A transformer internal temperature field deduction method based on multipoint transformation

Abstract

The application discloses a transformer internal temperature field deduction method based on multipoint transformation, and relates to the field of transformer state monitoring. The method selects surface temperature measuring points on a part of the surface of a transformer external box, and performs multipoint transformation on the surface temperature measuring points, so that the temperature of an internal point of the transformer which cannot be directly measured is deduced. After the modeling of the deduction model is completed, the temperature data of the internal temperature point to be deduced of the transformer can be obtained by processing the data of the surface temperature measuring points and combining the deduction model under any working condition. Compared with the existing direct measurement method of transformer temperature rise, the method has the advantages that the operation is simple, the pre-embedded sensor and the invasive temperature measurement experiment are not needed, and the temperature of each point in the transformer can be deduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of transformer state monitoring, and in particular to a transformer internal temperature field deduction method based on multipoint transformation. BACKGROUND

[0002] Transformer state monitoring technology is a technology for monitoring and evaluating the operating state of power transformers, which can timely discover potential faults and abnormal conditions of transformers, thereby preventing accidents and ensuring the safe and stable operation of power systems, and has important application scenarios in the preventive maintenance and intelligent management of smart grids. Transformer state monitoring technology collects data of indicators required for state monitoring and evaluation through various sensors, and combines various analysis methods and processing methods for maintaining the stable operation of transformers. In transformer state monitoring technology, temperature field monitoring plays an important role in avoiding overheat loss of transformers. Due to the complex structure inside the transformer and the complex electromagnetic effect during operation, it is difficult to directly measure the internal temperature of the transformer through sensors, which is the focus and difficulty of current research. SUMMARY

[0003] The present application proposes a transformer internal temperature field deduction method based on multipoint transformation to solve the problem that the specific distribution of the internal temperature field of the transformer is difficult to measure. A multipoint transformation deduction model is established by using the data provided by the transformer temperature field simulation model, so as to achieve the goal of deducing the internal temperature of the points to be deduced from the surface temperature of the points.

[0004] In the first aspect, the present application provides a transformer internal temperature field deduction method based on multipoint transformation, comprising the following steps:

[0005] A transformer temperature field simulation model is established by taking the structure of the transformer as a reference and combining the physical effects that have an impact on the temperature distribution law of the transformer during actual operation;

[0006] N surface temperature measurement points are selected on each part of the surface of the transformer tank, and m internal temperature points to be deduced are selected inside the transformer;

[0007] The transformer temperature field simulation model is used to calculate the temperature data of the N surface temperature measurement points and the m internal temperature points to be deduced under different working conditions;

[0008] A multipoint transformation deduction model is solved based on the calculated temperature data, which is used to calculate the temperature data of the internal temperature points to be deduced in the transformer under any working condition by combining the temperature data of the multiple surface temperature measurement points.

[0009] The further technical scheme is that the multipoint transformation deduction model is solved based on the calculated temperature data, comprising:

[0010] The temperature data corresponding to the N surface temperature measurement points under different working conditions is subjected to sensitivity analysis, and n points with high sensitivity are selected as surface characteristic temperature measurement points;

[0011] According to the temperature data of the m selected temperature points to be deduced and the temperature data of the n surface characteristic temperature measurement points, a corresponding multipoint transformation deduction model is established, and is solved based on the temperature data.

[0012] The further technical scheme is that the temperature data corresponding to the N surface temperature measurement points under different working conditions is subjected to sensitivity analysis, including:

[0013] The sensitivity is defined as the ratio of the relative change of the temperature of the surface temperature measurement point to the relative change of the temperature of the temperature point to be deduced, and is expressed as:

[0014]

[0015] In the formula, Y i is the temperature data of a certain temperature point to be deduced inside the transformer, ΔY i is the temperature change of the point under different working conditions, i=1, 2, …, m; X j is the temperature data of a certain surface temperature measurement point of the transformer, ΔX j is the temperature change of the point under different working conditions, j=1, 2, …, N.

[0016] The further technical scheme is that in order to meet the accuracy of the sensitivity analysis, the working conditions selected when calculating ΔX j and ΔY i should be similar.

