CN115966739B - Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm - Google Patents

Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm Download PDF

Info

Publication number
CN115966739B
CN115966739B CN202310231558.XA CN202310231558A CN115966739B CN 115966739 B CN115966739 B CN 115966739B CN 202310231558 A CN202310231558 A CN 202310231558A CN 115966739 B CN115966739 B CN 115966739B
Authority
CN
China
Prior art keywords
current density
fuel cell
hippocampal
individual
solid oxide
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
CN202310231558.XA
Other languages
Chinese (zh)
Other versions
CN115966739A (en
Inventor
付波
张万祥
何晗哲
陈登耀
黎祥程
李超顺
范秀香
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.)
Hubei University of Technology
Original Assignee
Hubei University of 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 Hubei University of Technology filed Critical Hubei University of Technology
Priority to CN202310231558.XA priority Critical patent/CN115966739B/en
Publication of CN115966739A publication Critical patent/CN115966739A/en
Application granted granted Critical
Publication of CN115966739B publication Critical patent/CN115966739B/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
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
    • Y02E60/30—Hydrogen technology
    • Y02E60/50—Fuel cells

Landscapes

  • Fuel Cell (AREA)

Abstract

The invention relates to the technical field of fuel cells, in particular to a fuel cell parameter identification method and system based on an improved hippocampal optimization algorithm. Collecting experimental values of the solid oxide fuel cell output voltage at a plurality of moments, and calculating a theoretical value of the solid oxide fuel cell output voltage at each moment; selecting constraint conditions of decision variables such as slope of a Tafil line and the like to construct parameters, and constructing a solid oxide fuel cell optimization model by taking the minimum mean square error of a calculated value of output voltage and an experimental measurement value as an objective function; and taking parameters such as the slope of the Tafil line and the like as variables to be solved, and solving the solid oxide fuel cell optimization model by improving a hippocampus optimization algorithm to obtain optimized parameters and optimized open-circuit voltage. The invention improves the local optimizing capability of the algorithm, helps the algorithm to more quickly locate the accurate position of the local optimal solution, and further improves the accuracy of fuel cell parameter identification.

