CN112364490B - Model parameter identification method for hysteresis characteristics of ferromagnetic material - Google Patents
Model parameter identification method for hysteresis characteristics of ferromagnetic material Download PDFInfo
- Publication number
- CN112364490B CN112364490B CN202011172745.8A CN202011172745A CN112364490B CN 112364490 B CN112364490 B CN 112364490B CN 202011172745 A CN202011172745 A CN 202011172745A CN 112364490 B CN112364490 B CN 112364490B
- Authority
- CN
- China
- Prior art keywords
- model
- values
- man
- value
- genetic algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 239000003302 ferromagnetic material Substances 0.000 title claims abstract description 37
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 56
- 230000002068 genetic effect Effects 0.000 claims abstract description 54
- 238000004088 simulation Methods 0.000 claims abstract description 29
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims abstract description 23
- 238000005259 measurement Methods 0.000 claims abstract description 15
- 230000005291 magnetic effect Effects 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims abstract description 10
- 230000005415 magnetization Effects 0.000 claims description 12
- 238000004804 winding Methods 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 3
- 230000004907 flux Effects 0.000 claims description 3
- 230000002427 irreversible effect Effects 0.000 claims description 3
- 230000035699 permeability Effects 0.000 claims description 3
- 230000002441 reversible effect Effects 0.000 claims description 3
- 230000008878 coupling Effects 0.000 claims description 2
- 238000010168 coupling process Methods 0.000 claims description 2
- 238000005859 coupling reaction Methods 0.000 claims description 2
- 230000001052 transient effect Effects 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 3
- 238000000691 measurement method Methods 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 8
- 238000005457 optimization Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 101100499229 Mus musculus Dhrsx gene Proteins 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 235000001968 nicotinic acid Nutrition 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Computer Hardware Design (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a model parameter identification method of a magnetic hysteresis effect of a ferromagnetic material, B-H data of a magnetic hysteresis loop of the iron core is obtained through a measurement method or a simulation method of the iron core, 9 parameters of the model are obtained in two steps through a genetic algorithm according to a description formula and a principle of a Jiles-Athereon model integrated by typical electromagnetic transient simulation software, B-H loops of the iron core are simulated according to the obtained model parameters, and the identification accuracy of the model parameters is improved by continuously narrowing a range of values of parameters to be obtained through the genetic algorithm in a parameter identification process compared with the B-H loops obtained through measurement or simulation.
Description
Technical Field
The invention relates to the field of electricity, in particular to a model parameter identification method for hysteresis characteristics of a ferromagnetic material.
Background
The power system transformer and the electromagnetic transformer generally adopt nonlinear ferromagnetic materials, but the current magnetic theory development is not complete, theoretical description models of hysteresis include Lucas model and Jiles-Atherent model, and the like, and the Jiles-Atherent model has clear physical meaning, can truly describe the nonlinear relation of hysteresis characteristics, is convenient to realize and the like, and is widely applied to hysteresis modeling and simulation analysis of ferromagnetic materials. However, the method also faces the problem that the model parameters are difficult to directly obtain through actual measurement and the parameter acquisition is difficult, and accurate and reliable parameter identification is required. Parameter identification is often used in situations where it is difficult to directly obtain parameters, where the theoretical parameters are approximated by fitting parameters, and when the error between the fitting parameters and the theoretical parameters is sufficiently small, the fitting parameters can be considered to be sufficiently accurate and can replace the theoretical parameters.
The current common algorithms applied to the field of parameter identification are mainly intelligent optimization algorithms based on bionics, such as genetic algorithms, neural networks, particle swarm optimization and the like, the Jiles-Atheren model is integrated into mainstream electromagnetic transient simulation software, such as EMTDS/PSCAD, RTDS, EMTP and the like, the models integrated in the current version of simulation software are improved versions, namely 9 parameter versions, the calculation of the parameters of the Jiles-Atheren model is actually a nonlinear optimization problem in a 9-dimensional space, and the problems that the calculation efficiency is low and the acquisition of an accurate global optimal solution is not easy are caused, so that more accurate results are difficult to obtain in the later calculation optimizing process and the calculation is ended.
Disclosure of Invention
Object of the invention
The invention aims to provide a model parameter identification method for hysteresis characteristics of ferromagnetic materials, which solves the complex optimization problem of processing the multiple targets of the Jiles-Athereton model parameters and improves the identification precision of the model parameters.