[0017] The further technical scheme is that according to the temperature data of the m selected temperature points to be deduced and the temperature data of the n surface characteristic temperature measurement points, a corresponding multipoint transformation deduction model is established and is expressed as follows:

[0018]

[0019] In the formula, Y matrix is the temperature matrix of the temperature points to be deduced, Y i represents the temperature data of the i-th selected temperature point to be deduced, i=1, 2, …, m; X matrix is the temperature matrix of the selected surface characteristic temperature measurement points, X j represents the temperature data of the j-th surface characteristic temperature measurement point, j=1, 2, …, n; K matrix is the parameter matrix of the established multipoint transformation deduction model, K ij represents the contribution of the j-th surface characteristic temperature measurement point X j to the i-th temperature point to be deduced Y i in the multipoint transformation deduction model.

[0020] Further technical solutions are as follows: solving the multi-point transformation deduction model based on temperature data, comprising:

[0021]

[0022] In the formula, Y matrix is a temperature matrix of a temperature point to be deduced, (Y i ) w represents specific temperature data of the i th temperature point to be deduced under the w th group of working conditions calculated by the simulation model, i = 1, 2,..., m; X matrix is a temperature matrix of a selected surface feature temperature measuring point, (X j ) w represents specific temperature data of the j th surface feature temperature measuring point under the w th group of working conditions calculated by the simulation model, j = 1, 2,..., n; K matrix is a parameter matrix of the multi-point transformation deduction model to be solved, (K ij ) w represents the j th surface feature temperature measuring point X j under the w th group of working conditions. i The corresponding contribution in the multi-point transformation deduction model.

[0023] Further technical solutions are as follows: taking the transformer structure as a reference, combining physical effects that exist in actual operation and have an influence on the temperature distribution law of the transformer, and establishing a transformer temperature field simulation model, comprising:

[0024] In the COMSOL simulation software, a transformer geometric model is established by simplifying each part of the transformer and selecting appropriate materials to fill in.

[0025] Reasonably setting related parameters of the domain effect and the heat transfer effect of the boundary part of the transformer temperature field part to simulate the temperature field change in the actual operation of the transformer; wherein the domain effect includes internal heat sources of the transformer and losses in the operation process, and the heat transfer effect includes heat conduction, heat convection and heat radiation between the transformer and the external environment.

[0026] The transformer geometric model is meshed, and frequency domain steady-state calculation of the temperature field is performed to obtain stable temperature field results of each coordinate point of the transformer geometric model.

[0027] Further technical solutions are as follows: the internal heat sources of the transformer and the losses in the operation process include:

[0028] Electromagnetic loss of the nonlinear material core, including hysteresis loss and eddy current loss;

[0029] Joule heat of the coil winding, the heat generation of which is related to the current size and the resistance value of the coil.

[0030] Further technical solutions are as follows: the method further comprises:

[0031] The temperature data of the surface feature temperature measurement points under different verification conditions are selected and substituted into the multi-point transformation deduction model to obtain the temperature data of the temperature points to be deduced.

[0032] The temperature data corresponding to the temperature points to be deduced under the corresponding verification conditions are calculated by using the transformer temperature field simulation model.

[0033] The two types of temperature data are compared to verify the reliability of the prediction performance of the multi-point transformation deduction model.

[0034] The verification condition is a condition that is not used to establish the multi-point transformation deduction model.

[0035] In a second aspect, the application also provides a computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the method of the first aspect when executing the computer program.

[0036] The beneficial technical effects of the application are:

[0037] (1) Easy operation. The internal structure of the transformer is complex, and there are many factors affecting the temperature field. Direct measurement of the internal temperature points to be measured of the transformer not only requires higher tolerance of the measuring equipment in harsh environments, but also requires intrusion into the transformer when changing the temperature measurement point position, which is tedious. In the transformer internal temperature field deduction method based on multi-point transformation, only the temperature data provided by the reliable simulation model is used to establish the deduction model, and then the temperature of part of the surface temperature measurement points is combined with the deduction model to calculate the internal temperature of the transformer during operation.

[0038] (2) Simple algorithm. Compared with the modeling process of other algorithms, the multi-point transformation deduction model is easy to implement and fast in calculation, and can achieve good prediction performance of the specific temperature of the internal temperature points to be deduced in the case of a large number of selected points. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 is the flowchart of the transformer internal temperature field deduction method based on multi-point transformation provided by the application.

[0040] Figure 2 is the schematic diagram of the simplified geometric structure of the transformer temperature field simulation model provided by the application.

