Pilot test monitoring data-based distribution transformer load loss prediction method and system

Abstract

The invention discloses a distribution transformer load loss prediction method and system based on pilot-scale test monitoring data. The method comprises the following steps: collecting pilot-scale test data of a distribution transformer load loss related variable; preprocessing the collected pilot plant test data, and performing correlation analysis on the preprocessed variable data to obtain an independent variable in a distribution transformer finished product load loss prediction function; performing multiple regression analysis on the selected independent variables to obtain a regression model, correcting the regression model to obtain a distribution transformer finished product load loss prediction function, and verifying the load loss prediction function; and performing load loss prediction on the distribution transformer finished product based on the verified load loss prediction function. According to the method, the predicted value of the load loss can be quickly obtained by analyzing pilot test data, the time and the cost of a traditional experiment are reduced, and the calculated load loss prediction precision can be effectively improved by analyzing variables and verifying functions.

Description

Technical Field

[0001] The present invention belongs to the technical field of numerical calculation of the load of distribution transformers, and more specifically, relates to a method and system for predicting the load loss of a distribution transformer based on pilot test monitoring data. Background Art

[0002] At present, distribution transformers have mature design schemes and process supports, but there will still be a large difference between the inspection data and the calculated theoretical values after the finished products are made. There are some factors that cannot be detected during the production of transformers, such as temperature, humidity, human factors, etc., resulting in excessive load losses of the manufactured finished transformers. Because these problems are generally not found during the production process, the difference between the actual value and the theoretical value of the load loss can only be found after the transformer finished product is completed.

[0003] In the prior art, the load data of the transformer finished product is usually predicted by collecting pilot test data. If there are problems with the pilot test data, the problems in the previous links can be directly investigated at the pilot test stage. Avoid investigating problems after the transformer finished product is unqualified.

[0004] The prior art lacks a precise calculation method for relevant process parameters and has an empirical dependence problem in parameter selection during the production process of distribution transformers, resulting in a large fluctuation range of the load loss performance of distribution transformers. Summary of the Invention

[0005] To solve the deficiencies in the prior art, the present invention provides a method and system for predicting the load loss of a distribution transformer based on pilot test monitoring data, which can solve the problems of low prediction accuracy and complex prediction process of the load loss of transformers in the prior art.

[0006] The present invention adopts the following technical solutions.

[0007] A method for predicting the load loss of a distribution transformer based on pilot test monitoring data includes the following steps:

[0008] Collect pilot test data of variables related to the load loss of the distribution transformer;

[0009] Preprocess the collected pilot test data, and perform a correlation analysis on the preprocessed variable data to obtain the independent variables in the load loss prediction function of the distribution transformer finished product;

[0010] Perform a multiple regression analysis on the selected independent variables to obtain a regression model, correct the regression model to obtain a load loss prediction function for the distribution transformer finished product, and verify the load loss prediction function;

[0011] Predict the load loss of the distribution transformer finished product based on the verified load loss prediction function.

[0012] Preferably, the relevant variables of the load loss of the distribution transformer include: low-voltage resistance value, low-voltage current, high-voltage resistance value, high-voltage current, high-voltage load loss, low-voltage load loss, and load loss during the in-process test of the transformer body.

[0013] Preferably, the preprocessing of the in-process test data includes: data screening and selection of variables for specific analysis based on the correlation of reference variables;

[0014] The data screening includes deleting data with abnormal or missing results caused by faults in the detection system or relevant parameter sensors.

[0015] Preferably, the correlation analysis of the preprocessed variable data to obtain the independent variables in the load loss prediction function of the distribution transformer finished product specifically includes:

[0016] Calculating the Pearson correlation index, significance coefficient, and scatter plot of the low-voltage resistance value, low-voltage current, high-voltage resistance value, high-voltage current, high-voltage load loss, low-voltage load loss, and load loss during the in-process test of the transformer body;

[0017] If the absolute value of the Pearson correlation meets the condition of being greater than 0.5 and the significance is less than 0.05, and at the same time the scatter plot shows a positive correlation and is close to a straight line, it indicates that the variable is an independent variable in the load loss prediction function of the distribution transformer finished product.

