CN113945299B - Transformer winding temperature distribution and residual life prediction method - Google Patents
Transformer winding temperature distribution and residual life prediction method Download PDFInfo
- Publication number
- CN113945299B CN113945299B CN202111078692.8A CN202111078692A CN113945299B CN 113945299 B CN113945299 B CN 113945299B CN 202111078692 A CN202111078692 A CN 202111078692A CN 113945299 B CN113945299 B CN 113945299B
- Authority
- CN
- China
- Prior art keywords
- temperature
- transformer
- winding
- cake
- neuron
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004804 winding Methods 0.000 title claims abstract description 126
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000000835 fiber Substances 0.000 claims abstract description 33
- 230000032683 aging Effects 0.000 claims abstract description 19
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000003878 thermal aging Methods 0.000 claims abstract description 11
- 210000002569 neuron Anatomy 0.000 claims description 47
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 11
- 238000005259 measurement Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 238000009413 insulation Methods 0.000 claims description 8
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 7
- 238000013461 design Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 4
- 238000010304 firing Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 claims description 2
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 claims description 2
- 229910052782 aluminium Inorganic materials 0.000 claims description 2
- 230000004907 flux Effects 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 claims description 2
- 238000012353 t test Methods 0.000 claims description 2
- 238000012546 transfer Methods 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 abstract description 7
- 230000009286 beneficial effect Effects 0.000 abstract description 7
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000000691 measurement method Methods 0.000 description 4
- 239000013307 optical fiber Substances 0.000 description 4
- 238000004088 simulation Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005485 electric heating Methods 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- -1 polytetrafluoroethylene Polymers 0.000 description 1
- 229920001343 polytetrafluoroethylene Polymers 0.000 description 1
- 239000004810 polytetrafluoroethylene Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K11/00—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00
- G01K11/32—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00 using changes in transmittance, scattering or luminescence in optical fibres
- G01K11/3206—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00 using changes in transmittance, scattering or luminescence in optical fibres at discrete locations in the fibre, e.g. using Bragg scattering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Housings And Mounting Of Transformers (AREA)
Abstract
The invention belongs to the technical field of power transformer state evaluation, and particularly relates to a method for predicting the temperature distribution and the residual life of a transformer winding.A plurality of groups of fiber bragg grating temperature sensors are used for directly measuring the discrete point temperature of the internal winding of the transformer, a neural network algorithm is combined to fit continuous data among windings, the temperature distribution of each cake of the transformer winding is predicted, the temperature prediction is carried out on the position of each region, and the nonlinear relation between the input temperature and the output temperature is extracted and approximated through learning; according to the obtained hot spot temperature of the transformer, the aging rate and the residual life of the transformer are predicted by using a thermal aging 6-degree method; the method is beneficial to adjusting the load condition of the transformer and warning the abnormal condition of the transformer, is convenient for finding the transformer under the abnormal working condition, is beneficial to ensuring the safe and stable operation of the power transformer, and has popularization prospect.
Description
Technical Field
The invention belongs to the technical field of power transformer state evaluation, and particularly relates to a method for predicting temperature distribution and residual life of a transformer winding.
Background
The transformer is used as one of the core devices of the power grid and plays an important role in electric energy transmission. When the transformer runs, the loss generated by the iron core, the winding and other parts of the transformer can generate temperature rise in the interior, so that the insulation aging of the transformer winding is accelerated, and the service life of the transformer is further influenced. The load capacity and the remaining life of the transformer are mainly dependent on its thermal characteristics. Therefore, in order to prevent the transformer from overheating and ensure safe and stable operation, it is necessary to study a method for predicting the temperature and the service life of the transformer.