[0041] Figure 3 is the distribution diagram of the transformer internal temperature points to be deduced and part of the surface temperature measurement points on the transformer simulation model selected by the application.

[0042] Figure 4 is the flowchart of the transformer temperature field calculation provided by the application.

[0043] Figure 5 is a schematic diagram of the grid division result of the transformer geometric model provided in the present application.

[0044] Figure 6 is a schematic diagram of the distribution of the selected surface feature temperature measurement points on the transformer simulation model provided in the present application.

[0045] Figure 7 is a diagram of the error size under different verification conditions obtained by the reliability verification of the deduction model provided in the present application.

[0046] Figure 8 is an internal structure diagram of the computer device provided in the present application. DETAILED DESCRIPTION

[0047] The specific embodiments of the present application will be further described in combination with the accompanying drawings.

[0048] Referring to Figure 1 , the present application provides a transformer internal temperature field deduction method based on multi-point transformation, which specifically includes the following steps:

[0049] Step 1: Taking the structure of the transformer as a reference, a transformer temperature field simulation model is established in combination with the physical effects that have an influence on the temperature distribution law of the transformer in actual operation. As shown in Figure 2 , the geometric structure of the transformer temperature field simulation model includes a transformer tank 1, a plurality of cooling fins 2 distributed on the outside of the tank 1, and an iron core 3, a primary winding 4 and a secondary winding 5 wound on the iron core 3 located inside the tank 1. This step can be implemented in COMSOL simulation software.

[0050] Step 2: Select N surface temperature measurement points at each part of the surface of the transformer tank, denoted as X1, X2, X3, …, X N , and select m internal temperature points to be deduced in the transformer, denoted as Y1, Y2, …, Y m , and the position distribution of each point on the transformer simulation model is as shown in Figure 3 .

[0051] Step 3: Calculate the temperature data of the N surface temperature measurement points and the m internal temperature points to be deduced under different conditions by using the transformer temperature field simulation model.

[0052] Step 4: Solve the multi-point transformation deduction model based on the calculated temperature data. The multi-point transformation deduction model is used to calculate the temperature data of the internal temperature points to be deduced in the transformer under any condition in combination with the temperature data of the plurality of surface temperature measurement points obtained.

[0053] In the embodiment, the above method steps are realized by selecting surface temperature measuring points on the transformer external box and performing multi-point transformation on the surface temperature measuring points, so as to realize temperature deduction of the internal points of the transformer which cannot be directly measured. Compared with the existing direct measurement method of transformer temperature rise, the method has the advantages of simple operation, no need of pre-embedded sensor and invasive temperature measurement experiment, and can deduce the temperature of each point in the transformer.

[0054] In step 1, the specific process of establishing the transformer temperature field simulation model is as shown in Figure 4 . First, simplify each part of the transformer and select appropriate materials to fill in to establish a transformer geometric model as shown in Figure 2 . The specific simplification method is: ignoring the structures such as insulation layer, pad, fan and the like which have little effect on the temperature field, equivalent the transformer external box to a cuboid box 1 with fins 2; equivalent the originally stacked silicon steel sheets to two pieces of low-loss soft magnetic material of the same size and ignore the gap of the core gasket; assume that the space density inside and outside the box is stable, the external environment is stable, there is no forced convection phenomenon, and the internal fluid flow rate is not considered; select the materials shown in Table 1 to fill in the corresponding geometric model.

[0055] Table 1 electromagnetic parameters and heat transfer parameters of oil-immersed transformer

[0056]

[0057] Then, the related parameters of the domain effect affecting the transformer temperature field and the heat transfer effect of the boundary part are reasonably set to accurately simulate the temperature field change in the actual operation of the transformer. Among them, the domain effect includes the internal heat source of the transformer and the loss in the running process, which mainly includes the electromagnetic loss of the nonlinear material core and the Joule heat of the coil winding, and the calculation formula of the electromagnetic loss is as follows:

[0058] P Fe =P h +P c +P e =K h ·B m 2 ·f+K c ·(f·B m ) 2 +K e ·(f·B m ) 2 (1)

[0059] In formula (1), B m is the maximum magnetic flux peak value, f is the frequency, K h is a constant reflecting the influence of hysteresis loss on power, K c is a constant reflecting the influence of eddy current loss on electromagnetic loss, and Ke is a constant representing the effect of stray loss on electromagnetic loss. Hysteresis loss (P h ) refers to the energy lost by the magnetic core material as the magnetic domains overcome the friction of the inter-domain walls. This part of the loss is converted into heat energy generated by the magnetic core. Eddy current loss (P c ) is caused by the excitation of eddy currents due to the finite resistivity of the core material at high frequencies. Stray loss (P e ) is the loss caused by the magnetic hysteresis effect. In this model, stray loss is ignored because it is relatively small.

