Transformer operation state monitoring method and system, electronic equipment and storage medium

Abstract

The invention discloses a transformer operation state monitoring method and system, electronic equipment and a storage medium, and relates to the technical field of urban carbon neutralization planning. According to the method, transformer internal parameters of which the influence degree on safe operation of a transformer exceeds a preset degree under a power flow switching working condition are acquired as target internal parameters, so that a transformer performance sample fine management database is constructed; based on the database, establishing a linearization relation of target external reference, including voltage and current, of the transformer participating in the target internal reference; and establishing a sensitivity equation of the target internal parameter relative to the target external parameter based on the relation, establishing a high-dimensional feasible region participating in the insulation life of the transformer in the target based on the sensitivity equation, and determining a low-order safe operation region of the transformer based on the high-dimensional feasible region so as to realize the evaluation of the operation state based on whether the working condition point of the transformer is located in the operation region or not. Therefore, the problem that the influence of frequent power flow switching on short-term operation characteristics, long-term operation life and load capacity of the transformer cannot be quantitatively analyzed is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban carbon neutrality planning, and more specifically, to a transformer operating status monitoring method, system, electronic device and storage medium. Background Art

[0002] Based on the operation experience, it is found that if the main grid power transformer is frequently in the reverse transmission working mode, a series of operation problems will appear. Taking the #1 main transformer of a 110 kV substation as an example (model SZ10-6300 / 110, rated voltage 110 / 10kV, on-load voltage regulation), it switched the power flow twice from 0:00 to 24:00 on a certain day, at 10:16 the power was switched from forward transmission to reverse transmission, and at 16:53 the power was switched from reverse transmission to forward transmission. During the period, the maximum reverse transmission load reached 66% at 13:00, and the maximum forward transmission load was 75% at 20:37. During the period, in order to achieve the maximum output of photovoltaic power, the photovoltaic power generation node continuously increased the output voltage. Monitoring data shows that during reverse transmission, the average voltage on the low-voltage side is about 2% higher than the rated voltage, and the high-voltage side needs to continuously adjust the tap switch to reduce the voltage. In addition, the operation and maintenance personnel found that the noise of the main transformer during reverse transmission was significantly higher than usual, and the oil temperature was 1-2°C higher than that of the same power and roughly equivalent ambient temperature. From a theoretical analysis, the reasons for the increase in transformer operating oil temperature and noise are roughly due to the reverse transmission of the power flow after the new energy is connected to the grid, the increase in harmonics, and the transformer excitation distortion and additional losses. However, due to the lack of relevant theoretical and experimental research, it is impossible to quantitatively analyze the impact of frequent power flow switching on the short-term operating characteristics, long-term operating life and load capacity of the transformer, and it is even more impossible to formulate targeted power grid dispatching and coordination management strategies.

[0003] In summary, in order to accurately judge the operating status of the transformer and lay the foundation for subsequent coordinated management, it is necessary to study a method for evaluating the performance of power transformers under frequent power flow switching, so as to efficiently and accurately evaluate the transformer status of bidirectional power flow. Summary of the invention

[0004] The object of the present invention is to provide a transformer operating status monitoring method, system, electronic device and storage medium to solve the problem of being unable to quantitatively analyze the impact of frequent power flow switching on the transformer's short-term operating characteristics, long-term operating life and load capacity.

[0005] In a first aspect, the present invention provides a method for monitoring a transformer operating state, comprising:

[0006] Obtaining target internal parameters to construct a transformer performance sample refined management database, wherein the target internal parameters are transformer internal parameters whose influence on the safe operation of the transformer exceeds a preset degree under power flow switching conditions;

[0007] Based on the transformer performance sample refined management database, a linear relationship between the target external parameters and the target internal parameters of the transformer is established, wherein the target external parameters are voltage and current;

[0008] Based on the linearized relationship, establishing a sensitivity equation of the target internal parameter relative to the target external parameter;

[0009] Based on the sensitivity equation, construct a high-dimensional feasible domain between the insulation life of the participating transformers within the target;

[0010] Based on the high-dimensional feasible domain, determining a low-order safe operation domain of the transformer;

[0011] Based on whether the operating point of the transformer is located in the low-order safe operation domain, the operating state of the transformer is evaluated.

[0012] In a preferred solution, the target internal parameters include: hot spot temperature, harmonic content and magnetic flux inside the transformer.

[0013] In a preferred solution, the step of obtaining the target internal parameter includes:

[0014] Acquire first historical sample data, wherein the first historical sample data includes attribute information generated during the design, manufacture, maintenance and test of the transformer, and operating parameters under power flow switching conditions, wherein the attribute information is inherent information describing the characteristics and / or state of the transformer;

[0015] Eliminate erroneous data in the first historical sample data to obtain second historical sample data;

[0016] Redundant data in the second historical sample data is eliminated to obtain the target internal reference.

[0017] In a preferred solution, the step of removing erroneous data from the first historical sample data includes:

[0018] The erroneous data in the first historical sample data is eliminated through the 3σ data analysis method.

[0019] In a preferred solution, before removing the erroneous data in the first historical sample data by the 3σ data analysis method, the method further includes:

[0020] The skewness of the distribution of the first historical sample data is transformed and corrected through the BOX-COX transformation.

[0021] In a preferred solution, the removing redundant data from the second historical sample data includes:

[0022] Determine the covariance matrix of the second historical sample data by principal component analysis;

[0023] Calculating eigenvalues ​​of the covariance matrix and eigenvectors corresponding to the eigenvalues;

[0024] Arrange the eigenvectors in rows according to a preset order, and take the first preset rows to form a target matrix, wherein the preset order is that the eigenvalues ​​corresponding to the eigenvectors are arranged in order from large to small and from top to bottom;

[0025] Based on the target matrix, the reserved data in the second historical sample data is determined, and the second historical sample data except the reserved data is removed as the redundant data.

[0026] In a preferred solution, the establishing of a linear relationship between the target external parameter and the target internal parameter of the transformer includes:

[0027] Establishing the display relational expressions of the target internal parameters under the condition of reverse power flow transmission;

[0028] Based on the display relationship expression, a linear relationship between the target external parameter and the target internal parameter is established by calculating the sensitivity coefficient.

[0029] In a preferred solution, the display relationship expression of each target internal parameter established under the reverse flow transmission condition includes:

[0030] Establishing a display relationship expression of the hot spot temperature under the condition of reverse flow transmission as the first expression;

[0031] The first expression is:

[0032]

[0033] Among them, F AA is the accelerated aging factor of the transformer insulation life, B is a constant, T is the hot spot temperature, is the reference value of hot spot temperature;

[0034] Establishing a display relationship expression of the harmonic content under the reverse power flow transmission condition as the second expression;

[0035] The second expression is:

[0036]

[0037] Among them, P h is the hysteresis loss, P CO,N is the power of the Nth harmonic, R r,h is a certain resistance element, h is the harmonic order, Ih is the hth harmonic current, U h is the voltage of the hth harmonic, and U1 is the fundamental voltage;

[0038] Establishing a display relationship expression of the magnetic flux under the condition of reverse power flow transmission as the third expression;

[0039] The third expression is:

[0040]

[0041] Among them, R a ,R b ,R c is the resistance, I ea is the primary current, N a is the number of turns on the original side.

[0042] In a preferred solution, the linearized relationship between the target external parameter and the target internal parameter is established by calculating the sensitivity coefficient based on the display relationship expression, including:

[0043] Based on the display relationship expression, combined with the transformer performance sample refined management database, under the condition of changing the low-voltage side voltage and the inflow current of the transformer under the reverse power transmission condition, the change amount of each target internal parameter of the transformer is collected;

[0044] By calculating the sensitivity coefficient of the change amount of each target internal parameter, a linear relationship between the target external parameter and the target internal parameter is established.

[0045] In a preferred solution, establishing a sensitivity equation of the target internal parameter relative to the target external parameter based on the linearized relationship includes:

[0046] Based on the linearized relationship, combined with the transformer performance sample refined management database, select the section with the closest Euclidean distance to the current operating state of the transformer as a reference, use the historical similar state as the initial search point of the step search algorithm, start the search from the current given power grid operation control variable, and gradually increase or decrease the preset given step size;

[0047] For each given search value obtained, calculate the approximation error between the actual value and the observed value;

[0048] With the goal of making the approximation error smaller than a preset allowable error, the sensitivity equation is output.

[0049] In a preferred embodiment, the sensitivity equation is:

[0050] Y C =Y 0 +K1*(VC -V 0 )+K2*(I C -I 0 ),Y=[T,G,ψ] T

[0051] Among them, Y C and Y 0 are the maximum and minimum values ​​of hot spot temperature rise, harmonic content or magnetic flux, respectively, V C and V 0 are the maximum and minimum voltage amplitudes, I C and I 0 are the maximum and minimum values ​​of the current respectively, K1 and K2 are the independent variable thresholds.

[0052] In a preferred solution, constructing a high-dimensional feasible domain between the transformer insulation life involved in the target includes:

[0053] Based on the sensitivity equation, combined with the transformer performance sample refined management database, the target internal parameter is used as the control variable, the insulation life of the transformer is used as the state variable, and the relationship between the control variable and the state variable is constructed in the global space by using the Monte Carlo sampling method to obtain a pre-selected high-dimensional feasible domain;

[0054] Based on actual grid planning data, the wind or light quota and access location in the grid system are continuously changed to correct the transformer performance sample refined management database and conduct offline Monte Carlo numerical simulation experiments;

[0055] Based on the revised transformer performance sample refined management database and the data obtained from the simulation experiment, the preselected high-dimensional feasible domain is updated to obtain the high-dimensional feasible domain.

[0056] In a preferred solution, determining the low-order safe operation domain of the transformer based on the high-dimensional feasible domain includes:

[0057] By using a hyperspace projection method, the high-dimensional feasible domain is subjected to dimensionality reduction processing, and the expected life of the transformer is used as a cutting plane;

[0058] Projecting the intersection line of the cutting plane and the high-dimensional feasible domain after the dimensionality reduction process onto the plane formed by the target internal parameters to form a critical life curve;

[0059] Based on the critical life curve, the low-order safe operation domain is established.

[0060] In a second aspect, the present invention further provides a transformer operation status monitoring system, which is used to run the transformer operation status monitoring method provided by the first aspect of the present invention, and the system comprises:

[0061] A collection unit, used to obtain target internal parameters to build a transformer performance sample refined management database, wherein the target internal parameters are transformer internal parameters whose influence on the safe operation of the transformer exceeds a preset degree under power flow switching conditions;

[0062] A first processing unit is used to establish a linear relationship between a target external parameter and a target internal parameter of the transformer based on the transformer performance sample refined management database, wherein the target external parameter is voltage and current;

[0063] A second processing unit, configured to establish a sensitivity equation of the target intrinsic parameter relative to the target extrinsic parameter based on the linearized relationship;

[0064] A third processing unit is used to construct a high-dimensional feasible domain between the insulation lives of the participating transformers within the target based on the sensitivity equation;

[0065] A fourth processing unit, configured to determine a low-order safe operation domain of the transformer based on the high-dimensional feasible domain;

[0066] An evaluation unit is used to evaluate the operating state of the transformer based on whether the operating point of the transformer is located in the low-order safe operation domain.

[0067] In a third aspect, the present invention provides an electronic device, the electronic device comprising:

[0068] processor;

[0069] a memory for storing instructions executable by the processor;

[0070] The processor is used to execute the transformer operating status monitoring method provided in the first aspect of the present invention.

[0071] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method provided by the first invention of the present invention are implemented.

[0072] In order to achieve the above-mentioned purpose, the transformer operation state monitoring method, system, electronic device and storage medium provided by the present invention are constructed by obtaining the transformer internal parameters whose influence on the safe operation of the transformer exceeds the preset degree under the power flow switching condition as the target internal parameters, so as to construct a transformer performance sample fine management database; then, based on the transformer performance sample fine management database, a linearized relationship between the target external parameters including voltage and current of the transformer and the target internal parameters is established; after that, based on the linearized relationship, a sensitivity equation of the target internal parameter relative to the target external parameter is established, and based on the sensitivity equation, a high-dimensional feasible domain between the target internal parameter and the insulation life of the transformer is constructed; based on the high-dimensional feasible domain, the low-order safe operation domain of the transformer is determined; thereby, based on whether the operating point of the transformer is located in the low-order safe operation domain, the transformer operation state is evaluated. On the one hand, the operating parameters and operation strategies of the equipment can be optimized based on the evaluation results, such as adjusting the load distribution of the transformer, reducing the loss of the equipment, etc., so as to improve the operation efficiency and economy of the equipment. On the other hand, by real-time monitoring of the transformer state, potential faults or abnormal conditions can be discovered in time, and preventive maintenance measures can be taken, thereby reducing the occurrence of sudden faults and improving the reliability and availability of the equipment. By properly evaluating the status of the transformer, equipment problems can be discovered and handled in a timely manner, thereby effectively extending the service life of the equipment and reducing downtime and maintenance costs caused by equipment failures. At the same time, it can help system operators better understand the system's load conditions and power flow, adjust system operation strategies in a timely manner, improve system responsiveness and stability, and ensure that the system can meet user needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A flowchart of a transformer operating status monitoring method provided by an embodiment of the present invention.

[0074] Figure 2 (a) is a structural diagram of an example of a sensitivity mathematical model provided in an embodiment of the present invention.

[0075] Figure 2 (b) is a distribution diagram of a sensitivity threshold example provided in an embodiment of the present invention.

[0076] Figure 3 A distribution diagram of an example of a transformer low-order safety operation domain provided in an embodiment of the present invention.

[0077] Figure 4 A distribution diagram of an example of a safety domain range corresponding to transformers with different lifespans provided in an embodiment of the present invention.

[0078] Figure 5 The embodiment of the present invention provides a method for replacing the linear boundary with Figure 4Schematic diagram of the original non-convex and complex lifetime boundary of the transformer provided.

[0079] Figure 6 A distribution diagram of an example of a low-level security operation domain provided in an embodiment of the present invention.

[0080] Figure 7 A structural diagram of a transformer operation status monitoring system provided in an embodiment of the present invention.

[0081] Figure 8 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0082] The present invention is further described in detail below through the accompanying drawings and specific embodiments.

[0083] The transformer operation status monitoring method provided by the present invention is executed on an electronic device, which may be an intelligent terminal device such as a notebook computer, a personal computer, and a tablet computer, or may be a server on the network side.

[0084] like Figure 1 As shown, the transformer operation status monitoring method provided by the embodiment of the present invention mainly includes the following steps:

[0085] 110. Obtain target internal parameters to build a transformer performance sample refined management database.

[0086] Among them, the target internal parameter is the internal parameter of the transformer whose impact on the safe operation of the transformer exceeds the preset level under the power flow switching condition.

[0087] 120. Based on the refined management database of transformer performance samples, a linear relationship between the target external parameters and target internal parameters of the transformer is established.

[0088] Among them, the target external parameters are voltage and current.

[0089] 130. Based on the linearization relationship, establish the sensitivity equation of the target internal parameter relative to the target external parameter;

[0090] 140. Based on the sensitivity equation, a high-dimensional feasible domain between the insulation life of participating transformers within the target is constructed;

[0091] 150. Based on the high-dimensional feasible domain, determine the low-order safe operation domain of the transformer;

[0092] 160. Evaluate the operating status of the transformer based on whether the operating point of the transformer is in the low-order safe operating domain.

[0093] Specifically, the target internal parameter is a parameter that has a certain influence on the safe operation of the transformer under frequent power flow switching. By establishing a linear relationship between the target external parameter and the target internal parameter of the transformer, a functional relationship and a calculation basis can be provided for calculating the sensitivity equation of the target internal parameter relative to the target external parameter in step 130, while simplifying the sensitivity calculation. The sensitivity equation of the target internal parameter relative to the target external parameter obtained in step 130 reveals the specific influence of the change of the external parameter on the internal parameter, thereby providing a basis for constructing a high-dimensional feasible domain in step 140.

[0094] More specifically, based on the content of the above embodiment, the target internal parameters include: hot spot temperature, harmonic content and magnetic flux inside the transformer.

[0095] Further, based on the content of the above embodiment, obtaining the target internal parameter includes:

[0096] Obtaining first historical sample data;

[0097] The first historical sample data includes attribute information generated during the design, manufacture, maintenance and testing of the transformer, as well as operating parameters under power flow switching conditions;

[0098] Eliminate erroneous data in the first historical sample data to obtain second historical sample data;

[0099] The redundant data in the second historical sample data is eliminated to obtain the target internal reference.

[0100] Specifically, attribute information refers to the information inherent in the transformer during various stages of design, manufacturing, maintenance and testing, which describes its characteristics or status. It mainly includes: excitation current, material properties, no-load loss, load loss, insulation form and other design parameters, manufacturing materials and manufacturing process information during the design and manufacturing process, as well as oil sample records, gas in oil and oil temperature during the maintenance and testing process.

[0101] When obtaining the attribute information generated by the transformer during the design, manufacturing, maintenance and testing process, as well as the operating parameters under the power flow switching condition, the following transformer load and life assessment model is considered:

[0102] 1. Transformer insulation aging model:

[0103] For insulation materials with high heat resistance, based on the relative aging rate formula shown in formula (1) (reference hot spot temperature is 110°C):

[0104]

[0105] Among them, θ h is the hot spot temperature of the transformer winding.

[0106] The relative aging rate of the insulation material at different hot spot temperatures can be calculated. The formula for calculating the insulation life loss based on the relative aging rate formula of the insulation is:

[0107]

[0108] In addition, it is known that the relationship between the transformer insulation life and the winding hot spot temperature follows the Arrhenius law and can be expressed as:

[0109]

[0110] Among them, L pu is the transformer insulation life standard value, which is the difference between the actual insulation life and the normal insulation life L n ratio; A and B are constants.

[0111] For a given load factor and ambient temperature, the accelerated aging factor F of the transformer insulation life is AA The calculation formula is:

[0112]

[0113] in, is the reference value of hot spot temperature.

[0114] In a certain time interval, the equivalent accelerated aging factor F EQA for:

[0115]

[0116] Among them, Δt n is the time interval; F AAn is the accelerated aging factor during this time interval.

[0117] It can be seen from the above transformer insulation aging model that the transformer can obtain normal insulation life by continuous operation under normal ambient temperature and rated conditions. However, when the load exceeds the nameplate rating or the ambient temperature is higher than the normal temperature, the transformer will be exposed to a certain degree of danger, which will accelerate the insulation aging.

[0118] 2. Transformer flux calculation model

[0119] Taking the single-phase four-column iron core as an example, it has a double main column side yoke structure, and the core magnetic circuit is symmetrical on the left and right. Windings are set on the two middle main columns, and the windings are connected in parallel or in series. The magnetic flux directions in the two main columns are opposite.

[0120] In engineering calculations, it is assumed that the magnetic flux of the upper and lower yokes and the side yokes is equal to half of the magnetic flux of the main column, and the cross section of the iron yoke is selected to be 1 / 2 of the cross section of the main column, that is:

[0121]

[0122] Among them, S1, S2 and S3 are the cross-sectional areas of the main column, upper yoke and side yoke respectively (unit: m 2 ). Thus we can conclude:

[0123]

[0124] Among them, B1, B2 and B3 are the magnetic flux densities of the main column, upper yoke and side yoke respectively (unit: T).

[0125] Furthermore, the no-load loss and the magnitude of the excitation current can be calculated based on the magnetic flux density.

[0126] 3. Hot spot temperature calculation model

[0127] Taking a type of natural oil circulation ONAN oil-immersed transformer as an example, it is generally believed that the final hot spot temperature under any load K is equal to the sum of the ambient temperature, the top oil temperature rise, and the temperature gradient of the hot spot temperature relative to the top oil temperature:

[0128]

[0129] in, is the temperature rise of the oil temperature at the top of the winding relative to the ambient temperature when the load factor K = 1, θ amb is the ambient temperature, θ hst is the final hot spot temperature, R is the loss ratio, Hg r is the temperature gradient of the hot spot temperature to the top oil temperature when the load factor K=1, and x and y represent the indexes related to the transformer cooling structure.

[0130] Further, based on the content of the above embodiment, removing erroneous data in the first historical sample data includes:

[0131] Through the 3σ data analysis method, the erroneous data in the first historical sample data are eliminated.

[0132] It can be understood that the 3σ principle is: if a variable follows a normal distribution, and its mean is u and its standard deviation is σ, then there will be:

[0133] 1) 68% of the data will fall within u±σ, that is, the probability that the data is distributed in (u-σ,u+σ) is 0.68;

[0134] 2) 95% of the data will fall within u±2σ, that is, the probability that the data is distributed in (u-2σ,u+2σ) is 0.95;

[0135] 2) 99% of the data will fall within u±3σ, that is, the probability that the data is distributed in (u-3σ,u+3σ) is 0.99.

[0136] Therefore, when there is a data falling outside the mean u±three times the standard deviation 3σ, it can be initially regarded as abnormal data. This is because the probability of the data falling outside the mean u±three times the standard deviation 3σ is only 1%, which is a low-probability event. Therefore, this data is not considered normal data. In other words, it is an abnormal data.

[0137] Furthermore, based on the content of the above embodiment, before removing the erroneous data in the first historical sample data by the 3σ data analysis method, the method further includes:

[0138] Through the BOX-COX transformation, the skewness of the distribution of the first historical sample data is transformed and corrected.

[0139] It should be noted that BOX-COX transformation is a general term for multiple transformations, and the expression is:

[0140]

[0141] Among them, y(λ) is the new variable obtained after BOX-COX transformation, y is the original continuous dependent variable, and λ is the transformation parameter.

[0142] Specifically, before eliminating erroneous data in the first historical sample data through the 3σ data analysis method, the non-normal data is first processed into normal data through BOX-COX transformation, which is more conducive to eliminating erroneous historical sample data through the 3σ principle provided by the above embodiment.

[0143] In an optional embodiment, removing redundant data from the second historical sample data includes:

[0144] Determine the covariance matrix of the second historical sample data by principal component analysis;

[0145] Calculate the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to the eigenvalues;

[0146] Arrange the feature vectors in rows according to the preset order, and take the first preset rows to form the target matrix;

[0147] The preset order is that the eigenvalues ​​corresponding to the eigenvectors are arranged in order from large to small and from top to bottom;

[0148] Based on the target matrix, the retained data in the second historical sample data is determined, and the second historical sample data except the retained data is removed as redundant data.

[0149] In this embodiment, the principal component analysis method is used to calculate the covariance matrix to achieve data transformation, which can show the differences in multi-dimensional data, so as to filter out key information from complex data, thereby eliminating redundant historical data samples.

[0150] Specifically, taking the second historical sample data obtained after removing the erroneous data as m pieces of n-dimensional historical sample data as an example, by forming it into an n-row and m-column matrix A by column, and then centering each row of A to obtain the matrix B, the covariance matrix C can be calculated:

[0151]

[0152] After that, after finding the eigenvalues ​​of the covariance matrix C and its corresponding eigenvectors, the eigenvectors are arranged into a matrix from top to bottom according to the corresponding eigenvalues, and the first k rows are taken to form the target matrix P. Then, the data after dimensionality reduction to k can be obtained:

[0153] Y=PB (12)

[0154] Finally, the corresponding second historical sample data is retained according to the target matrix P to eliminate redundant data in the second historical sample data.

[0155] In an optional embodiment, step 120 establishes a linear relationship between a target external parameter and a target internal parameter of the transformer, specifically including:

[0156] Establish the display relation expression of each target internal parameter under the condition of reverse flow transmission;

[0157] Based on the explicit relationship expression, the linear relationship between the target external parameter and the target internal parameter is established by calculating the sensitivity coefficient.

[0158] It is understandable that, on the one hand, for transformers, the monitoring and measurement of voltage and current are relatively simple and accurate. Therefore, by establishing the display relationship expression of target internal parameters such as hot spot temperature and harmonic content, it is possible to indirectly obtain the size and change trend of the target internal parameters by measuring the changes in voltage and current. In other words, by establishing the display relationship expression of each target internal parameter under the reverse power transmission condition, the observability of the key internal parameters that have a significant impact on the insulation life of the transformer is achieved.

[0159] On the other hand, considering that the internal parameters are essentially attribute information of the transformer, although there is an internal connection between the power grid operation information data and the transformer attribute information, they belong to two different data types. Based on this, by establishing an explicit relationship expression, the two data can be integrated to form a more comprehensive and accurate data set, thereby providing a more reliable basis for transformer condition monitoring and performance evaluation. At the same time, in the process of establishing an explicit relationship expression, it is usually necessary to use mathematical models to describe the relationship between the internal parameters of the transformer. These models can be constructed based on the attribute information of the transformer (such as design parameters, structural characteristics, etc.), and verified and optimized in combination with the power grid operation information data. Therefore, establishing an explicit relationship expression helps to build a more accurate and reliable transformer condition monitoring and performance evaluation model.

[0160] Furthermore, after establishing the linear relationship between the target external parameters and the target internal parameters by calculating the sensitivity coefficient based on the display relationship expression of each target internal parameter, the key indicator parameter system for the transformer attribute information can be obtained according to the response sensitivity coefficient of each target internal parameter.

[0161] Furthermore, in an optional embodiment, under the condition of reverse power flow transmission, the display relationship expression of each target internal parameter includes:

[0162] Establishing a display relation expression of hot spot temperature under the condition of reverse flow transport as the first expression;

[0163] The first expression is:

[0164]

[0165] Among them, F AA is the accelerated aging factor of transformer insulation life, B is a constant, T is the hot spot temperature, is the reference value of hot spot temperature;

[0166] Establishing a display relation expression of harmonic content under the condition of reverse power flow transmission as the second expression;

[0167] The second expression is:

[0168]

[0169] Among them, P h is the hysteresis loss, P CO,N is the power of the Nth harmonic, R r,h is a certain resistance element, h is the harmonic order, I h is the hth harmonic current, U h is the voltage of the hth harmonic, and U1 is the fundamental voltage;

[0170] Establishing a display relationship expression of magnetic flux under the condition of reverse power flow transmission as the third expression;

[0171] The third expression is:

[0172]

[0173] Among them, R a ,R b ,R c is the resistance, I ea is the primary current, N a is the number of turns on the original side.

[0174] In an optional embodiment, based on the display relationship expression, a linear relationship between the target external parameter and the target internal parameter is established by calculating the sensitivity coefficient, including:

[0175] Based on the display relational expression and combined with the transformer performance sample refined management database, the changes in the transformer's target internal parameters are collected when the transformer's low-voltage side voltage and inflow current are changed under the reverse power transmission condition;

[0176] By calculating the sensitivity coefficient of the change of each target internal parameter, the linear relationship between the target external parameter and the target internal parameter is established.

[0177] In this embodiment, based on the display relationship expression and combined with the transformer performance sample refined management database, the sensitivity analysis method is used to change the low-voltage side voltage V and the inflow current I of the transformer under the reverse power transmission condition, collect the changes in the hot spot temperature, harmonic content and magnetic flux inside the transformer, and establish the linear relationship between the internal parameters of the transformer and the external voltage and current by calculating the sensitivity coefficient K.

[0178] In an optional embodiment, step 130 establishes a sensitivity equation of the target internal parameter relative to the target external parameter based on the linearized relationship, including:

[0179] Based on the linearization relationship and combined with the transformer performance sample refined management database, the section with the closest Euclidean distance to the current operating state of the transformer is selected as a reference, and the historical similar state is used as the initial search point of the step search algorithm. The search starts from the current given power grid operation control variable, and the preset given step size is gradually increased or decreased;

[0180] For each given search value obtained, calculate the approximation error between the actual value and the observed value;

[0181] With the goal of making the approximation error smaller than the preset allowable error, the sensitivity equation is output.

[0182] In this embodiment, a step search algorithm is used to approximate the error to meet the minimum error requirement, that is, less than the preset allowable error as the threshold range of the target output control variable and its corresponding sensitivity equation. The specific sensitivity mathematical model and threshold can be respectively as follows Figure 2 As shown in (a) and (b).

[0183] That is, first, the database is refined through the transformer performance sample management, and the section with the closest Euclidean distance to the current operating state is selected as a reference. Then, the historical similar state is used as the initial search point of the step search algorithm, and the search starts from the current given power grid operation control variable, and gradually increases or decreases a given step size. After that, for each given search value, the approximation error between the actual value and the observed value is calculated, and finally the sensitivity equation of the control variable is output with the approximation error meeting the minimum error requirement as the target.

[0184] In an optional embodiment, the sensitivity equation is:

[0185] Y C =Y 0 +K1*(V C -V 0 )+K2*(I C -I 0 ),Y=[T,G,ψ] T (16)

[0186] Among them, Y C and Y 0 is the maximum and minimum value of hot spot temperature rise\harmonic content\magnetic flux; V C and V 0 is the maximum and minimum value of the voltage amplitude; I C and I 0 are the maximum and minimum values ​​of the current; K1 and K2 are the independent variable thresholds.

[0187] In an optional embodiment, step 140 constructs a high-dimensional feasible domain between the insulation life of the participating transformers within the target, including:

[0188] Based on the sensitivity equation and combined with the transformer performance sample refined management database, the target internal parameter is used as the control variable, and the insulation life of the transformer is used as the state variable. The relationship between the control variable and the state variable is constructed in the global space using the Monte Carlo sampling method to obtain the pre-selected high-dimensional feasible domain.

[0189] Based on the actual grid planning data, the wind or solar quota and access location in the grid system are continuously changed to correct the transformer performance sample fine management database and conduct offline Monte Carlo numerical simulation experiments;

[0190] Based on the refined management database of the corrected transformer performance samples and the data obtained from simulation experiments, update the preselected high-dimensional feasible region to obtain the high-dimensional feasible region.

[0191] In this embodiment, first, based on the sensitivity equation, combined with the refined management database of transformer performance samples, using the hot spot temperature, harmonic content, and magnetic flux as control variables and the transformer insulation life as the state variable, the Monte Carlo sampling method is used to construct the relationship between the control variable and the state variable in the global space, accurately depicting the high-dimensional feasible region of the transformer with a complex mechanism including insulation life. Then, taking the actual power grid planning data as an example, continuously changing the wind / solar quota and access location of the power grid system, correcting the refined management database of transformer performance samples and conducting a large number of offline Monte Carlo numerical simulation experiments, so as to update and obtain the high-dimensional feasible region applicable to different wind / solar penetration scenarios.

[0192] In an alternative embodiment, step 150 determines the low-order safe operating region of the transformer based on the high-dimensional feasible region, including:

[0193] By using the hyperspace projection method, perform dimensionality reduction processing on the high-dimensional feasible region, and use the expected life of the transformer as the cutting plane;

[0194] Project the intersection line of the cutting plane and the dimensionally reduced high-dimensional feasible region onto the plane formed by the target internal parameters to form the critical life curve;

[0195] Based on the critical life curve, establish the low-order safe operating region.

[0196] In this embodiment, the hyperspace projection method is used to perform dimensionality reduction processing on the high-dimensional feasible region, using the expected life of the transformer as the cutting plane, projecting the intersection line of the transformer life expectancy plane and the high-dimensional feasible region onto the plane formed by the temperature rise, harmonic content, and magnetic flux to form the critical life curve, and establishing the low-order safe operating region of the transformer as shown in Figure 3 shown.

[0197] Specifically, taking transformers A, B, and C with different lives as examples, they respectively correspond to different safe region ranges as shown in Figure 4 shown, where the life relationship is A < B < C. Further, by making the inner tangent of the critical life curve or performing piecewise linearization processing, a linearized boundary as shown in Figure 5 shown is formed to replace the original non-convex and complex boundary, thereby reducing the computational complexity and realizing the accurate assessment of the transformer's safe operating state level. Based on this, for power transformers under frequent power flow switching, the low-order safe operating region as shown in Figure 6 shown can be finally established, and then use whether the transformer operating condition points (voltage, current) are located within the low-order safe operating region to evaluate its state.

[0198] A transformer operation status monitoring system provided in an embodiment of the present invention is introduced below. The transformer operation status monitoring system described below can be considered as a module architecture for implementing the transformer operation status monitoring method provided in an embodiment of the present invention; the content described below can be cross-referenced with the above.

[0199] Optional, see Figure 7 , Figure 7 : is a structural block diagram of a transformer operation status monitoring system provided by an embodiment of the present invention for running a transformer operation status monitoring method provided by any of the above embodiments of the present invention, and the system may include:

[0200] The acquisition unit 10 is used to obtain target internal parameters to construct a transformer performance sample refined management database, wherein the target internal parameters are transformer internal parameters whose influence on the safe operation of the transformer exceeds a preset degree under power flow switching conditions;

[0201] A first processing unit 20 is used to establish a linear relationship between target external parameters and target internal parameters of the transformer based on a transformer performance sample refined management database, where the target external parameters are voltage and current;

[0202] A second processing unit 30 is used to establish a sensitivity equation of the target internal parameter relative to the target external parameter based on the linearized relationship;

[0203] The third processing unit 40 is used to construct a high-dimensional feasible domain between the insulation life of the participating transformers within the target based on the sensitivity equation;

[0204] A fourth processing unit 50 is used to determine a low-order safe operation domain of the transformer based on the high-dimensional feasible domain;

[0205] The evaluation unit 60 is used to evaluate the operating state of the transformer based on whether the operating point of the transformer is located in the low-order safe operating domain.

[0206] Optionally, the target internal parameters include: hot spot temperature, harmonic content and magnetic flux inside the transformer.

[0207] Optionally, the collection unit 10 is specifically used for:

[0208] Acquire first historical sample data, the first historical sample data including attribute information generated during the design, manufacture, maintenance and test of the transformer, and operating parameters under power flow switching conditions, the attribute information being inherent information describing characteristics and / or status of the transformer;

[0209] Eliminate erroneous data in the first historical sample data to obtain second historical sample data;

[0210] The redundant data in the second historical sample data is eliminated to obtain the target internal reference.

[0211] Optionally, the acquisition unit 10 is more specifically used for:

[0212] Through the 3σ data analysis method, the erroneous data in the first historical sample data are eliminated.

[0213] Optionally, the collection unit 10 is further specifically used for:

[0214] Through the BOX-COX transformation, the skewness of the distribution of the first historical sample data is transformed and corrected.

[0215] Optionally, the collection unit 10 is further specifically used for:

[0216] Determine the covariance matrix of the second historical sample data by principal component analysis;

[0217] Calculate the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to the eigenvalues;

[0218] The eigenvectors are arranged in rows according to a preset order, and the first preset rows are taken to form a target matrix. The preset order is that the eigenvalues ​​corresponding to the eigenvectors are arranged in order from large to small and from top to bottom;

[0219] Based on the target matrix, the retained data in the second historical sample data is determined, and the second historical sample data except the retained data is removed as redundant data.

[0220] Optionally, the first processing unit 20 is specifically configured to:

[0221] Establish the display relation expression of each target internal parameter under the condition of reverse flow transmission;

[0222] Based on the explicit relationship expression, the linear relationship between the target external parameter and the target internal parameter is established by calculating the sensitivity coefficient.

[0223] Optionally, the first processing unit 20 is further specifically configured to:

[0224] Based on the display relational expression and combined with the transformer performance sample refined management database, the changes in the transformer's target internal parameters are collected when the transformer's low-voltage side voltage and inflow current are changed under the reverse power transmission condition;

[0225] By calculating the sensitivity coefficient of the change of each target internal parameter, the linear relationship between the target external parameter and the target internal parameter is established.

[0226] Optionally, the second processing unit 30 is specifically configured to:

[0227] Based on the linearization relationship and combined with the transformer performance sample refined management database, the section with the closest Euclidean distance to the current operating state of the transformer is selected as a reference, and the historical similar state is used as the initial search point of the step search algorithm. The search starts from the current given power grid operation control variable, and the preset given step size is gradually increased or decreased;

[0228] For each given search value obtained, calculate the approximation error between the actual value and the observed value;

[0229] With the goal of making the approximation error smaller than the preset allowable error, the sensitivity equation is output.

[0230] Optionally, the third processing unit 40 is specifically configured to:

[0231] Based on the sensitivity equation and combined with the transformer performance sample refined management database, the target internal parameter is used as the control variable, and the insulation life of the transformer is used as the state variable. The relationship between the control variable and the state variable is constructed in the global space using the Monte Carlo sampling method to obtain the pre-selected high-dimensional feasible domain.

[0232] Based on the actual grid planning data, the wind or solar quota and access location in the grid system are continuously changed to correct the transformer performance sample fine management database and conduct offline Monte Carlo numerical simulation experiments;

[0233] Based on the revised transformer performance sample refined management database and the data obtained from the simulation experiment, the pre-selected high-dimensional feasible domain is updated to obtain the high-dimensional feasible domain.

[0234] Optionally, the fourth processing unit 50 is specifically configured to:

[0235] The high-dimensional feasible domain is reduced in dimension through the hyperspace projection method, and the expected life of the transformer is used as the cutting plane;

[0236] The intersection line of the cutting plane and the high-dimensional feasible region after dimensionality reduction is projected onto the plane formed by the target internal parameters to form a critical life curve;

[0237] Based on the critical life curve, a low-order safe operation domain is established.

[0238] Below, reference Figure 8 To describe the electronic device provided by the embodiment of the present application, the electronic device provided by the embodiment may include: at least one processor 100, at least one communication interface 200, at least one memory 300 and at least one communication bus 400;

[0239] In the embodiment of the present invention, the number of the processor 100, the communication interface 200, the memory 300, and the communication bus 400 is at least one, and the processor 100, the communication interface 200, and the memory 300 communicate with each other through the communication bus 400; obviously, Figure 8 The communication connections shown for the processor 100, the communication interface 200, the memory 300, and the communication bus 400 are merely optional;

[0240] Optionally, the communication interface 200 may be an interface of a communication module, such as an interface of a GSM module; the processor 100 may be a central processing unit CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0241] The memory 300 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0242] The processor 100 is specifically used to execute the application program in the memory to implement the above-mentioned transformer operation status monitoring steps.

[0243] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the transformer operating status monitoring method provided by the above methods is implemented.

[0244] The above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A transformer operating status monitoring method, characterized in that: include: Obtaining target internal parameters to construct a transformer performance sample refined management database, wherein the target internal parameters are transformer internal parameters whose influence on the safe operation of the transformer exceeds a preset degree under power flow switching conditions; Based on the transformer performance sample refined management database, a linear relationship between the target external parameters and the target internal parameters of the transformer is established, wherein the target external parameters are voltage and current; Based on the linearized relationship, establishing a sensitivity equation of the target internal parameter relative to the target external parameter; Based on the sensitivity equation, construct a high-dimensional feasible domain between the insulation life of the participating transformers within the target; Based on the high-dimensional feasible domain, determining a low-order safe operation domain of the transformer; Based on whether the operating point of the transformer is located in the low-order safe operation domain, the operating state of the transformer is evaluated.

2. The transformer operating status monitoring method according to claim 1, characterized in that: The target internal parameters include: hot spot temperature, harmonic content and magnetic flux inside the transformer.

3. The transformer operating status monitoring method according to claim 1, characterized in that: The obtaining of target internal parameters includes: Acquire first historical sample data, wherein the first historical sample data includes attribute information generated during the design, manufacture, maintenance and test of the transformer, and operating parameters under power flow switching conditions, wherein the attribute information is inherent information describing the characteristics and / or state of the transformer; Eliminate erroneous data in the first historical sample data to obtain second historical sample data; Redundant data in the second historical sample data is eliminated to obtain the target internal reference.

4. The transformer operating status monitoring method according to claim 3, characterized in that: The removing of erroneous data from the first historical sample data includes: The erroneous data in the first historical sample data is eliminated through the 3σ data analysis method.

5. The transformer operating status monitoring method according to claim 4, characterized in that: Before removing the erroneous data in the first historical sample data by the 3σ data analysis method, the method further includes: The skewness of the distribution of the first historical sample data is transformed and corrected through the BOX-COX transformation.

6. The transformer operating status monitoring method according to claim 3, characterized in that: The removing redundant data from the second historical sample data includes: Determine the covariance matrix of the second historical sample data by principal component analysis; Calculating eigenvalues ​​of the covariance matrix and eigenvectors corresponding to the eigenvalues; Arrange the eigenvectors in rows according to a preset order, and take the first preset rows to form a target matrix, wherein the preset order is that the eigenvalues ​​corresponding to the eigenvectors are arranged in order from large to small and from top to bottom; Based on the target matrix, the reserved data in the second historical sample data is determined, and the second historical sample data except the reserved data is removed as the redundant data.

7. The transformer operating status monitoring method according to claim 2, characterized in that: The establishing of a linearized relationship between a target external parameter and a target internal parameter of the transformer includes: Establishing the display relational expressions of the target internal parameters under the condition of reverse power flow transmission; Based on the display relationship expression, a linear relationship between the target external parameter and the target internal parameter is established by calculating the sensitivity coefficient.

8. The transformer operating status monitoring method according to claim 7, characterized in that: The display relationship expressions of the target internal parameters established under the reverse flow transmission condition include: Establishing a display relationship expression of the hot spot temperature under the condition of reverse flow transmission as the first expression; The first expression is: Among them, F AA is the accelerated aging factor of the transformer insulation life, B is a constant, T is the hot spot temperature, is the reference value of hot spot temperature; Establishing a display relationship expression of the harmonic content under the reverse power flow transmission condition as the second expression; The second expression is: Among them, P h is the hysteresis loss, P CO,N is the power of the Nth harmonic, R r,h is a certain resistance element, h is the harmonic order, I h is the hth harmonic current, U h is the voltage of the hth harmonic, and U1 is the fundamental voltage; Establishing a display relationship expression of the magnetic flux under the condition of reverse power flow transmission as the third expression; The third expression is: Among them, R a ,R b ,R c is the resistance, I ea is the primary current, N a is the number of turns on the original side.

9. The transformer operating status monitoring method according to claim 7, characterized in that: The step of establishing a linear relationship between the target external parameter and the target internal parameter by calculating a sensitivity coefficient based on the display relationship expression includes: Based on the display relationship expression, combined with the transformer performance sample refined management database, under the condition of changing the low-voltage side voltage and the inflow current of the transformer under the reverse power transmission condition, the change amount of each target internal parameter of the transformer is collected; By calculating the sensitivity coefficient of the change amount of each target internal parameter, a linear relationship between the target external parameter and the target internal parameter is established.

10. The transformer operating status monitoring method according to claim 1, characterized in that: The step of establishing a sensitivity equation of the target internal parameter relative to the target external parameter based on the linearized relationship includes: Based on the linearized relationship, combined with the transformer performance sample refined management database, select the section with the closest Euclidean distance to the current operating state of the transformer as a reference, use the historical similar state as the initial search point of the step search algorithm, start the search from the current given power grid operation control variable, and gradually increase or decrease the preset given step size; For each given search value obtained, calculate the approximation error between the actual value and the observed value; With the goal of making the approximation error smaller than a preset allowable error, the sensitivity equation is output.

11. The transformer operating status monitoring method according to claim 10, characterized in that: The sensitivity equation is: Y C =Y 0 +K1*(V C -V 0 )+K2*(I C -I 0 ),Y=[T,G,ψ] T Among them, Y C and Y 0 are the maximum and minimum values ​​of hot spot temperature rise, harmonic content or magnetic flux, respectively, V C and V 0 are the maximum and minimum voltage amplitudes, I C and I 0 are the maximum and minimum values ​​of the current respectively, K1 and K2 are the independent variable thresholds.

12. The transformer operating status monitoring method according to claim 1, characterized in that: The constructing of a high-dimensional feasible domain between the transformer insulation life involved in the target includes: Based on the sensitivity equation, combined with the transformer performance sample refined management database, the target internal parameter is used as the control variable, the insulation life of the transformer is used as the state variable, and the relationship between the control variable and the state variable is constructed in the global space by using the Monte Carlo sampling method to obtain a pre-selected high-dimensional feasible domain; Based on actual grid planning data, the wind or light quota and access location in the grid system are continuously changed to correct the transformer performance sample refined management database and conduct offline Monte Carlo numerical simulation experiments; Based on the revised transformer performance sample refined management database and the data obtained from the simulation experiment, the preselected high-dimensional feasible domain is updated to obtain the high-dimensional feasible domain.

13. The transformer operating status monitoring method according to claim 1, characterized in that: The step of determining the low-order safe operation domain of the transformer based on the high-dimensional feasible domain includes: By using a hyperspace projection method, the high-dimensional feasible domain is subjected to dimensionality reduction processing, and the expected life of the transformer is used as a cutting plane; Projecting the intersection line of the cutting plane and the high-dimensional feasible domain after the dimensionality reduction process onto the plane formed by the target internal parameters to form a critical life curve; Based on the critical life curve, the low-order safe operation domain is established.

14. A transformer operation status monitoring system, used for running the transformer operation status monitoring method according to any one of claims 1 to 13, characterized in that: include: A collection unit, used to obtain target internal parameters to build a transformer performance sample refined management database, wherein the target internal parameters are transformer internal parameters whose influence on the safe operation of the transformer exceeds a preset degree under power flow switching conditions; A first processing unit is used to establish a linear relationship between a target external parameter and a target internal parameter of the transformer based on the transformer performance sample refined management database, wherein the target external parameter is voltage and current; A second processing unit, configured to establish a sensitivity equation of the target intrinsic parameter relative to the target extrinsic parameter based on the linearized relationship; A third processing unit is used to construct a high-dimensional feasible domain between the insulation lives of the participating transformers within the target based on the sensitivity equation; A fourth processing unit, configured to determine a low-order safe operation domain of the transformer based on the high-dimensional feasible domain; An evaluation unit is used to evaluate the operating state of the transformer based on whether the operating point of the transformer is located in the low-order safe operation domain.

15. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing instructions executable by the processor; The processor is used to execute the transformer operation status monitoring method described in any one of claims 1 to 13 above.

16. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

Images