[0018] Preferably, the regression model includes the F-test of the forced regression model, the sample determination coefficient test, and stepwise regression of the low-voltage resistance value, low-voltage current, high-voltage resistance value, high-voltage current, high-voltage load loss, and low-voltage load loss to obtain a basic prediction function.

[0019] Preferably, the verification of the load loss prediction function specifically includes:

[0020] Verifying the load loss prediction function to judge its applicability: Import the load loss prediction function of the oil distribution transformer finished product into the original data table, and calculate the amount of data whose error value between the actual value and the theoretical value exceeds the preset error tolerance interval. If the error value is lower than 5% of the total data volume, it indicates applicability, otherwise it indicates inapplicability.

[0021] Preferably, the load loss prediction function is specifically as follows:

[0022] P = 0.2057I l 2 R l +0.0.176I h 2 R h +0.7183P t

[0023] Among them, I l is the low-voltage current, and R l is the total low-voltage three-phase resistance. I h is the high-voltage current, and R h is the total high-voltage three-phase resistance. P t is the pilot test load loss, and P is the finished product load loss of the distribution transformer.

[0024] The present invention also provides a distribution transformer load loss prediction system based on pilot test monitoring data for implementing the distribution transformer load loss prediction method based on pilot test monitoring data, including: an acquisition module, a processing module, an analysis module, a function construction module, a verification module, and a prediction module;

[0025] The acquisition module is used to obtain the pilot test data of the variables related to the distribution transformer load loss;

[0026] The processing module is used to preprocess the collected pilot test data;

[0027] The analysis module is used to perform a correlation analysis on the preprocessed variable data to obtain the independent variables in the distribution transformer finished product load loss prediction function;

[0028] The function construction module is used to perform a multiple regression analysis on the selected independent variables to obtain a regression model, and after correcting the regression model, obtain the distribution transformer finished product load loss prediction function.

[0029] The verification module is used to verify the load loss prediction function;

[0030] The prediction module is used to predict the load loss of the distribution transformer finished product based on the load loss prediction function that has passed the verification.

[0031] The present invention also provides a terminal, including a processor and a storage medium;

[0032] The storage medium is used to store instructions;

[0033] The processor is used to operate according to the instructions to execute the steps of the load loss prediction method based on the distribution transformer pilot test data.

[0034] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps of the load loss prediction method based on the distribution transformer pilot test data.

[0035] The beneficial effects of the present invention are as follows. Compared with the prior art, the present invention proposes a method for predicting load loss based on pilot test data of distribution transformers, which can improve the prediction accuracy of load loss and simplify the prediction process. The present invention uses the data obtained from the pilot test to predict the value of the load loss of the finished transformer, so as to prevent the finished transformer from not meeting the theoretical calculation value after the test. This method helps to improve the qualified rate of the finished transformer and reduce the waste of costs. The present invention enriches the reliable ways for users to calculate and analyze product performance. In terms of flexibility, the present invention is relatively easy to operate and has a fast result realization speed when dealing with a large amount of data, which can greatly shorten the time for technicians to handle sudden abnormalities in product performance and facilitate enterprises to stabilize product quality and improve production efficiency. In terms of iterability, the present invention supports users to continue to optimize the original model based on the subsequently optimized product data, so as to continuously reduce the fluctuation of the load loss performance of oil-immersed amorphous three-dimensional wound distribution transformers.

[0036] The present invention has at least the following technical effects:

[0037] 1. Improve efficiency: By analyzing the pilot test data, the predicted value of the load loss can be quickly obtained, reducing the time and cost of traditional experiments.

[0038] 2. Enhance accuracy: Using the actual test data for calculation can effectively improve the prediction accuracy of load loss.

[0039] 3. Easy to operate: The process of this method is simple and clear, and it is easy to implement in the manufacturing and testing processes of transformers;

[0040] 4. The present invention constructs a load loss prediction function for the test requirements of distribution transformers, and this function is not publicly disclosed in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is the flow chart of the method for predicting load loss based on pilot test data of distribution transformers in the present invention;

[0042] Figure 2 is the structural diagram of the system for predicting load loss based on pilot test data of distribution transformers in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the spirit of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0044] Data analysis refers to the analysis of a large amount of collected data using appropriate statistical analysis methods, summarizing, understanding, and digesting them to maximize the development of the data's functions and play its role. Data analysis is a process of detailed study and summary of data to extract useful information and form conclusions. The mathematical foundation of data analysis was established in the early 20th century, but it was not until the emergence of computers that practical operations became possible and data analysis was popularized. Data analysis is the product of the combination of mathematics and computer science. The present invention combines big data analysis with the production of distribution transformers, and the finished transformers need to undergo finished product tests to ensure that the load losses of the products reach the standard values.

[0045] As Figure 1 shown, the present invention provides a method for predicting the load loss based on the pilot test data of a distribution transformer, which predicts the load loss of the finished transformer through the pilot test data to prevent the production of unqualified transformers in advance. The method includes the following steps:

[0046] Step 1, collect various relevant data of the pilot test of the distribution transformer and summarize them;

[0047] In a preferred but non-limiting embodiment of the present invention, the data of the pilot test of the distribution transformer includes low-voltage resistance value, low-voltage current, high-voltage resistance value, high-voltage current, high-voltage load loss, low-voltage load loss, and core pilot test load loss.

[0048] Preferably, as much relevant data as possible needs to be collected in terms of sample size and type, including the information recorded by the system during the production process and the product inspection reports, to ensure the accuracy and fitting degree of the model.

[0049] Step 2, preprocess the data obtained by summarizing in Step 1 to obtain the preprocessed data for analysis;

[0050] Specifically, the preprocessing of the collected data includes: screening processing;

[0051] The screening processing includes deleting the data with abnormal or missing results caused by the failure of the detection system or related parameter sensors.

[0052] During the data collection process, various abnormal situations may occur, including but not limited to sensor failures, storage device failures, etc., resulting in the generation of incorrect data and empty data. The abnormal data is deleted through data screening and the missing data is filled.

[0053] Step 3, perform variable correlation analysis on the preprocessed data to determine the possible independent variables in the load loss prediction function of the finished distribution transformer, and perform multiple regression analysis on the selected independent variables to obtain a regression model;

[0054] The variable correlation analysis specifically includes:

[0055] The variable correlation analysis includes: calculating the Pearson correlation index, significance coefficient and scatter plot of the low-voltage resistance value, low-voltage current, high-voltage resistance value, high-voltage current, high-voltage load loss, low-voltage load loss, and in-body pilot load loss respectively.

[0056] Specifically, determine the variables for subsequent specific analysis: determine the possible independent variables of the acoustic load loss according to the Pearson correlation absolute value greater than 0.5 and significance less than 0.05 and referring to the scatter plot.

[0057] The multiple regression model analysis includes the F-test of the load loss forced regression model, the coefficient of determination of the sample, and the stepwise regression model.

[0058] Specifically, the regression model includes the F-test of the forced regression model of the low-voltage resistance value, low-voltage current, high-voltage resistance value, high-voltage current, high-voltage load loss, and low-voltage load loss, the coefficient of determination test of the sample, and the basic prediction function obtained by stepwise regression.

[0059] Step 4, after the regression model is corrected for multicollinearity and heteroscedasticity problems, obtain the load loss prediction function of the distribution transformer finished product, and verify the load loss prediction function and judge its applicability;

[0060] After the regression model is corrected for multicollinearity and heteroscedasticity problems, obtain the load loss prediction function of the distribution transformer finished product, which specifically includes:

[0061] Determine the variables for subsequent specific analysis by referring to the variable correlation analysis;

[0062] The correlation analysis specifically includes: if the absolute value of the Pearson correlation of a certain variable is greater than 0.5 and the significance level is less than 0.05, and at the same time the scatter plot shows a positive correlation and approaches a straight line, it means that the variable can be used for specific analysis.

[0063] In actual operation, the above process shows that there is a significant correlation between the load loss and the low-voltage resistance value, low-voltage current, high-voltage resistance value, and high-voltage current.

[0064] Judging the applicability of the load loss prediction function specifically includes:

[0065] Import the load loss prediction function of the oil distribution transformer finished product into the original data table, calculate the error value between the actual value and the theoretical value, and obtain the data volume exceeding the preset error tolerance interval. If the data volume is less than 5% of the total data volume, it means it has actual applicability.

[0066] Step 5: Based on the verified load loss prediction function, during operation, collect new samples and process them using the applicable sound pressure level noise performance prediction function to analyze the load loss of the finished distribution transformer.

[0067] The load loss prediction function finally obtained through the above experimental data is as follows:

[0068] P = 0.2057I l 2 R l + 0.0.176I h 2 R h + 0.7183P t

[0069] Wherein, I l is the low - voltage current, R l is the total low - voltage three - phase resistance, I h is the high - voltage current, R h is the total high - voltage three - phase resistance, P t is the pilot - scale load loss, and P is the load loss of the finished transformer.

[0070] As Figure 2 shown, the present invention also proposes a distribution transformer load loss prediction system based on pilot - scale monitoring data, which includes: an acquisition module, a processing module, an analysis module, a function construction module, a verification module, and a prediction module;

[0071] Among them, the acquisition module is used to obtain the pilot - scale data of the variables related to the load loss of the distribution transformer;

[0072] The processing module is used to pre - process the collected pilot - scale data;

[0073] The analysis module is used to perform a correlation analysis on the pre - processed variable data to obtain the independent variables in the load loss prediction function of the finished distribution transformer;

[0074] The function construction module is used to perform a multiple regression analysis on the selected independent variables to obtain a regression model, and after correcting the regression model, obtain the load loss prediction function of the finished distribution transformer,

[0075] The verification module is used to verify the load loss prediction function;

[0076] The prediction module predicts the load loss of the finished distribution transformer based on the verified load loss prediction function.

[0077] The beneficial effects of the present invention are as follows. Compared with the prior art, the present invention starts from the perspective of big data, provides a method and system for analyzing the load loss performance of distribution transformers, mines potential influencing factors by collecting a large amount of production data and performing program calculations, and establishes a relevant linear regression mathematical model.

[0078] The present disclosure can be a system, a method, and / or a computer program product. The computer program product can include a computer-readable storage medium having thereon computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0079] The computer-readable storage medium can be a tangible device that can retain and store instructions used by an instruction execution device. The computer-readable storage medium can be, for example, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (non-exhaustive list) of the computer-readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanically encoded device, such as a punched card or raised structure in a groove having instructions stored thereon, and any suitable combination of the foregoing. The computer-readable storage medium used herein is not construed as an instantaneous signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated through a waveguide or other transmission medium (e.g., an optical pulse through an optical fiber cable), or an electrical signal transmitted through a wire.

[0080] The computer-readable program instructions described herein can be downloaded from the computer-readable storage medium to various computing / processing devices, or downloaded to an external computer or external storage device through a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include a copper transmission cable, an optical fiber transmission, a wireless transmission, a router, a firewall, a switch, a gateway computer, and / or an edge server. The network adapter or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0081] Computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine - related instructions, microcode, firmware instructions, state - setting data, or source code or object code written in any combination of one or more programming languages, including object - oriented programming languages such as Smalltalk, C++, etc., and conventional procedural programming languages such as the "C" language or similar programming languages. The computer - readable program instructions may be executed entirely on the user's computer, partially on the user's computer, executed as a stand - alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider). In some embodiments, by using the state information of computer - readable program instructions to customize an electronic circuit, such as a programmable logic circuit, a field - programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit can execute the computer - readable program instructions to implement various aspects of the present disclosure.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: it is still possible to modify the specific embodiments of the present invention or make equivalent substitutions, and any modification or equivalent substitution that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A method for predicting load loss of distribution transformers based on pilot monitoring data, characterized in that: The following steps are involved: Collect pilot data on distribution transformer load loss related variables; Preprocess the collected pilot data, and perform correlation analysis on the preprocessed variable data to obtain the independent variables in the distribution transformer finished product load loss prediction function; The selected independent variables are subjected to multivariate regression analysis to obtain a regression model, and the load loss prediction function of the distribution transformer finished product is obtained after the regression model is modified, and the load loss prediction function is verified; The load loss prediction function is used to predict the load loss of the finished distribution transformer.

2. A method for predicting load loss of distribution transformers based on pilot monitoring data according to claim 1, characterized in that: The variables related to the load loss of the distribution transformer include: low voltage resistance value, low voltage current, high voltage resistance value, high voltage current, high voltage load loss, low voltage load loss, and body pilot load loss.

3. A method for predicting load loss of distribution transformers based on pilot monitoring data according to claim 2, characterized in that: The preprocessing of the pilot data includes: data screening, and selecting variables for specific analysis based on the correlation of reference variables; The data screening includes deleting data with abnormal or missing results due to failure of the detection system or related parameter sensors.

4. A method for predicting load loss of distribution transformers based on pilot monitoring data according to claim 3, characterized in that: The correlation analysis of the preprocessed variable data is performed to obtain the independent variables in the distribution transformer finished product load loss prediction function, specifically including: Calculate the Pearson correlation index, significance coefficient and scatter plot of low voltage resistance, low voltage current, high voltage resistance, high voltage current, high voltage load loss, low voltage load loss and device body pilot load loss; According to the fact that the absolute value of the Pearson correlation is greater than 0.5 and the significance is less than 0.05, and the scatter plot is positively correlated and close to a straight line, it means that the variable is the independent variable in the distribution transformer finished product load loss prediction function.

5. The method for predicting load loss of distribution transformer based on pilot monitoring data according to claim 1 is characterized in that: The regression model includes a forced regression model F test of low voltage resistance value, low voltage current, high voltage resistance value, high voltage current, high voltage load loss, low voltage load loss, a sample determination coefficient test and a stepwise regression to obtain a basic prediction function.

6. A method for predicting load loss of distribution transformers based on pilot monitoring data according to claim 1, characterized in that: The verification of the load loss prediction function specifically includes: The load loss prediction function is verified to determine its applicability: the finished load loss prediction function of the oil distribution transformer is imported into the original data table to calculate the error between the actual value and the theoretical value to determine whether the data exceeds the preset error tolerance range. If the error value is less than 5% of the total data volume, it indicates applicability, otherwise it indicates inapplicability.

7. The method for predicting load loss of distribution transformers based on pilot monitoring data according to claim 1 is characterized in that: The load loss prediction function is as follows: P=0.2057I l 2 R l +0.0.176I h 2 R h +0.7183P t Among them, I l is the low voltage current, R l is the total resistance of the low voltage three-phase, I h is the high voltage current, R h is the total resistance of high voltage three-phase, P t is the pilot load loss, and P is the finished product load loss of the distribution transformer.

8. A distribution transformer load loss prediction system based on pilot monitoring data, executing a distribution transformer load loss prediction method based on pilot monitoring data according to any one of claims 1 to 7, characterized in that: include: Acquisition module, processing module, analysis module, function building module, verification module, prediction module; Acquisition module for pilot data of distribution transformer load loss related variables; The processing module is used to pre-process the collected pilot data; The analysis module is used to perform correlation analysis on the preprocessed variable data to obtain the independent variables in the distribution transformer finished product load loss prediction function; The function construction module is used to perform multivariate regression analysis on the selected independent variables to obtain a regression model, and then the regression model is modified to obtain the distribution transformer finished product load loss prediction function. The verification module is used to verify the load loss prediction function; The prediction module predicts the load loss of the distribution transformer product based on the verified load loss prediction function.

9. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images