(II) technical scheme
In order to solve the above problems, an aspect of the present invention provides a method for identifying model parameters of hysteresis characteristics of ferromagnetic materials, wherein a genetic algorithm is adopted to obtain a part of model parameters of a Jiles-Atherton model of the ferromagnetic materials, and then according to the obtained model parameters, the genetic algorithm is adopted to obtain the remaining model parameters of the Jiles-Atherton model of the ferromagnetic materials; and according to the obtained model parameters, the hysteresis loop of the ferromagnetic material iron core is simulated, and compared with the hysteresis loop obtained by measurement or simulation, the identification precision of the model parameters is improved by continuously narrowing the range of the parameter to be obtained by a genetic algorithm in the parameter identification process.
According to one aspect of the invention, formulas 1-5 of the ferromagnetic material Jiles-Atherton model are respectively:
B＝μ 0 (M+H) (equation 1)
Wherein B is magnetic flux density; h is the magnetic field strength; m is magnetization intensity; alpha is the average of the inter-domain couplingA field parameter; m is M an Is hysteresis-free magnetization; h e Is the effective magnetic field strength; m is M s Is saturation magnetization; mu (mu) 0 Vacuum permeability, its value is mu 0 ＝4π×10 -7 H/m; c is a parameter reflecting the relationship between reversible and irreversible magnetization, and 0 < c < 1; k is a parameter reflecting the hysteresis strength, and k is more than 0; k (k) mod The k value after correction; delta is a symbol representing H, H is greater than or equal to zero, delta is 1, H is less than zero, delta is-1; a, a 1 、a 2 、a 3 Beta, b are coefficients, and a 2 >a 1 ,a 1 >0,a 3 >0，0<β<1，b>1；
Part of model parameters of the ferromagnetic material Jiles-Athereton model are Ms, a1, a2, a3 and b, and the rest of model parameters of the ferromagnetic material Jiles-Athereton model are alpha, beta, c and k.
According to one aspect of the invention, the obtaining model parameters Ms, a1, a2, a3, b using a genetic algorithm comprises the steps of:
s1: hysteresis loop data of the iron core made of ferromagnetic materials are obtained through measurement or simulation;
s2: calculating Ms value and Man-He curve according to the hysteresis loop data;
s3: according to a Man-He model formula, a first objective function of a genetic algorithm is written;
s4: setting the value ranges of a1, a2, a3 and b;
s5: calling a genetic algorithm to calculate values a1, a2, a3 and b;
s6: narrowing the value ranges of a1, a2, a3 and b, and calling a genetic algorithm again to calculate the values of a1, a2, a3 and b;
s7: generating a Man-He curve according to the calculated Ms, a1, a2, a3 and b values, if the fitting degree is poor, switching to S8, and if the fitting degree is good, switching to S9;
s8: adopting an arrangement and combination method to further reduce the value ranges of three variables in four variables a1, a2, a3 and b, expanding the value ranges of the remaining one variable, and calling a genetic algorithm to calculate the values a1, a2, a3 and b again to obtain the values a1, a2, a3 and b of the Man-He curve with good fitting degree;
s9: and outputting the calculated result values of Ms, a1, a2, a3 and b.
According to one aspect of the present invention, in S1, a B-H loop of a ferromagnetic material core is obtained by obtaining a hysteresis loop of the ferromagnetic material core or an electromagnetic transient simulation method by measuring a winding of the ferromagnetic material core to which an exciting current is applied;
in S2, obtaining the hysteresis loop through the obtained hysteresis loop data according to the formula 1 of the ferromagnetic material Jiles-Athereon model, obtaining the maximum value of M in the hysteresis loop, namely obtaining the Ms, and after obtaining the hysteresis loop, obtaining the hysteresis loop according to the formula H e (n)＝0.5×(H e (n) R +H e (n) L ) Calculating He value, wherein H e (n) R And H e (n) L And (3) respectively hysteresis the left and right side values of He corresponding to M (n) on the loop, and obtaining a Man-He curve.
