A Correction Method for Transformer Core Noise Calculation System
By establishing the feedback correction logic of transformer core noise and multiple correction calculation formulas, the problem of large errors and low accuracy of noise recognition and control logic in the prior art is solved, and high accuracy recognition and control of transformer core noise is realized. Reduce verification costs and improve technical level.
Patent Information
- Application Number
- CN202311489079.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-09
- Publication Date
- 2024-12-31
- Estimated Expiration
- 2043-11-09
AI Technical Summary
In the prior art, the identification, control and calculation logic of transformer core noise have problems such as large errors and low accuracy, and the core noise performance cannot be effectively identified and optimized.
By establishing the feedback correction logic for transformer core noise, including the input of electromagnetic parameters, structural parameters and process parameters, the testing conditions and environmental identification are carried out, the theoretical calculations of standard and non-standard environments are carried out, and the difference analysis is carried out based on the noise test results. Correct the calculation formula and fitting curves multiple times to establish a transformer core noise database to achieve accurate noise identification and control.
It improves the theoretical calculation accuracy of transformer core noise, reduces verification costs, provides an effective noise design basis, forms a relatively complete control technology process, reduces the additional cost of the design backend, and improves the overall technical level.
Smart Images
Figure CN117494443B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a correction method for a transformer core noise calculation system, and in particular to a method for guiding the identification, control and theoretical calculation of noise through feedback correction logic of transformer core noise. Background Art
[0002] As one of the important performance indicators of transformer core, transformer core noise has long lacked mature and complete identification, control and calculation logic, and there is a lack of systematic basis for optimizing core noise performance. Since the overall transformer noise involves multiple components and is the comprehensive result of multi-field coupling, it is difficult to conduct a single-aspect control analysis, so comprehensive consideration is required.
[0003] In the prior art, the calculation logic is generally to perform simple noise calculations based on an empirical system and to perform theoretical calculations based on a variety of design parameters. This is used as the design basis for transformer noise, which has problems such as large errors, low accuracy, and inability to perform effective identification. Summary of the invention
[0004] The present invention proposes a correction method for a transformer core noise calculation system, aiming to overcome the above-mentioned deficiencies in the prior art, realize the full logical relationship of transformer core noise identification control and calculation, guide the noise design process, reduce errors and improve accuracy.
[0005] The technical solution of the present invention is to establish feedback correction logic of transformer core noise, guide the identification, control and theoretical calculation of noise, and provide a correction method for transformer core noise calculation system, including the following steps:
[0006] S101, taking electromagnetic parameters, structural parameters and process parameters designed based on the transformer as an input parameter group;
[0007] S102, perform test condition and environment identification to determine whether it is an IEC standard environment. If so, it is a standard environment, and the parameter group input in step S101 is input into the standard environment theoretical calculation system. If not, it is a non-standard environment, and the parameter group input in step S101 is input into the non-standard environment theoretical calculation system.
[0008] S103, performing theoretical noise calculations under two theoretical calculation systems in step S102, and outputting theoretical noise calculation results A or theoretical noise calculation results B, corresponding to standard environments and non-standard environments, respectively;
[0009] S201. Perform noise test on finished transformer core, including noise test under designed magnetic flux density and noise test under full magnetic flux density. Perform standardized environment test or non-standard environment test according to test conditions and environment. After the test, obtain the sound pressure level noise test result under corresponding magnetic flux density.
[0010] S202, identifying the test environment for the finished transformer core noise test to determine whether it is an IEC standard environment test;
[0011] S203. If the judgment result of S202 is yes, then compare the noise test result under the designed magnetic density with the theoretical calculation result A outputted by S103; if S202 is no, then compare the noise test result under the designed magnetic density with the theoretical calculation result B outputted by S103;
[0012] The above tests for test conditions and environmental identification can reduce the verification cost of transformer core noise, avoid noise verification differences caused by differences in experimental conditions, and provide a theoretical benchmark for experimental noise control.
[0013] S301, performing difference analysis between the noise theoretical calculation result and the test result to determine whether the difference range is within the target value, if so, proceed to S302, if not, proceed to S303;
[0014] S302, incorporating the noise test data of the full magnetic field segment into the theoretical calculation system for system output, the theoretical calculation system is constructed by combining the noise data in the full magnetic field segment with the calculation formula and the fitting curve, and re-correcting the noise data in the full magnetic field segment N times in combination with the calculation formula and the fitting curve, N represents the number of correctable data items in the test and represents the number of corrections of the theoretical calculation system, N≥2, and continuously outputting a new theoretical calculation system based on the original primary calculation formula to establish and accumulate a transformer core noise database;
[0015] S303, classifying the full magnetic field noise test data into a special database, temporarily not using it for the correspondence of the calculation system, and not performing the calculation system correction.
[0016] The above method can improve the theoretical calculation accuracy of transformer core noise, realize specific correction and precise analysis, and provide an effective basis for transformer noise design; and establish an open-loop system to control transformer core noise with the help of multi-channel measures, forming a relatively complete control technology process.
[0017] Preferably, the electromagnetic parameters, structural parameters and process parameters in S101 are derived from the transformer core design drawing, the electromagnetic parameters include the designed magnetic flux density, the structural parameters include the core window height, the core center distance and the joint structure, and the process parameters include the silicon steel weight, the core thickness and the paint properties.
[0018] Preferably, the result data of the noise test under the designed magnetic density in S201 is a fixed-point test performed at 12 standard measurement points, and the single-point data is collected using a 5s integration, and then the 12 single-point data are A-weighted to obtain the final noise value.
[0019] Preferably, in the S201, the core magnetic density value interval based on the transformer design is distributed in the range of 1.2T-1.8T, and the division value is a basic theory about the change of noise and magnetic density, that is, for every 0.1T increase in magnetic density, the noise increases by about 2-3dB, so the full magnetic density segment test covers 1.2T-1.8T, and the division value is ≤0.1T.
[0020] Preferably, the judgment error in S301 is within ±5%.
[0021] Preferably, the calculation formula in S302 involves structural parameters, geometric parameters, process parameters and design parameters, and the weight of silicon steel, magnetic density and operating frequency are substituted into the calculation to obtain the theoretical sound pressure level noise value, and the initial calculation formula is used as the starting calculation system.
[0022] Preferably, the fitting curve in S302 is a factor-free fitting of the original noise value based on a calculation formula, including test real points and fitting virtual points. The test real points and fitting virtual points are connected in series in ascending order according to the corresponding noise to obtain a fitting curve. The purpose is to eliminate variables, realize control variables, confirm the relationship between noise and magnetic density separately, and obtain a reference noise value. The noise data in the full magnetic density segment is corrected N times to correct the test real points, and the fitting virtual points are automatically updated.
[0023] Preferably, it also includes:
[0024] 1) First, perform the corresponding magnetic field density discreteness rounding on the reference noise value to eliminate the error.
[0025] 2) Then perform arithmetic averaging on the corresponding single point of magnetic density noise data; find the mean point of multiple groups of noise data to fit the curve.
[0026] 3) Perform the same magnetic density point processing on the full magnetic density segment noise data of the historical data group and the new data group. After processing, the data is arranged into a new full magnetic density segment noise data to obtain the test real points on the fitting curve. The fitting virtual points are the interpolation between the test real points achieved with the help of the fitting tool.
[0027] Preferably, the structural parameters, geometric parameters, process parameters and design parameters of the N+1th data are stripped by the Nth calculation formula, and then the N+1th group of benchmark noise data is curve fitted to obtain the N+1th fitting curve, and the fitting method is polynomial fitting.
[0028] The advantages of the present invention are as follows: 1) The establishment and accumulation of the transformer core noise database is completed by utilizing the cyclic correction process, which can provide an effective database for core noise control and identification;
[0029] 2) By using multiple cycles of comparison and correction, a transformer core noise calculation system with high accuracy is provided;
[0030] 3) It can realize the noise prediction of the transformer design end and provide a numerical basis for the overall noise of the transformer, avoiding the additional cost caused by noise correction at the back end of the design, which is conducive to improving the overall technical level;
[0031] 4) By identifying the parameters, test conditions and environment and conducting corresponding tests respectively, the verification cost of the transformer core noise can be reduced, the difference in noise verification caused by differences in experimental conditions can be avoided, and a theoretical benchmark can be provided for experimental noise control. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is a principle diagram of an embodiment of a correction method for a transformer core noise calculation system of the present invention. DETAILED DESCRIPTION
[0033] The present invention is further described in detail below with reference to examples and specific implementation modes. Example
[0034] Design magnetic flux density: refers to the average magnetic field density generated in the transformer core under the specified number of turns and excitation voltage based on design requirements.
[0035] Total magnetic flux density: refers to the gradient average magnetic flux density generated in the transformer core according to the ladder voltage excitation based on the specified number of turns required by the design, usually covering 1.2T-1.8T.
[0036] like Figure 1 As shown, a method for correcting a transformer core noise calculation system includes:
[0037] S101. The electromagnetic parameters, structural parameters and process parameters based on the transformer design are used as the input parameter group, wherein the electromagnetic parameters, structural parameters and process parameters are derived from the transformer core design drawing provided by the customer. The electromagnetic parameters include the designed magnetic flux density, the structural parameters include the core window height, the core center distance and the joint structure, and the process parameters include the silicon steel weight, the core thickness and the paint properties.
[0038] S102, perform test condition and environment identification to determine whether it is an IEC standard environment. If so, it is a standard environment, and the parameter group input in step S101 is input into the standard environment theoretical calculation system. If not, it is a non-standard environment, and the parameter group input in step S101 is input into the non-standard environment theoretical calculation system.
[0039] S103, performing theoretical noise calculations under the two theoretical calculation systems in step S102, and outputting theoretical noise calculation values of the independent systems respectively, that is, outputting theoretical noise calculation results A or theoretical noise calculation results B, corresponding to the standard environment and the non-standard environment respectively;
[0040] S201. Perform noise test on finished transformer core, including noise test under designed magnetic flux density and noise test under full magnetic flux density. Perform standardized environment test or non-standard environment test according to test conditions and environment. After the test, obtain the sound pressure level noise test result under corresponding magnetic flux density.
[0041] The result data of the noise test under designed magnetic density is fixed-point test at 12 standard measuring points. The single-point data is collected by 5s integration, and then the 12 single-point data are A-weighted to obtain the final noise value.
[0042] The core magnetic flux density value range based on transformer design is distributed in the range of 1.2T-1.8T, and the division value is a basic theory about the change of noise and magnetic flux density, that is, for every 0.1T increase in magnetic flux density, the noise increases by about 2-3dB. Therefore, the full magnetic flux density segment test covers 1.2T-1.8T, and the division value is ≤0.1T.
[0043] There is usually a marking voltage under the design number of coil turns for the design magnetic density. By using this voltage for excitation, the magnetic density of the core can reach the design magnetic density, provided that the product processing and assembly meet the design requirements. Therefore, the design magnetic density is excited by the marking voltage based on the design specifications.
[0044] The above tests for test conditions and environmental identification can reduce the verification cost of transformer core noise, avoid noise verification differences caused by differences in experimental conditions, and provide a theoretical benchmark for experimental noise control.
[0045] S202, identifying the test environment for the finished transformer core noise test to determine whether it is an IEC standard environment test;
[0046] S203. If the judgment result of S202 is yes, then compare the noise test result under the designed magnetic density with the theoretical calculation result A outputted by S103; if S202 is no, then compare the noise test result under the designed magnetic density with the theoretical calculation result B outputted by S103;
[0047] S301, performing difference analysis between the noise theoretical calculation result and the test result to determine whether the difference range is within the target value, if so, proceed to S302, if not, proceed to S303;
[0048] S302, incorporating the noise test data of the full magnetic field density segment into a theoretical calculation system for system output, wherein the theoretical calculation system is constructed by combining the noise data within the full magnetic field density segment with a calculation formula and a fitting curve.
[0049] The calculation formula involves structural parameters, geometric parameters, process parameters and design parameters. By substituting parameters such as silicon steel weight, magnetic density, operating frequency, etc. into the calculation, the theoretical sound pressure level noise value is obtained.
[0050] The fitting curve is a factor-free fitting of the original noise value based on the calculation formula, including test real points and fitting virtual points. According to the corresponding noise, the test real points and fitting virtual points are connected in series in order from small to large to obtain the fitting curve. The purpose is to eliminate variables, realize control variables, confirm the relationship between noise and magnetic density separately, and obtain the reference noise value. The corresponding magnetic density discreteness is rounded off for the reference noise value to eliminate errors. Then the corresponding magnetic density noise data single point is arithmetic averaged to find the mean point of multiple groups of noise data to fit the curve. The historical data group and the new data group full magnetic density segment noise data are processed with the same magnetic density point. After processing, the data is arranged as a new full magnetic density segment noise data to obtain the test real points on the fitting curve. The fitting virtual points are interpolated between the test real points with the help of the fitting tool. Achieve quadratic fitting of the theoretical calculation system, and perform N (N≥2) corrections to the theoretical calculation system (N refers to the number of correctable data items in the test, and also represents the number of corrections to the theoretical calculation system), correct the actual test points, and automatically update the fitting virtual points, and continuously output a new theoretical calculation system based on the original first-generation calculation formula.
[0051] The structural parameters, geometric parameters, process parameters and design parameters of the N+1th data are stripped by the N-time calculation formula, and then the N+1th group of benchmark noise data is curve fitted to obtain the N+1th fitting curve, and the fitting method is polynomial fitting.
[0052] S303, classifying the full magnetic field noise test data into a special database, temporarily not using it for the correspondence of the calculation system, and not performing the calculation system correction.
[0053] The above cyclic correction process realizes the establishment and accumulation of the transformer core noise database, which can provide an effective database for core noise control and identification.
