CN110633494B - Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm - Google Patents

Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm Download PDF

Info

Publication number
CN110633494B
CN110633494B CN201910729666.3A CN201910729666A CN110633494B CN 110633494 B CN110633494 B CN 110633494B CN 201910729666 A CN201910729666 A CN 201910729666A CN 110633494 B CN110633494 B CN 110633494B
Authority
CN
China
Prior art keywords
population
individuals
diode
igbt
volume
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
Application number
CN201910729666.3A
Other languages
Chinese (zh)
Other versions
CN110633494A (en
Inventor
颜景斌
沈云森
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin University of Science and Technology
Original Assignee
Harbin University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Harbin University of Science and Technology filed Critical Harbin University of Science and Technology
Priority to CN201910729666.3A priority Critical patent/CN110633494B/en
Publication of CN110633494A publication Critical patent/CN110633494A/en
Application granted granted Critical
Publication of CN110633494B publication Critical patent/CN110633494B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/60—Other road transportation technologies with climate change mitigation effect
    • Y02T10/72—Electric energy management in electromobility

Landscapes

  • Rectifiers (AREA)

Abstract

The invention provides a Swiss rectifier multi-objective optimization design method based on NSGA-II algorithm, belonging to the technical field of rectifier parameter optimization design. Step 1, determining optimized components and modeling power and volume; step 2, defining a target function; step 3, initializing a population; step 4, selecting; step 5, selecting a binary tournament; step 6, crossing; step 7, mutation; step 8, merging the parent population and the child population to obtain a first generation population; step 9, forming a population by utilizing the parent population and the child population; step 10, reaching a maximum evolution algebra gen; and 11, outputting the Pareto optimal solution front edge and the parameter matrix, obtaining corresponding decision variable values, namely solutions of optimization problems, according to the selected target performance indexes, and selecting and designing components according to the parameters.

Description

Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm
Technical Field
The invention relates to a Swiss rectifier multi-objective optimization design method based on an NSGA-II algorithm, and belongs to the technical field of rectifier parameter optimization design.
Background
In recent years, with the wide application of power electronic devices, more and more researches and optimization technologies about converters are paid high attention by researchers at home and abroad, and various high-efficiency and novel rectifier topological structures and digital control are generated at the same time, so that electric vehicles and new energy technologies are rapidly developed; the rectifier system has the obvious performance characteristics of high frequency, high efficiency, high power density, high power factor, high reliability and the like, and along with the appearance and application of various novel electromagnetic materials, electronic components and transformation technologies, the rectifier system develops towards the direction of small volume, high efficiency and low cost.
However, since the Swiss rectifier has nonlinear elements such as inductors, the rectifier system is a strong nonlinear system, and the parameters and performances of components to be optimized are more, and the performances conflict with each other and are restricted with each other, and the improvement of one performance is usually at the cost of the reduction of the other performance; therefore, the Swiss rectifier which is a complex optimization problem with multiple objectives, uncertainty, nonlinearity and multiple parameters is difficult to realize in the traditional optimization method, and the accuracy of each parameter can not be ensured by judging each parameter by experience; how to select a proper device by a rigorous mathematical method and process the multi-objective optimization problem with discrete variables, such as a rectifier, so that the comprehensive performance of the system is optimal, and the method has important significance on the reliable operation of the system and the sustainable development of energy; therefore, a multi-objective optimization method for the Swiss rectifier is imperative.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a Swiss rectifier multi-objective optimization design method based on an NSGA-II algorithm, which is beneficial to the optimization design of Swiss rectifier parameters, so that the efficiency and the power density of the rectifier are improved. The technical scheme is as follows:
a Swiss rectifier multi-objective optimization design method based on NSGA-II algorithm comprises the following steps:
step 1, taking a direct current inductor L, an output capacitor C, a switching tube IGBT and a Diode in a basic topology circuit of a Swiss rectifier as optimization variables influencing performance indexes of the rectifier, and carrying out power and volume modeling on the direct current inductor L, the output capacitor C, the switching tube IGBT and the Diode;
step 2, defining an objective function: measuring the performance index of the Swiss rectifier by taking the efficiency and the power density of the Swiss rectifier as objective functions;
step 3, initializing a population: setting the population size as N, the maximum evolution algebra as gen, and the number of decision variables as V; then, real number coding is carried out on the decision variables, and the upper limit and the lower limit of each parameter variable are input by taking the established component database as a constraint condition; randomly generating an N-dimensional decision matrix, taking a V +1 th column and a V +2 th column of the decision matrix as objective function values, taking the last two columns as non-dominant layer numbers and crowding degree distances, and enabling one parameter vector to correspond to one individual or be called a chromosome;
Step 4, selection: setting two parameters n i And S i ，n i Governing the number of individuals i, S, for all individuals in a population i For an individual set dominated by the individual i, judging the quality of a solution of the matrix and sequencing operation by adopting rapid non-dominated sequencing;
step 5, selecting the binary championship: setting the championship size to be 2 and the matching pool size to be N/2; randomly selecting two individuals in an initial population, comparing the non-dominant grades of the two individuals, and putting the low-grade individual into a matching pool; if the grades are the same, comparing the crowdedness distance, and keeping the distance larger; if the non-domination grade and the crowding degree distance are the same, randomly selecting one individual to be reserved, and repeating the step until the number of the individuals in the matching pool is N/2;
step 6, crossing: firstly, randomly selecting two individuals in a current generation population as parent individuals of cross operation, carrying out cross operation on genes on matched chromosomes to generate a pair of new chromosomes, and repeating the cross operation to form a new generation population; wherein the current generation population is a parent generation population; the specific operation process is as follows: setting the gene length of chromosome as L, randomly selecting an integer value K as a cross position within the range of [0, L ], and mutually exchanging genes of [ K, L ] parts at the cross position by matching chromosomes, thereby forming a new pair of chromosomes, namely individuals;
And 7, mutation: defining a mutation operator, carrying out small-probability replacement on the gene of an individual, and acting the mutation operator on a population to change the gene of a part of individuals in the population to generate a new allele, and marking the population at the moment as an offspring population;
step 8, merging the parent population in the step 6 with the child population obtained in the step 7, and marking the merged population as a first generation population (gen is 1) with the population size of N;
step 9, taking the first generation population obtained in the step 8 as a parent population, then carrying out cross variation on the parent population to obtain a child population, and then combining the parent population and the child population in the step to form a population, wherein the size of the population is 2N;
step 10, processing the population formed in the step 9 by using the operation process of the step 4, selecting N individuals as a new parent population, and generating a new child population P through the steps 6 and 7 t+1 (ii) a Repeating the step 9 until the maximum evolution algebra gen is reached;
and 11, outputting the Pareto optimal solution front edge and the parameter matrix, obtaining corresponding decision variable values, namely solutions of optimization problems, according to the selected target performance indexes, and selecting and designing components according to the parameters.
Further, the power and volume modeling process in step 1 includes:
calculating the power of an inductor, the power P of the inductor L Comprises the following steps:
Figure BDA0002160101270000021
wherein u is L Is the input voltage of an inductor, I rms Effective value of current flowing through the inductor, f sw To the switching frequency, V L Is the volume of the inductor;
calculating the coil volume of the inductor; the coil volume V cl Comprises the following steps:
Figure BDA0002160101270000022
wherein, ω is w Is the winding width, d w Is the winding depth;
calculating the volume V of the conductor cd Comprises the following steps: v cd ＝k pf V cl Wherein k is pf Is the fill factor of the coil and,
Figure BDA0002160101270000031
n is the number of turns of the coil,
Figure BDA0002160101270000032
is the cross-sectional area of the conductor,
calculating core volume
Figure BDA0002160101270000033
Comprises the following steps:
Figure BDA0002160101270000034
then the volume of the inductor V L Comprises the following steps:
Figure BDA0002160101270000035
wherein l represents the core length;
and calculating the on-state loss of the IGBT of the switching tube as follows:
Figure BDA0002160101270000036
wherein, V GE Is the threshold voltage of IGBT, I IGBT Is the magnitude of the current flowing through the IGBT, r CE Is an on-state equivalent resistance, M is a modulation ratio,
Figure BDA0002160101270000037
expressed as a voltage current phase angle;
the switching losses of the IGBT are:
Figure BDA0002160101270000038
wherein E is sw(on) And E sw(off) Energy lost for switching on and off the IGBT once, respectively, I N Rated operating current for the rectifier, u CEN At a rated operating voltage of u pn Outputting a voltage for the rectifier;
and (3) calculating the total loss of the IGBT as follows: p IGBT ＝P cond,IGBT +P sw,IGBT ；
The conduction loss of the diode is:
P on,Diode ＝V F I F D
wherein, V F Is the forward conduction voltage drop of the diode, I F The current is positive on-state current, D is duty ratio, and the duty ratios of different diodes are obtained according to 4 conducting circuit states;
The off-state loss of the diode is:
P off,Diode ＝V R I R (1-D)
wherein, V R For its reverse pressure drop, I R Is a diode reverse leakage current;
the switching losses of the diodes are:
Figure BDA0002160101270000039
wherein, V fp And V rp Forward and reverse peak voltages, I, of the diode, respectively fp And I rp Respectively the forward and reverse peak currents through the diode, t fp For its forward recovery time, t b Is its reverse current fall time;
the total diode loss is then: p Diode ＝P on,Diode +P off,Diode +P sw,Diode 。
Further, the specific process of defining the objective function in step 2 includes:
defining an objective function, taking the efficiency and the power density of the Swiss rectifier as the objective function, wherein the efficiency eta is as follows:
Figure BDA0002160101270000041
wherein, P 0 For output power, the power density ρ is:
Figure BDA0002160101270000042
wherein, P 0 To output power, P loss For total loss, P l For total loss of inductance, P IGBT For total IGBT losses, P Diode Is Diode total loss; v General assembly Is the total volume of the component, V L Is the volume of the inductor, V IGBT Is IGBT volume, V Diode Is a diode volume, V C Is the output capacitance volume, C is the inductance; the power density rho is an objective function
Further, the specific process of judging the quality and the ranking of the solution in the fourth step includes:
step 1, selecting n in the population i Individuals of 0, i.e. non-dominant optimal individuals, and put into the set F 1 Performing the following steps;
Step 2, pair set F 1 Respectively finding out a dominant individual set S corresponding to i i To S i Of the population, the number n of dominant individuals j dominating all individuals in the population j Performing a subtraction operation, i.e. n j ＝n j -1, excluding the influence of the individual i dominating j if n j 0, then put the individual j into the set F 2 Performing the following steps;
then set F 1 For the first layer non-dominant set, for F 1 The number of non-dominant layers of all individuals in (A) is recorded as Rank1, F 2 The Chinese is marked as Rank 2;
step 3, repeating the operations of the step 1 and the step 2, and performing non-dominated sorting on all individuals in the population, wherein the number of non-dominated layers is from low to high, and the smaller the Rank value is, the better the individuals are;
the population size is always N in the iteration process of the NSGA-II algorithm, so that individuals in the same Rank layer need to be chosen and rejected in the selection operation, a crowding degree distance C is defined, namely the sum of the distance differences of two adjacent individuals on each sub-objective function, and C is k ＝(f k+1,1 -f k-1,1 )+(f k-1,2 -f k+1,2 ) Wherein f is k+1,1 、f k-1,1 、f k-1,2 And f k+1,2 Respectively expressed as target functions corresponding to (k +1,1), (k-1,2) and (k +1, 2);
the individuals with large crowding degree distance are better than the individuals with small crowding degree distance, the non-dominated layer number and the crowding degree distance are used as the last two elements of the decision vector, and the quality of the solution and the sorting operation are judged according to the non-dominated layer number and the crowding degree distance.
The invention has the beneficial effects that:
the multi-objective optimization design method of the Swiss rectifier based on the NSGA-II algorithm, provided by the invention, is used for carrying out multi-objective optimization on the complex problems of the Swiss rectifier, such as multi-objective, uncertainty, nonlinearity and multi-parameter, solves the problem of device selection of each element selected by experience, can select a performance index through a Pareto frontier obtained by the optimization algorithm, improves the efficiency to 95.28-96.86%, and has the power density of 4.48-6.05 KW/dm 3 The overall performance of the rectifier is optimized.
Drawings
FIG. 1 is a circuit topology of a Swiss rectifier;
FIG. 2 is a diagram of four turn-on circuits of a Swiss rectifier;
FIG. 3 is a flow chart of the NSGA-II algorithm;
FIG. 4 is a schematic diagram of an elite selection strategy;
FIG. 5 is a diagram of a DC inductor UI core structure;
FIG. 6 is a diagram of an output capacitor structure;
fig. 7 is a Pareto optimal solution front.
Detailed Description
The present invention will be further described with reference to the following specific examples, but the present invention is not limited to these examples.
Example 1:
step 1, performing power and volume modeling on the optimized component; FIG. 5 is a diagram of the UI core structure of a DC inductor L, the power P of the inductor L Comprises the following steps:
Figure BDA0002160101270000051
wherein u L Is the input voltage of an inductor, I rms Effective value of current flowing through the inductor, f sw To the switching frequency, V L Is an inductive volume.
Coil volume V of inductor cl Comprises the following steps:
Figure BDA0002160101270000052
volume of conductor V cd Comprises the following steps: v cd ＝k pf V cl Wherein k is pf Is the fill factor of the coil and,
Figure BDA0002160101270000053
n is the number of turns of the coil,
Figure BDA0002160101270000054
is conductor cross section area, magnetic core volume
Figure BDA0002160101270000055
Comprises the following steps:
Figure BDA0002160101270000056
then V L Comprises the following steps:
Figure BDA0002160101270000057
the on-state loss of the switching tube IGBT is as follows:
Figure BDA0002160101270000061
wherein, V GE Is the threshold voltage of IGBT, I IGBT Is the magnitude of the current flowing through the IGBT, r CE M is the modulation ratio for the on-state equivalent resistance. The switching losses of the IGBT are:
Figure BDA0002160101270000062
E sw(on) and E sw(off) The energy lost for switching on and off the IGBT once, respectively.
Then the total IGBT losses are: p IGBT ＝P cond,IGBT +P sw,IGBT 。
The conduction loss of the diode is:
P on,Diode ＝V F I F D
V F is the forward conduction voltage drop of the diode, I F For forward on-state current, D is the duty cycle, and the duty cycles of the different diodes can be obtained from the 4 conducting circuit states shown in fig. 2.
The off-state loss of the diode is:
P off,Diode ＝V R I R (1-D)
wherein, V R For its reverse pressure drop, I R Is a diode reverse leakage current.
The switching losses of the diodes are:
Figure BDA0002160101270000063
V fp and V rp Respectively forward and reverse peak voltage of the diode, I fp And I rp Respectively the forward and reverse peak currents through the diode, t fp For its forward recovery time, t b Its reverse current fall time.
The total diode loss is then: p Diode ＝P on,Diode +P off,Diode +P sw,Diode 。
Step 2, defining an objective function, taking the efficiency and the power density of the Swiss rectifier as the objective function, wherein the efficiency eta is as follows:
Figure BDA0002160101270000064
wherein, P 0 For output power, set to 5KW, the power density ρ is:
Figure BDA0002160101270000065
step 3, initializing a population, setting the population size to be 200, setting the maximum evolution algebra to be 200, setting the number of decision variables to be 39, carrying out real number coding on the decision variables, taking the established component database as a constraint condition, inputting the upper limit and the lower limit of each parameter variable, randomly generating a 200 x 43 decision matrix, setting the 40 th column and the 41 th column of the matrix to be target function values, setting the last two columns to be non-dominated layer numbers and crowding degree distances, and setting one parameter vector to correspond to one individual or be called a chromosome.
Step 4, selecting, adopting rapid non-dominated sorting, and setting two parameters n i And S i ，n i Governing the number of individuals i, S, for all individuals in a population i A set of individuals governed by an individual i;
firstly, n in the population is selected i Individuals of 0, i.e. non-dominant optimal individuals, and put into the set F 1 Performing the following steps;
for F 1 Respectively finding out a dominant individual set S corresponding to i i To S i Individual j, pair n of j Performing a subtraction operation, i.e. n j ＝n j -1, excluding the influence of the individual i dominating j if n j 0, then put the individual j into the set F 2 Performing the following steps;
then set F 1 For the first layer non-dominant set, for F 1 The number of non-dominant layers of all individuals in the population is recorded as Rank1, F 2 The Chinese is marked as Rank 2;
repeating the above operations, and performing non-dominated sorting on all individuals in the population, wherein the number of non-dominated layers is from low to high, and the smaller the Rank value is, the better the individuals are;
in the iterative process of the NSGA-II algorithm, the population size is always N, so that individuals in the same Rank layer need to be selected and rejected in the selection operation, and a crowdedness distance C is defined, namely the sum of the distance differences of two adjacent individuals on each sub-target function, namely: c k ＝(f k+1,1 -f k-1,1 )+(f k-1,2 -f k+1,2 )；
The individuals with large crowding degree distance are better than the individuals with small crowding degree distance, and the non-dominated layer number and the crowding degree distance are used as the last two elements of a decision vector and are used for judging the quality of a solution and carrying out sorting operation.
Step 5, selecting a binary tournament, setting the tournament size to be 2 and the matching pool size to be 100; randomly selecting two individuals in an initial population, comparing the non-dominant grades, and putting the low-grade individuals into a matching pool; if the grades are the same, comparing the crowdedness distance, and keeping the distance larger; if the non-dominant grade and the crowding degree distance are the same, one individual is randomly selected for reservation, and the operations are repeated until 100 individuals in the matching pool are obtained.
Step 6, crossing, namely randomly selecting two individuals from the current generation population as parent individuals for crossing operation, and performing crossing operation on genes on matched chromosomes to generate a pair of new chromosomes; the specific operation process is as follows: assuming that the gene length of the chromosome is L, in the range of [0, L ], an integer value K is randomly selected as a crossing position, and the matching chromosomes mutually exchange the genes of the [ K, L ] part at the crossing position, thereby forming a new pair of chromosomes, namely individuals.
And 7, mutation, namely defining a mutation operator, carrying out small-probability replacement on the genes of the individuals, and acting the mutation operator on the population to change the genes of some individuals in the population to generate new alleles.
And 8, combining the parent-child population subjected to the operations in the steps 5 to 7, and marking as a first generation population (gen is 1).
Step 9, fig. 4 is a schematic diagram of elite selection strategy, the offspring population and the parent population are combined through step 6 and step 7, the population size is 400, 200 individuals are selected as a new parent population through step 4, and a new offspring population P is generated through step 6 and step 7 t+1 (ii) a And repeating the step 9 until the maximum evolution algebra is reached to 200.
Step 10, as shown in FIG. 7, the Pareto optimal solution front edge is output after 200 generations, the front edge and the parameter matrix are output, the efficiency eta is 95.28-96.86%, and the power density rho is 4.48-6.05 KW/dm 3 The performance index in the range can be selected to obtain corresponding decision variable values, namely the solution of the optimization problem, according to whichAnd selecting and designing components according to the parameters.
Although the present invention has been described with reference to the preferred embodiments, it should be understood that various changes and modifications can be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (2)

1. A Swiss rectifier multi-objective optimization design method based on NSGA-II algorithm is characterized by comprising the following steps:
step 1, taking a direct current inductor L, an output capacitor C, a switching tube IGBT and a Diode in a basic topology circuit of a Swiss rectifier as optimization variables influencing performance indexes of the rectifier, and carrying out power and volume modeling on the direct current inductor L, the output capacitor C, the switching tube IGBT and the Diode;
step 2, defining an objective function: measuring the performance index of the Swiss rectifier by taking the efficiency and the power density of the Swiss rectifier as objective functions;
Step 3, initializing a population: setting the population size as N, the maximum evolution algebra as gen, and the number of decision variables as V; then, real number coding is carried out on the decision variables, and the upper limit and the lower limit of each parameter variable are input by taking the established component database as a constraint condition; randomly generating an N-dimensional decision matrix, taking a V +1 th column and a V +2 th column of the decision matrix as objective function values, taking the last two columns as non-dominant layer numbers and crowding degree distances, and enabling one parameter vector to correspond to one individual or be called a chromosome;
step 4, selection: setting two parameters n i And S i ，n i Governing the number of individuals i, S, for all individuals in a population i Judging the quality of the solution of the matrix and the sorting operation for the individual set dominated by the individual i by adopting rapid non-dominated sorting;
step 5, selecting the binary championship: setting the championship size as 2 and the matching pool size as N/2; randomly selecting two individuals in an initial population, comparing the non-dominant grades of the two individuals, and putting the low-grade individual into a matching pool; if the grades are the same, comparing the crowdedness distance, and keeping the distance larger; if the non-domination grade and the crowding degree distance are the same, randomly selecting one individual to be reserved, and repeating the step until the number of the individuals in the matching pool is N/2;
Step 6, crossing: firstly, randomly selecting two individuals in a current generation population as parent individuals of cross operation, carrying out cross operation on genes on matched chromosomes to generate a pair of new chromosomes, and repeating the cross operation to form a new generation population;
wherein the current generation population is a parent generation population;
and 7, mutation: defining a mutation operator, carrying out small-probability replacement on the gene of an individual, and acting the mutation operator on a population to change the gene of a part of individuals in the population to generate a new allele, and marking the population at the moment as an offspring population;
step 8, merging the parent population obtained in the step 6 and the child population obtained in the step 7, wherein the merged population is marked as a first generation population gen which is equal to 1, and the population size is N;
step 9, taking the first generation population obtained in the step 8 as a parent population, then carrying out cross variation on the parent population to obtain a child population, and then combining the parent population and the child population in the step to form a population, wherein the size of the population is 2N;
step 10, processing the population formed in the step 9 by using the operation process of the step 4, selecting N individuals as a new parent population, and generating a new child population P through the steps 6 and 7 t+1 (ii) a Repeating the step 9 until the maximum evolution algebra gen is reached;
step 11, outputting Pareto optimal solution front edges and parameter matrixes, obtaining corresponding decision variable values, namely solutions of optimization problems, according to the selected target performance indexes, and selecting and setting components according to the parameters;
step 1 the process of power and volume modeling comprises:
calculating the power of an inductor, the power P of the inductor L Comprises the following steps:
Figure FDA0003669565550000021
wherein u is L Is the input voltage of an inductor, I rms Effective value of current flowing through the inductor, f sw To the switching frequency, V L Is the volume of the inductor; calculating the coil volume of the inductor; the coil volume V cl Comprises the following steps:
Figure FDA0003669565550000022
wherein, ω is w Is the winding width, d w Is the winding depth;
calculating the volume V of the conductor cd Comprises the following steps: v cd ＝k pf V cl Wherein k is pf Is the fill factor of the coil and,
Figure FDA0003669565550000023
n is the number of turns of the coil,
Figure FDA0003669565550000024
is the cross-sectional area of the conductor,
calculating core volume
Figure FDA0003669565550000025
Comprises the following steps:
Figure FDA0003669565550000026
then the volume of the inductor V L Comprises the following steps:
Figure FDA0003669565550000027
wherein l represents the core length;
and calculating the on-state loss of the IGBT of the switching tube as follows:
Figure FDA0003669565550000028
wherein, V GE Is the threshold voltage of IGBT, I IGBT Is the magnitude of the current flowing through the IGBT, r CE Is an on-state equivalent resistance, M is a modulation ratio,
Figure FDA0003669565550000029
expressed as a voltage current phase angle;
the switching losses of the IGBT are:
Figure FDA00036695655500000210
wherein E is sw(on) And E sw(off) Energy lost for switching on and off the IGBT once, respectively, I N Rated operating current for the rectifier, u CEN At a rated operating voltage of u pn Outputting a voltage for the rectifier;
and (3) calculating the total loss of the IGBT as follows: p IGBT ＝P cond,IGBT +P sw,IGBT ；
The conduction loss of the diode is:
P on,Diode ＝V F I F D
wherein, V F Is the forward conduction voltage drop of the diode, I F The current is positive on-state current, D is duty ratio, and the duty ratios of different diodes are obtained according to 4 conducting circuit states;
the off-state loss of the diode is:
P off,Diode ＝V R I R (1-D)
wherein, V R For its reverse pressure drop, I R Is a diode reverse leakage current;
the switching losses of the diodes are:
Figure FDA0003669565550000031
wherein, V fp And V rp Respectively forward direction of the diodeAnd inverse peak voltage, I fp And I rp Respectively the forward and reverse peak currents through the diode, t fp For its forward recovery time, t b Is its reverse current fall time;
the total diode loss is then: p Diode ＝P on,Diode +P off,Diode +P sw,Diode ；
The specific process of defining the objective function in step 2 comprises:
defining an objective function, taking the efficiency and the power density of the Swiss rectifier as the objective function, wherein the efficiency eta is as follows:
Figure FDA0003669565550000032
wherein, P 0 To output power, P L For inductive losses, the power density ρ is:
Figure FDA0003669565550000033
wherein, P 0 To output power, P loss For total loss, P IGBT For total IGBT losses, P Diode Is the total Diode loss; v General (1) Is the total volume of the component, V L Is the volume of the inductor, V IGBT Is the IGBT volume, V Diode Is the diode volume, V C And the power density rho is an objective function for outputting the capacitance volume.
2. The optimal design method according to claim 1, wherein the specific process of judging the quality and the ranking of the solutions in the fourth step comprises:
step 1, selecting n in the population i Individuals of 0, i.e. non-dominant optimal individuals, and put into the set F 1 Performing the following steps;
step 2, pair set F 1 Respectively finding out a dominant individual set S corresponding to i i To S i Individual j in (1), pairThe number n of all individuals in the population dominating an individual j j Performing a subtraction operation, i.e. n j ＝n j -1, excluding the influence of the individual i dominating j if n j 0, then put the individual j into the set F 2 Performing the following steps;
then set F 1 For the first layer non-dominant set, for F 1 The number of non-dominant layers of all individuals in (A) is recorded as Rank1, F 2 The Chinese is marked as Rank 2;
step 3, repeating the operations of the step 1 and the step 2, and performing non-dominated sorting on all individuals in the population, wherein the number of non-dominated layers is from low to high, and the smaller the Rank value is, the better the individuals are;
the population size is always N in the iteration process of the NSGA-II algorithm, so that individuals in the same Rank layer need to be chosen and rejected in the selection operation, a crowding degree distance C is defined, namely the sum of the distance differences of two adjacent individuals on each sub-objective function, and C is k ＝(f k+1,1 -f k-1,1 )+(f k-1,2 -f k+1,2 ) Where k is any point in the solution front edge, f k+1,1 、f k-1,1 、f k-1,2 And f k+1,2 Respectively expressed as target functions corresponding to (k +1,1), (k-1,2) and (k +1, 2);
the individuals with large crowding degree distance are better than the individuals with small crowding degree distance, the non-dominated layer number and the crowding degree distance are used as the last two elements of the decision vector, and the quality of the solution and the sorting operation are judged according to the non-dominated layer number and the crowding degree distance.
CN201910729666.3A 2019-08-08 2019-08-08 Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm Active CN110633494B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910729666.3A CN110633494B (en) 2019-08-08 2019-08-08 Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910729666.3A CN110633494B (en) 2019-08-08 2019-08-08 Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm

Publications (2)

Publication Number Publication Date
CN110633494A CN110633494A (en) 2019-12-31
CN110633494B true CN110633494B (en) 2022-08-12

Family

ID=68969614

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910729666.3A Active CN110633494B (en) 2019-08-08 2019-08-08 Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm

Country Status (1)

Country Link
CN (1) CN110633494B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111461911B (en) * 2020-01-17 2022-08-16 华中科技大学 Mixed NSGA-II-based method for constructing virtual unit of machined workpiece
CN112084706B (en) * 2020-09-02 2023-11-07 合肥工业大学 NSGA-II-based multi-objective optimization method for capacitor module component configuration
CN112968474B (en) * 2021-03-30 2022-12-30 合肥工业大学 Multi-target optimization method for photovoltaic off-grid inverter system
CN114465477A (en) * 2021-09-26 2022-05-10 天津大学 Comprehensive design method for parameters of interleaved parallel Boost converter
CN114065512B (en) * 2021-11-16 2024-08-20 吉林大学 NSGA-II algorithm-based induction magnetometer coil parameter multi-objective optimization method
CN113987891B (en) * 2021-11-26 2024-07-09 国网江苏省电力有限公司电力科学研究院 Multi-objective optimization method for IGBT module

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103793758A (en) * 2014-01-23 2014-05-14 华北电力大学 Multi-objective optimization scheduling method for electric vehicle charging station including photovoltaic power generation system
CN104952067A (en) * 2015-05-13 2015-09-30 西安电子科技大学 Method for segmenting color images on basis of NSGA-II (non-dominated sorting genetic algorithm-II) evolution algorithms
CN108493997A (en) * 2018-04-13 2018-09-04 哈尔滨理工大学 Rotary inertia optimal control method based on virtual synchronous generator

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107423529A (en) * 2017-08-30 2017-12-01 同济大学 Metamaterial Precise spraying method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103793758A (en) * 2014-01-23 2014-05-14 华北电力大学 Multi-objective optimization scheduling method for electric vehicle charging station including photovoltaic power generation system
CN104952067A (en) * 2015-05-13 2015-09-30 西安电子科技大学 Method for segmenting color images on basis of NSGA-II (non-dominated sorting genetic algorithm-II) evolution algorithms
CN108493997A (en) * 2018-04-13 2018-09-04 哈尔滨理工大学 Rotary inertia optimal control method based on virtual synchronous generator

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
On the performance of multiobjective evolutionary algorithms in automatic parameter extraction of power diodes;Daniele parada等;《2014 IEEE Transactions on power electronics》;20140930;第30卷(第9期);4986-4997 *
VIENNA整流器电容设计;刘平等;《装备制造》;20090801(第8期);1 *
基于NSGA-Ⅱ算法Vienna整流器多目标优化;常伟等;《中国电机工程学报》;20160930;第36卷(第S1期);179-185 *
混合整流器多目标优化研究;常伟;《中国优秀博硕士学位论文全文数据库(硕士)(工程科技Ⅱ辑)》;20180115;C042-283 *

Also Published As

Publication number Publication date
CN110633494A (en) 2019-12-31

Similar Documents

Publication Publication Date Title
CN110633494B (en) Multi-objective optimization design method of Swiss rectifier based on NSGA-II algorithm
Arsalan et al. MPPT for photovoltaic system using nonlinear backstepping controller with integral action
Lee et al. Niching genetic algorithm with restricted competition selection for multimodal function optimization
Eldahab et al. Enhancing the maximum power point tracking techniques for photovoltaic systems
CN107612016B (en) Planning method of distributed power supply in power distribution network based on maximum voltage correlation entropy
KR20110086520A (en) Controller for a rotating electrical machine
Mejbri et al. Bi-objective sizing optimization of power converter using genetic algorithms: application to photovoltaic systems
Li et al. Fuzzy inference NSGA-III algorithm-based multi-objective optimization for switched reluctance generator
CN110532630B (en) Switch block, parasitic switch state and electromagnetic transient simulation method and device
Falehi Novel harmonic elimination strategy based on multi-objective grey wolf optimizer to ameliorate voltage quality of odd-nary multi-level structure
Karthikeyan et al. Application of self-adaptive differentialevolution algorithm to generation expansion planning problem
Jeong et al. Optimal IPT Core Design for Wireless Electric Vehicles by Reinforcement Learning
CN117335667A (en) Fractional order control method of fractional order phase-shifting full-bridge converter of power electronic transformer
Delaforge et al. Design automation of power electronic converters a grid elitist multiobjective genetic algorithm
Zhang et al. A multi-objective analytical design approach of switched reluctance machines with integrated active current profile optimization
Sahu et al. Design paradigm for modular multilevel converter-based generator rectifier systems
CN110189230A (en) A kind of construction method of the parsingization model of dynamic partition
CN115221787A (en) Cylindrical permanent magnet linear motor multi-objective optimization method and system based on NSGA-II
Lin The dual active bridge converter design with artificial intelligence
Rao et al. Optimal design of a furnace transformer by intelligent evolutionary methods
Samajpati Distributed generation allocation for power loss minimization and voltage improvement of radial distribution systems using genetic algorithm
Sredenšek et al. Parameter determination of a solar cell model using differential evolution algorithm
Zapata et al. Optimal Design of Inductors Based on Homothetic Shape and using Geometric Programming
Hyeon et al. A Multi-Objective PFC Boost Inductor Optimal Design Algorithm Based on Pareto Front
Shen et al. Artificial Intelligence Applications in High-Frequency Magnetic Components Design for Power Electronics Systems: An Overview

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