In S3, the first objective function is an error function obtained by calling a Man-He curve generated by a Man-He model formula according to the acquired Ms, a1, a2, a3, and b values, and making a difference between the Man-He curve calculated according to the hysteresis loop data in S2, where the error function formula is:
wherein H (i) model To call the He value, H (i), in the Man-He curve generated by the Man-He model formula according to the acquired Ms, a1, a2, a3 and b values meas And (2) calculating the He value in the obtained Man-He curve according to the hysteresis loop data in the step (S2).
According to one aspect of the invention, in S7, a Man-He curve generated by calling a Man-He model formula according to the obtained Ms, a1, a2, a3, b values is compared with the Man-He curve calculated according to the hysteresis loop data in S2, if the difference between the corresponding points selected by the two sets of Man-He curves is smaller than a predetermined value, the calculated Ms, a1, a2, a3, b values have good fitting degree, otherwise, the fitting degree is poor.
According to one aspect of the invention, the acquisition of model parameters α, β, c, k using a genetic algorithm comprises the steps of:
s10: writing a second objective function of the genetic algorithm according to the Jiles-Athereton model formula;
s11: setting the value ranges of alpha, beta, c and k;
s12: calling a genetic algorithm to calculate alpha, beta, c and k values;
s13: reducing the value ranges of alpha, beta, c and k, and calling a genetic algorithm again to calculate the values of alpha, beta, c and k;
s14: generating hysteresis loops according to the acquired Ms, a1, a2, a3, b, alpha, beta, c and k, if the fitting degree is poor, switching to S15, and if the fitting degree is good, switching to S16;
s15: adopting an arrangement and combination method to further reduce the value range of three variables in the four variables of alpha, beta, c and k, expanding the value range of one residual variable, and calling a genetic algorithm to calculate the values of alpha, beta, c and k again to obtain the values of alpha, beta, c and k with good fitting degree;
s16: and outputting calculated result values of Ms, a1, a2, a3, b, alpha, beta, c and k.
According to one aspect of the present invention, in S10, the objective function is an error function obtained by calling Jiles-Atherton model function formula according to the calculated Ms, a1, a2, a3, b, α, β, c, k values, and by subtracting the hysteresis loop obtained by measurement or simulation in S1, where the error function formula is:
wherein B (i) model To call the B value, B (i), in the B-H loop of the Jiles-Athereton model function formula according to the calculated Ms, a1, a2, a3, B, alpha, beta, c, k values meas Is the B value in the hysteresis loop obtained by measurement or simulation according to S1.
According to one aspect of the present invention, in S14, the hysteresis loop generated by Jiles-Atherton model function formula is called according to the calculated values Ms, a1, a2, a3, b, α, β, c, k, and compared with the hysteresis loop obtained by measurement or simulation in S1, if the difference between the corresponding points selected by the two sets of hysteresis loops is smaller than a predetermined value, the calculated values α, β, c, k are better, otherwise, the fitting degree is poor.
(III) beneficial effects
The technical scheme of the invention has the following beneficial technical effects:
according to the invention, 9 parameters of the model are acquired in two steps by adopting a genetic algorithm according to the Jiles-Athereton model integrated by the conventional typical electromagnetic transient simulation software and the description formula and principle of the model, and B-H loop wires of the iron core are simulated according to the acquired model parameters, and compared with the B-H loop wires acquired by measurement or simulation, the identification precision of the model parameters is improved by continuously narrowing the range of the parameter values to be obtained by the genetic algorithm in the parameter identification process.
Drawings
FIG. 1 is a flow chart of a method for Jiles-Athereon model parameter identification according to one embodiment of the invention.
FIG. 2 is a comparison of B-H loop identification results of a run-time genetic algorithm according to one embodiment of the present invention.
FIG. 3 is a graph comparing B-H loop identification results of two runs of a genetic algorithm according to one embodiment of the present invention.
Detailed Description
The objects, technical solutions and advantages of the present invention will become more apparent by the following detailed description of the present invention with reference to the accompanying drawings. It should be understood that the description is only illustrative and is not intended to limit the scope of the invention. In addition, in the following description, descriptions of well-known structures and techniques are omitted so as not to unnecessarily obscure the present invention.
According to the model modeling principle of the Jiles-Athereon model, the model parameters with higher precision can be obtained through a conventional genetic algorithm. The power system components with ferromagnetic materials, such as transformers, electromagnetic transformers and the like, are modeled by adopting a Jiles-Atherton model, the accuracy and high precision of model parameters are the basis of modeling simulation and simulation analysis, and the simulation analysis capability of the power system of the transformers, the electromagnetic transformers and the like is promoted.
The technical scheme of the invention is further specifically described below through an embodiment and with reference to fig. 1.
S1: and measuring or simulating to obtain the B-H data of the hysteresis loop of the iron core.
And B-H loop of the ferromagnetic material iron core is obtained by applying exciting current to the ferromagnetic material iron core winding and using tools such as an oscilloscope. Or the B-H loop of the ferromagnetic material iron core is obtained by an electromagnetic transient simulation method.
Selecting a typical parameter of a current transformer as an example, wherein the parameters of the current transformer are as follows: the average sectional area of the transformer iron core is 0.00119532 square meters, the magnetic path length is 0.4987m, the secondary winding is 0.253 Ω, the inductance is 0.0008H, the rated transformation ratio of one secondary winding is 180:1, the secondary load impedance is 2 Ω, and the measured B-H single value curve is shown in the following table:
table 1 shows the corresponding point values of the B-H single value curve
The 9 parameter values corresponding to the selected current transformer description by adopting a Jiles-Athereton model are shown in the following table:
table 2 theoretical values of model parameters
The method comprises the steps that a 'general CT Model' Model element integrated in RSCAD software of an RTDS technology company real-time simulation system, which is also called conventional CT, is adopted, and the conventional CT only needs to input B-H single-value curve parameters shown in a table 1 and obtained through iron core measurement, the iron core sectional area, the magnetic path length, a secondary winding rated transformation ratio, secondary winding impedance, load impedance and loss branch percent (set to 20%) of the transformers, and the nameplate parameters of the transformers are subjected to operation simulation, so that B-H loop data are obtained as simulation results;
s2: ms values and Man-He curve data were calculated from the B-H data.
According to the principle formula of the current mainstream electromagnetic transient simulation software, the current transformer Jiles-Athereon model adopts the following calculation formulas 1-5:
B＝μ 0 (M+H) (equation 1)
Wherein, in the formulas 1 to 5, B is magnetic flux density; h is the magnetic field strength; m is magnetization intensity; alpha is the average field parameter of the inter-domain coupling; m is M an Is hysteresis-free magnetization; h e Is the effective magnetic field strength; m is M s Is saturation magnetization; mu (mu) 0 Vacuum permeability, its value is mu 0 ＝4π×10 -7 H/m; c is a parameter reflecting the relationship between reversible and irreversible magnetization, and 0 < c < 1; k is a parameter reflecting the hysteresis strength, and k is more than 0; k (k) mod The k value after correction; delta is a symbol representing H, H is greater than or equal to zero, delta is 1, H is less than zero, delta is-1; a, a 1 、a 2 、a 3 Beta, b are coefficients, and a 2 >a 1 ,a 1 >0,a 3 >0，0<β<1，b>1；
According to the formula, a model function code is written, and a group of B-H loop values can be generated by inputting 9 parameters of the current transformer.
The Ms value and the Man-He curve data are calculated according to the hysteresis loop B-H data obtained by simulation in the embodiment S1, specifically, the B-H loop is obtained by calculation in the formula 1, and the maximum value of M in the B-H is obtained, thereby obtaining the Ms. After obtaining the B-H loop, according to formula H e (n)＝0.5×(H e (n) R +H e (n) L ) Calculating He value, wherein H e (n) R And H e (n) L And (5) obtaining the Man-He curve by respectively obtaining the values of the left and right sides of the He corresponding to M (n) on the B-H loop.
S3: and according to the Man-He model formula, compiling a genetic algorithm objective function.
The genetic algorithm mainly completes the solution of the minimum value of an objective function, wherein the objective function is an error function obtained by substituting values Ms, a1, a2, a3 and b and then taking the difference between a Man-He curve generated according to formulas 2 and 3 and a Man-He curve calculated by S2, and the error function formula is as followsWherein H (i) model For simulation of the He value, H (i), in the obtained Man-He curve meas The He value in the Man-He curve generated in S2.
S4: the value ranges of a1, a2, a3 and b are set.
And setting the value ranges of the four variables a1, a2, a3 and b in the Man-He model formula by adopting a genetic algorithm function or a genetic algorithm tool box integrated in MATLAB.
S5: and calling a genetic algorithm to calculate a1, a2, a3 and b variable values.
S6: and (5) narrowing the value ranges of a1, a2, a3 and b, and calling the genetic algorithm again to calculate the values of a1, a2, a3 and b.
S7: and obtaining the values of Ms, a1, a2, a3 and b to generate a Man-He curve. If the fitting degree is poor, the process proceeds to S8, and if the fitting degree is good, the process proceeds to S9.
And calling a Man-He model formula according to the obtained Ms, a1, a2, a3 and B values to generate a Man-He curve, and comparing the Man-He curve with a Man-He curve obtained by simulating an iron core B-H loop calculation, wherein if the difference value of the corresponding points selected by the two groups of Man-He curves is smaller, the accuracy of the calculated a1, a2, a3 and B values is high, the Man-He curve can be better fitted, and otherwise, the fitting degree is poor.
S8: and (3) adopting an arrangement and combination method to further reduce the value ranges of three variables in the four variables a1, a2, a3 and b, expanding the value ranges of the remaining variables, calling a genetic algorithm to calculate the values a1, a2, a3 and b again, and finally obtaining the values a1, a2, a3 and b which can be better fit with the Man-He curve.
S9: and outputting the calculated result values of Ms, a1, a2, a3 and b.
S10: and writing a genetic algorithm objective function according to the J-A model formula.
According to B-H loop data acquired in S1, a Jiles-Athereton model formula and Ms, a1, a2, a3 and B values obtained by S9 calculation, a genetic algorithm objective function is written, the genetic algorithm mainly completes the solution of the minimum value of the objective function, the objective function is an error function obtained by substituting the values of Ms, a1, a2, a3, B, alpha, beta, c and k and then differencing B-H loops generated according to formulas 1-5 in S2 and B-H loops obtained by simulation, and the error function formula is thatWherein B (i) model For the B value, B (i), in the B-H loop generated according to equations 1-5 in S2 meas The B value in the B-H loop obtained by simulation is obtained.
S11: and setting the value ranges of alpha, beta, c and k.
S12: and calling a genetic algorithm to calculate alpha, beta, c and k values.
S13: and (5) narrowing the value ranges of alpha, beta, c and k, and calling a genetic algorithm again to calculate the alpha, beta, c and k values.
S14: the 9 obtained variable values generate a B-H loop. If the fitting degree is poor, the process proceeds to S15. If the fitting degree is good, the process proceeds to S16.
And according to the calculated values of Ms, a1, a2, a3, B, alpha, beta, c and k, invoking a Jiles-Athereton model function formula to generate a B-H loop of the ferromagnetic material iron core, and comparing the B-H loop with a B-H loop obtained by simulation, if the difference value of the selected corresponding points of the two groups of B-H loops is smaller, the calculated values of alpha, beta, c and k are high in precision, the B-H loop can be better fitted, and otherwise, the fitting degree is poor.
S15: the method of permutation and combination is adopted, the value range of three variables in the four variables of alpha, beta, c and k is further reduced, the value range of the rest variables is enlarged, the alpha, beta, c and k values are calculated again by calling a genetic algorithm, and finally the alpha, beta, c and k values which can be well fit with the B-H loop are obtained.
S16: and outputting calculated result values of Ms, a1, a2, a3, b, alpha, beta, c and k.
According to the above 16 steps, the calculated parameter identification results are shown in fig. 2 and 3, wherein fig. 2 is a comparison graph of the identification results of the loop of the genetic algorithm B-H operated once, and fig. 3 is a comparison graph of the identification results of the loop of the genetic algorithm B-H operated again after the variable value range is narrowed.
For the common ferromagnetic material of the transformer, the S6 and the S13 are repeated in a mode of reducing the variable value range, and the parameter identification result with higher precision can be obtained by running for two to three times. And for the condition that the operation result is not ideal, adopting S8 and S15, and searching for a proper result again by a method of amplifying the value range of the parameter.
It will be readily appreciated that the genetic algorithm herein may be a replacement genetic algorithm or a modified genetic algorithm, and that the error function may be further optimized.
In summary, the invention discloses a method for identifying parameters of a Jiles-Atherton model of a hysteresis effect of a ferromagnetic material, which specifically obtains B-H data of a hysteresis loop of the iron core by a measurement method or a simulation method of the iron core, obtains 9 parameters of the model in two steps by adopting a genetic algorithm according to a description formula and a principle of the Jiles-Atherton model integrated by the conventional typical electromagnetic transient simulation software, and compares the obtained parameters of the model with the B-H loop of the iron core obtained by measurement or simulation, and improves the identification accuracy of the model parameters by continuously narrowing the range of the parameter to be obtained by the genetic algorithm in the parameter identification process.
It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explanation of the principles of the present invention and are in no way limiting of the invention. Accordingly, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention should be included in the scope of the present invention. Furthermore, the appended claims are intended to cover all such changes and modifications that fall within the scope and boundary of the appended claims, or equivalents of such scope and boundary.
Claims (5)
1. Firstly, acquiring a part of model parameters of a ferromagnetic material Jiles-Athereton model by adopting a genetic algorithm, and then acquiring the rest model parameters of the ferromagnetic material Jiles-Athereton model by adopting the genetic algorithm according to the acquired model parameters; according to the obtained model parameters, the hysteresis loop of the ferromagnetic material iron core is simulated, and compared with the hysteresis loop obtained by measurement or simulation, the identification precision of the model parameters is improved in a mode of continuously narrowing the range of the parameter to be obtained by a genetic algorithm in the parameter identification process;
wherein, formulas 1-5 of the ferromagnetic material Jiles-Athereton model are respectively:
B＝μ 0 (M+H) (equation 1)
Wherein B is magnetic flux density; h is the magnetic field strengthThe method comprises the steps of carrying out a first treatment on the surface of the M is magnetization intensity; alpha is the average field parameter of the inter-domain coupling; m is M an Is hysteresis-free magnetization; h e Is the effective magnetic field strength; m is M s Is saturation magnetization; mu (mu) 0 Vacuum permeability, its value is mu 0 ＝4π×10 -7 H/m; c is a parameter reflecting the relationship between reversible and irreversible magnetization, and 0 < c < 1; k is a parameter reflecting the hysteresis strength, and k is more than 0; k (k) mod The k value after correction; delta is a symbol representing H, H is greater than or equal to zero, delta is 1, H is less than zero, delta is-1; a, a 1 、a 2 、a 3 Beta, b are coefficients, and a 2 >a 1 ,a 1 >0,a 3 >0，0<β<1，b>1；
Part of model parameters of the ferromagnetic material Jiles-Athereton model are Ms, a1, a2, a3 and b, and the rest of model parameters of the ferromagnetic material Jiles-Athereton model are alpha, beta, c and k;
the genetic algorithm for obtaining the model parameters Ms, a1, a2, a3, b comprises the following steps:
s1: hysteresis loop data of the iron core made of ferromagnetic materials are obtained through measurement or simulation;
s2: calculating Ms value and Man-He curve according to the hysteresis loop data;
s3: according to a Man-He model formula, a first objective function of a genetic algorithm is written;
s4: setting the value ranges of a1, a2, a3 and b;
s5: calling a genetic algorithm to calculate values a1, a2, a3 and b;
s6: narrowing the value ranges of a1, a2, a3 and b, and calling a genetic algorithm again to calculate the values of a1, a2, a3 and b;
s7: generating a Man-He curve according to the calculated Ms, a1, a2, a3 and b values, if the fitting degree is poor, switching to S8, and if the fitting degree is good, switching to S9;
s8: adopting an arrangement and combination method to further reduce the value ranges of three variables in four variables a1, a2, a3 and b, expanding the value ranges of the remaining one variable, and calling a genetic algorithm to calculate the values a1, a2, a3 and b again to obtain the values a1, a2, a3 and b of the Man-He curve with good fitting degree;
s9: outputting calculated result values of Ms, a1, a2, a3 and b;
the genetic algorithm is adopted to obtain model parameters alpha, beta, c and k, and the method comprises the following steps:
s10: writing a second objective function of the genetic algorithm according to the Jiles-Athereton model formula;
s11: setting the value ranges of alpha, beta, c and k;
s12: calling a genetic algorithm to calculate alpha, beta, c and k values;
s13: reducing the value ranges of alpha, beta, c and k, and calling a genetic algorithm again to calculate the values of alpha, beta, c and k;
s14: generating hysteresis loops according to the acquired Ms, a1, a2, a3, b, alpha, beta, c and k, if the fitting degree is poor, switching to S15, and if the fitting degree is good, switching to S16;
s15: adopting an arrangement and combination method to further reduce the value range of three variables in the four variables of alpha, beta, c and k, expanding the value range of one residual variable, and calling a genetic algorithm to calculate the values of alpha, beta, c and k again to obtain the values of alpha, beta, c and k with good fitting degree;
s16: and outputting calculated result values of Ms, a1, a2, a3, b, alpha, beta, c and k.
2. The method for identifying model parameters according to claim 1, wherein,
in S1, a hysteresis loop of a ferromagnetic material iron core or a B-H loop of the ferromagnetic material iron core is obtained by a method of measuring a winding of the ferromagnetic material iron core to which exciting current is applied;
in S2, obtaining the hysteresis loop through the obtained hysteresis loop data according to the formula 1 of the ferromagnetic material Jiles-Athereon model, obtaining the maximum value of M in the hysteresis loop, namely obtaining the Ms, and after obtaining the hysteresis loop, obtaining the hysteresis loop according to the formula H e (n)＝0.5×(H e (n) R +H e (n) L ) Calculating He value, wherein H e (n) R And H e (n) L The left and right side values of He corresponding to M (n) on the hysteresis loop respectively,then a Man-He curve is obtained;
in S3, the first objective function is an error function obtained by calling a Man-He curve generated by a Man-He model formula according to the acquired Ms, a1, a2, a3, and b values, and making a difference between the Man-He curve calculated according to the hysteresis loop data in S2, where the error function formula is:
wherein H (i) model To call the He value, H (i), in the Man-He curve generated by the Man-He model formula according to the acquired Ms, a1, a2, a3 and b values meas And (2) calculating the He value in the obtained Man-He curve according to the hysteresis loop data in the step (S2).
3. The method for identifying model parameters according to claim 1, wherein,
in S7, a Man-He curve generated by calling a Man-He model formula according to the obtained Ms, a1, a2, a3, and b values is compared with the Man-He curve calculated according to the hysteresis loop data in S2, if the difference value of the corresponding points selected by the two sets of Man-He curves is smaller than a predetermined value, the calculated Ms, a1, a2, a3, and b values have good fitting degree, otherwise, the fitting degree is poor.
4. The method for identifying model parameters according to claim 1, wherein,
in S10, the objective function is an error function obtained by calling a hysteresis loop generated by Jiles-Atherton model function formula according to the calculated Ms, a1, a2, a3, b, α, β, c, k values and making a difference with the hysteresis loop obtained by measurement or simulation in S1, where the error function formula is:
wherein B (i) model Is based onThe calculated values of Ms, a1, a2, a3, B, alpha, beta, c and k are called the B value, B (i), in the B-H loop of the Jiles-Athereton model function formula meas Is the B value in the hysteresis loop obtained by measurement or simulation according to S1.
5. The method for identifying model parameters according to claim 1, wherein,
in S14, according to the calculated values Ms, a1, a2, a3, b, α, β, c, and k, the hysteresis loop generated by Jiles-Atherton model function formula is called, and compared with the hysteresis loop obtained by measurement or simulation in S1, if the difference between the corresponding points selected by the two sets of hysteresis loops is smaller than a predetermined value, the calculated values α, β, c, and k are better, otherwise, the fitting degree is poor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011172745.8A CN112364490B (en) | 2020-10-28 | 2020-10-28 | Model parameter identification method for hysteresis characteristics of ferromagnetic material |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011172745.8A CN112364490B (en) | 2020-10-28 | 2020-10-28 | Model parameter identification method for hysteresis characteristics of ferromagnetic material |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112364490A CN112364490A (en) | 2021-02-12 |
| CN112364490B true CN112364490B (en) | 2024-02-20 |
Family
ID=74510875
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011172745.8A Active CN112364490B (en) | 2020-10-28 | 2020-10-28 | Model parameter identification method for hysteresis characteristics of ferromagnetic material |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112364490B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113158432B (en) * | 2021-03-25 | 2023-03-21 | 潍柴动力股份有限公司 | Stress and strain-containing JA hysteresis model parameter identification method |
| CN113343616B (en) * | 2021-04-29 | 2023-05-23 | 西安交通大学 | Optimal design method and system for current transformer power supply |
| CN114236433B (en) * | 2021-11-23 | 2022-08-30 | 浙江大学 | Ferromagnetic material magnetic induction intensity online estimation method |