[0060] The Joule heat of the coil winding is mainly related to the current size and the resistance value of the coil. Considering the coil Joule heat as a heat source and considering its geometric structure, the formula for the coil Joule heat is derived as follows:

[0061]

[0062] In equation (2), Q is the total Joule heat of the coil, q0 is the unit Joule heat generated by the coil, r1 is the outer diameter of the ring-shaped coil, r2 is the inner diameter of the coil, I is the current flowing through the coil, p is the electrical conductivity, N is the number of turns of the coil, and h is the convective heat transfer coefficient.

[0063] The heat transfer effect includes heat conduction, heat convection, and heat radiation between the transformer and the external environment. The heat conduction equation is expressed as:

[0064]

[0065] In equation (3), AT is the temperature difference between the transformer and the external environment, l is the thermal conductivity, S is the heat transfer contact area of the transformer, and L is the thickness of the core.

[0066] The heat convection equation is approximately expressed as:

[0067] q = h · AT = Q / S (4)

[0068] In equation (4), q is the heat flux density per unit area.

[0069] Heat radiation is obtained by solving the Stefan-Boltzmann equation, which can obtain the energy distribution and transmission rate in the radiation heat transfer process, and is expressed as:

[0070]

[0071] In equation (5), e is the emissivity (0-1), d is the Stefan-Boltzmann constant, T1 and T2 are the temperatures of radiation surfaces 1 and 2, respectively.

[0072] Finally, the transformer geometry model is meshed and the temperature field in the frequency domain is calculated to obtain the stable temperature field results of each coordinate point of the transformer geometry model. In this embodiment, free tetrahedral mesh is selected in meshing, the maximum cell size is 0.366 m, the minimum cell size is 0.0457 m, the maximum cell growth rate is 1.45, the curvature factor is 0.5, the narrow area resolution is 0.6, and the meshing result is as shown in Figure 5 It should be noted that the temperature field calculation in the COMSOL simulation model is based on the above heat source and three heat transfer equations, and the form of expression may be different, but the principle and description are the same.

[0073] In step 4, solving the multi-point transformation deduction model based on the calculated temperature data specifically includes: first, performing sensitivity analysis on the temperature data of the N surface temperature measurement points corresponding to different working conditions, selecting n points (n < N) with higher sensitivity, and renumbering them as X1, X2, …, Xn, and then establishing a multi-point transformation deduction model based on the temperature data of the m temperature points to be deduced and the n surface characteristic temperature measurement points, and the model is expressed as follows: n As the surface feature temperature measurement points used for modeling the deduction model, the position distribution on the transformer simulation model is as shown in Figure 6 In this embodiment, the sensitivity is defined as the ratio of the relative change of the temperature of the surface temperature measurement point to the relative change of the temperature of the temperature point to be deduced, that is, the sensitivity of the temperature point X to Y should be expressed as:

[0074]

[0075] In formula (6), Y i is the temperature data of a certain temperature point to be deduced inside the transformer, ΔY i is the temperature change of the point under different working conditions, i = 1, 2, …, m. X j is the temperature data of a certain surface temperature measurement point of the transformer, ΔX j is the temperature change of the point under different working conditions, j = 1, 2, …, N. In order to ensure the accuracy of the sensitivity analysis, the working conditions selected when calculating ΔX j and ΔY i should be similar, so that ΔX j and ΔY i are as small as possible, and the final selection result will comprehensively consider the sensitivity values of each surface temperature measurement point X to each Y.

