RU82867U1 - DIAGNOSTIC SYSTEM FOR OIL-FILLED MEASURING TRANSFORMERS - Google Patents

Abstract

1. A diagnostic system for oil-filled measuring transformers for the concentrations of gases dissolved in the oil, comprising a chromatograph equipped with a sample preparation unit and connected via a data transmission interface to a data processing module, which is equipped with a common bus and digital processing, memory, input and output units connected to it, the chromatograph is configured to analyze the concentrations of hydrogen, ethylene, acetylene, methane, ethane, carbon monoxide and dioxide dissolved in a transformer oil sample, and the module data processing - with the possibility of coordinate-wise comparison of the vector of measured concentrations generated from the data received from the chromatograph with the vector of boundary concentrations selected from the memory block in accordance with the values of at least six characteristics of the diagnosed transformer entered through the input block, alternately sampling from the block memory vectors of abnormal concentrations corresponding to the purpose of the diagnosed measuring transformer, calculating the proximity measure of each choice nnogo abnormal concentrations vector to the vector of the measured concentrations exceeded, at least in one coordinate vector boundary concentration and outputting a signal corresponding to the at least one vector abnormal concentrations closest to the vector of the measured concentrations. ! 2. The system according to claim 1, characterized in that the data processing module is configured to select the boundary concentration vector from the memory block in accordance with the values entered through the input block characterizing the diagnosed

Description

Technical field

The utility model relates to systems (hardware and software complexes) designed for the diagnosis of oil-filled high-voltage equipment according to the results of chromatographic analysis of gases dissolved in oil (the common abbreviation KhARG), and can be used to diagnose oil-filled measuring transformers (current transformers and voltage transformers).

State of the art

One of the types of determination of the state of oil-filled high-voltage equipment is its diagnosis according to the results of HARG, which is carried out periodically during operation. Known diagnostic systems for oil-filled high-voltage equipment according to the results of HARG, containing a gas chromatograph equipped with a sample preparation unit, and a programmable data processing module connected by a data transmission interface [1-3]. Known diagnostic systems based on HARG results, as a rule, have been implemented (as reported in their advertising and operational documentation) with the possibility of diagnosing oil-filled transformers in accordance with the standard [4] governing the diagnosis of power transformers and their inputs.

As for the diagnostics of oil-filled measuring transformers (IT) according to the results of HARG, there are currently no Russian standards and norms for its implementation, but there are international IEC standards [5] and CIGRE recommendations [6]. Therefore, in practice, for the diagnosis of domestic transformers operating in the energy systems of Russia and the CIS countries

industries use systems that are executed (programmed) with the possibility of performing HARG diagnostics either by analogy with power transformers according to the Russian standard [4], or in accordance with international documents - IEC standard [5] or CIGRE recommendations [6].

IT diagnostic systems based on the standard [4] do not take into account the differences between IT and power transformers in operating conditions and in design (including the type of solid (paper) insulation and the ratio of its volume to oil volume), which result in differences in manifestations and causes of developing defects detected during diagnosis according to the results of HARG. IT diagnostic systems based on standard [5] or recommendations [6] do not differentiate diagnostic criteria depending on the type and performance of IT. As a result, the well-known IT diagnostic systems based on the results of HARG do not provide the reliability necessary for practice to identify a developing defect and do not allow to determine its type.

Utility Model Essence

The technical result of the utility model is to increase the reliability and information content of IT diagnostics. The proposed diagnostic system makes it possible to more reliably detect the presence of a developing defect in the measuring transformer and determine its type.

The subject of the utility model is a diagnostic system for oil-filled measuring transformers by the concentration of gases dissolved in oil, containing a chromatograph equipped with a sample preparation unit and connected via a data transmission interface to a data processing module, which is equipped with a common bus and digital processing, memory, input and output, while the chromatograph is configured to analyze the concentrations of hydrogen, ethylene, acetylene, methane, ethane, oxide and carbon dioxide,

dissolved in a transformer oil sample, and the data processing module - with the possibility of coordinate-wise comparison of the vector of measured concentrations, formed according to the data received from the chromatograph, with the boundary concentration vector selected from the memory block in accordance with at least six values entered through the input block characteristics of the diagnosed transformer, alternately sampling from the memory block vectors of abnormal concentrations corresponding to the destination characteristics of the diagnosed measuring about a transformer, calculating a measure of proximity of each selected vector of anomalous concentrations to a vector of measured concentrations exceeding at least one coordinate vector of boundary concentrations, and outputting a signal corresponding to at least one vector of anomalous concentrations closest to the vector of measured concentrations .

This allows you to get the above technical result.

The utility model has development, which consists in the fact that the data processing module is executed:

- with the possibility of sampling the vector of boundary concentrations from the memory unit in accordance with the values entered through the input unit characterizing the diagnosed transformer according to one of the five voltage classes, one of the three time ranges of the service life, one of two types of oil protection, one of the three types of oil, one of two types of solid insulation and one of two types of purpose;

- with the ability to calculate the specified proximity measure according to the rms criterion, weighted in accordance with the vector of abnormal concentrations;

- with the possibility of forming, storing and processing scaled concentration vectors with a scale in coordinates,

corresponding to carbon monoxide and dioxide, 100-200 times less than other gases, and

- with the ability to normalize and denormalize scaled concentration vectors by the sum of the coordinates.

The implementation of the utility model, taking into account its developments.

The block diagram of the diagnostic system is shown in figure 1. It contains: a chromatograph 1, equipped with a sample preparation unit 2 and an interface unit 3, and a programmable data processing module 4, equipped with a common bus 5 and connected to it digital processing unit 6, a memory unit 7, an interface unit 8, an input unit 9, and a unit 10 output. Chromatograph 1 is connected to module 4 via a data transfer interface, which is formed by interconnected interface units 3 and 8.

Chromatograph 1 is configured to analyze the concentrations of hydrogen, ethylene, acetylene, methane, ethane, carbon monoxide and dioxide dissolved in a transformer oil sample. The programmable module 4 is capable of coordinatewise comparing the vector of measured concentrations generated from the data received from chromatograph 1 with the boundary concentration vector selected from block 7 in accordance with the values of at least six characteristics of the diagnosed IT entered through input block 9. In addition, module 4 is capable of alternately extracting from the block 7 vectors of abnormal concentrations corresponding to the characteristic of the diagnosed IT, assigning a measure of the proximity of each selected vector to the measured concentration vector, which exceeds the boundary concentration vector by at least one coordinate, and outputs through block 10 of the signal corresponding to at least one vector of abnormal concentrations closest to the vector of measured concentrations.

Each boundary concentration vector corresponds to one possible combination of values of IT characteristics. Vector border

concentrations can have two levels: permissible and maximum permissible. In addition to boundary concentration vectors and anomalous concentration vectors, block 7 can store weight vectors. Each anomalous concentration vector and each weight vector correspond to one type of developing defect characteristic of IT with one of two values of the destination characteristic. For this, the vectors of abnormal concentrations and weight vectors are stored in block 7 in two groups, and block 4 selects the indicated vectors from the group that corresponds to the destination characteristic entered through block 9, i.e. selects a group depending on whether the current transformer (CT) or voltage transformer (VT) diagnosed by the IT is. The number of stored vectors of abnormal concentrations in these two groups may not be the same, for example, 9 vectors for TT and 3 vectors for VT.

The diagnostic system operates as follows.

Block 2 prepares for chromatographic analysis a sample of transformer oil taken from the oil-filled cavity of the diagnosed measuring transformer (block 2 can be performed, for example, according to utility model patent RU37831U1 IPC G01 25/14, 2004. “A device for preparing liquid for analysis on a chromatograph ").

Chromatograph 1 analyzes the concentrations of at least seven of these gases dissolved in a transformer oil sample using various detectors (for example, a thermal conductivity detector and a flame ionization detector). It measures (for example, sequentially) the concentrations of hydrogen, ethylene, acetylene, methane, ethane, carbon monoxide and dioxide dissolved in a transformer oil sample, and through the interface unit 3 provides the corresponding data (results of chromatographic analysis) to

data processing module 4. The interface unit 8 receives data from the chromatograph 1 and through the common bus 5 transmits them to block 6.

As a result of processing the obtained data in block 6, a vector of measured concentrations is formed, the coordinates of which are the values of the concentrations of these seven gases dissolved in transformer oil.

Through block 9, the following values of six characteristics of the diagnosed IT, from which the analyzed oil sample is taken, are entered (for example, by the operator) into module 4.

One input value characterizes the IT voltage class (this characteristic can have one of five values: 110, 220, 330, 500, 750 kV). The second value entered characterizes the time range of the IT operating life (this characteristic can, for example, have one of three values: up to 1.5 years, 1.5-19 years, more than 19 years). The third, fourth and fifth input values characterize one of two types of oil protection (IT with film protection or with free breathing), one of three types of oil (TKP, GK or T-750) and one of two types of IT purpose (TT or TN ) respectively. The sixth input value characterizes the type of solid insulation (paper-oil or paper-oil condenser type).

In accordance with the characteristic values entered through block 9, module 4 selects one of the boundary concentration vectors of the permissible level stored in block 7. Then block 6 carries out coordinate-wise comparison with the selected vector of boundary concentrations of the previously formed vector of measured concentrations.

If, in such a comparison, block 6 fixes that the measured concentration does not exceed the boundary concentration in any of the seven coordinates (i.e., for none of the seven analyzed gases), module 4 through block 10 outputs a signal informing the operator about the health of the diagnosed IT ( i.e. the absence of a developing defect in it).

If, in the indicated comparison, block 6 detects that the vector of measured concentrations exceeds the selected vector of boundary concentrations by at least one of the seven coordinates, module 4 outputs through block 10 a signal informing the operator about the presence of a developing defect in the IT being diagnosed, and proceeds to determine the type of such defect.

For this, module 4, in accordance with the value entered through block 9, characterizing one of the two types of destination of the diagnosed IT, determines one of two groups of abnormal concentration vectors stored in block 7 (a group of vectors for CTs or a group of vectors for VTs), selects the vectors of abnormal concentrations included in it and calculates in block 6 a measure of the proximity of each of the selected vectors to the vector of measured concentrations for which the above excess of the boundary concentration is recorded.

The proximity measure of two vectors can be calculated by block 6, for example, according to the rms criterion, i.e. in the form of a weighted sum of squares of the differences of the coordinates of the same name. Weights before the terms of this sum form a weight vector. Different weight vectors can be used for each vector of abnormal concentrations (and therefore for each diagnosed defect). Weight vectors are stored in block 7 and are alternately selected from it together with the corresponding vector of abnormal concentrations.

Remembering the results obtained and comparing them with each other, module 4 determines the vector of abnormal concentrations closest to the vector of measured concentrations obtained by chromatographic analysis. It is also possible to determine the next closest vector of abnormal concentrations. In this case, module 4 generates a corresponding signal. This signal, carrying information about the most probable defect (or, in difficult cases, about two such defects) is output via block 10 to indicate the name of the developing defect (for example, “partial discharges”, “discharges”, “high discharges”

energies accompanied by heating ”,“ partial discharges accompanied by medium heating ”,“ partial discharges accompanied by weak heating ”,“ aging, etc.) as a result of diagnostics.

After determining the type of developing defect, the vector of measured concentrations can be further compared with the vector of boundary concentrations of the maximum permissible level. If the concentration of at least one of the seven gases exceeds the maximum permissible level, then the message “dangerous stage of development of the defect” is added to the name of the developing defect, which is displayed through block 10.

In the process of performing the described functions (during the formation, storage, processing of vectors), module 4 can operate with scaled concentration vectors (measured, boundary, anomalous). Concentrations of carbon monoxide and dioxide in scaled vectors are reduced 100–200 times, and the concentrations of other gases are left unchanged. With this scaling, the coordinates of the vectors become values of the same order (the actual concentration of carbon oxides is 2 orders of magnitude higher than the concentrations of the other gases), which simplifies digital processing.

When performing the described functions, module 4 can operate with scaled concentration vectors presented in normalized or non-normalized form and convert vectors of one type to another. The coordinates of normalized vectors are presented in relative, and non-normalized - in absolute units of concentration (for example, in ppm or in µl / l). Normalization is carried out by the sum of all coordinates of the scaled vector. For this, module 4 can calculate the sum of the coordinates of the scaled concentration vectors and normalize them by dividing the coordinates of the non-normalized vector by the specified sum. Multiplying by

the corresponding amount of normalized vectors can be denormalized.

From the above it follows that the set of features of the proposed diagnostic system, which is described in the independent clause of the utility model formula, provides IT diagnostics:

- with extended differentiation of diagnosed parameters (according to six characteristics of measuring transformers) when determining boundary and abnormal concentrations (known systems for diagnostics according to the results of HARG take into account differences in power transformers [4] for three characteristics, and differences in IT characteristics [5], [6] not taken into account at all).

- with the determination of the type of defect by the vector of abnormal concentrations closest to the vector of measured concentrations (known systems do not determine the type of defect in measuring transformers);

This allows the proposed system to identify IT with a developing defect, with sufficient reliability for practice, and to provide information about the type of developing defect, which, in turn, allows us to assess the possibility of continued operation or the need to reject the diagnosed IT if the defect is in a dangerous stage.

The developments stipulated by the dependent clauses of the utility model formula characterize particular cases of its implementation and are aimed at making the diagnostic system more convenient for long-term operation, in which statistical data on dissolved gas concentrations and types of defects in various types of IT are accumulated and periodically taken into account.

Information sources

1. Calambet Yu.A. Software and hardware complex "Multichrom" for the automation of chromatographic analysis. Methods and Tools

assessment of the state of power equipment. 2001, issue 16, St. Petersburg, PEIPK, pp. 69-75.

2. Research and production company "Meta-chrome". Transformer oil analysis. Hardware-software complex based on an automated multi-channel gas chromatograph Kristallux-4000M. (Electronic resource). URL: http://www.meta-chrom.ru/chromatograms/transformer_oil.php (accessed December 1, 2008).

3. "MP Diagnostics." Diagnostic and measuring instruments. Instruments for monitoring transformer oil. Model TFGA-P200. (Electronic resource). URL: http://www.diagnost.ru/oil_l.htm (accessed December 1, 2008).

4. RD 153-34.0-46.302-00. “Guidelines for the diagnosis of transformer equipment defects based on the results of chromatographic analysis of gases dissolved in transformer oil” M., VNIIE OJSC, 2001

5. IEC standard. IEC 60599. Mineral oil-impregnated electrical equipment in service - Guide to the interpretation of dissolved and free gases analysis. 1999-03.

6. CIGRE Recommendations. Lewis K.G. et al. "Diagnostic methods for I / O CTs and inputs" Annual conference of clients of the American company DOBLE "Proceedings Failure Prediction and Prevention", Symposium, Oregon, pp. III-1-1-III-1-16, 2000

Claims (5)

1. A diagnostic system for oil-filled measuring transformers for the concentrations of gases dissolved in the oil, comprising a chromatograph equipped with a sample preparation unit and connected via a data transmission interface to a data processing module, which is equipped with a common bus and digital processing, memory, input and output units connected to it, the chromatograph is configured to analyze the concentrations of hydrogen, ethylene, acetylene, methane, ethane, carbon monoxide and dioxide dissolved in a transformer oil sample, and the module data processing - with the possibility of coordinate-wise comparison of the vector of measured concentrations generated from the data received from the chromatograph with the vector of boundary concentrations selected from the memory block in accordance with the values of at least six characteristics of the diagnosed transformer entered through the input block, alternately sampling from the block memory vectors of abnormal concentrations corresponding to the purpose of the diagnosed measuring transformer, calculating the proximity measure of each choice nnogo abnormal concentrations vector to the vector of the measured concentrations exceeded, at least in one coordinate vector boundary concentration and outputting a signal corresponding to the at least one vector abnormal concentrations closest to the vector of the measured concentrations. 2. The system according to claim 1, characterized in that the data processing module is configured to select the boundary concentration vector from the memory block in accordance with the values entered through the input block characterizing the diagnosed measuring transformer according to one of five voltage classes, one of three time ranges lifetime, one of two types of oil protection, one of three types of oil, one of two types of solid insulation and one of two types of purpose. 3. The system according to claim 1, characterized in that the data processing module is configured to calculate the indicated proximity measure according to the rms criterion weighted in accordance with the vector of abnormal concentrations. 4. The system according to claim 1, characterized in that the data processing module is configured to form, store and process scaled concentration vectors with a scale in the coordinates corresponding to carbon monoxide and dioxide, 100-200 times less than for other gases. 5. The system according to claim 4, characterized in that the data processing module is configured to normalize and denormalize scaled concentration vectors by the sum of the coordinates.