| CN114325516B (en) * | 2021-12-16 | 2024-03-01 | 华中科技大学 | Method and system for measuring magnetic characteristics of permanent magnet material |
| CN117272683B (en) * | 2023-11-13 | 2024-01-26 | 江苏南方永磁科技有限公司 | Model parameter optimization method and system for magnetic hysteresis effect simulation of ferromagnetic material |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7502723B1 (en) * | 2005-06-01 | 2009-03-10 | Linear Technology Corporation | Asymmetric minor hysteresis loop model and circuit simulator including the same |
| CN107609298A (en) * | 2017-09-26 | 2018-01-19 | 广东电网有限责任公司电力科学研究院 | A kind of Jiles Atherton model parameter identification methods and device |
-
2020
- 2020-10-28 CN CN202011172745.8A patent/CN112364490B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7502723B1 (en) * | 2005-06-01 | 2009-03-10 | Linear Technology Corporation | Asymmetric minor hysteresis loop model and circuit simulator including the same |
| CN107609298A (en) * | 2017-09-26 | 2018-01-19 | 广东电网有限责任公司电力科学研究院 | A kind of Jiles Atherton model parameter identification methods and device |
Non-Patent Citations (2)
| Title |
|---|
| 变压器铁芯磁滞回线模型参数辨识;袁中琛;刘连光;;电网与清洁能源(第09期);全文 * |
| 电压互感器铁芯J-A磁滞参数特性研究;周求宽;龙国华;刘衍;周友武;王羽;陈斌;;武汉大学学报(工学版)(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112364490A (en) | 2021-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112364490B (en) | Model parameter identification method for hysteresis characteristics of ferromagnetic material | |
| Toman et al. | Parameter identification of the Jiles–Atherton hysteresis model using differential evolution | |
| CN105740569B (en) | Current transformer engineering model method for building up based on transient heavy current measuring technology | |
| Saitz | Newton-Raphson method and fixed-point technique in finite element computation of magnetic field problems in media with hysteresis | |
| Li et al. | Hysteresis modeling for electrical steel sheets using improved vector Jiles-Atherton hysteresis model | |
| CN107656221A (en) | A kind of transformer core remanent magnetism evaluation method based on minor loop's slope | |
| CN109444776A (en) | Three-phase full-bridge converter iron core remanent magnetism measuring method, system and storage medium | |
| CN106777836A (en) | Transformer direct-current magnetic bias simulation method and device | |
| CN109884564A (en) | A kind of magnetic core of transformer magnetic characteristic measurement method and device | |
| CN114417555A (en) | Transformer modeling method and device, computer equipment and storage medium | |
| Ruderman et al. | Identification of soft magnetic BH characteristics using discrete dynamic Preisach model and single measured hysteresis loop | |
| CN115267295B (en) | Determination method and system of fluxgate current sensor | |
| CN110188480B (en) | System and method for simulating and analyzing magnetic hysteresis characteristics of ferromagnetic material under direct-current magnetic biasing condition | |
| CN110399677A (en) | Based on the transformer DC magnetic bias simulation method for improving J-A formula under bias state | |
| Liu et al. | Analytical inverse Preisach model and its comparison with inverse Jiles–Atherton model in terms of accuracy and computational speed | |
| Cale et al. | An improved magnetic characterization method for highly permeable materials | |
| Lesniewska et al. | Influence of the number of core air gaps on transient state parameters of TPZ class protective current transformers | |
| Shafieipour et al. | Principle of duality with normalized core concept for modeling multi-limb transformers | |
| CN113255285B (en) | Modeling method of single-phase four-column transformer based on JA model and dual principle | |
| CN109543240B (en) | Current transformer modeling method based on dynamic region saturation J-A theory | |
| Mendes et al. | An improved method for acquisition of the parameters of Jiles-Atherton hysteresis scalar model using integral calculus | |
| CN113033067A (en) | Current transformer J-A model parameter identification method based on improved particle swarm optimization | |
| Detka et al. | Validation of selected models of the magnetization curve of ferrite cores | |
| Wang et al. | 3-D FEM analysis in electromagnetic system considering vector hysteresis and anisotropy | |
| Xueyan et al. | Identification of Hysteresis Parameters of Current Transformer JA Model Based on Improved Whale Algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |