis performance, and the accuracy of transformer fault diagnosis is effectively improved.

The invention discloses a transformer fault diagnosis method based on a deep forest model. The method comprises the steps that a non-coding ratio of analysis data of dissolved gas in transformer oil is taken as a characteristic parameter of the deep forest model, and sample data is divided into a training set and a test set; and then a deep forest model DF is constructed, the deep forest model DFextracts more feature information from multi-dimensional data of a transformer fault through multi-granularity scanning, and the effect of diagnosing and identifying the fault type of the transformeris optimal through cascade forest training. According to the method, the fault diagnosis accuracy of the transformer is effectively improved, and a reliable basis is provided for operation and maintenance personnel to correctly judge the operation condition of the transformer.

A system and a process for detecting early stage electrical components problems with an analysis system is disclosed. The system process includes determining an oil quality of the oil of an electrical component using the analysis system. The system the process further including determining dissolved gases in the oil of the electrical component using the analysis system, processing and analyzing the oil quality and the dissolved gases using the analysis system, and determining whether there are problems in the electrical component using the analysis system.

The invention discloses a method for evaluating a health state of a reactor based on Bayesian theory. The method comprises the following steps: step 1, performing oiling test, dissolved gas analysis test in oil, and vibration noise measurement test on the reactor to obtain true values of obvious types of experimental data; step 2, establishing a reactor fault diagnosis model based on the Na ve Bayesian network; step 3, selecting a rising-half-ladder model to normalize the real values; step 4, determining evaluation indexes and state levels; step 5, determining membership functions of differentevaluation indexes about different state levels; step 6, establishing an unbiased gray prediction model to evaluate the future state of the reactor; and step 7, establishing a comprehensive health state diagnosis model of the reactor based on Bayesian network. The evaluating method can provide technical support for the transition of reactors from periodic preventive maintenance to state maintenance.

The invention discloses a transformer electrical fault gas simulation device, and relates to the field of the transformer fault diagnosis. The device comprises a tank body, a pressure balance releasing device, a temporary storage tank and a tee valve arranged at the top of the tank body; the tee valve is located at the top of the tank body, the tee valve is used for connecting the tank body, the pressure balance releasing device and the temporary storage device or any two of them; an electrode system for applying the electric field is arranged in the tank body, and the electrode layer comprises a high-voltage electrode and a ground electrode. The device can be used for electric fault simulation of the transformer to guarantee that the test process is completely isolated from the atmosphereinterference and the simulation test of different insulation mediums and different electric fields can be performed under the same test environment. The device provides basic reference data for gas dissolved in oil analysis of the novel transformer oil, thereby prejudging whether fault potential is existent when the novel transformer oil transformer is in operation through the gas dissolved in oil analysis technology, and quickly diagnosing the fault reason after the fault occurs.

A method for analysing gas dissolved in power transformer oil based on Raman technique includes measuring out Raman spectrum generated by exciting various gas with laser, calibrating methematical relation of each gas characteristic spectrum line position and concentration to spectrum intensity to obtain Raman spectrum wavelength and intensity as wavelength being a set valve and intensity being changed direct ratio with changing of concentration confirming gas concentration by comparing Raman spectrum line chart with characteristic spectrum line.

The invention relates to a thermal conductivity detector and a gas chromatograph dedicated for analysis of gas dissolved in insulating oil, wherein the thermal conductivity detector comprises a thermal conductivity cell, a thermal conductivity wire, as well as a lead frame, a work frame and an insulating base that are used for fixing and mounting the thermal conductivity wire; the thermal conductivity wire is composed by a rhenium tungsten filament and a nickel-plated rhenium tungsten filament that are connected in series and arranged in two testing arms respectively; and the thickness of the nickel plate of the nickel-plated rhenium tungsten filament is 4.2 micrometers to 4.6 micrometers. The thermal conductivity detector disclosed by the invention can realize the low efficient nickel catalyst conversion effect, converts portions of carbon monoxide and carbon dioxide into methane, and is normally used for determination of dissolved gas components (carbon monoxide and carbon dioxide) in the insulating oil in an oil-immersed electrical apparatus. The invention can significantly simplify air paths of the gas chromatograph, and requires no longer the configuration of a nickel catalyst converter.