At present, the conventional method for acquiring the temperature of the internal winding of the transformer mainly comprises a direct measurement method, a thermal simulation measurement method, an indirect calculation method and the like. According to the direct measurement method, the sensors are embedded in the transformer winding, the temperature of a winding measurement point is obtained by using the temperature measuring instrument, and the more the embedded temperature sensors are, the more accurate the measurement result is. However, due to uneven heat dissipation inside the transformer, it is difficult to ensure that the sensor is installed at the hot spot temperature position of the winding a priori. The thermal simulation measurement method superimposes the temperature rise of the additional current on the electric heating element on the top layer oil temperature of the transformer to obtain the hot spot temperature of the transformer winding. The method has larger measurement error and poorer practicability. The indirect calculation method is mainly used for calculating the winding hot spot temperature by calculating each key parameter and the transformer load current value according to the transformer and the environment parameters which are easy to measure in actual operation. Since the simplified thermal path model does not take into account many influencing factors, such as: viscosity coefficient of oil, load dynamic loss, etc., so that accuracy of calculation results cannot be ensured. Therefore, it is desirable to provide a method for predicting the temperature distribution and estimating the residual life of the transformer winding accurately.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a method for predicting the temperature distribution and the residual life of a transformer winding, which combines a conventional optical fiber temperature measurement method with a temperature distribution prediction algorithm, has high calculation speed and high temperature distribution prediction precision, accurately predicts the aging rate and the life of the transformer, is convenient for quickly finding out the transformer under abnormal working conditions, and is beneficial to ensuring the safe and stable operation of the power transformer. The specific technical scheme of the invention is as follows:
A method for predicting temperature distribution and residual life of a transformer winding comprises the following steps:
S1, collecting operation parameters and temperature rise test data of a transformer with the same model as a transformer to be tested;
S2, after the transformer to be tested completes iron core manufacturing and winding, burying a plurality of fiber bragg grating temperature sensors among windings of the transformer according to a specific rule, and measuring temperature data of discrete positions of the windings in real time;
s3, based on temperature data of each discrete position of the winding obtained through real-time measurement, predicting temperature data of each region inside the winding by combining a neural network algorithm, and predicting temperature distribution of each cake of the transformer winding;
S4, predicting the aging rate and the residual life of the transformer by using a thermal aging 6-degree method according to the predicted value of the temperature distribution of each cake of the transformer winding obtained in the S3.
Preferably, the transformers of the same type in the step S1 are power transformers of the same specification and already put into operation.
Preferably, the operation parameters and the temperature rise test data in the step S1 include parameters of the transformer itself, top layer oil temperature, inlet and outlet temperatures of the radiator, and winding hot spot temperature information.
Preferably, the parameters of the transformer include the height and diameter of the transformer coil, the length, width and height of the iron core from the side yoke, and the three-dimensional size of the oil tank.
Preferably, the specific rule in step S2 is: the part 10% -20% below the top of the transformer winding is a local hot spot area, and the number of the fiber bragg grating temperature sensors embedded in each cake coil in the local hot spot area is more than that of the fiber bragg grating temperature sensors embedded in each cake coil in other conventional temperature areas.
Preferably, the embedding method of the fiber bragg grating temperature sensor comprises the following steps: embedding and fixing the tail fiber of the temperature measuring channel in the insulation block groove along the gaps among cakes at the top of the transformer winding, and connecting the tail fiber of the temperature measuring channel into the through plate from the top of the winding.
Preferably, the step S3 specifically includes the following steps:
Let x 1,x2,...,xn be the temperature data of each discrete position of the winding measured by the fiber grating temperature sensor, as the input of the neural network, the total number of the temperature data input into the neural network is n; let y 1,y2,...,ym be the pre-measured temperature of the transformer winding output after the calculation of the neuron, and the total number output after the calculation of the neural network is m; w ij is the connection weight between the ith neuron of the input layer and the jth neuron of the hidden layer, and w jk is the connection weight between the jth neuron of the hidden layer and the kth neuron of the output layer;
The weighted sum of the j-th temperature inputs of the intermediate hidden layer is:
for hidden layer neurons, the firing function is:
for the hidden layer jth neuron there are:
H j is the output of the jth neuron of the hidden layer, and its calculation formula is as follows:
Hj＝f(netj)；(3)
The output error is:
l is the total number of hidden layer neurons.
Wherein Y k is neuron output expectation, and is obtained based on the operation parameters and the temperature rise test data of the transformers with the same type acquired in the step S1;
Based on the output error, the output layer and the hidden layer are modified as follows:
and (3) correcting the weight value of an output layer:
wjk2＝wjk+ηHjek；(5)
wherein η is a learning rate and is a network parameter of the neural network; w jk2 is the correction value of the connection weight between the jth neuron of the hidden layer and the kth neuron of the output layer;
hidden layer weight correction:
wherein w ij2 is a correction value of a connection weight between the ith neuron of the input layer and the jth neuron of the hidden layer;
based on the corrected output layer weight and hidden layer weight, the predicted temperature distribution of each cake of the transformer winding can be obtained:
wherein H j2 denotes the modified hidden layer jth neuron,
Preferably, the predicting the aging rate and the remaining life of the transformer by using the thermal aging 6 degree method in the step S4 specifically includes:
The thermal aging 6-degree method means that the aging rate is doubled every time the internal temperature of the transformer rises by 6K, and based on the method, the relative aging rate at each hot spot temperature is obtained as follows:
V＝2(t-98)/6；(8)
Wherein t is the internal temperature of the transformer, and the unit is the temperature of the transformer;
the remaining life of the transformer is obtained by using the lost life:
Wherein L is the residual life of the transformer; s is the design life of the transformer; v is the relative aging rate of the transformer operating during the period t 1～t2.
The beneficial effects of the invention are as follows: on one hand, the invention directly measures the discrete point temperature of the internal winding of the transformer through a plurality of groups of fiber bragg grating temperature sensors, fits continuous data among windings by combining a neural network algorithm, predicts the temperature distribution of each cake of the winding of the transformer, wherein the neural network algorithm is used for fitting the temperature data of the discrete points, predicting the continuous temperature distribution of each area of the winding of the internal winding of the transformer, each area of the winding refers to equally dividing each cake of the winding into a plurality of areas, carrying out temperature prediction on the positions of each area, and extracting and approximating the nonlinear relation between the input temperature and the output temperature through learning; on the other hand, according to the obtained hot spot temperature of the transformer, the aging rate and the residual life of the transformer are predicted by using a thermal aging 6-degree method; the method is beneficial to adjusting the load condition of the transformer and warning the abnormal condition of the transformer, is convenient for finding the transformer under the abnormal working condition, is beneficial to ensuring the safe and stable operation of the power transformer, and has popularization prospect.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. Like elements or portions are generally identified by like reference numerals throughout the several figures. In the drawings, elements or portions thereof are not necessarily drawn to scale.
FIG. 1 is a flow chart of a method for predicting temperature distribution and residual life of a transformer winding based on optical fiber sensing according to an embodiment of the invention;
FIG. 2 is a detailed view of the embedded measurement points of the optical fiber temperature sensor according to the present invention;
FIG. 3 is a schematic diagram of a conventional temperature zone array and a hot spot temperature zone array of an optical fiber temperature sensor according to the present invention;
FIG. 4 is a neural network architecture of the present invention;
reference numerals illustrate: 1. a tenth winding cake, 2, a ninth winding cake, 3, an eighth winding cake, 4, a seventh winding cake, 5, a sixth winding cake, 6, a fifth winding cake, 7, a fourth winding cake, 8, a third winding cake, 9, a second winding cake, 10, a first winding cake, 11, a fiber bragg grating sensor, 12, a conventional temperature area array of the sensor, 13, a hot spot temperature area array of the sensor, 14 and an array connector.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As shown in fig. 1, a method for predicting temperature distribution and residual life of a transformer winding includes the following steps:
S1, collecting operation parameters and temperature rise test data of a transformer with the same model as a transformer to be tested; the transformers with the same type are power transformers with the same specification and already put into operation. The operation parameters and the temperature rise test data comprise parameters of the transformer, top layer oil temperature, inlet and outlet temperatures of the radiator and winding hot spot temperature information. The parameters of the transformer include the height and diameter of the transformer coil, the length, width and height of the iron core from the side yoke, and the three-dimensional size of the oil tank.
S2, after the transformer to be tested completes iron core manufacturing and winding, a plurality of fiber bragg grating temperature sensors are buried among windings of the transformer according to a specific rule, and temperature data of discrete positions of the windings are measured in real time. The specific rules are: the part 10% -20% below the top of the transformer winding is a local hot spot area, and the number of the fiber bragg grating temperature sensors embedded in each cake coil in the local hot spot area is more than that of the fiber bragg grating temperature sensors embedded in each cake coil in other conventional temperature areas. As shown in fig. 2, a plurality of fiber grating sensors 11 are buried between windings of a transformer according to the specific rule, and temperature data of discrete positions of the windings are measured in real time; the winding 10 is wound together and comprises a winding tenth cake 1, a winding ninth cake 2, a winding eighth cake 3, a winding seventh cake 4, a winding sixth cake 5, a winding fifth cake 6, a winding fourth cake 7, a winding third cake 8, a winding second cake 9 and a winding first cake 10. Specifically, 2 sensors are embedded in each cake in a local hot spot area, 1 sensor is embedded in each cake in a conventional temperature area, a sensor array for measuring the temperature of a high/low voltage winding is formed by the fiber bragg grating sensors, the high/low voltage winding corresponds to 1 conventional temperature area array 12 of the sensor and 1 hot spot temperature area array 13 of the sensor respectively, as shown in fig. 3, the diameter of the fiber bragg grating sensor is 3mm, a tail fiber is made of polytetrafluoroethylene, FC/APC connectors are adopted, and connectors of each array are array connectors 14. The embedding method of the fiber bragg grating temperature sensor comprises the following steps: embedding and fixing the tail fiber of the temperature measuring channel in the insulation block groove along the gaps among cakes at the top of the transformer winding, and connecting the tail fiber of the temperature measuring channel into the through plate from the top of the winding.
