CN107463995B - Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA - Google Patents

Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA Download PDF

Info

Publication number
CN107463995B
CN107463995B CN201710648027.5A CN201710648027A CN107463995B CN 107463995 B CN107463995 B CN 107463995B CN 201710648027 A CN201710648027 A CN 201710648027A CN 107463995 B CN107463995 B CN 107463995B
Authority
CN
China
Prior art keywords
fuel cell
individuals
rna
model
parameters
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
CN201710648027.5A
Other languages
Chinese (zh)
Other versions
CN107463995A (en
Inventor
张丽
王宁
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.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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 Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201710648027.5A priority Critical patent/CN107463995B/en
Publication of CN107463995A publication Critical patent/CN107463995A/en
Application granted granted Critical
Publication of CN107463995B publication Critical patent/CN107463995B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation

Abstract

The invention discloses a fuel cell optimization modeling method with an adaptive genetic strategy RNA-GA. 1) Obtaining sampling data of input current and output voltage of the proton exchange membrane fuel cell through field operation or experiments; 2) taking the sum of squares of errors of the estimated output of the fuel cell model and the actually output sampling data as an objective function in the RNA-GA optimizing search; 3) setting algorithm operation parameters; 4) and operating the RNA-GA to estimate unknown parameters in the fuel cell model, obtaining an estimated value of the unknown parameters in the model by minimizing an objective function, and substituting the estimated value of the unknown parameters into the fuel cell model to form a mathematical model. The method uses the self-adaptive genetic strategy to decide and execute the crossover or mutation operation, thereby effectively keeping the population diversity, accelerating the convergence speed of the algorithm towards the global optimal solution, obtaining reliable parameters of the fuel cell model, and being suitable for the optimization modeling of other complex chemical reaction processes.