Description

Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm
Technical Field
The invention relates to the technical field of fuel cells, in particular to a fuel cell parameter identification method and system based on an improved hippocampal optimization algorithm.
Background
With the continuous development of renewable energy technology, future worldwide electrical energy consumption will be provided by renewable energy sources. The fuel cell is a device capable of directly converting chemical energy into electrochemical energy, can provide required electric energy for external loads, and has the advantages of good performance, high efficiency, noiseless operation, high power-heat ratio, simple maintenance service, high operating efficiency of partial load, no pollution gas emission and the like in the working process. However, the solid oxide fuel cell is a high temperature fuel cell and the working conditions are relatively closed, which brings certain challenges to the analysis and dynamic control of the internal parameters of the solid oxide fuel cell.
The parameter analysis of the solid oxide fuel cell is complex, and is a nonlinear non-convex problem, and the traditional optimization method such as a linear programming method, a Newton-Lawson method and the like is difficult to obtain accurate solid oxide fuel cell parameters due to serious dependence on gradient information and initialization conditions. In this context, intelligent algorithms provide a more efficient solution to the solution of solid oxide fuel cell parameters, which does not require gradient information or specific initial conditions. If the intelligent algorithm is adopted to optimize the parameters of the solid oxide fuel cell, the cost and the operation time are greatly saved.
Hippocampus optimization algorithms (SHO) use population wisdom to search for optimal solutions in solution space by mimicking common and familiar social behaviors such as movement, predation, and reproduction of the hippocampus. The device has the advantages of simple structure, good development performance and the like, but the defects of low searching efficiency and premature convergence in practical application still need to be improved.
Disclosure of Invention
The invention aims to provide a parameter identification method of a solid oxide fuel cell, which is used for obtaining an optimal solution of the solid oxide fuel cell parameters, so that the theoretical value and the measured value of the output voltage of the solid oxide fuel cell are as close as possible, and the accurate identification of the solid oxide fuel cell parameters is realized, thereby facilitating the further performance prediction and fault analysis of the solid oxide fuel cell.
The technical problems of the invention are mainly solved by the following technical proposal:
a fuel cell parameter identification method based on improved Hippocampus optimization algorithm comprises
Collecting experimental values of the solid oxide fuel cell output voltage at a plurality of moments, and calculating a theoretical value of the solid oxide fuel cell output voltage at each moment;
selecting the slope of a tafel line, anode current density, cathode current density, ion resistance, constant factor, limiting current density and open-circuit voltage as constraint conditions of decision variable construction parameters, and constructing a solid oxide fuel cell optimization model by taking the minimum mean square error of a calculated value and an experimental measurement value of an output voltage as an objective function;
and solving the solid oxide fuel cell optimization model by taking the slope, anode current density, cathode current density, ion resistance, constant factor, limiting current density and open-circuit voltage of the Tafil line as variables to be solved through improving a Hippocampus optimization algorithm to obtain the slope, the optimized anode current density, the optimized cathode current density, the optimized ion resistance, the optimized constant factor, the optimized limiting current density and the optimized open-circuit voltage of the Tafil line.
In the fuel cell parameter identification method based on the improved hippocampal optimization algorithm, the theoretical value calculation formula of the output voltage is as follows:
Figure SMS_1
；
the measured output voltages of the solid oxide fuel cells at the various times are:
Figure SMS_2
，/>
Figure SMS_3
；
wherein ,
Figure SMS_4
for the slope of the Tafil line, +.>
Figure SMS_6
For anodic current density>
Figure SMS_8
For cathode current density, ">
Figure SMS_10
Is an ionic resistance>
Figure SMS_11
Is a constant factor, ++>
Figure SMS_13
For limiting current density, +.>
Figure SMS_15
Is an open circuit voltage>
Figure SMS_5
For the total number of cells in the solid oxide fuel cell stack, etc.>
Figure SMS_7
For load current density>
Figure SMS_9
For outputting voltage +.>
Figure SMS_12
Indicate->
Figure SMS_14
Measurement output voltage of solid oxide fuel cell at each instant +.>
Figure SMS_16
Indicating the number of moments.
In the above fuel cell parameter identification method based on the improved hippocampal optimization algorithm, the objective function is based on the following formula:
Figure SMS_17
；
wherein ,
Figure SMS_19
representing a minimization function +.>
Figure SMS_21
For the mean square error of the theoretical value and the experimental measurement value of the output voltage of the solid oxide fuel cell, +.>
Figure SMS_22
For the slope of the Tafil line, +.>
Figure SMS_23
For anodic current density>
Figure SMS_24
For the cathode current density,
Figure SMS_25
is an ionic resistance>
Figure SMS_26
Is a constant factor, ++>
Figure SMS_18
For limiting current density, +.>
Figure SMS_20
Is an open circuit voltage.
In the above-described fuel cell parameter identification method based on the improved hippocampal optimization algorithm,
Figure SMS_27
the calculation formula is as follows:
Figure SMS_28
；
wherein ：
Figure SMS_29
indicate->
Figure SMS_30
Theoretical value of solid oxide fuel cell output voltage at each instant +.>
Figure SMS_31
Indicate->
Figure SMS_32
Experimental measurement output voltage of solid oxide fuel cell at each instant +.>
Figure SMS_33
Indicating the number of moments.
In the fuel cell parameter identification method based on the improved hippocampal optimization algorithm, the constraint conditions of the parameters are as follows:
Figure SMS_34
；/>
wherein ,
Figure SMS_36
=0/>
Figure SMS_37
is the lower limit of the open circuit voltage, +.>
Figure SMS_39
=1.2/>
Figure SMS_41
Is the upper limit of the open circuit voltage, +.>
Figure SMS_44
=0/>
Figure SMS_46
For the lower limit of the tafel-line slope, < >>
Figure SMS_49
=1/>
Figure SMS_51
Is the upper limit of the slope of the tafel line, < >>
Figure SMS_53
=0/>
Figure SMS_55
As a lower limit of the current density of the anode,
Figure SMS_57
=100/>
Figure SMS_59
is the upper limit of the anode current density, +.>
Figure SMS_60
=0/>
Figure SMS_61
Is the lower limit of the cathode current density, +.>
Figure SMS_62
=100
Figure SMS_35
Is the upper limit of the cathode current density, +.>
Figure SMS_38
=0/>
Figure SMS_40
Is the lower limit of the constant factor, +.>
Figure SMS_42
=1/>
Figure SMS_43
Is the upper limit of the constant factor,
Figure SMS_45
=0/>
Figure SMS_47
is the lower limit of the limiting current density, +.>
Figure SMS_48
=10000/>
Figure SMS_50
Is the upper limit of the limiting current density, +.>
Figure SMS_52
=0
Figure SMS_54
Is the lower limit of ionic resistance, +.>
Figure SMS_56
=1/>
Figure SMS_58
Is the upper limit of ionic resistance.
The specific process for solving the improved Hippocampus optimization algorithm based on the fuel cell parameter identification method based on the improved Hippocampus optimization algorithm is as follows:
step 3.1, initializing a hippocampal optimization algorithm, wherein the specific process is as follows:
setting search space of sea horse according to constraint condition of parameter
Figure SMS_63
Storing the lower limits of the 7 parameters to be determined in a dimension-by-dimension manner +.>
Figure SMS_64
In which the upper limits of the 7 parameters to be determined are stored in dimension by dimension +.>
Figure SMS_65
In (I)>
Figure SMS_66
For the lower limit of the hippocampal search space, +.>
Figure SMS_67
An upper limit for the hippocampal search space;
setting the quantity of Hippocampus in the population as
Figure SMS_68
The maximum iteration number is->
Figure SMS_69
Action judging factor->
Figure SMS_70
Predation judgment factor->
Figure SMS_71
Number of reproducing individuals->
Figure SMS_72
The method comprises the steps of carrying out a first treatment on the surface of the Randomly generating a hippocampal population in a hippocampal activity interval, and setting the current iteration number +.>
Figure SMS_73
；
Step 3.2, the sea horse searches for hunting objects in two movement behaviors including spiral movement of the sea horse along with vortexes in the sea and Brownian movement of the sea horse along with sea waves,
step 3.3, selecting different routes for preying on prey objects according to Chebyshev distance between the sea horse and the optimal individual, wherein the specific steps are as follows:
calculating chebyshev distances between the remaining Hippocampus individuals and elite individuals
Figure SMS_74
Judging whether the distance satisfies +.>
Figure SMS_75
If so, the hippocampal individual surrounds the prey according to the formula; if not, the sea horse preys prey, wherein the chebyshev distance calculation formula is as follows:
Figure SMS_76
；
wherein ,
Figure SMS_78
is->
Figure SMS_79
Iterative->
Figure SMS_80
Chebyshev distance seen by the individual from the optimal individual,>
Figure SMS_81
first->
Figure SMS_82
Iterative->
Figure SMS_83
The location of individual hippocampus, the +.>
Figure SMS_84
First->
Figure SMS_77
Iterating the position of the optimal sea horse individual;
step 3.4, performing reproduction and mutation operations on the sea horse individuals, wherein the steps are as follows:
performing fitness sequencing on current hippocampal individuals according to an objective function in the solid oxide fuel cell optimization model, performing role allocation on the current hippocampal population, and then propagating to generate offspring hippocampal individuals;
role assignment is specifically defined before fitness ranking