S3, based on temperature data of each discrete position of the winding obtained through real-time measurement, temperature data of each region inside the winding are predicted by combining a neural network algorithm, and temperature distribution of each cake of the transformer winding is predicted. The method specifically comprises the following steps:
Let x 1,x2,...,xn be the temperature data of each discrete position of the winding measured by the fiber grating temperature sensor, as the input of the neuron network, the number is n; let y 1,y2,...,ym be the pre-measured temperature of the transformer winding output after the calculation of the neuron, and the number is m; w ij is the connection weight between the ith neuron of the input layer and the jth neuron of the hidden layer, and w jk is the connection weight between the jth neuron of the hidden layer and the kth neuron of the output layer;
The weighted sum of the j-th temperature inputs of the intermediate hidden layer is:
for hidden layer neurons, the firing function is:
for the hidden layer jth neuron there are:
H j is the output of the jth neuron of the hidden layer, and its calculation formula is as follows:
Hj＝f(netj)；(3)
The output error is:
l is the total number of hidden layer neurons.
Wherein Y k is neuron output expectation, and is obtained based on the operation parameters and the temperature rise test data of the transformers with the same type acquired in the step S1;
Based on the output error, the output layer and the hidden layer are modified as follows:
and (3) correcting the weight value of an output layer:
wjk2＝wjk+ηHjek；(5)
wherein η is a learning rate and is a network parameter of the neural network; w jk2 is the correction value of the connection weight between the jth neuron of the hidden layer and the kth neuron of the output layer;
hidden layer weight correction:
wherein w ij2 is a correction value of a connection weight between the ith neuron of the input layer and the jth neuron of the hidden layer;
based on the corrected output layer weight and hidden layer weight, the predicted temperature distribution of each cake of the transformer winding can be obtained:
y k is the predicted temperature of the internal winding of the transformer after calculation by a neural network, each cake of the winding is equally divided into a plurality of areas, and the temperature prediction is carried out on the positions of each area; k is the number of each region of the transformer winding, such as: dividing each cake into 4 areas equally, wherein k=1 of a first area of a first cake, k=2 of a second area of the first cake, k=3 of a third area of the first cake and k=4 of a fourth area of the first cake;
K=5 in the first block region of the second cake, k=6 in the second block region of the second cake, k=7 in the third block region of the second cake, k=8 in the fourth block region of the second cake, and so on.
Wherein H j2 denotes the modified hidden layer jth neuron,
In the algorithm, the neural network algorithm is used for fitting temperature data of discrete points and predicting continuous temperature distribution of each region of the internal winding of the transformer. The winding areas are characterized in that each cake of the winding is divided into a plurality of areas, the temperature of each area is predicted, and the nonlinear relation between the input temperature and the output temperature is extracted and approximated through learning. The continuous temperature distribution of each position of the winding means that each cake of the winding is divided into a plurality of areas uniformly, and the temperature prediction is carried out on the position of each area of each cake.
S4, predicting the aging rate and the residual life of the transformer by using a thermal aging 6-degree method according to the predicted value of the temperature distribution of each cake of the transformer winding obtained in the S3.
The thermal aging 6-degree method means that the aging rate is doubled every time the internal temperature of the transformer rises by 6K, and based on the method, the relative aging rate at each hot spot temperature is obtained as follows:
V＝2(t-98)/6；(8)
wherein t is the internal temperature of the transformer in degrees celsius.
The remaining life of the transformer is obtained by using the lost life:
Wherein L is the residual life of the transformer; s is the design life of the transformer; v is the relative aging rate of the transformer operating during the period t 1～t2.
In addition, the embedded position of the fiber bragg grating temperature sensor is outside the winding insulation, and the measured temperature is the temperature of the insulation layer close to the wire. According to the heat conduction mechanism of heat transfer, the surface of the copper wire or the aluminum wire and the outer surface of the insulating paper have temperature difference. For this, the measurement value correction formula is as follows:
wherein q is the heat flux density of the surface of the winding; lambda is the heat conductivity coefficient of the insulating layer; delta is the thickness of the insulating layer; t real is a temperature true value; t test is the measured value of the fiber grating temperature sensor.
On one hand, the invention directly measures the temperature of discrete points of the internal winding of the transformer through a plurality of groups of fiber bragg grating sensors, fits continuous data among windings by combining a neural network algorithm, predicts the temperature distribution of each cake of the winding of the transformer, wherein the neural network algorithm is used for fitting the temperature data of the discrete points, predicting the continuous temperature distribution of each area of the winding of the internal winding of the transformer, each area of the winding refers to equally dividing each cake of the winding into a plurality of areas, carrying out temperature prediction on the positions of each area, and extracting and approximating the nonlinear relation between the input temperature and the output temperature through learning; on the other hand, according to the obtained hot spot temperature of the transformer, the aging rate and the residual life of the transformer are predicted by using a thermal aging 6-degree method; the method and the device have the advantages that the obtained result is beneficial to adjusting the load condition of the transformer, warning the abnormal condition of the transformer, conveniently finding the transformer under the abnormal working condition, ensuring the safe and stable operation of the power transformer, and having popularization prospect.
Those of ordinary skill in the art will appreciate that the elements of the examples described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the elements of the examples have been described generally in terms of functionality in the foregoing description to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present application, it should be understood that the division of the units is merely a logic function division, and there may be other division manners in actual implementation, for example, multiple units may be combined into one unit, one unit may be split into multiple units, or some features may be omitted.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention, and are intended to be included within the scope of the appended claims and description.
Claims (6)
1. A method for predicting temperature distribution and residual life of a transformer winding is characterized by comprising the following steps of: the method comprises the following steps:
S1, collecting operation parameters and temperature rise test data of a transformer with the same model as a transformer to be tested;
S2, after the transformer to be tested completes iron core manufacturing and winding, burying a plurality of fiber bragg grating temperature sensors among windings of the transformer according to a specific rule, and measuring temperature data of discrete positions of the windings in real time; the specific rule is as follows: the part 10% -20% below the top of the transformer winding is a local hot spot area, and the number of the fiber bragg grating temperature sensors embedded in each cake coil in the local hot spot area is more than that of the fiber bragg grating temperature sensors embedded in each cake coil in other conventional temperature areas; the embedding method of the fiber bragg grating temperature sensor comprises the following steps: embedding and fixing the tail fiber of the temperature measuring channel in the insulation block groove along the gaps among cakes at the top of the transformer winding, and connecting the tail fiber of the temperature measuring channel into the through plate from the top of the winding; the fiber bragg grating temperature sensor is buried outside the winding insulation, and the measured temperature is the temperature of an insulation layer close to the wire; according to the heat conduction mechanism of heat transfer, the surface of the copper wire or the aluminum wire and the outer surface of the insulating paper have temperature difference; for this, the measurement value correction formula is as follows:
wherein q is the heat flux density of the surface of the winding; lambda is the heat conductivity coefficient of the insulating layer; delta is the thickness of the insulating layer; t real is a temperature true value; t test is the measured value of the fiber grating temperature sensor;
S3, based on temperature data of each discrete position of the winding obtained through real-time measurement, predicting temperature data of each region inside the winding by combining a neural network algorithm, and predicting temperature distribution of each cake of the transformer winding; the neural network algorithm is used for fitting temperature data of discrete points and predicting continuous temperature distribution of each region of the internal winding of the transformer; the winding areas are characterized in that each cake of the winding is divided into a plurality of areas, the temperature of each area is predicted, and the nonlinear relation between the input temperature and the output temperature is extracted and approximated through learning; the continuous temperature distribution of each position of the winding means that each cake of the winding is divided into a plurality of areas uniformly, and the temperature prediction is carried out on the position of each area of each cake;
S4, predicting the aging rate and the residual life of the transformer by using a thermal aging 6-degree method according to the predicted value of the temperature distribution of each cake of the transformer winding obtained in the S3.