Description

Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA
Technical Field
The invention relates to an intelligent optimization modeling method, in particular to a fuel cell optimization modeling method with an adaptive genetic strategy RNA-GA.
Background
Fuel cells are one of the high and new technologies that have a great impact on the human society in the 21 st century, and among them, Proton Exchange Membrane Fuel Cells (PEMFCs) are receiving wide attention because of their advantages of environmental protection, high efficiency, low emission, etc. The PEMFC is a nonlinear, multi-input and strongly-coupled complex system, and the research on the modeling problem of the PEMFC has important significance on the theoretical research and engineering application of the fuel cell. To build a reliable model, the first solution is the parameter estimation problem in the model. The parameter estimation problem is essentially an optimization problem, and researchers have adopted some traditional optimization methods, such as Levenberg-Marquardt (L-M), Gauss-Newton, etc., to solve the problem. However, these deterministic optimization algorithms are prone to be trapped in local minima during the search, or are difficult to achieve satisfactorily. The Genetic Algorithm (GA) is a global optimization searching method formed by simulating the genetic and evolutionary processes of organisms in a natural environment, can solve the complex optimization problem which is difficult to solve by the traditional optimization method, can perform effective search in a complex search space, and has strong robustness and adaptability. However, the traditional genetic algorithm is easy to have the defects of premature convergence, poor local search capability and the like.
The invention provides a novel intelligent optimization modeling method based on RNA biological characteristics and evolutionary computation ideas, establishes a high-precision model of a proton exchange membrane fuel cell, and researches the effectiveness of the intelligent optimization modeling method through examples.
The invention content is as follows:
the invention aims to provide a fuel cell optimization modeling method with an adaptive genetic strategy RNA-GA aiming at the defects of the prior art.
The fuel cell optimization modeling method with the adaptive genetic strategy RNA-GA comprises the following steps:
1) obtaining sampling data of current input and voltage output of the fuel cell through field operation or experiments; inputting the sampled data for the current of each group of fuel cells, and taking the sum of squares of errors of the voltage estimation output of the fuel cell model and the voltage actual output of the fuel cell as an objective function in the RNA-GA optimizing search;
2) setting algorithm operating parameters, including: the population scale N, the maximum evolution algebra MaxGen, the value range of the estimated parameters and the coding length l of each unknown parameter;
3) the termination criteria for the algorithm are set as: the algebra is operated to reach the maximum evolution algebra;
4) estimating unknown parameters in the fuel cell model by RNA-GA with an adaptive genetic strategy, obtaining an estimated value of the unknown parameters in the fuel cell model by minimizing an objective function, and substituting the estimated value into the fuel cell model to obtain a mathematical model of the fuel cell for estimating the output voltage of the fuel cell.
The steps can adopt the following specific implementation modes:
the RNA-GA with the adaptive genetic strategy comprises the following steps of estimating unknown parameters in a fuel cell model:
4.1) randomly generating N individuals with the length L ═ nxl to form an initial population, wherein N is the number of unknown parameters in the fuel cell model, and converting A, U, G, C four bases into quaternary codes of 0,1,2 and 3 by adopting an RNA coding mode for each individual in the population (for convenience of expression, the corresponding codes are still expressed by adopting the bases below);
4.2) calculating the fitness value of individuals in the population, randomly selecting two individuals as parents, and simultaneously deciding whether to execute cross operation or mutation operation according to a self-adaptive genetic strategy to generate new offspring individuals;
4.3) repeating the step 4.2) until the number of the generated filial generation units is 3N/2;
4.4) selecting N individuals to form a new population by using a proportion selection strategy;
4.5) if the set termination criterion is met, the algorithm ends, otherwise, steps 4.2) to 4.4) are repeated.
In 4.2), the decision of performing the crossover operation or the mutation operation is made according to the adaptive genetic strategy, and the specific operation process of generating the new offspring individuals is as follows:
(a) constructing an evaluation function SC (x, y) of the similarity degree, wherein the evaluation function is in the form of:
Figure GDA0002272444340000021
evaluation function x and y are two randomly selected quaternary coded dye color body strings, xpkAnd ypk⊙ is the same or operation for the kth code value of the p-th substring in two chromosome strings, in the evolution process of each generation, the evaluation function between the randomly selected x and y individuals is calculated, and the average similarity coefficient SC of the initial population is calculated at the same timeinitial；
(b) Calculating similarity coefficient SC (t) ═ SC of the t generation populationinitial×(βt)，
Figure GDA0002272444340000022
(c) If SC (x, y) < SC (t) is satisfied, performing a crossover operation to generate a new child individual; if SC (x, y) ≧ SC (t), respectively performing anticodon variation operation and rare base variation operation on the two individuals to generate new filial generation individuals;
(d) repeating steps (a) to (c) until 3N/2 offspring individuals are generated.
The cross operation method in 4.2) comprises the following steps:
(A) in the evolution process of the tth generation, if SC (x, y) < SC (t) is satisfied, the two-point intersection operation is performed with a probability of 1;
(B) randomly selecting one segment from the coding sequences of the two individuals as a subsequence, wherein the number of bases of the two subsequences is equal, and exchanging the two subsequences, thereby generating two new individuals.
The mutation operation method in 4.2) comprises the following steps:
(A) in the mutation operation, calculating the fitness values of two individuals satisfying SC (x, y) ≧ SC (t), and taking the individual with the larger fitness value as a father of the anticodon mutation operation; dividing the parent into n subsequences according to the number of coding parameters, randomly selecting a certain number of continuous bases in each subsequence to form codons, generating a base sequence which is complementary with the codons according to the Watton-Crick complementary principle, namely an anti-codon, and inverting the anti-codon to replace the position of the codons to form a new filial generation;
(B) selecting an individual with a smaller fitness value as a father of rare base variation operation for two individuals satisfying SC (x, y) being not less than SC (t); counting the number of four bases in the father, and replacing the base with the highest use frequency with the base with the lowest appearance frequency to generate a new filial individual.
The invention combines RNA calculation and genetic algorithm, adopts an RNA coding method, and introduces the decision of adopting cross operation or variation operation according to the similarity degree between two individuals in the population, thereby effectively increasing the diversity of the population, reserving the excellent genes of the original population, improving the global search capability of the algorithm and avoiding premature convergence. The method is used as an intelligent optimization algorithm, is successfully used for model parameter estimation of the proton exchange membrane fuel cell, and has good application prospect.
Drawings
FIG. 1 is a flow chart of RNA-GA with adaptive genetic strategy in the present invention;
FIG. 2 is a schematic diagram of the anti-codon usage of the present invention;
FIG. 3 is a schematic diagram of the rare base mutation operation in the present invention;
FIG. 4(a) is an i-V curve of a PEM fuel cell according to the present invention under the identified group of data;
FIG. 4(b) is an i-V curve of a PEM fuel cell under validation set data in accordance with the present invention.
Detailed Description
The method of the present invention is used in the estimation of proton exchange membrane fuel cell model parameters, and is described in detail as follows:
example (b):
a proton exchange membrane fuel cell is an electrochemical device that generates electricity by utilizing the reverse reaction of electrolyzed water.
The electrochemical reaction equation that occurs for the 2 electrodes is as follows:
anode: h2→2H++2e-
Cathode:
Figure GDA0002272444340000031
the total reaction of the battery:
Figure GDA0002272444340000032
the reaction products are direct current electrical energy, liquid water and heat of reaction formation.
The application object model of the embodiment adopts a single cell voltage V-I model proposed by J.C.Amphlett, and the basic expression of the output voltage of a single cell is as follows:
Vcell＝ENernst-Vact-Vohmic-Vcon(1)
in the formula VcellIs the terminal voltage (V), E of the batterynernst,Vact,VohmicAnd VconRespectively, the cell thermodynamic voltage, the activated polarization electromotive force, the ohmic voltage drop and the concentration overvoltage (V).
Figure GDA0002272444340000041
Figure GDA0002272444340000042
Vohmic＝i(RM+RC) (4)
Figure GDA0002272444340000043
Wherein T is the absolute temperature (K) of the battery environment;
Figure GDA0002272444340000044
and
Figure GDA0002272444340000045
partial pressures (atm) of hydrogen and oxygen, respectively ξk(k ═ 1,2,3,4) for activated polarization electromotive force coefficient; i is the load current (A);
Figure GDA0002272444340000046
oxygen concentration (mol cm) at the interface between the cathode film and the reaction gas-3)；RMIs the equivalent membrane impedance of the proton membrane; rCA fuel cell internal resistance that is an ohmic voltage drop; b is concentration polarization overpotential coefficient (V), and I represents the actual current density (A cm) of the cell-2)；ImaxRepresents the maximum current density (A cm) of the cell under the maximum fuel supply condition-2)。
According to Henry's law, the interfacial dissolved oxygen concentration of the cathode catalyst is known as follows:
Figure GDA0002272444340000047
equivalent membrane impedance RMCan be derived from ohm's law, namely:
Figure GDA0002272444340000048
wherein l is the membrane thickness (cm) of the proton exchange membrane, A is the activation area of the membrane, and the resistivity rho of the Nafion series proton exchange membraneM(Ω cm) can be represented by the following formula, i.e.
Figure GDA0002272444340000049
Wherein, lambda is the water content of proton exchange membrane, is an adjustable parameter, reflects gas humidity state, and the value range is between 10 ~ 23 usually.
Pressure of saturated water vapor
Figure GDA00022724443400000410
Can be expressed as a function of temperature T:
Figure GDA0002272444340000051
when the anode inlet of the PEMFC is hydrogen and the cathode inlet is air, the following components are adopted:
Figure GDA0002272444340000052
Figure GDA0002272444340000053
Figure GDA0002272444340000054
wherein
Figure GDA0002272444340000055
Is the pressure of the nitrogen in the tank,
Figure GDA0002272444340000056
and
Figure GDA0002272444340000057
the oxygen and hydrogen partial pressures in the anode and cathode cells, respectively. PaAnd PcPressure at the anode and cathode inlets (atm), RH, respectivelyaAnd RHcAnode and cathode relative steam humidity, respectively.
Battery pack consisting of n series-connected cells, the output voltage V of whichsCan be calculated by the following formula:
Vs＝n×Vcell(13)
in the model of PEFMC, ξ are parameters to be estimated1,ξ2,ξ3,ξ4,λ,RcAnd b is seven in total.
As shown in FIG. 1, modeling proton exchange membrane fuel cells based on RNA-GA with adaptive genetic strategy has the steps 1) to 4):
1) the current input and voltage output sampling data of 60 groups of fuel cells were measured by field experiments according to four conditions (condition 1: 3/5bar, 353.15K; working condition 2: 1/1bar, 343.15K; working condition 3: 2.5/3bar, 353.15K; working condition 4: 1.5/1.5bar,353.15K) the data were divided into four groups. And selecting 30 groups of data of the working condition 1 and the working condition 2 as training samples for parameter estimation, and verifying the accuracy of the model by using the data of the working condition 3 and the working condition 4. For current input sampling data of 60 groups of fuel cells, taking the sum of squares of errors of voltage estimation output of a fuel cell model and voltage actual output of the fuel cell as an objective function in the RNA-GA optimization search, wherein the optimization objective function is specifically as follows:
Figure GDA0002272444340000058
where M is the sample volume, VsIs the model output value, V, of the ith samplesmAre experimental data. The optimization index is used as an objective function in RNA-GA optimizing search;
extrinsic characteristic sampling data of fuel cells are papers: morgans, Chongzhuang, Cao Guangyi, Parametric optimization for a PEMFC model with a hybrid genetic algorithm.International journal of Energy Research 2006; volume 30. page: 585-.
2) Setting the maximum evolution algebra MaxGen of algorithm operation as 1000, the population scale N as 60, the coding length l of each parameter as 20, and presetting the value range of the estimated parameter according to the table 1;
table 1 value ranges of model parameters:
Figure GDA0002272444340000061
3) setting the termination criteria of the algorithm: the algebra is operated to reach the maximum evolution algebra;
4) using RNA-GA with adaptive genetic strategy for unknown parameters ξ in pem fuel cells1,ξ2,ξ3,ξ4,λ,RcAnd b, estimating, namely obtaining an estimated value of an unknown parameter in the fuel cell model by minimizing the objective function, and substituting the estimated value of the unknown parameter into the fuel cell model to form a mathematical model of the proton exchange membrane fuel cell model for estimating the output voltage of the fuel cell.
Unknown parameters in the PEMFC model using RNA-GA with adaptive genetic strategy as described above (ξ)1,ξ2,ξ3,ξ4,λ,RcB) the estimation step is as follows:
4.1) randomly generated containing 60 individuals with different RNA sequences, making up the initial population. Each RNA sequence represents a set of possible solutions for the unknown parameter to be estimated. In this example, each parameter to be estimated is coded by an RNA subsequence with length L equal to 20 using quaternary {0,1,2,3}, and the coding length L equal to n × L equal to 7 × 20 equal to 140 for each individual in the population, where n is the number of unknown parameters in the fuel cell model, which is 7;
4.2) decoding each RNA sequence in the population into a group of unknown parameters to be estimated of the proton exchange membrane fuel cell model, and calculating a target function value corresponding to the group of parameters and a corresponding fitness value;
4.3) randomly selecting two individuals as parents, and simultaneously deciding whether to execute cross operation or mutation operation according to a self-adaptive genetic strategy to generate new filial individuals;
4.4) repeating the step 4.3) until the number of the generated filial generation units is 3N/2;
4.5) selecting N individuals to form a new population by using a proportion selection strategy;
4.6) if the termination criterion is met, the algorithm ends, otherwise, steps 4.2) to 4.5) are repeated.
The adaptive genetic strategy comprises the following operation steps:
(a) constructing an evaluation function SC (x, y) of the similarity degree, wherein the evaluation function is in the form of:
Figure GDA0002272444340000062
evaluation function in which x and y are two quaternary coded chromosome strings, xpkAnd ypk⊙ is the same or operation for the kth coding value of the pth substring in two chromosome strings (0 ⊙ 0 ═ 1,0 ⊙ 1 ═ 0,1 ⊙ 0 ═ 0,1 ⊙ 1 ═ 1). in the course of evolution of each generation, two individuals x and y were randomly selected in the population, the evaluation function of the similarity degree between the two individuals was calculated
Figure GDA0002272444340000071
(b) Calculating similarity coefficient SC (t) ═ SC of the t generation populationinitial×(βt)，
Figure GDA0002272444340000072
(c) If SC (x, y) < SC (t) is satisfied, performing a crossover operation to generate a new child individual; if SC (x, y) is more than or equal to SC (t), respectively carrying out mutation operation on the two individuals to generate new filial generation individuals;
(d) repeating steps (a) to (c) until 3N/2 offspring individuals are generated.
The specific method for performing the selecting operation in the above 4.5) is:
in each generation of evolution, after crossover and mutation operations, the replica population was selected proportionally among the populations consisting of 3N/2 individuals until a new population containing N different RNA sequences was generated.
The specific method for performing the crossover operation comprises the following steps:
(A) in the evolution process of the tth generation, if SC (x, y) < SC (t) is satisfied, the two-point intersection operation is performed with a probability of 1;
(B) randomly selecting one segment from the coding sequences of the two individuals as a subsequence, wherein the number of bases of the two subsequences is equal, and exchanging the two subsequences, thereby generating two new individuals.
The specific method for performing mutation operation comprises the following steps:
(A) in the mutation operation, the sizes of the fitness values of two individuals satisfying SC (x, y) ≧ SC (t) are calculated, and the individual with the larger fitness value is taken as a father of the anticodon mutation operation. Dividing the parent into n subsequences according to the number of coding parameters, randomly selecting a certain number of continuous bases in each subsequence to form codons, generating a base sequence which is complementary with the codons according to the Watton-Crick complementary principle, namely an anti-codon, and inverting the anti-codon to replace the position of the codons to form a new filial generation;
(B) for two individuals satisfying SC (x, y) ≧ SC (t), the individual with smaller fitness value is selected as the father of rare base variation operation. Counting the number of 4 bases in the father, and replacing the base with the highest use frequency with the base with the lowest appearance frequency to generate a new filial individual.
According to the method, the parameter estimation value in the proton exchange membrane fuel cell model is obtained as follows:
table 3: the modeling method has the following parameter optimization results:
Figure GDA0002272444340000073
and substituting the estimation parameters into a proton exchange membrane fuel cell model to obtain a corresponding mathematical model. For the identification group and the verification group, fitted curves of the voltage and the current density of the galvanic pile calculated by the model and the voltage and the current density of the galvanic pile acquired in the actual PEMFC galvanic pile experiment are respectively shown in fig. 4(a) and 4 (b). The result shows that the RNA-GA with the adaptive genetic strategy can be used for parameter estimation of a proton exchange membrane fuel cell model, and the obtained model can accurately reflect the system characteristics. The output voltage of the fuel cell can be predicted by the model, and can be used as a reference for cell design and performance analysis or used for adjusting the working state and relevant characteristics of a later-stage power electronic device.

Claims (3)

1. A fuel cell optimization modeling method with an adaptive genetic strategy RNA-GA is characterized by comprising the following steps:
1) obtaining sampling data of current input and voltage output of the fuel cell through field operation or experiments; inputting the sampled data for the current of each group of fuel cells, and taking the sum of squares of errors of the voltage estimation output of the fuel cell model and the voltage actual output of the fuel cell as an objective function in the RNA-GA optimizing search;
2) setting algorithm operating parameters, including: the population scale N, the maximum evolution algebra MaxGen, the value range of the estimated parameters and the coding length l of each unknown parameter;
3) the termination criteria for the algorithm are set as: the algebra is operated to reach the maximum evolution algebra;
4) estimating unknown parameters in the fuel cell model through RNA-GA with a self-adaptive genetic strategy, obtaining an estimated value of the unknown parameters in the fuel cell model through minimizing an objective function, substituting the estimated value into the fuel cell model to obtain a mathematical model of the fuel cell, and estimating the output voltage of the fuel cell;
the RNA-GA with the adaptive genetic strategy comprises the following steps of estimating unknown parameters in a fuel cell model:
4.1) randomly generating N individuals with the length L being equal to N multiplied by L to form an initial population, wherein N is the number of unknown parameters in the fuel cell model, adopting an RNA coding mode for each individual in the population, and converting A, U, G, C four bases into quaternary codes of 0,1,2 and 3;
4.2) calculating the fitness value of individuals in the population, randomly selecting two individuals as parents, and simultaneously deciding whether to execute cross operation or mutation operation according to a self-adaptive genetic strategy to generate new offspring individuals;
4.3) repeating the step 4.2) until the number of the generated filial generation units is 3N/2;
4.4) selecting N individuals to form a new population by using a proportion selection strategy;
4.5) if the set termination criterion is met, the algorithm is ended, otherwise, the steps 4.2) to 4.4) are repeated;
in 4.2), the decision of performing the crossover operation or the mutation operation is made according to the adaptive genetic strategy, and the specific operation process of generating the new offspring individuals is as follows:
(a) constructing an evaluation function SC (x, y) of the similarity degree, wherein the evaluation function is in the form of:
Figure FDA0002272444330000011
evaluation function x and y are two randomly selected quaternary-coded chromosome strings, xpkAnd ypk⊙ is the same or operation for the kth code value of the p-th substring in two chromosome strings, during the evolution process of each generation, the evaluation function between the randomly selected x and y individuals is calculated, and at the same time, the PopSize group parent individuals are randomly selected, and the average similarity coefficient SC of the initial population is calculatedinitial；
(b) Calculating similarity coefficient SC (t) ═ SC of the t generation populationinitial×(βt)，
Figure FDA0002272444330000021
(c) If SC (x, y) < SC (t) is satisfied, performing a crossover operation to generate a new child individual; if SC (x, y) ≧ SC (t), the anti-codon variation operation and the rare base variation operation are respectively carried out on the two individuals, and a new filial generation individual is generated.
2. The fuel cell optimization modeling method with adaptive genetic strategy RNA-GA of claim 1, characterized in that the crossover operation method in 4.2) is:
(A) in the evolution process of the tth generation, if SC (x, y) < SC (t) is satisfied, the two-point intersection operation is performed with a probability of 1;
(B) randomly selecting one segment from the coding sequences of the two individuals as a subsequence, wherein the number of bases of the two subsequences is equal, and exchanging the two subsequences, thereby generating two new individuals.
3. The fuel cell optimization modeling method with adaptive genetic strategy RNA-GA of claim 2, characterized in that the mutation operation method in 4.2) is:
(A) in the mutation operation, calculating the fitness values of two individuals satisfying SC (x, y) ≧ SC (t), and taking the individual with the larger fitness value as a father of the anticodon mutation operation; dividing the parent into n subsequences according to the number of coding parameters, randomly selecting a certain number of continuous bases in each subsequence to form codons, generating a base sequence which is complementary with the codons according to the Watton-Crick complementary principle, namely an anti-codon, and inverting the anti-codon to replace the position of the codons to form a new filial generation;
(B) selecting an individual with a smaller fitness value as a father of rare base variation operation for two individuals satisfying SC (x, y) being not less than SC (t); counting the number of four bases in the father, and replacing the base with the highest use frequency with the base with the lowest appearance frequency to generate a new filial individual.
CN201710648027.5A 2017-08-01 2017-08-01 Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA Active CN107463995B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710648027.5A CN107463995B (en) 2017-08-01 2017-08-01 Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710648027.5A CN107463995B (en) 2017-08-01 2017-08-01 Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA

Publications (2)

Publication Number Publication Date
CN107463995A CN107463995A (en) 2017-12-12
CN107463995B true CN107463995B (en) 2020-04-17

Family

ID=60547293

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710648027.5A Active CN107463995B (en) 2017-08-01 2017-08-01 Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA

Country Status (1)

Country Link
CN (1) CN107463995B (en)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102017127276A1 (en) * 2017-08-30 2019-02-28 Taiwan Semiconductor Manufacturing Co., Ltd. STANDARD CELLS AND ADAPTATIONS FROM THEREOF WITHIN A STANDARD CELL LIBRARY
CN108341342B (en) * 2018-01-23 2019-04-12 浙江大学 The overhead crane model construction of SVM method of prominent ring crossover operation RNA-GA
CN108711955B (en) * 2018-04-24 2021-10-08 国网电力科学研究院武汉南瑞有限责任公司 Method for tracking maximum output power point of laser energy supply system
CN109657348B (en) * 2018-12-18 2023-08-08 安徽江淮汽车集团股份有限公司 Modeling method for proton exchange membrane fuel cell
CN111781819B (en) * 2020-07-08 2021-05-04 浙江大学 Greenhouse temperature PID control method of RNA cross-operation symbiotic algorithm
CN112968515B (en) * 2021-02-04 2022-06-03 武汉理工大学 Energy management strategy and system for emergency power supply of fuel cell
CN114297942A (en) * 2021-12-27 2022-04-08 天津大学 Proton exchange membrane electrolytic cell model parameter estimation method based on random optimization algorithm
CN115000471A (en) * 2022-05-18 2022-09-02 西安交通大学 Fuel cell catalyst layer prediction-analysis-optimization method based on machine learning
CN116258089B (en) * 2023-05-15 2023-07-18 中汽研新能源汽车检验中心(天津)有限公司 Fuel cell simulation optimization method and system based on sensitivity and genetic analysis

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102521654A (en) * 2011-11-22 2012-06-27 浙江大学 Supercritical water oxidation reaction kinetic model parameter estimation method employing RNA (Ribonucleic Acid) genetic algorithm
CN103279659A (en) * 2013-05-22 2013-09-04 浙江大学 Fuel cell optimization modeling method of double loop crossover operation RNA genetic algorithm
KR101498391B1 (en) * 2013-11-27 2015-03-03 제주대학교 산학협력단 Method and apparatus of scheduling a multi-day visiting schedule over multiple places for an electric vehicle

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102521654A (en) * 2011-11-22 2012-06-27 浙江大学 Supercritical water oxidation reaction kinetic model parameter estimation method employing RNA (Ribonucleic Acid) genetic algorithm
CN103279659A (en) * 2013-05-22 2013-09-04 浙江大学 Fuel cell optimization modeling method of double loop crossover operation RNA genetic algorithm
KR101498391B1 (en) * 2013-11-27 2015-03-03 제주대학교 산학협력단 Method and apparatus of scheduling a multi-day visiting schedule over multiple places for an electric vehicle

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
DNA遗传算法及其在燃料电池中的应用研究;吕慧珍;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20160415;第38-55页 *
RNA genetic algorithm with adaptive crossover probability for estimating parameters of heavy oil thermal cracking model;Li Zhang etal.;《Proceedings of the 32nd Chinese Control Conference》;20130728;第1866-1870页 *
基于改进遗传算法的渔港规划问题研究;冯艳红 等;《辽宁科技大学学报》;20151031;第38卷(第5期);第396-399页 *

Also Published As

Publication number Publication date
CN107463995A (en) 2017-12-12

Similar Documents

Publication Publication Date Title
CN107463995B (en) Fuel cell optimization modeling method with adaptive genetic strategy RNA-GA
Mo et al. Parameter optimization for a PEMFC model with a hybrid genetic algorithm
Miao et al. Parameter estimation of PEM fuel cells employing the hybrid grey wolf optimization method
CN103336867B (en) Proton Exchange Membrane Fuel Cells model optimization disposal route
Tafaoli-Masoule et al. Optimum design parameters and operating condition for maximum power of a direct methanol fuel cell using analytical model and genetic algorithm
CN106407590A (en) Membrane calculation-based proton exchange membrane fuel cell model optimization method
CN110210071B (en) All-vanadium redox flow battery energy storage system operation optimization method based on genetic algorithm
Sultan et al. Parameter identification of proton exchange membrane fuel cell stacks using bonobo optimizer
CN110929451A (en) Fuel cell single voltage consistency prediction method
CN114447378A (en) Parameter optimization method of proton exchange membrane fuel cell
CN102521654B (en) Supercritical water oxidation reaction kinetic model parameter estimation method employing RNA (Ribonucleic Acid) genetic algorithm
CN116151114A (en) Method and system for predicting service life of fuel cell under meta-universe based on hybrid framework
CN114300713B (en) Cross flow SOEC pile two-dimensional dynamic model construction method and application thereof
CN104778315B (en) Merge the fuel cell Optimization Modeling method of krill behavior artificial bee colony algorithm
CN102521523B (en) Autophagy membrane-computing optimization modeling method for fuel cell
CN103279659B (en) A kind of fuel cell Optimization Modeling method of dicyclo interlace operation RNA genetic algorithm
CN116108680A (en) Fuel cell voltage model parameter identification method based on uncertainty quantitative evaluation
CN113078335B (en) Performance analysis method and device of proton exchange membrane fuel cell and terminal equipment
CN110993035B (en) Two-dimensional steady-state model establishment method for proton exchange membrane electrolytic cell characteristic analysis
Ohenoja et al. Identification of electrochemical model parameters in PEM fuel cells
Menon et al. Realisation of optimal parameters of PEM fuel cell using simple genetic algorithm (SGA) and simulink modeling
CN111859807A (en) Initial pressure optimizing method, device, equipment and storage medium for steam turbine
Sun et al. Inverse control optimization of voltage and temperature parameters of electric vehicle fuel cell based on genetic algorithm
CN116053536B (en) Proton exchange membrane fuel cell estimation method and computer readable medium
Khajuria et al. Model Parameter Extraction for PEM Electrolyzer Using Honey Badger 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