Figure SMS_85
Is the parent Hippocampus ++>
Figure SMS_86
After definition of fitness row->
Figure SMS_87
Is mother generation Hippocampus +.>
Figure SMS_88
The formula is as follows:
Figure SMS_89
；/>
first, the
Figure SMS_90
The calculation formula of individual offspring hippocampal individuals is as follows:
Figure SMS_91
；
wherein ,
Figure SMS_94
for the parent hippocampal population, cryptocarya>
Figure SMS_96
For parent sea horse population, ->
Figure SMS_99
Indicate->
Figure SMS_102
Hippocampus population in ascending order of fitness for multiple iterations,/->
Figure SMS_104
For the number of Hippocampus populations, < >>
Figure SMS_106
For interval->
Figure SMS_107
Random number between->
Figure SMS_93
Indicate->
Figure SMS_95
The>
Figure SMS_97
Individual offspring hippocampal individuals, < >>
Figure SMS_98
Indicate->
Figure SMS_100
The>
Figure SMS_101
The individual male parent, the individual Hippocampus,
Figure SMS_103
indicate->
Figure SMS_105
The>
Figure SMS_92
Individuals with mother-generation hippocampus;
after the breeding process is finished, the fitness of the hippocampal population is recalculated, and the hunting of the natural enemies is simulated by carrying out position mutation on individuals with better fitness based on a variable-scale mutation strategy, wherein the variable-scale mutation strategy has the following update formula:
Figure SMS_108
；
wherein ,
Figure SMS_109
is->
Figure SMS_112
Updated +.>
Figure SMS_114
Individual hippocampal individuals,/->
Figure SMS_115
For the maximum number of iterations to be performed,
Figure SMS_116
to meet mathematical expectations->
Figure SMS_117
Variance is->
Figure SMS_118
Is a normal distribution of random numbers, wherein mathematical expectation +.>
Figure SMS_110
Is equal to zero, variance->
Figure SMS_111
Initial set to 1, then every time the history optimal solution changes, let +.>
Figure SMS_113
，
Step 3.5: repeating the steps 3.2-3.4 until
Figure SMS_119
Is greater than->
Figure SMS_120
And outputting an optimized open-circuit voltage, an optimized tafel line slope, an optimized anode current density, an optimized cathode current density, an optimized constant factor, an optimized limiting current density and an optimized ion resistance.
In the above fuel cell parameter identification method based on the improved hippocampal optimization algorithm, in step 3.1, the definition of the initialized hippocampal population is as follows:
Figure SMS_121
；
and satisfies the following:
Figure SMS_122
；
wherein ,
Figure SMS_134
indicate->
Figure SMS_136
The->
Figure SMS_138
Individual hippocampal individuals,/->
Figure SMS_139
Indicate->
Figure SMS_141
The->
Figure SMS_144
Open circuit voltage of individual solution vector, +.>
Figure SMS_146
Indicate->
Figure SMS_148
The->
Figure SMS_149
The tafel-line slope of the individual solution vectors,
Figure SMS_151
indicate->
Figure SMS_154
The->
Figure SMS_155
Anodic current density of individual solution vector, +.>
Figure SMS_156
Indicate->
Figure SMS_157
The->
Figure SMS_158
Cathode current density of individual solution vector, +.>
Figure SMS_123
Indicate->
Figure SMS_126
The->
Figure SMS_127
Constant factor of individual solution vector, +.>
Figure SMS_129
Indicate->
Figure SMS_132
The->
Figure SMS_133
Poles of individual solution vectorsCurrent limiting density->
Figure SMS_135
Indicate->
Figure SMS_137
The->
Figure SMS_140
Ion resistance of individual solution vectors, +.>
Figure SMS_142
Indicate->
Figure SMS_143
The->
Figure SMS_145
No. H of individual Hippocampus>
Figure SMS_147
Vector of dimensions>
Figure SMS_150
For the dimension of the solution, <' > for>
Figure SMS_152
Indicating +.>
Figure SMS_153
Lower bound of dimension solution vector parameters +.>
Figure SMS_124
Indicating +.>
Figure SMS_125
Upper limit of vector parameters of dimension solution, action judgment factor +.>
Figure SMS_128
And predatory judgment factor->
Figure SMS_130
All are interval +.>
Figure SMS_131
Constant of the same.
In the above fuel cell parameter identification method based on improved hippocampal optimization algorithm, in step 3.2, the fitness of all hippocampal individuals in the population is calculated according to the objective function of the solid oxide fuel cell optimization model, and the hippocampal individuals are ranked according to the fitness and the first is selected
Figure SMS_159
Elite individuals in the course of a second iteration +.>
Figure SMS_160
；
Setting an action factor for each individual Hippocampus
Figure SMS_161
And judging action factor->
Figure SMS_162
Whether or not to meet->
Figure SMS_163
If the sea wave is satisfied, the current sea horse individual moves in a novel spiral way along with the vortex in the sea, if the sea wave is not satisfied, the current sea horse individual moves in a Brownian way along with the sea wave,
the updated formula of the novel spiral motion of the sea horse individual along with the vortex in the sea is as follows:
Figure SMS_164
；
parameters (parameters)
Figure SMS_165
For the first shape adjustment factor, parameter +.>
Figure SMS_166
For the second form factor, parameter +.>
Figure SMS_167
For the third shape adjustment factor, the specific calculation formulas are as follows:
Figure SMS_168
；
the specific update formula of the Brownian motion of the sea horse individual along with sea waves is as follows:
Figure SMS_169
；
wherein ,
Figure SMS_171
is->
Figure SMS_172
Updated +.>
Figure SMS_174
Individual hippocampal individuals,/->
Figure SMS_176
Is->
Figure SMS_179
The +.f. before updating in the iterative process>
Figure SMS_181
Individual hippocampal individuals,/->
Figure SMS_183
To obey mathematical expectations +.>
Figure SMS_185
Variance is->
Figure SMS_187
Is used for the scaling factor of the gaussian distribution,
Figure SMS_190
is->
Figure SMS_192
Optimal individual in the course of the second iteration, +.>
Figure SMS_193
To round-down function->
Figure SMS_194
For the step length adjustment factor, for
Figure SMS_195
Random number between->
Figure SMS_196
The iteration times; />
Figure SMS_170
Is->
Figure SMS_173
Updated +.>
Figure SMS_175
The individual sea horse,
Figure SMS_177
is->
Figure SMS_178
The +.f. before updating in the iterative process>
Figure SMS_180
Individual hippocampal individuals,/->
Figure SMS_182
For interval->
Figure SMS_184
Random number between->
Figure SMS_186
Is a constant coefficient>
Figure SMS_188
For the Brownian motion random walk coefficient following normal distribution, +.>
Figure SMS_189
Is->
Figure SMS_191
Optimal individuals after spiral movement in the iterations.
In the fuel cell parameter identification method based on the improved hippocampal optimization algorithm, the position update formula of the hippocampal individual surrounding the hunting object is specifically as follows:
Figure SMS_197
；
the update formula of the prey position of the sea horse is specifically as follows:
Figure SMS_198
；
wherein ,
Figure SMS_199
for maximum number of iterations +.>
Figure SMS_200
For the number of iterations->
Figure SMS_201
Is->
Figure SMS_202
Elite individuals in the course of a second iteration, +.>
Figure SMS_203
For interval->
Figure SMS_204
Random number between->
Figure SMS_205
Is the number of iterations.
A fuel cell parameter identification system based on improved Hippocampus optimization algorithm comprises
A first module: configured to collect experimental values of solid oxide fuel cell output voltages at a plurality of times, and calculate theoretical values of the solid oxide fuel cell output voltages at each time;
a second module: the solid oxide fuel cell optimization model is configured to select the slope of the tafel line, the anode current density, the cathode current density, the ion resistance, the constant factor, the limiting current density and the open-circuit voltage as constraint conditions of decision variable construction parameters, and construct the solid oxide fuel cell optimization model by taking the minimum mean square error of the calculated value and the experimental measurement value of the output voltage as an objective function;
and a third module: the method is configured to take the slope, anode current density, cathode current density, ion resistance, constant factor, limiting current density and open-circuit voltage of the Tariff line as variables to be solved, and solve a solid oxide fuel cell optimization model by improving a Hippocampus optimization algorithm to obtain the slope, the optimized anode current density, the optimized cathode current density, the optimized ion resistance, the optimized constant factor, the optimized limiting current density and the optimized open-circuit voltage of the Tariff line.
Therefore, the invention has the following advantages:
(1) The novel Archimedes spiral is adopted to replace a spiral formula in the original algorithm to update the sea horse population, so that the action route of the sea horse is enriched, the spatial distribution of the sea horse is optimized, and the global optimizing capability of the algorithm is improved.
(2) The predation mode is judged according to the Chebyshev distance between the sea horse and the optimal individual, and an adaptive position updating strategy is adopted to improve the updating formula of predation and surrounding of prey, so that the local optimizing capability of the algorithm is improved, the algorithm is helped to locate the accurate position of the local optimal solution more quickly, and the accuracy of fuel cell parameter identification is improved.
(3) On the basis of the original algorithm, a variable-scale variation strategy is introduced to simulate the hunting of natural enemies of the sea horse, so that the survival characteristic of the sea horse is better matched, the algorithm can jump out of a local optimal solution faster during optimizing, and premature convergence of the algorithm due to the search of the local optimal solution is avoided.
(4) The sea horse optimization algorithm is improved so that the sea horse optimization algorithm can adapt to a more complex operation model.
Drawings
FIG. 1 is an overall flow chart of a fuel cell parameter identification method based on an improved hippocampal optimization algorithm of the present invention.
FIG. 2 is a flow chart of a fuel cell parameter identification method based on an improved hippocampal optimization algorithm.
Fig. 3 is a graph of theoretical output voltage-current versus measured output voltage-current for a solid oxide fuel cell of the present invention (where the graph represents calculated data and the circle represents measured data).
Fig. 4 is a graph of theoretical output power versus current versus measured for a solid oxide fuel cell of the present invention.
FIG. 5 is a graph of improved hippocampal optimization algorithm (New-SHO), particle Swarm Optimization (PSO), simulated annealing algorithm (SA), and hippocampal optimization algorithm (SHO) versus optimization model employed in simulation experiments of the present invention.
Detailed Description
The technical scheme of the invention is further specifically described below through examples and with reference to the accompanying drawings.
Examples:
the present invention will be further described in detail with reference to the following detailed description and the accompanying drawings.
A fuel cell parameter identification method based on an improved hippocampal optimization algorithm as shown in fig. 1, the method comprising the following steps:
step 1: by analyzing the working principle and the structural characteristics of the solid oxide fuel cell, a mathematical model of the solid oxide fuel cell is established, experimental values of the output voltage of the solid oxide fuel cell at a plurality of moments are collected, and theoretical values of the output voltage of the solid oxide fuel cell at each moment are calculated.
The anode reaction equation of the solid oxide fuel cell in step 1 is as follows:
Figure SMS_206
the cathode reaction equation of the solid oxide fuel cell in step 1 is as follows:
Figure SMS_207
the total chemical reaction equation of the solid oxide fuel cell described in step 1 is as follows:
Figure SMS_208
wherein ：
Figure SMS_209
under the action of a catalyst, the catalyst is decomposed into protons and electrons; />
Figure SMS_210
Combines with electrons and protons under the action of a catalyst to produce water oxygen. />
Figure SMS_211
Proton(s)>
Figure SMS_212
Is an electron.
The theoretical value calculation formula for calculating the output voltage of the solid oxide fuel cell at each moment in step 1 is as follows:
Figure SMS_213
wherein ：
Figure SMS_214
indicate->
Figure SMS_216
Output voltage of solid oxide fuel cell at each instant, +.>
Figure SMS_218
For the number of units in the solid oxide fuel cell stack, for example>
Figure SMS_220
Indicate->
Figure SMS_222
Open circuit voltage of solid oxide fuel cell at each instant, +.>
Figure SMS_223
Indicate->
Figure SMS_225
The activation voltage drop of the solid oxide fuel cell at a single instant,/->
Figure SMS_215
Indicate->
Figure SMS_217
Ohmic drop of solid oxide fuel cell at various moments,/->
Figure SMS_219
Indicate->
Figure SMS_221
Concentration drop of solid oxide fuel cell at each moment,/->
Figure SMS_224
=600 denotes the number of times.
The specific calculation formula after the theoretical value of the output voltage of the solid oxide fuel cell at each moment is brought into each voltage calculation formula in the step 1 is as follows:
Figure SMS_226
fixing at multiple moments as described in step 1The measured output voltage of a bulk oxide fuel cell is defined as:
Figure SMS_227
，/>
Figure SMS_228
；
wherein ,
Figure SMS_230
for the slope of the Tafil line, +.>
Figure SMS_231
For anodic current density>
Figure SMS_233
For cathode current density, ">
Figure SMS_235
Is an ionic resistance>
Figure SMS_236
Is a constant factor, ++>
Figure SMS_239
For limiting current density, +.>
Figure SMS_240
Is an open circuit voltage>
Figure SMS_229
For the total number of cells in the solid oxide fuel cell stack, etc.>
Figure SMS_232
For load current density>
Figure SMS_234
For outputting voltage +.>
Figure SMS_237
Indicate->
Figure SMS_238
Measurement output voltage of solid oxide fuel cell at each instant +.>
Figure SMS_241
Indicating the number of moments.
Step 2: and constructing an output voltage optimization target of the solid oxide fuel cell, and selecting the slope of a tafel line, the anode current density, the cathode current density, the ion resistance, the constant factor, the limiting current density and the open-circuit voltage as constraint conditions of decision variable construction parameters. And constructing a solid oxide fuel cell optimization model by taking the minimum mean square error of the calculated value of the output voltage and the experimental measurement value as an objective function.
The solid oxide fuel cell output voltage optimization target is constructed in the step 2, and the method specifically comprises the following steps:
Figure SMS_242
；
the calculation formula is as follows:
Figure SMS_243
wherein ,
Figure SMS_245
representing a minimization function +.>
Figure SMS_246
For the mean square error of the theoretical value and the experimental measurement value of the output voltage of the solid oxide fuel cell, +.>
Figure SMS_248
For the slope of the Tafil line, +.>
Figure SMS_249
For anodic current density>
Figure SMS_250
Is a cathodeThe current density is such that,
Figure SMS_251
is an ionic resistance>
Figure SMS_252
Is a constant factor, ++>
Figure SMS_244
For limiting current density, +.>
Figure SMS_247
Is an open circuit voltage.
The constraint conditions of the parameters in the step 2 are specifically as follows:
Figure SMS_253
；
wherein ,
Figure SMS_255
=0/>
Figure SMS_256
is the lower limit of the open circuit voltage, +.>
Figure SMS_258
=1.2/>
Figure SMS_260
Is the upper limit of the open circuit voltage, +.>
Figure SMS_262
=0/>
Figure SMS_264
For the lower limit of the tafel-line slope, < >>
Figure SMS_266
=1/>
Figure SMS_268
Is the upper limit of the slope of the tafel line, < >>
Figure SMS_270
=0/>
Figure SMS_272
Is the lower limit of the anode current density, +.>
Figure SMS_273
=100/>
Figure SMS_276
Is the upper limit of the anode current density, +.>
Figure SMS_278
=0/>
Figure SMS_280
Is the lower limit of the cathode current density, +.>
Figure SMS_281
=100/>
Figure SMS_254
Is the upper limit of the cathode current density, +.>
Figure SMS_257
=0/>
Figure SMS_259
Is the lower limit of the constant factor, +.>
Figure SMS_261
=1/>
Figure SMS_263
Is the upper limit of the constant factor, +.>
Figure SMS_265
=0
Figure SMS_267
Is the lower limit of the limiting current density, +.>
Figure SMS_269
=10000/>
Figure SMS_271
Is the upper limit of the limiting current density, +.>
Figure SMS_274
=0/>
Figure SMS_275
Is the lower limit of ionic resistance, +.>
Figure SMS_277
=1/>
Figure SMS_279
Is the upper limit of ionic resistance.
Step 3: and combining constraint conditions of an optimization target and parameters of the output voltage of the solid oxide fuel cell, taking the slope, the anode current density, the cathode current density, the ionic resistance, the constant factor, the limiting current density and the open-circuit voltage of the Tafil line as variables to be solved, and solving by improving a Hippocampus optimization algorithm to obtain the slope, the optimized anode current density, the optimized cathode current density, the optimized ionic resistance, the optimized constant factor, the optimized limiting current density and the optimized open-circuit voltage of the Tafil line.
As shown in fig. 2, the solution by using the modified hippocampal optimization algorithm is described in step 3, and the specific procedure is as follows:
step 3.1: the sea horse optimization algorithm is initialized, and the specific process is as follows:
setting search space of sea horse according to constraint condition of parameter
Figure SMS_282
The method comprises the steps of carrying out a first treatment on the surface of the The lower limit of the 7 parameters to be determined is stored in dimension by dimension +.>
Figure SMS_283
In which the upper limits of the 7 parameters to be determined are stored in dimension by dimension +.>
Figure SMS_284
In (a) and (b); the method comprises the following steps: />
Figure SMS_286
Is the lower limit of the open circuit voltage, +.>
Figure SMS_288
For the lower limit of the tafel-line slope, < >>
Figure SMS_291
Is the lower limit of the anode current density, +.>
Figure SMS_293
Is the lower limit of the cathode current density, +.>
Figure SMS_296
Is the lower limit of the constant factor, +.>
Figure SMS_297
Is the lower limit of the limiting current density, +.>
Figure SMS_298
Is the lower limit of ionic resistance; />
Figure SMS_285
Is the upper limit of the open circuit voltage, +.>
Figure SMS_287
Is the upper limit of the slope of the tafel line, < >>
Figure SMS_289
Is the upper limit of the anode current density, +.>
Figure SMS_290
Is the upper limit of the cathode current density, +.>
Figure SMS_292
Is the upper limit of the constant factor, +.>
Figure SMS_294
Is the upper limit of the limiting current density, +.>
Figure SMS_295
Is the upper limit of ionic resistance.
Setting the quantity of Hippocampus in the population as
Figure SMS_299
The maximum iteration number is->
Figure SMS_300
Action judging factor->
Figure SMS_301
Predation judgment factor->
Figure SMS_302
Number of reproducing individuals->
Figure SMS_303
；
Wherein: action judgment factor
Figure SMS_304
And predatory judgment factor->
Figure SMS_305
All are interval +.>
Figure SMS_306
Constant of the same.
Randomly generating a hippocampal population in a hippocampal activity interval, and setting the current iteration times
Figure SMS_307
；
Wherein the initialized hippocampal population is defined as follows:
Figure SMS_308
and satisfies the following:
Figure SMS_309
；
wherein ,
Figure SMS_320
indicate->
Figure SMS_322
The->
Figure SMS_323
Individual hippocampal individuals,/->
Figure SMS_325
Indicate->
Figure SMS_327
The->
Figure SMS_329
Open circuit voltage of individual solution vector, +.>
Figure SMS_331
Indicate->
Figure SMS_332
The->
Figure SMS_335
The tafel-line slope of the individual solution vectors,
Figure SMS_336
indicate->
Figure SMS_338
The->
Figure SMS_340
Anodic current density of individual solution vector, +.>
Figure SMS_342
Indicate->
Figure SMS_343
In the second iteration process
Figure SMS_345
Cathode current density of individual solution vector, +.>
Figure SMS_310
Indicate->
Figure SMS_313
The->
Figure SMS_315
Constant factor of individual solution vector, +.>
Figure SMS_318
Indicate->
Figure SMS_319
The->
Figure SMS_321
Limiting current density of individual solution vector, +.>
Figure SMS_324
Indicate->
Figure SMS_326
The->
Figure SMS_328
Ion resistance of individual solution vectors, +.>
Figure SMS_330
Indicate->
Figure SMS_333
The->
Figure SMS_334
No. H of individual Hippocampus>
Figure SMS_337
Vector of dimensions>
Figure SMS_339
For the dimension of the solution, <' > for>
Figure SMS_341
Indicating +.>
Figure SMS_344
Lower bound of dimension solution vector parameters +.>
Figure SMS_311
Indicating +.>
Figure SMS_312
Upper limit of vector parameters of dimension solution, action judgment factor +.>
Figure SMS_314
And predatory judgment factor->
Figure SMS_316
All are interval +.>
Figure SMS_317
Constant of the same.
Step 3.2: the hippocampus searches for prey in two motor activities; the two motor activities include spiral movement of the hippocampus with the vortexes in the sea and brownian movement of the hippocampus with the waves. The method comprises the following steps:
calculating the fitness of all the sea horse individuals in the population according to the objective function of the solid oxide fuel cell optimization model in the step 2, sequencing the sea horse individuals according to the fitness, and selecting the first sea horse individual
Figure SMS_346
Elite individuals in the course of a second iteration +.>
Figure SMS_347
。
Setting an action factor for each individual Hippocampus
Figure SMS_348
And judging action factor->
Figure SMS_349
Whether or not to meet->
Figure SMS_350
If the sea wave is satisfied, the current sea horse individual moves in a novel spiral mode along with the vortex in the sea, and if the sea wave is not satisfied, the current sea horse individual moves in a Brownian mode along with the sea wave.
The updated formula of the novel spiral motion of the sea horse individual along with the vortex in the sea in the step 3.2 is as follows:
Figure SMS_352
wherein ：/>
Figure SMS_353
Is->
Figure SMS_355
Updated +.>
Figure SMS_356
Individual hippocampal individuals,/->
Figure SMS_358
To obey mathematical expectations +.>
Figure SMS_359
Variance is->
Figure SMS_360
Is a scaling factor of the gaussian distribution, +.>
Figure SMS_351
Is->
Figure SMS_354
Optimal individual in the iterative process, parameter ∈ ->
Figure SMS_357
The specific calculation formula of the shape adjustment factor is as follows: />
Figure SMS_361
wherein ：
Figure SMS_362
to round-down function->
Figure SMS_363
Is a regulatory factor.
The specific updating formula of the brownian motion of the sea horse individuals along with sea waves in the step 3.2 is as follows:
Figure SMS_364
wherein ：
Figure SMS_365
is->
Figure SMS_366
Updated +.>
Figure SMS_367
Individual hippocampal individuals,/->
Figure SMS_368
For interval->
Figure SMS_369
Random number between->
Figure SMS_370
Is a constant coefficient>
Figure SMS_371
To follow the normal distribution of the Brownian motion random walk coefficient.
Step 3.3: the hippocampus selects different routes to predate the prey according to its chebyshev distance from the optimal individual. The method comprises the following steps:
calculating the cuts between the other Hippocampus individuals and elite individualsDistance from snow
Figure SMS_372
Judging whether the distance is satisfied
Figure SMS_373
If so, the hippocampal individual surrounds the prey according to the formula; if not, the hippocampus predates the prey. Wherein the chebyshev distance calculation formula is as follows:
Figure SMS_374
wherein ：
Figure SMS_375
is->
Figure SMS_376
Iterative->
Figure SMS_377
The chebyshev distance seen by the individual from the optimal individual.
Step 3.3, the position update formula of the hippocampal individual surrounding the prey is specifically as follows:
Figure SMS_378
wherein ：
Figure SMS_379
is the maximum number of iterations.
The updated formula of the position of the prey of the sea horse individual in the step 3.3 is specifically as follows:
Figure SMS_380
wherein ：
Figure SMS_381
for interval->
Figure SMS_382
Random number between
Step 3.4: the sea horse individuals perform breeding and mutation operations, specifically as follows:
step 3.4, the specific process of the propagation process of the Hippocampus individuals is that the current Hippocampus individuals are adaptively ordered according to the objective function in the solid oxide fuel cell optimization model in the step 2, the role of the current Hippocampus population is allocated, and then the offspring Hippocampus individuals are propagated;
role assignment is specifically defined before fitness ranking
Figure SMS_383
Is the parent Hippocampus ++>
Figure SMS_384
After definition of fitness row->
Figure SMS_385
Is mother generation Hippocampus +.>
Figure SMS_386
The formula is as follows: />
Figure SMS_387
wherein ：
Figure SMS_388
for the parent hippocampal population, cryptocarya>
Figure SMS_389
For parent sea horse population, ->
Figure SMS_390
Indicate->
Figure SMS_391
Hippocampus population in ascending order of fitness for multiple iterations,/->
Figure SMS_392
Is the sea horse population quantity.
Step 3.4 the first
Figure SMS_393
The calculation formula of individual offspring hippocampal individuals is as follows:
Figure SMS_394
wherein ：
Figure SMS_396
for interval->
Figure SMS_397
Random number between->
Figure SMS_398
Indicate->
Figure SMS_400
The>
Figure SMS_403
Individual offspring hippocampal individuals, < >>
Figure SMS_404
Indicate->
Figure SMS_405
The>
Figure SMS_395
Individual male hippocampal individuals, < >>
Figure SMS_399
Indicate->
Figure SMS_401
The>
Figure SMS_402
Personal motherIndividuals with Hippocampus.
And (3) after the breeding process is finished, recalculating the fitness of the hippocampal population, introducing a variable-scale mutation strategy, and carrying out position mutation on individuals with better fitness to simulate natural enemies to kill.
The variable scale mutation strategy updating formula in the step 3.4 is as follows:
Figure SMS_406
wherein ：
Figure SMS_408
is->
Figure SMS_410
Updated +.>
Figure SMS_412
Individual hippocampal individuals,/->
Figure SMS_413
For the maximum number of iterations to be performed,
Figure SMS_414
to meet mathematical expectations->
Figure SMS_415
Variance is->
Figure SMS_416
Is a normal distribution of random numbers, wherein mathematical expectation +.>
Figure SMS_407
Is equal to zero, variance->
Figure SMS_409
Initial set to 1, then every time the history optimal solution changes, let +.>
Figure SMS_411
。
And 3, step 3.5: repeating the steps 3.2-3.4 until
Figure SMS_417
Is greater than->
Figure SMS_418
And outputting an optimized open-circuit voltage, an optimized tafel line slope, an optimized anode current density, an optimized cathode current density, an optimized constant factor, an optimized limiting current density and an optimized ion resistance.
The solid oxide fuel cell parameter identification method of the invention is analyzed by a simulation experiment.
Solid oxide fuel cell in simulation experiment at 1073
Figure SMS_419
Carrying in relevant parameters, solving the solid oxide fuel cell model by adopting a simulated annealing algorithm (SA), a particle swarm algorithm (PSO), a hippocampal optimization algorithm (SHO) and a modified hippocampal optimization algorithm (New-SHO), wherein the optimal parameters and RMSE solved by each algorithm are shown in table 1:
table 1: results of each algorithm on solid oxide fuel cell model parameters
Figure SMS_420
As can be seen from table 1: compared with other algorithms, the improved hippocampal optimization algorithm obtains the minimum value of RMSE, and compared with the improved hippocampal optimization algorithm, the improved hippocampal optimization algorithm has greatly optimized RMSE. The result shows that the New-SHO algorithm has better optimizing effect and higher robustness, and the improved Hippocampus optimization algorithm has more remarkable advantages for optimizing the parameters of the solid oxide fuel cell compared with other algorithms.
From fig. 3 and fig. 4, it can be known that the simulation result of the parameter of the solid oxide fuel cell improved by the hippocampal optimization algorithm is very close to the experimental result, which indicates that the New-SHO algorithm has better optimizing capability, the simulation curve is smoother and basically has no fluctuation, and the superiority of the New-SHO algorithm is further verified.
As can be seen from fig. 5, the improved hippocampal optimization algorithm (New-SHO) cloud has stronger capability than other algorithms, so that the global optimizing and local optimizing capabilities of the original algorithm are effectively improved, the solving efficiency of the algorithm is improved, and the identification parameters of the better solid oxide fuel cell are obtained.
The embodiment also provides a fuel cell parameter identification system based on the improved Hippocampus optimization algorithm, which comprises
A first module: configured to collect experimental values of solid oxide fuel cell output voltages at a plurality of times, and calculate theoretical values of the solid oxide fuel cell output voltages at each time;
a second module: the solid oxide fuel cell optimization model is configured to select the slope of the tafel line, the anode current density, the cathode current density, the ion resistance, the constant factor, the limiting current density and the open-circuit voltage as constraint conditions of decision variable construction parameters, and construct the solid oxide fuel cell optimization model by taking the minimum mean square error of the calculated value and the experimental measurement value of the output voltage as an objective function;
and a third module: the method is configured to take the slope, anode current density, cathode current density, ion resistance, constant factor, limiting current density and open-circuit voltage of the Tariff line as variables to be solved, and solve a solid oxide fuel cell optimization model by improving a Hippocampus optimization algorithm to obtain the slope, the optimized anode current density, the optimized cathode current density, the optimized ion resistance, the optimized constant factor, the optimized limiting current density and the optimized open-circuit voltage of the Tariff line.
The specific embodiments described herein are offered by way of example only to illustrate the spirit of the invention. Those skilled in the art may make various modifications or additions to the described embodiments or substitutions thereof without departing from the spirit of the invention or exceeding the scope of the invention as defined in the accompanying claims.

Claims (5)

1. A fuel cell parameter identification method based on improved Hippocampus optimization algorithm is characterized by comprising the following steps of
Collecting experimental values of the solid oxide fuel cell output voltage at a plurality of moments, and calculating a theoretical value of the solid oxide fuel cell output voltage at each moment;
selecting the slope of a tafel line, anode current density, cathode current density, ion resistance, constant factor, limiting current density and open-circuit voltage as constraint conditions of decision variable construction parameters, and constructing a solid oxide fuel cell optimization model by taking the minimum mean square error of a calculated value and an experimental measurement value of an output voltage as an objective function;
the slope of the Tafil line, the anode current density, the cathode current density, the ion resistance, the constant factor, the limiting current density and the open-circuit voltage are used as variables to be solved, and the solid oxide fuel cell optimization model is solved through an improved hippocampus optimization algorithm, so that the slope of the optimized Tafil line, the optimized anode current density, the optimized cathode current density, the optimized ion resistance, the optimized constant factor, the optimized limiting current density and the optimized open-circuit voltage are obtained;
the theoretical value of the output voltage is calculated as follows:
Figure QLYQS_1
；
the measured output voltages of the solid oxide fuel cells at the various times are:
Figure QLYQS_2
，/>
Figure QLYQS_3
；
wherein ,
Figure QLYQS_5
for the slope of the Tafil line, +.>
Figure QLYQS_7
For anodic current density>
Figure QLYQS_9
For cathode current density, ">
Figure QLYQS_12
In the form of an ionic resistance,
Figure QLYQS_13
is a constant factor, ++>
Figure QLYQS_15
For limiting current density, +.>
Figure QLYQS_16
Is an open circuit voltage>
Figure QLYQS_4
For the total number of cells in the solid oxide fuel cell stack, etc.>
Figure QLYQS_6
For load current density>
Figure QLYQS_8
For outputting voltage +.>
Figure QLYQS_10
Indicate->
Figure QLYQS_11
Measurement output voltage of solid oxide fuel cell at each instant +.>
Figure QLYQS_14
Indicating the number of moments;
wherein the objective function is based on the following formula:
Figure QLYQS_17
；
wherein ,
Figure QLYQS_19
representing a minimization function +.>
Figure QLYQS_20
For the mean square error of the theoretical value and the experimental measurement value of the output voltage of the solid oxide fuel cell, +.>
Figure QLYQS_22
For the slope of the Tafil line, +.>
Figure QLYQS_23
For anodic current density>
Figure QLYQS_24
For cathode current density, ">
Figure QLYQS_25
Is an ionic resistance>
Figure QLYQS_26
Is a constant factor, ++>
Figure QLYQS_18
For limiting current density, +.>
Figure QLYQS_21
Is an open circuit voltage;
wherein ,
Figure QLYQS_27
the calculation formula is as follows:
Figure QLYQS_28
；
wherein ：
Figure QLYQS_29
indicate->
Figure QLYQS_30
Theoretical value of solid oxide fuel cell output voltage at each instant +.>
Figure QLYQS_31
Indicate->
Figure QLYQS_32
Experimental measurement output voltage of solid oxide fuel cell at each instant +.>
Figure QLYQS_33
Indicating the number of moments;
the constraint conditions of the parameters are specifically as follows:
Figure QLYQS_34
；
wherein ,
Figure QLYQS_36
=0/>
Figure QLYQS_38
is the lower limit of the open circuit voltage, +.>
Figure QLYQS_39
=1.2/>
Figure QLYQS_42
Is the upper limit of the open circuit voltage, +.>
Figure QLYQS_44
=0/>
Figure QLYQS_46
For the lower limit of the tafel-line slope, < >>
Figure QLYQS_47
=1/>
Figure QLYQS_49
Is the upper limit of the slope of the tafel line, < >>
Figure QLYQS_50
=0/>
Figure QLYQS_52
Is the lower limit of the anode current density, +.>
Figure QLYQS_53
=100
Figure QLYQS_56
Is the upper limit of the anode current density, +.>
Figure QLYQS_57
=0/>
Figure QLYQS_60
Is the lower limit of the cathode current density, +.>
Figure QLYQS_62
=100/>
Figure QLYQS_35
Is the upper limit of the cathode current density, +.>
Figure QLYQS_37
=0/>
Figure QLYQS_40
Is the lower limit of the constant factor, +.>
Figure QLYQS_41
=1/>
Figure QLYQS_43
Is the upper limit of the constant factor, +.>
Figure QLYQS_45
=0
Figure QLYQS_48
Is the lower limit of the limiting current density, +.>
Figure QLYQS_51
=10000/>
Figure QLYQS_54
Is the upper limit of the limiting current density, +.>
Figure QLYQS_55
=0
Figure QLYQS_58
Is the lower limit of ionic resistance, +.>
Figure QLYQS_59
=1/>
Figure QLYQS_61
Is the upper limit of ionic resistance;
the specific process for solving by improving the hippocampal optimization algorithm is as follows:
step 3.1, initializing a hippocampal optimization algorithm, wherein the specific process is as follows:
setting search space of sea horse according to constraint condition of parameter
Figure QLYQS_63
Storing the lower limit of the 7 parameters to be determined into the memory in a dimension-by-dimension manner
Figure QLYQS_64
In which the upper limits of the 7 parameters to be determined are stored in dimension by dimension +.>
Figure QLYQS_65
In (I)>
Figure QLYQS_66
For the lower limit of the hippocampal search space, +.>
Figure QLYQS_67
An upper limit for the hippocampal search space;
setting the quantity of Hippocampus in the population as
Figure QLYQS_68
The maximum iteration number is->
Figure QLYQS_69
Action judging factor->
Figure QLYQS_70
Predation judgment factor->
Figure QLYQS_71
Number of reproducing individuals->
Figure QLYQS_72
The method comprises the steps of carrying out a first treatment on the surface of the Randomly generating a hippocampal population in a hippocampal activity interval, and setting the current iteration number +.>
Figure QLYQS_73
；
Step 3.2, the sea horse searches for hunting objects in two movement behaviors including spiral movement of the sea horse along with vortexes in the sea and Brownian movement of the sea horse along with sea waves,
step 3.3, selecting different routes for preying on prey objects according to Chebyshev distance between the sea horse and the optimal individual, wherein the specific steps are as follows:
calculating chebyshev distances between the remaining Hippocampus individuals and elite individuals
Figure QLYQS_74
JudgingWhether or not the distance is satisfied->
Figure QLYQS_75
If so, the hippocampal individual surrounds the prey according to the formula; if not, the sea horse preys prey, wherein the chebyshev distance calculation formula is as follows:
Figure QLYQS_76
；
wherein ,
Figure QLYQS_78
is->
Figure QLYQS_79
Iterative->
Figure QLYQS_80
Chebyshev distance seen by the individual from the optimal individual,>
Figure QLYQS_81
first->
Figure QLYQS_82
Iterative->
Figure QLYQS_83
The location of individual hippocampus, the +.>
Figure QLYQS_84
First->
Figure QLYQS_77
Iterating the position of the optimal sea horse individual;
step 3.4, performing reproduction and mutation operations on the sea horse individuals, wherein the steps are as follows:
performing fitness sequencing on current hippocampal individuals according to an objective function in the solid oxide fuel cell optimization model, performing role allocation on the current hippocampal population, and then propagating to generate offspring hippocampal individuals;
role assignment is specifically defined before fitness ranking
Figure QLYQS_85
Is the parent Hippocampus ++>
Figure QLYQS_86
After defining fitness measures
Figure QLYQS_87
Is mother generation Hippocampus +.>
Figure QLYQS_88
The formula is as follows:
Figure QLYQS_89
；
first, the
Figure QLYQS_90
The calculation formula of individual offspring hippocampal individuals is as follows:
Figure QLYQS_91
；
wherein ,
Figure QLYQS_95
for the parent hippocampal population, cryptocarya>
Figure QLYQS_96
For parent sea horse population, ->
Figure QLYQS_98
Indicate->
Figure QLYQS_100
Hippocampus population in ascending order of fitness for multiple iterations,/->
Figure QLYQS_103
For the number of Hippocampus populations, < >>
Figure QLYQS_105
For interval->
Figure QLYQS_106
Random number between->
Figure QLYQS_92
Indicate->
Figure QLYQS_94
The>
Figure QLYQS_97
Individual offspring hippocampal individuals, < >>
Figure QLYQS_99
Indicate->
Figure QLYQS_101
The>
Figure QLYQS_102
Individual male hippocampal individuals, < >>
Figure QLYQS_104
Indicate->
Figure QLYQS_107
The>
Figure QLYQS_93
Individuals with mother-generation hippocampus;
after the breeding process is finished, the fitness of the hippocampal population is recalculated, and the hunting of the natural enemies is simulated by position mutation on individuals with better fitness based on a variable-scale mutation strategy, wherein the variable-scale mutation strategy has the following update formula:
Figure QLYQS_108
；
wherein ,
Figure QLYQS_109
is->
Figure QLYQS_112
Updated +.>
Figure QLYQS_114
Individual hippocampal individuals,/->
Figure QLYQS_115
For maximum number of iterations +.>
Figure QLYQS_116
To meet mathematical expectations->
Figure QLYQS_117
Variance is->
Figure QLYQS_118
Is a normal distribution of random numbers, wherein mathematical expectation +.>
Figure QLYQS_110
Is equal to zero, variance->
Figure QLYQS_111
Initial set to 1, then every time the history optimal solution changes, let +.>
Figure QLYQS_113
，
Step 3.5: repeating the steps 3.2-3.4 until
Figure QLYQS_119
Is greater than->
Figure QLYQS_120
And outputting an optimized open-circuit voltage, an optimized tafel line slope, an optimized anode current density, an optimized cathode current density, an optimized constant factor, an optimized limiting current density and an optimized ion resistance.
2. The method for identifying parameters of a fuel cell based on an improved hippocampal optimization algorithm according to claim 1, wherein in step 3.1, the definition of the initialized hippocampal population is as follows:
Figure QLYQS_121
；
and satisfies the following:
Figure QLYQS_122
；
wherein ,
Figure QLYQS_134
indicate->
Figure QLYQS_136
The->
Figure QLYQS_137
Individual hippocampal individuals,/->
Figure QLYQS_139
Indicate->
Figure QLYQS_141
The->
Figure QLYQS_142
Open circuit voltage of individual solution vector, +.>
Figure QLYQS_144
Indicate->
Figure QLYQS_146
The->
Figure QLYQS_148
The tafel-line slope of individual solution vectors,/>
Figure QLYQS_150
Indicate->
Figure QLYQS_152
The->
Figure QLYQS_155
Anodic current density of individual solution vector, +.>
Figure QLYQS_156
Indicate->
Figure QLYQS_157
The->
Figure QLYQS_158
Cathode current density of individual solution vector, +.>
Figure QLYQS_123
Indicate->
Figure QLYQS_126
The->
Figure QLYQS_127
Constant factor of individual solution vector, +.>
Figure QLYQS_130
Indicate->
Figure QLYQS_132
The->
Figure QLYQS_133
Limiting current density of individual solution vector, +.>
Figure QLYQS_135
Indicate->
Figure QLYQS_138
The->
Figure QLYQS_140
Ion resistance of individual solution vectors, +.>
Figure QLYQS_143
Indicate->
Figure QLYQS_145
The->
Figure QLYQS_147
No. H of individual Hippocampus>
Figure QLYQS_149
Vector of dimensions>
Figure QLYQS_151
For the dimension of the solution, <' > for>
Figure QLYQS_153
Indicating +.>
Figure QLYQS_154
Lower bound of dimension solution vector parameters +.>
Figure QLYQS_124
Indicating +.>
Figure QLYQS_125
Upper limit of vector parameters of dimension solution, action judgment factor +.>
Figure QLYQS_128
And predatory judgment factor->
Figure QLYQS_129
All are interval +.>
Figure QLYQS_131
Constant of the same.
3. The method for identifying parameters of fuel cell based on improved hippocampal optimization algorithm as recited in claim 1, wherein in step 3.2, the fitness of all the hippocampal individuals in the population is calculated according to the objective function of the solid oxide fuel cell optimization model, and the hippocampal individuals are ranked according to the fitness and the first is selected
Figure QLYQS_159
Elite individuals in the course of a second iteration +.>
Figure QLYQS_160
；
Setting an action factor for each individual Hippocampus
Figure QLYQS_161
And judging action factor->
Figure QLYQS_162
Whether or not to meet->
Figure QLYQS_163
If the sea wave is satisfied, the current sea horse individual moves in a novel spiral way along with the vortex in the sea, if the sea wave is not satisfied, the current sea horse individual moves in a Brownian way along with the sea wave,
the updated formula of the novel spiral motion of the sea horse individual along with the vortex in the sea is as follows:
Figure QLYQS_164
；
parameters (parameters)
Figure QLYQS_165
For the first shape adjustment factor, parameter +.>
Figure QLYQS_166
For the second form factor, parameter +.>
Figure QLYQS_167
For the third shape adjustment factor, the specific calculation formulas are as follows:
Figure QLYQS_168
；
the specific update formula of the Brownian motion of the sea horse individual along with sea waves is as follows:
Figure QLYQS_169
；
wherein ,
Figure QLYQS_171
is->
Figure QLYQS_172
Updated +.>
Figure QLYQS_174
Individual hippocampal individuals,/->
Figure QLYQS_177
Is->
Figure QLYQS_178
The +.f. before updating in the iterative process>
Figure QLYQS_180
Individual hippocampal individuals,/->
Figure QLYQS_182
To obey mathematical expectations +.>
Figure QLYQS_184
Variance is->
Figure QLYQS_187
Is a scaling factor of the gaussian distribution, +.>
Figure QLYQS_188
Is->
Figure QLYQS_190
Optimal individual in the course of the second iteration, +.>
Figure QLYQS_193
To round-down function->
Figure QLYQS_194
For step size adjustment factor +.>
Figure QLYQS_195
Random number between->
Figure QLYQS_196
The iteration times; />
Figure QLYQS_170
Is->
Figure QLYQS_173
Updated +.>
Figure QLYQS_175
Individual hippocampal individuals,/->
Figure QLYQS_176
Is->
Figure QLYQS_179
The +.f. before updating in the iterative process>
Figure QLYQS_181
Individual hippocampal individuals,/->
Figure QLYQS_183
For interval->
Figure QLYQS_185
Random number between->
Figure QLYQS_186
Is a constant coefficient of the number of the pieces of the material,
Figure QLYQS_189
for the Brownian motion random walk coefficient following normal distribution, +.>
Figure QLYQS_191
Is->
Figure QLYQS_192
Optimal individuals after spiral movement in the iterations.
4. The method for identifying parameters of a fuel cell based on an improved hippocampal optimization algorithm according to claim 1, wherein the location update formula of the surrounding prey of the hippocampal individual is specifically as follows:
Figure QLYQS_197
；
the update formula of the prey position of the sea horse is specifically as follows:
Figure QLYQS_198
；
wherein ,
Figure QLYQS_199
for maximum number of iterations +.>
Figure QLYQS_200
For the number of iterations->
Figure QLYQS_201
Is->
Figure QLYQS_202
Elite individuals in the course of a second iteration, +.>
Figure QLYQS_203
For interval->
Figure QLYQS_204
Random number between->
Figure QLYQS_205
Is the number of iterations. />
5. A system adapted for use in the improved hippocampal optimization algorithm based fuel cell parameter identification method of claim 1, comprising
A first module: configured to collect experimental values of solid oxide fuel cell output voltages at a plurality of times, and calculate theoretical values of the solid oxide fuel cell output voltages at each time;
a second module: the solid oxide fuel cell optimization model is configured to select the slope of the tafel line, the anode current density, the cathode current density, the ion resistance, the constant factor, the limiting current density and the open-circuit voltage as constraint conditions of decision variable construction parameters, and construct the solid oxide fuel cell optimization model by taking the minimum mean square error of the calculated value and the experimental measurement value of the output voltage as an objective function;
and a third module: the method is configured to take the slope, anode current density, cathode current density, ion resistance, constant factor, limiting current density and open-circuit voltage of the Tariff line as variables to be solved, and solve a solid oxide fuel cell optimization model by improving a Hippocampus optimization algorithm to obtain the slope, the optimized anode current density, the optimized cathode current density, the optimized ion resistance, the optimized constant factor, the optimized limiting current density and the optimized open-circuit voltage of the Tariff line.
CN202310231558.XA 2023-03-13 2023-03-13 Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm Active CN115966739B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310231558.XA CN115966739B (en) 2023-03-13 2023-03-13 Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310231558.XA CN115966739B (en) 2023-03-13 2023-03-13 Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm

Publications (2)

Publication Number Publication Date
CN115966739A CN115966739A (en) 2023-04-14
CN115966739B true CN115966739B (en) 2023-06-02

Family

ID=85888669

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310231558.XA Active CN115966739B (en) 2023-03-13 2023-03-13 Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm

Country Status (1)

Country Link
CN (1) CN115966739B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107122827A (en) * 2017-04-28 2017-09-01 安徽工程大学 A kind of RatSLAM algorithms based on DGSOM neutral nets
CN108372857A (en) * 2017-01-31 2018-08-07 通用汽车环球科技运作有限责任公司 Autonomous driving system is occurred and the effective context aware of episodic memory review progress by event
CN111506856A (en) * 2020-03-10 2020-08-07 燕山大学 Photovoltaic cell parameter identification method based on improved Harris eagle optimization algorithm
CN112801163A (en) * 2021-01-22 2021-05-14 安徽大学 Multi-target feature selection method of mouse model hippocampal biomarker based on dynamic graph structure
CN114492087A (en) * 2022-04-02 2022-05-13 国网浙江省电力有限公司电力科学研究院 Fault diagnosis method and device for proton exchange membrane fuel cell of hydrogen energy storage power station
CN114976129A (en) * 2022-05-13 2022-08-30 南京工业大学 Application of novel crow search algorithm in proton exchange membrane fuel cell model parameter identification
CN115392139A (en) * 2022-10-27 2022-11-25 西北工业大学 Fuel cell parameter identification method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108372857A (en) * 2017-01-31 2018-08-07 通用汽车环球科技运作有限责任公司 Autonomous driving system is occurred and the effective context aware of episodic memory review progress by event
CN107122827A (en) * 2017-04-28 2017-09-01 安徽工程大学 A kind of RatSLAM algorithms based on DGSOM neutral nets
CN111506856A (en) * 2020-03-10 2020-08-07 燕山大学 Photovoltaic cell parameter identification method based on improved Harris eagle optimization algorithm
CN112801163A (en) * 2021-01-22 2021-05-14 安徽大学 Multi-target feature selection method of mouse model hippocampal biomarker based on dynamic graph structure
CN114492087A (en) * 2022-04-02 2022-05-13 国网浙江省电力有限公司电力科学研究院 Fault diagnosis method and device for proton exchange membrane fuel cell of hydrogen energy storage power station
CN114976129A (en) * 2022-05-13 2022-08-30 南京工业大学 Application of novel crow search algorithm in proton exchange membrane fuel cell model parameter identification
CN115392139A (en) * 2022-10-27 2022-11-25 西北工业大学 Fuel cell parameter identification method

Also Published As

Publication number Publication date
CN115966739A (en) 2023-04-14

Similar Documents

Publication Publication Date Title
Priya et al. A comprehensive review on parameter estimation techniques for Proton Exchange Membrane fuel cell modelling
Rezaie et al. Model parameters estimation of the proton exchange membrane fuel cell by a Modified Golden Jackal Optimization
Ismaeel et al. Gradient-based optimizer for parameter extraction in photovoltaic models
Özdemir Optimal parameter estimation of polymer electrolyte membrane fuel cells model with chaos embedded particle swarm optimization
Bigdeli Optimal management of hybrid PV/fuel cell/battery power system: A comparison of optimal hybrid approaches
Niu et al. A biogeography-based optimization algorithm with mutation strategies for model parameter estimation of solar and fuel cells
Shi et al. Multi-objective optimal design of hybrid renewable energy systems using preference-inspired coevolutionary approach
Wang et al. An optimal configuration for hybrid SOFC, gas turbine, and Proton Exchange Membrane Electrolyzer using a developed Aquila Optimizer
Wang et al. Application of co-evolution RNA genetic algorithm for obtaining optimal parameters of SOFC model
CN110210071B (en) All-vanadium redox flow battery energy storage system operation optimization method based on genetic algorithm
Isa et al. PEM fuel cell model parameters optimization using modified particle swarm optimization algorithm
Yakout et al. Comparison among different recent metaheuristic algorithms for parameters estimation of solid oxide fuel cell: Steady-state and dynamic models
Rezk et al. A novel strategy based on recent equilibrium optimizer to enhance the performance of PEM fuel cell system through optimized fuzzy logic MPPT
CN115392139B (en) Fuel cell parameter identification method
Beshr et al. Novel optimization technique of isolated microgrid with hydrogen energy storage
Zhang et al. Application of coRNA-GA based RBF-NN to model proton exchange membrane fuel cells
Wang et al. The utilization of adaptive African vulture optimizer for optimal parameter identification of SOFC
Fathy et al. Proton exchange membrane fuel cell model parameters identification using Chaotically based-bonobo optimizer
Wai et al. Installed capacity selection of hybrid energy generation system via improved particle‐swarm‐optimisation
CN110688762A (en) Novel solid oxide fuel cell stack model construction method
Zhou et al. Improved fish migration optimization method to identify PEMFC parameters
CN115966739B (en) Fuel cell parameter identification method and system based on improved Hippocampus optimization algorithm
Erlin et al. Parameter derivation of a proton exchange membrane fuel cell based on coevolutionary ribonucleic acid genetic algorithm
Jabbary et al. Artificial intelligence-assisted optimization and multiphase analysis of polygon PEM fuel cells
Celtek Estimation of PEMFC design parameters with social learning-based optimization

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