2. The method for predicting temperature distribution and remaining life of a transformer winding of claim 1, wherein: and in the step S1, the transformers with the same type are power transformers with the same specification and already put into operation.
3. The method for predicting temperature distribution and remaining life of a transformer winding of claim 1, wherein: the operation parameters and the temperature rise test data in the step S1 comprise parameters of the transformer, top layer oil temperature, inlet and outlet temperatures of the radiator and winding hot spot temperature information.
4. A method for predicting temperature distribution and remaining life of a transformer winding according to claim 3, wherein: the parameters of the transformer include the height and diameter of the transformer coil, the length, width and height of the iron core from the side yoke, and the three-dimensional size of the oil tank.
5. The method for predicting temperature distribution and remaining life of a transformer winding of claim 1, wherein: the step S3 specifically comprises the following steps:
Let x 1,x2,...,xn be the temperature data of each discrete position of the winding measured by the fiber grating temperature sensor, as the input of the neuron network, the number is n; let y 1,y2,...,ym be the pre-measured temperature of the transformer winding output after the calculation of the neuron, and the number is m; w ij is the connection weight between the ith neuron of the input layer and the jth neuron of the hidden layer, and w jk is the connection weight between the jth neuron of the hidden layer and the kth neuron of the output layer;
The weighted sum of the j-th temperature inputs of the intermediate hidden layer is:
for hidden layer neurons, the firing function is:
for the hidden layer jth neuron there are:
H j is the output of the jth neuron of the hidden layer, and its calculation formula is as follows:
Hj＝f(netj)；(3)
The output error is:
l is the total number of hidden layer neurons;
Wherein Y k is neuron output expectation, and is obtained based on the operation parameters and the temperature rise test data of the transformers with the same type acquired in the step S1;
Based on the output error, the output layer and the hidden layer are modified as follows:
and (3) correcting the weight value of an output layer:
wjk2＝wjk+ηHjek；(5)
wherein η is a learning rate and is a network parameter of the neural network; w jk2 is the correction value of the connection weight between the jth neuron of the hidden layer and the kth neuron of the output layer;
hidden layer weight correction:
wherein w ij2 is a correction value of a connection weight between the ith neuron of the input layer and the jth neuron of the hidden layer;
based on the corrected output layer weight and hidden layer weight, the predicted temperature distribution of each cake of the transformer winding can be obtained:
Wherein y k is the predicted internal winding temperature of the transformer after calculation by the neural network, k is the number of each region of the transformer winding, H j2 represents the j-th neuron of the modified hidden layer,
6. The method for predicting temperature distribution and remaining life of a transformer winding of claim 1, wherein: in the step S4, the method for predicting the aging rate and the residual life of the transformer by using the thermal aging 6 degree method specifically includes: the thermal aging 6-degree method means that the aging rate is doubled every time the internal temperature of the transformer rises by 6K, and based on the method, the relative aging rate at each temperature is obtained as follows:
V＝2(t-98)/6；(8)
Wherein t is the internal temperature of the transformer, and the unit is the temperature of the transformer;
the remaining life of the transformer is obtained by using the lost life:
Wherein L is the residual life of the transformer; s is the design life of the transformer; v is the relative aging rate of the transformer operating during the period t 1～t2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111078692.8A CN113945299B (en) | 2021-09-15 | 2021-09-15 | Transformer winding temperature distribution and residual life prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111078692.8A CN113945299B (en) | 2021-09-15 | 2021-09-15 | Transformer winding temperature distribution and residual life prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113945299A CN113945299A (en) | 2022-01-18 |
| CN113945299B true CN113945299B (en) | 2024-07-02 |
Family
ID=79328441
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111078692.8A Active CN113945299B (en) | 2021-09-15 | 2021-09-15 | Transformer winding temperature distribution and residual life prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113945299B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109598061A (en) * | 2018-12-03 | 2019-04-09 | 西南交通大学 | A kind of monitoring method of transformer group mean life loss |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101447048B (en) * | 2008-12-30 | 2011-08-03 | 上海发电设备成套设计研究院 | Method for predicting life of transformer insulation and management system thereof |
| CN102096030B (en) * | 2010-12-10 | 2013-04-17 | 西安交通大学 | Method for estimating residual insulation service life of power transformer based on operating data |
| CN103176058A (en) * | 2013-03-27 | 2013-06-26 | 国家电网公司 | Device for measuring transformer oil paper insulation test piece |
| CN103364658A (en) * | 2013-06-28 | 2013-10-23 | 国网电力科学研究院武汉南瑞有限责任公司 | Method for predicting service life of transformer based on fiber grating temperature measurement system |
| CN103376166B (en) * | 2013-06-28 | 2017-06-16 | 山西省电力公司太原供电分公司 | The arrangement and method for embedding of a kind of inside transformer fiber-optical grating temperature sensor |
| CN106874534A (en) * | 2016-12-28 | 2017-06-20 | 国网内蒙古东部电力有限公司检修分公司 | A kind of transformer overload capability assessment method |
| CN206339317U (en) * | 2017-01-10 | 2017-07-18 | 青岛数能电气工程有限公司 | A kind of Transformer Winding temperature on-line monitoring device |
| CN108120521B (en) * | 2017-12-08 | 2020-06-26 | 囯网河北省电力有限公司电力科学研究院 | Transformer winding hot spot temperature prediction method and terminal equipment |
| CN109668651B (en) * | 2019-02-28 | 2024-08-02 | 广东中鹏电气有限公司 | Pre-burying method of fiber bragg grating of transformer and transformer temperature measurement system manufactured by pre-burying method |
| CN110132447A (en) * | 2019-04-17 | 2019-08-16 | 上海电力学院 | A kind of coiling hot point of transformer temperature online monitoring system based on fiber grating |
-
2021
- 2021-09-15 CN CN202111078692.8A patent/CN113945299B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109598061A (en) * | 2018-12-03 | 2019-04-09 | 西南交通大学 | A kind of monitoring method of transformer group mean life loss |
Non-Patent Citations (1)
| Title |
|---|
| "基于GA-BP 神经网络的变压器绕组热点温度预测研究";王兴桐等;《东北电力技术》;第42卷(第2期);第1.1节、第3节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113945299A (en) | 2022-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110083908B (en) | Cable core temperature prediction method based on finite element analysis | |
| CN112115628B (en) | Hot spot temperature detection method based on distribution calculation of temperature field of oil-immersed transformer | |
| CN103383433B (en) | The status monitoring of stator core of ship generator and fault early warning method | |
| CN106066212A (en) | A kind of cable conductor temperature indirect measurement method | |
| CN106874534A (en) | A kind of transformer overload capability assessment method | |
| CN109598061A (en) | A kind of monitoring method of transformer group mean life loss | |
| CN104198068A (en) | Temperature monitoring device and temperature monitoring method for winding of oil immersed transformer | |
| CN115270622A (en) | Method for predicting transformer hot spot temperature based on numerical calculation and linear regression model | |
| CN109344559A (en) | A kind of transformer temperature rise of hot spot prediction technique comparing optical fiber temperature-measurement | |
| CN111623884A (en) | Transformer hot spot temperature identification method and system based on improved heat network model | |
| Srinivasan et al. | Prediction of transformer insulation life with an effect of environmental variables | |
| CN110333443B (en) | Temperature rise test method for stator winding of induction motor | |
| Li et al. | Test and analysis on extended temperature rise of 110 kV transformer based on distributed temperature sensing | |
| Abdali et al. | Magnetic-thermal analysis of distribution transformer: Validation via optical fiber sensors and thermography | |
| CN113945299B (en) | Transformer winding temperature distribution and residual life prediction method | |
| CN110487844A (en) | A kind of appraisal procedure of power cable insulation layer temperature and failure of insulation | |
| Abdali et al. | Liquid-immersed distribution transformers’ thermal analysis with consideration of unbalanced load current effect | |
| CN111177907B (en) | Automatic assessment method and device for service life of reactor | |
| Radakovic et al. | Loading of transformers in conditions of controlled cooling system | |
| Lai et al. | Prediction of top oil temperature for oil-immersed transformer based on Kalman filter algorithm | |
| Rezaeealam et al. | Real‐time monitoring of transformer hot‐spot temperature based on nameplate data | |
| CN110319953B (en) | Cable conductor temperature prediction system, method and device and readable storage medium | |
| Gong et al. | Distribution transformer tests for PEV smart charging control | |
| Luo et al. | A method for hot spot temperature monitoring of oil-immersed transformers combining physical simulation and intelligent neural network | |
| Shiravand et al. | New Thermal Model for Accurate Prediction of Top Oil Temperature of Distribution Transformer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |