Monitoring system for dissolved gas in transformer oil in alpine region

Abstract

A monitoring system for dissolved gas in transformer oil in an alpine region belongs to the technical field of power grid equipment state monitoring, aims to solve the problem that all gas in the transformer oil in the alpine region cannot be quickly and efficiently detected in the prior art, and comprises the following steps: a key characteristic gas screening module selects two gas components as key characteristic gas; the low-temperature data error compensation module is used for building a dissolved gas concentration simulation platform, collecting key characteristic gas concentrations in a non-low-temperature environment and a current operation temperature environment respectively, and calculating to obtain a parameter deviation correction coefficient; the dissolved gas concentration monitoring module collects the concentrations of all dissolved gases, and adopts a parameter deviation correction coefficient for compensation to obtain the corrected monitoring results of all dissolved gases; and the dissolved gas concentration prediction module constructs a BP neural network by taking the corrected monitoring results of all the dissolved gases as input data, and outputs concentration prediction results of all the dissolved gases. The method is used for monitoring transformer oil dissolved gas in a low-temperature environment.

Description

Technical Field

[0001] The invention relates to a monitoring system for dissolved gas in transformer oil in high-cold areas, belonging to the technical field of power grid equipment status monitoring. Background Art

[0002] In order to meet the growing demand for electricity, the development and utilization of power resources is gradually accelerating. As a key equipment for power transmission, the installation volume of transformers is also increasing. Oil-immersed transformers, as a new type of high-performance transformer with a more reasonable structure and better performance, are also suitable for power grid transformation in urban and rural areas, industrial and mining enterprises. Therefore, their demand is increasing day by day.

[0003] When an oil-immersed transformer is operating normally, gas will dissolve in the transformer oil. When a transformer fails, the dissolved gas composition in the transformer oil will change significantly. Therefore, by monitoring the dissolved gas composition and concentration in the transformer, it is possible to maintain the transformer operation and predict its status.

[0004] In the process of detecting dissolved gas in transformer oil in high-altitude cold areas, the following problems exist:

[0005] 1. The transformer oil and gas monitoring environment is complex. The gases in the transformer oil include H2, CO, CO2, CH4, C2H2, C2H4, C2H6, etc., and also other high-order organic gases. The gas composition is complex, and the content of some gases is relatively small, which makes it difficult to detect the target gas and the detection accuracy is relatively low;

[0006] 2. In high-cold areas, the sensitivity of the sensing material is reduced and the stability is poor. How to achieve a rapid and sensitive response of the gas dissolved in the transformer oil in the low-temperature oil and gas environment of the transformer is a problem that is difficult to solve in the existing technology.

[0007] Therefore, how to detect all dissolved gases in transformer oil in high-altitude and cold areas is the key to transformer operation, maintenance and status prediction. Summary of the invention

[0008] The purpose of the present invention is to solve the problem that the prior art cannot solve the rapid and efficient detection of all gases in transformer oil in high-cold areas, and to provide a monitoring system for dissolved gases in transformer oil in high-cold areas.

[0009] The present invention discloses a monitoring system for dissolved gas in transformer oil in high-cold areas, which comprises:

[0010] The key characteristic gas screening module is used to detect the concentration of all dissolved gases in transformer oil in a non-low temperature environment, sort them from high to low according to the concentration ratio, and select the gas components corresponding to the top two concentrations as key characteristic gases;

[0011] Temperature monitoring module, used to monitor the current operating temperature of the transformer;

[0012] The low-temperature data error compensation module is used to build a dissolved gas concentration simulation platform, collect the concentration of key characteristic gases in a non-low-temperature environment and the current operating temperature environment, and calculate the parameter deviation correction coefficient;

[0013] The dissolved gas concentration monitoring module is used to collect the concentration of all dissolved gases and use the parameter deviation correction coefficient to compensate and obtain the corrected monitoring results of all dissolved gases;

[0014] The dissolved gas concentration prediction module is used to take the corrected monitoring results of all dissolved gases as input data, construct a BP neural network, and output the concentration prediction results of all dissolved gases.

[0015] Preferably, all dissolved gases include: H2, CO, CO2, O2, Ar, N2, CH4, C2H2, C2H4, C2H6, C2H8, C3H6, C4H8 and C4H 10 .

[0016] Preferably, the calculation to obtain the parameter deviation correction coefficient specifically includes:

[0017]

[0018] Among them, C i1 Indicates the concentration of the gas component corresponding to the first concentration collected in a non-low temperature environment as the concentration of the key characteristic gas.

[0019] C i2 Indicates the concentration of the gas component corresponding to the first concentration collected under the current operating temperature environment as the concentration of the key characteristic gas.

[0020] C j1 Indicates the concentration of the gas component corresponding to the second concentration collected in a non-cryogenic environment as the key characteristic gas.

[0021] C j2 Indicates the concentration of the gas component corresponding to the second-ranked concentration collected under the current operating temperature environment as the concentration of the key characteristic gas.

[0022] δ represents the parameter deviation correction coefficient, m represents the amount of transformer oil, and ρ represents the density of transformer oil.

[0023] Preferably, the obtaining of the corrected monitoring results of all dissolved gases comprises:

[0024] C p =δK pC p0 ;

[0025] Among them, C p It indicates the monitoring result after correction of a certain dissolved gas, δ indicates the parameter deviation correction coefficient, K p Indicates the solubility of the dissolved gas, C p0 Indicates the concentration collected and obtained by the dissolved gas concentration monitoring module corresponding to this dissolved gas.

[0026] Preferably, the solubility K corresponding to the dissolved gas p They are:

[0027] Solubility of H2 Solubility of CO K CO =9, solubility of CO2

[0028] Solubility of O2 Solubility of Ar Ar =15, solubility of N2

[0029] Solubility of CH4 Solubility of C2H2

[0030] Solubility of C2H4 Solubility of C2H6

[0031] Solubility of C2H8 Solubility of C3H6

[0032] Solubility of C4H8 C4H 10 Solubility

[0033] Preferably, the step of constructing a BP neural network and outputting the concentration prediction results of all dissolved gases specifically includes:

[0034] The corrected monitoring results of all dissolved gases are used as input data D = {x, t}, and the input data are normalized;

[0035] Initialize the threshold θ and weight ω of the BP neural network;

[0036] Update the weight ω according to the input data D = {x, t}:

[0037]

[0038] Among them, ω k represents the weight after the kth iteration, η represents the learning rate, represents the error of the kth iteration, J(ω k ) represents the input of the kth iteration;

[0039] Calculate ΔJ=J(ω k )-J(ω k-1 );

[0040] Where ΔJ represents the input error, J(ω k-1 ) represents the input of the k-1th iteration;

[0041] Determine whether ΔJ is less than 0. When ΔJ < 0, the training ends. At this time, the corresponding weight ω is the optimal weight, and the corresponding threshold θ is the optimal threshold;

[0042] Determine whether the weights meet the convergence conditions. If so, output the concentration prediction results.

[0043] Preferably, the non-low temperature environment is: greater than or equal to -10°C.

[0044] Advantages of the present invention: The monitoring system for dissolved gas in transformer oil in high-cold areas proposed by the present invention can quickly detect dissolved gas in transformer oil with high precision in ultra-low temperature environment, realize rapid and sensitive response of transformer dissolved gas at ppb level in transformer low-temperature oil and gas environment, compensate for parameter deviation during field operation in low-temperature environment, and improve the accuracy of data output. In low-temperature environment, the detection accuracy of H2, a key characteristic gas of transformer, can reach 2ppm, and the detection accuracy of C2H2 can reach 0.2ppm. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a principle block diagram of the monitoring system for dissolved gas in transformer oil in high-cold areas of the present invention;

[0046] Figure 2 It is the BP neural network architecture diagram of the present invention;

[0047] Figure 3 It is a flowchart of the BP neural network of the present invention outputting the concentration prediction results of all dissolved gases. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0050] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0051] Embodiment 1:

[0052] Combine the following Figure 1-Figure 3 This embodiment is described. The embodiment described in this embodiment is a monitoring system for dissolved gas in transformer oil in a high-cold area, which includes:

[0053] The key characteristic gas screening module is used to detect the concentration of all dissolved gases in transformer oil in a non-low temperature environment, sort them from high to low according to the concentration ratio, and select the gas components corresponding to the top two concentrations as key characteristic gases;

[0054] Temperature monitoring module, used to monitor the current operating temperature of the transformer;

[0055] The low-temperature data error compensation module is used to build a dissolved gas concentration simulation platform, collect the concentration of key characteristic gases in a non-low-temperature environment and the current operating temperature environment, and calculate the parameter deviation correction coefficient;

[0056] The dissolved gas concentration monitoring module is used to collect the concentration of all dissolved gases and use the parameter deviation correction coefficient to compensate and obtain the corrected monitoring results of all dissolved gases;

[0057] The dissolved gas concentration prediction module is used to take the corrected monitoring results of all dissolved gases as input data, construct a BP neural network, and output the concentration prediction results of all dissolved gases.

[0058] Furthermore, all dissolved gases include: H2, CO, CO2, O2, Ar, N2, CH4, C2H2, C2H4, C2H6, C2H8, C3H6, C4H8 and C4H 10 .

[0059] Furthermore, the calculation to obtain the parameter deviation correction coefficient specifically includes:

[0060]

[0061] Among them, C i1 Indicates the concentration of the gas component corresponding to the first concentration collected in a non-low temperature environment as the concentration of the key characteristic gas.

[0062] C i2 Indicates the concentration of the gas component corresponding to the first concentration collected under the current operating temperature environment as the concentration of the key characteristic gas.

[0063] C j1 Indicates the concentration of the gas component corresponding to the second concentration collected in a non-cryogenic environment as the key characteristic gas.

[0064] C j2 Indicates the concentration of the gas component corresponding to the second-ranked concentration collected under the current operating temperature environment as the concentration of the key characteristic gas.

[0065] δ represents the parameter deviation correction coefficient, m represents the amount of transformer oil, and ρ represents the density of transformer oil.

[0066] Furthermore, the obtaining of the corrected monitoring results of all dissolved gases includes:

[0067] C p =δK p C p0 ;

[0068] Among them, C p It indicates the monitoring result after correction of a certain dissolved gas, δ indicates the parameter deviation correction coefficient, K p Indicates the solubility of the dissolved gas, C p0 Indicates the concentration collected and obtained by the dissolved gas concentration monitoring module corresponding to this dissolved gas.

[0069] Furthermore, the solubility K corresponding to the dissolved gas p They are:

[0070] Solubility of H2 Solubility of CO K CO =9, solubility of CO2

[0071] Solubility of O2 Solubility of Ar Ar =15, solubility of N2

[0072] Solubility of CH4 Solubility of C2H2

[0073] Solubility of C2H4 Solubility of C2H6

[0074] Solubility of C2H8 Solubility of C3H6

[0075] Solubility of C4H8 C4H 10 Solubility

[0076] Furthermore, the BP neural network is constructed to output the concentration prediction results of all dissolved gases, specifically including:

[0077] The corrected monitoring results of all dissolved gases are used as input data D = {x, t}, and the input data are normalized;

[0078] Initialize the threshold θ and weight ω of the BP neural network;

[0079] Update the weight ω according to the input data D = {x, t}:

[0080]

[0081] Among them, ω k represents the weight after the kth iteration, η represents the learning rate, represents the error of the kth iteration, J(ω k ) represents the input of the kth iteration;

[0082] Calculate ΔJ=J(ω k )-J(ω k-1 );

[0083] Where ΔJ represents the input error, J(ω k-1 ) represents the input of the k-1th iteration;

[0084] Determine whether ΔJ is less than 0. When ΔJ < 0, the training ends. At this time, the corresponding weight ω is the optimal weight, and the corresponding threshold θ is the optimal threshold;

[0085] Determine whether the weights meet the convergence conditions. If so, output the concentration prediction results.

[0086] Furthermore, the non-low temperature environment is: greater than or equal to -10°C.

[0087] In the present invention, Figure 1 As shown in the figure, the monitoring system of dissolved gas in transformer oil in high-cold areas of the present invention includes a key characteristic gas screening module, a temperature monitoring module, a low temperature data error compensation module, a dissolved gas concentration monitoring module and a dissolved gas concentration prediction module. 10 Conduct testing.

[0088] First, the key characteristic gas screening module is used to detect the concentration of all dissolved gases in the transformer oil in a non-low temperature environment (greater than or equal to -10°C), and the concentration ratio is sorted from high to low, and the gas components corresponding to the top two concentrations are selected as key characteristic gases. Then the temperature monitoring module is used to monitor the current operating temperature of the transformer. The low-temperature data error compensation module is used to build a dissolved gas concentration simulation platform, and the concentration of key characteristic gases is collected in a non-low temperature environment and the current operating temperature environment, and the parameter deviation correction coefficient is calculated. Specifically including:

[0089] Parameter deviation correction coefficient δ:

[0090]

[0091] Among them, C i1 Indicates the concentration of the gas component corresponding to the first concentration collected in a non-cryogenic environment as the key characteristic gas, C i2 Indicates the concentration of the gas component corresponding to the first concentration collected under the current operating temperature environment as the concentration of the key characteristic gas, C j1 Indicates the concentration of the gas component corresponding to the second concentration collected in a non-cryogenic environment as the key characteristic gas, C j2 It represents the concentration of the gas component corresponding to the second concentration collected under the current operating temperature environment as the key characteristic gas, m represents the oil volume of the transformer oil, and ρ represents the density of the transformer oil.

[0092] The dissolved gas concentration monitoring module is used to collect the concentration of all dissolved gases, and the parameter deviation correction coefficient is used for compensation to obtain the corrected monitoring results of all dissolved gases. Specifically including:

[0093] Monitoring results after correction for a certain dissolved gas C p It is expressed as:

[0094] C p =δK p C p0 ;

[0095] Among them, K p Indicates the solubility of the dissolved gas, C p0 Indicates the concentration collected and obtained by the dissolved gas concentration monitoring module corresponding to this dissolved gas.

[0096] Solubility K corresponding to dissolved gas p They are:

[0097] Dissolved Gas Solubility Dissolved Gas Solubility Dissolved Gas Solubility <![CDATA[H2]]> 7 CO 9 <![CDATA[CO2]]> 120 <![CDATA[O2]]> 16 Ar 15 <![CDATA[N2]]> 8.6 <![CDATA[CH4]]> 30 <![CDATA[C2H2]]> 400 <![CDATA[C2H4]]> 280 <![CDATA[C2H6]]> 230 <![CDATA[C2H8]]> 1000 <![CDATA[C3H6]]> 1200 <![CDATA[C4H8]]> 72000 <![CDATA[C4H 10 ]]> 72000

[0098] The dissolved gas concentration prediction module is used to take the corrected monitoring results of all dissolved gases as input data to construct a BP neural network. The architecture of the BP neural network is shown in the figure below. Figure 2 As shown in the figure, the concentration prediction results of all dissolved gases are output. The method process flow chart is as follows Figure 3 As shown, specifically including:

[0099] The corrected monitoring results of all dissolved gases are used as input data D = {x, t}, and the input data are normalized;

[0100] Initialize the threshold θ and weight ω of the BP neural network;

[0101] Update the weight ω according to the input data D = {x, t}:

[0102]

[0103] Among them, ω k represents the weight after the kth iteration, η represents the learning rate, represents the error of the kth iteration, J(ω k ) represents the input of the kth iteration;

[0104] Calculate ΔJ=J(ω k )-J(ω k-1 );

[0105] Where ΔJ represents the input error, J(ω k-1 ) represents the input of the k-1th iteration;

[0106] Determine whether ΔJ is less than 0. When ΔJ < 0, the training ends. At this time, the corresponding weight ω is the optimal weight, and the corresponding threshold θ is the optimal threshold;

[0107] Determine whether the weights meet the convergence conditions. If so, output the concentration prediction results.

[0108] The monitoring system for dissolved gas in transformer oil in high-cold areas proposed in the present invention can perform high-precision rapid detection of dissolved gas in transformer oil in an ultra-low temperature environment, achieve rapid and sensitive response of transformer dissolved gas at the ppb level in the transformer low-temperature oil and gas environment, compensate for parameter deviations during on-site operation in a low-temperature environment, and improve the accuracy of data output.

[0109] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. A monitoring system for dissolved gas in transformer oil in high-cold areas, characterized in that: It includes: The key characteristic gas screening module is used to detect the concentration of all dissolved gases in transformer oil in a non-low temperature environment, sort them from high to low according to the concentration ratio, and select the gas components corresponding to the top two concentrations as key characteristic gases; Temperature monitoring module, used to monitor the current operating temperature of the transformer; The low-temperature data error compensation module is used to build a dissolved gas concentration simulation platform, collect the concentration of key characteristic gases in a non-low-temperature environment and the current operating temperature environment, and calculate the parameter deviation correction coefficient; The dissolved gas concentration monitoring module is used to collect the concentration of all dissolved gases and use the parameter deviation correction coefficient to compensate and obtain the corrected monitoring results of all dissolved gases; The dissolved gas concentration prediction module is used to take the corrected monitoring results of all dissolved gases as input data, construct a BP neural network, and output the concentration prediction results of all dissolved gases.

2. The monitoring system for dissolved gas in transformer oil in high-cold areas according to claim 1 is characterized in that: All dissolved gases include: H2, CO, CO2, O2, Ar, N2, CH4, C2H2, C2H4, C2H6, C2H8, C3H6, C4H8 and C4H 10 .

3. The monitoring system for dissolved gas in transformer oil in high-cold areas according to claim 1 is characterized in that: The calculation to obtain the parameter deviation correction coefficient specifically includes: Among them, C i1 Indicates the concentration of the gas component corresponding to the first concentration collected in a non-low temperature environment as the concentration of the key characteristic gas. C i2 Indicates the concentration of the gas component corresponding to the first concentration collected under the current operating temperature environment as the concentration of the key characteristic gas. C j1 Indicates the concentration of the gas component corresponding to the second concentration collected in a non-cryogenic environment as the key characteristic gas. C j2 Indicates the concentration of the gas component corresponding to the second-ranked concentration collected under the current operating temperature environment as the concentration of the key characteristic gas. δ represents the parameter deviation correction coefficient, m represents the amount of transformer oil, and ρ represents the density of transformer oil.

4. The monitoring system for dissolved gas in transformer oil in high-cold areas according to claim 1 is characterized in that: The corrected monitoring results of all dissolved gases are obtained including: C p =δK p C p0 ; Among them, C p It indicates the monitoring result after correction of a certain dissolved gas, δ indicates the parameter deviation correction coefficient, K p Indicates the solubility of the dissolved gas, C p0 Indicates the concentration collected and obtained by the dissolved gas concentration monitoring module corresponding to this dissolved gas.

5. The monitoring system for dissolved gas in transformer oil in high-cold areas according to claim 4 is characterized in that: The solubility K corresponding to the dissolved gas p They are: Solubility of H2 Solubility of CO K CO =9, solubility of CO2 Solubility of O2 Solubility of Ar Ar =15, solubility of N2 Solubility of CH4 Solubility of C2H2 Solubility of C2H4 Solubility of C2H6 Solubility of C2H8 Solubility of C3H6 Solubility of C4H8 C4H 10 Solubility 6. The monitoring system for dissolved gas in transformer oil in high-cold areas according to claim 1 is characterized in that: The construction of the BP neural network and the output of the concentration prediction results of all dissolved gases specifically include: The corrected monitoring results of all dissolved gases are used as input data D = {x, t}, and the input data are normalized; Initialize the threshold θ and weight ω of the BP neural network; Update the weight ω according to the input data D = {x, t}: Among them, ω k represents the weight after the kth iteration, η represents the learning rate, represents the error of the kth iteration, J(ω k ) represents the input of the kth iteration; Calculate ΔJ=J(ω k )-J(ω k-1 ); Where ΔJ represents the input error, J(ω k-1 ) represents the input of the k-1th iteration; Determine whether ΔJ is less than 0. When ΔJ < 0, the training ends. At this time, the corresponding weight ω is the optimal weight, and the corresponding threshold θ is the optimal threshold; Determine whether the weights meet the convergence conditions. If so, output the concentration prediction results.

7. The monitoring system for dissolved gas in transformer oil in high-cold areas according to claim 1 is characterized in that: The non-low temperature environment is: greater than or equal to -10°C.

Images