[0054] The above is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several modifications and improvements without departing from the inventive concept of the present invention, which all belong to the protection scope of the present invention.
Claims
1. A method for correcting a transformer core noise calculation system, characterized in that: The following steps are involved: S101, taking some electromagnetic parameters, structural parameters and process parameters designed based on the transformer as an input parameter group; S102, perform test condition and environment identification to determine whether it is an IEC standard environment. If so, it is a standard environment, and the parameter group input in step S101 is input into the standard environment theoretical calculation system. If not, it is a non-standard environment, and the parameter group input in step S101 is input into the non-standard environment theoretical calculation system. S103, performing theoretical noise calculations under two theoretical calculation systems in step S102, and outputting theoretical noise calculation results A or theoretical noise calculation results B, corresponding to standard environments and non-standard environments, respectively; S201. Perform noise test on finished transformer core, including noise test under designed magnetic flux density and noise test under full magnetic flux density. Perform standardized environment test or non-standard environment test according to test conditions and environment. After the test, obtain the sound pressure level noise test result under corresponding magnetic flux density. S202, identifying the test environment for the finished transformer core noise test to determine whether it is an IEC standard environment test; S203. If the judgment result of S202 is yes, then compare the noise test result under the designed magnetic density with the theoretical calculation result A outputted by S103; if S202 is no, then compare the noise test result under the designed magnetic density with the theoretical calculation result B outputted by S103; S301, performing difference analysis between the noise theoretical calculation result and the test result to determine whether the difference range is within the target value, if so, proceed to S302, if not, proceed to S303; S302, incorporating the noise test data of the full magnetic field segment into the theoretical calculation system for system output, the theoretical calculation system is constructed by combining the noise data in the full magnetic field segment with the calculation formula and the fitting curve, and re-correcting the noise data in the full magnetic field segment N times in combination with the calculation formula and the fitting curve, N represents the number of correctable data items in the test and represents the number of corrections of the theoretical calculation system, N≥2, and continuously outputting a new theoretical calculation system based on the original primary calculation formula to establish and accumulate a transformer core noise database; S303, classifying the full magnetic field noise test data into a special database, temporarily not used for the correspondence of the calculation system, and not performing the calculation system correction.
2. The method for correcting a transformer core noise calculation system according to claim 1, characterized in that: The electromagnetic parameters, structural parameters and process parameters in S101 are derived from the transformer core design drawing. The electromagnetic parameters include the designed magnetic flux density, the structural parameters include the core window height, the core center distance and the joint structure, and the process parameters include the silicon steel weight, the core thickness and the paint properties.
3. The method for correcting a transformer core noise calculation system according to claim 1, characterized in that: The result data of the noise test under the designed magnetic density in S201 is a fixed-point test performed at 12 standard measuring points, and the single-point data is collected using a 5s integration, and then the 12 single-point data are A-weighted to obtain the final noise value.
4. The method for correcting a transformer core noise calculation system according to claim 1, characterized in that: In the above-mentioned S201, the full magnetic density segment test covers 1.2T-1.8T, and the division value is ≤0.1T.
5. The method for correcting a transformer core noise calculation system according to claim 1, characterized in that: The judgment error in S301 is within ±5%.
6. The method for correcting a transformer core noise calculation system according to claim 1, characterized in that: The calculation formula in S302 involves structural parameters, geometric parameters, process parameters and design parameters. The weight of silicon steel, magnetic flux density and operating frequency are substituted into the calculation to obtain the theoretical sound pressure level noise value. The initial calculation formula is used as the starting calculation system.
7. The method for correcting a transformer core noise calculation system according to claim 1, characterized in that: The fitting curve in S302 is a factor-free fitting of the original noise value based on a calculation formula, including test real points and fitting virtual points. The test real points and fitting virtual points are connected in series in ascending order according to the corresponding noise to obtain a fitting curve and obtain a reference noise value.
8. The method for correcting a transformer core noise calculation system according to claim 7, characterized in that: Also includes: 1) First, perform the corresponding magnetic field density dispersion rounding on the reference noise value; 2) Then perform arithmetic averaging on the corresponding magnetic density noise data points; 3) Perform the same magnetic density point processing on the full magnetic density segment noise data of the historical data group and the new data group. After processing, the data is arranged into a new full magnetic density segment noise data to obtain the test real points on the fitting curve. The fitting virtual points are the interpolation between the test real points achieved with the help of the fitting tool.
9. A method for correcting a transformer core noise calculation system as claimed in claim 8, characterized in that: The structural parameters, geometric parameters, process parameters and design parameters of the N+1th data are stripped by the Nth calculation formula, and then the N+1th group of benchmark noise data is subjected to curve fitting to obtain the N+1th fitting curve, and the fitting method is polynomial fitting.