[0076] Secondly, according to the temperature data of the m temperature points to be deduced and the temperature data of the n surface characteristic temperature measurement points, a corresponding multi-point transformation deduction model is established, which is expressed as follows:

[0077]

[0078] In formula (7), Y matrix is the temperature matrix of the temperature points to be deduced, Y i represents the temperature data of the selected i-th temperature point to be deduced, i = 1, 2, …, m. X matrix is the temperature matrix of the selected surface feature temperature measurement points, X j represents the temperature data of the j-th surface feature temperature measurement point, j = 1, 2, …, n. K matrix is the parameter matrix of the established multi-point transformation deduction model, K ij represents the temperature data of the j-th surface feature temperature measurement point X j to the i-th temperature point to be deduced Y i corresponding contribution in the multi-point transformation deduction model.

[0079] Finally, the process of solving the multi-point transformation deduction model based on the temperature data includes:

[0080]

[0081] In formula (8), the specific temperature data in X and Y matrices are calculated based on the transformer temperature field simulation model in step 1 in this solving process, and it is required to select reasonable working conditions to avoid singular matrix affecting the calculation results. Y matrix is the temperature matrix of the temperature points to be deduced, (Y i ) w represents the specific temperature data of the i-th temperature point to be deduced under the w-th group of working conditions calculated by the simulation model, i = 1, 2, …, m; X matrix is the temperature matrix of the selected surface feature temperature measurement points, (X j ) w represents the specific temperature data of the j-th surface feature temperature measurement point calculated by the simulation model under the w-th group of working conditions, j = 1, 2, …, n; K matrix is the parameter matrix of the multi-point transformation deduction model to be solved, (K ij ) w represents the temperature data of the j-th surface feature temperature measurement point X j to the i-th temperature point to be deduced Y i corresponding contribution in the multi-point transformation deduction model.

[0082] Optionally, the transformer internal temperature field derivation method based on multipoint transformation further comprises step 5: selecting the temperature data of the surface characteristic temperature measurement points under different verification conditions, substituting the temperature data into the solved multipoint transformation derivation model to obtain the temperature data of the temperature points to be derived, and using the transformer temperature field simulation model to calculate the temperature data corresponding to the temperature points to be derived under the corresponding verification conditions, and comparing the two types of temperature data to verify the reliability of the prediction performance of the multipoint transformation derivation model. The verification condition is a condition that is not used to establish the multipoint transformation derivation model. It should be noted that in actual application, after selecting N measurable surface temperature measurement points on each part of the transformer tank, the multipoint transformation derivation model also needs to select several surface characteristic temperature measurement points with higher sensitivity according to the sensitivity analysis method in step 4, and then substitute the temperature data into the multipoint transformation derivation model to calculate the temperature data of the transformer internal temperature points to be derived under any condition. The derived transformer internal temperature data is more accurate.

[0083] The following takes a transformer of model SF11-M-8000 / 110 as an example to introduce the multipoint transformation derivation process of the temperature points to be derived on each three-phase winding in the transformer in combination with the specific contents of each step of the above method. First, according to step 1, simplify the structure of the transformer which has little effect on the temperature field and is difficult to model, and establish a transformer temperature field simulation model taking the actual transformer of model SF11-M-8000 / 110 as a reference. According to the selection principle in step 2, select N = 14 transformer surface temperature measurement points and m = 3 transformer internal temperature points to be derived as shown in Table 2. Figure 3

[0084] Table 2 Coordinates of transformer surface temperature measurement points and internal temperature points to be derived

[0085]

[0086] After completing the selection of measurable points, the sensitivity analysis is performed on the 14 transformer surface temperature measurement points shown in Table 1. Figure 3

[0087] (1) Select five conditions with little difference, and set the condition as: ① h = 10, T = 20℃, I = 60A; ② h = 10, T = 20.5℃, I = 60A; ③ h = 9, T = 20℃, I = 60A; ④ h = 10, T = 20℃, I = 61A; ⑤ h = 15, T = 20℃, I = 62A. Take condition ① as a reference, calculate the sensitivity corresponding to conditions ⑤-①, ④-①, ③-①, ②-① according to formula (6), and form a 3 × 14 matrix corresponding to the surface 14 points and the internal 3 points to obtain sensitivity matrices S1, S2, S3, S4. Add and average the corresponding elements in the four sensitivity matrices to obtain matrix S​​a , S a in the element S aij represents S(X j , Y i ) is the sensitivity of X j to Y i , wherein S a Each element value is shown in Table 3 (to two decimal places). It is worth noting that the working conditions of ⑤ and ① seem to be quite different, but the increase in current is actually equivalent to increasing the heat source heat, and the increase in the convective heat transfer coefficient h is to increase the heat exchange with the external environment, which is equivalent to more heat dissipation, so the calculated ΔX and ΔY can also be guaranteed to be smaller.

[0088] Table 3 Sensitivity matrix S a Each element value

[0089]

[0090] (2) Because each column of matrix S a is the value of the surface temperature measurement point to each internal point to be deduced, in order to comprehensively consider the sensitivity of each point, the average value of the sum of each column element should be selected, and the sensitivity is arranged from large to small as shown in Table 4 (the sensitivity size is kept to two decimal places).

[0091] Table 4 Sensitivity matrix S a Each column element average value is sorted from large to small

[0092]

[0093] (3) In summary, take n = 7, select Figure 3 the 7 points with the highest sensitivity in S Figure 6 , X2, X3, X5, X6, X7, X8, X9, and renumber them as X1, X2, …, X7, to get the surface feature temperature measurement point distribution diagram as shown in

[0094] After completing the sensitivity analysis and selecting the points, calculate the temperature field size under several different working conditions to get the specific temperature data of each selected point under the corresponding working condition. Select 7 groups to solve the multi-point transformation deduction model, and finally get the specific value of K matrix as shown in Table 5 (to two decimal places).

[0095] Table 5 K matrix element value

[0096]

[0097] After obtaining the model, in order to verify the reliability of the model and its internal temperature prediction performance under different working conditions, 15 working conditions shown in Table 6 are selected, and the internal extrapolation temperature of the surface feature temperature measurement point is calculated based on the extrapolation model, and the actual temperature data of the corresponding point in the simulation model is compared, and the specific data shown in the following table is obtained:

[0098] Table 6 Simulation calculation and model calculation comparison data of internal extrapolation temperature

[0099]

[0100]

[0101] The error size of each internal extrapolation point under each working condition in Table 6 is fitted into a curve, and the error size corresponding to each working condition is obtained as shown in Figure 7 It can be seen that after selecting the seven surface feature temperature measurement points with the highest sensitivity, the multi-point transformation extrapolation model formed by the temperature error is mostly within 1%, and a few forced convection conditions (i.e. h is very large) have relatively large error due to the selection of similar working conditions, which is within 3%. The above analysis shows that the multi-point transformation extrapolation model has good prediction performance. For other arbitrary temperature points inside the transformer, the same as the above Y1, Y2, Y3 multi-point transformation extrapolation model establishment method can be obtained, and finally the extrapolation function relationship of each point is obtained, that is, the temperature field distribution of any position inside the transformer can be deduced.

[0102] Based on the same inventive concept, the application also provides a computer device, which can be a terminal, and its internal structure diagram can be as shown in Figure 8 The computer device includes a processor, a memory, a communication interface, a display unit and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through WIFI, mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement the above-mentioned transformer internal temperature field extrapolation method based on multi-point transformation. The display unit of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device. It can also be an external keyboard, touchpad or mouse, etc.

[0103] Those skilled in the art can understand that Figure 8 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0104] The above is only the preferred embodiment of the present application, and the present application is not limited to the above embodiments. It can be understood that other improvements and changes directly derived or thought of by those skilled in the art without departing from the spirit and concept of the present application should be considered to be within the protection scope of the present application.

Claims

1. A method for extrapolating the internal temperature field of a transformer based on multi-point transformation, characterized in that, The method includes: Taking the transformer structure as a reference and combining the physical effects that affect the temperature distribution of the transformer in actual operation, a transformer temperature field simulation model is established. Select various locations on the surface of the transformer tank. N Several surface temperature measurement points, and selected inside the transformer m One temperature point to be calculated; Calculation using the transformer temperature field simulation model N Surface temperature measuring points and m Temperature data corresponding to the temperature points to be simulated under different operating conditions; The multi-point transformation model is solved based on the calculated temperature data. The multi-point transformation model is used to combine the temperature data of multiple surface temperature measurement points to calculate the temperature data of the temperature point to be deduced inside the transformer under any operating condition. The step of solving the multi-point transformation derivation model based on the calculated temperature data includes: For the calculated N Sensitivity analysis was performed on the temperature data of each surface temperature measuring point under different operating conditions, and the points with higher sensitivity were selected. n Each point is used as a surface feature temperature measurement point; the sensitivity is defined as the ratio of the relative temperature change at the surface temperature measurement point to the relative temperature change at the point to be calculated, expressed as: In the formula, Y i For the temperature data of a certain temperature point inside the transformer to be estimated, Δ Y i This represents the temperature change at that point under different operating conditions. i =1,2,…, m ; X j Δ is the temperature data of a selected surface temperature measurement point on the transformer surface. X j This represents the temperature change at that point under different operating conditions. j =1,2,…, N ; According to the selected m Temperature data of the temperature points to be predicted and n Temperature data from several surface feature temperature measurement points are used to establish a corresponding multi-point transformation model, which is then solved based on the temperature data. The multi-point transformation model is expressed as follows: In the formula, Y The matrix is ​​the temperature matrix for the temperature point to be derived. Y i Indicates the selected first i Temperature data for the temperature points to be extrapolated i =1,2,…, m ; X The matrix is ​​the temperature matrix of the selected surface feature temperature measurement points. X j Indicates the first j Temperature data from one surface feature temperature measurement point. j =1,2,…, n ; K The matrix represents the parameter matrix of the established multi-point transformation derivation model. K ij Indicates the first j Temperature measurement points on surface features X j For the i One temperature point to be calculated Y i The corresponding contribution in the multi-point transformation derivation model.

2. The method for extrapolating the internal temperature field of a transformer based on multi-point transformation according to claim 1, characterized in that, To ensure the accuracy of sensitivity analysis, Δ is calculated. X j Δ Y i The operating conditions selected at that time should be similar.

3. The method for extrapolating the internal temperature field of a transformer based on multi-point transformation according to claim 1, characterized in that, Solving a multi-point transformation model based on the temperature data includes: In the formula, Y The matrix is ​​the temperature matrix of the temperature point to be derived. Y i ) w Represents the first calculated by the simulation model i The temperature point to be calculated is at the first w Specific temperature data under operating conditions. i =1,2,…, m ; X The matrix is ​​the temperature matrix of the selected surface feature temperature measurement points, ( X j ) w Represents the first calculated by the simulation model j The surface feature temperature measurement point at the first w Specific temperature data under operating conditions. j =1,2,…, n ; K The matrix is ​​the parameter matrix of the multi-point transformation derivation model to be solved. K ij ) w Indicates the first w The first working condition j Temperature measurement points on surface features X j For the i One temperature point to be calculated Y i The corresponding contribution in the multi-point transformation derivation model.

4. The method for extrapolating the internal temperature field of a transformer based on multi-point transformation according to claim 1, characterized in that, The transformer temperature field simulation model is established, taking the transformer structure as a reference and considering the physical effects that affect the temperature distribution of the transformer during actual operation. This model includes: In COMSOL simulation software, the various parts of the transformer are simplified and appropriate materials are selected for filling to establish the transformer's geometric model; The relevant parameters affecting the temperature field of the transformer, including the intra-domain effect and the heat transfer effect at the boundary, are reasonably set to simulate the temperature field changes during actual operation of the transformer. The intra-domain effect includes the heat source inside the transformer and the losses during operation, while the heat transfer effect includes heat conduction, heat convection, and heat radiation between the transformer and the external environment. The transformer geometric model is meshed, and steady-state temperature field calculations are performed in the frequency domain to obtain the stable temperature field results at each coordinate point of the transformer geometric model.

5. The method for extrapolating the internal temperature field of a transformer based on multi-point transformation according to claim 4, characterized in that, The internal heat source of the transformer and the losses during operation include: Electromagnetic losses in nonlinear material cores include hysteresis losses and eddy current losses. Joule heating of coil windings is related to the magnitude of the current and the resistance of the coil.

6. The method for extrapolating the internal temperature field of a transformer based on multi-point transformation according to claim 1, characterized in that, The method further includes: Temperature data from surface feature temperature measurement points under different verification conditions are selected and substituted into the multi-point transformation deduction model to calculate the temperature data of the temperature point to be deduced. The temperature data corresponding to the temperature point to be simulated under the corresponding verification working condition are calculated using the transformer temperature field simulation model. The two types of temperature data were compared to verify the reliability of the prediction performance of the multi-point transformation inference model; The verification condition refers to the condition that was not used to establish the multi-point transformation deduction model.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

Images