Solid oxide fuel cell parameter identification method, computer equipment and computer readable storage medium
By combining the mountain gazelle optimizer and electrochemical model, the SOFC parameter recognition technology is optimized, and the defects in SOFC parameter recognition accuracy and speed are solved, and the accurate identification of SOFC electrochemical model parameters is achieved, which improves the accuracy and stability of the recognition .
Patent Information
- Application Number
- CN202411100332.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-12
- Publication Date
- 2024-12-03
AI Technical Summary
The existing solid oxide fuel cell (SOFC) parameter identification technology has defects in accuracy and speed, and has failed to effectively consider the impact of insufficient measurement data and interfere with data on parameter identification, resulting in inaccurate and fast identification.
The mountain gazelle optimizer is used to combine electrochemical models, and the voltage loss expression and objective function are constructed, and the unknown parameter set is optimized using the iterative mechanism of the mountain gazelle optimizer until the convergence conditions are met, and the recognition results are output.
It realizes accurate identification of unknown parameters of SOFC electrochemical model, is simple to operate, low cost, has stronger global search capabilities, and improves the accuracy, speed, accuracy and stability of parameter recognition.
Smart Images

Figure CN119064787A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the technical field of power system optimization, and in particular, relates to a solid oxide fuel cell parameter identification method, a computer device, and a computer-readable storage medium. Background Art
[0002] Solid oxide fuel cell (SOFC) has the advantages of high power generation efficiency, strong fuel adaptability, and high-temperature waste heat recovery. It has broad application prospects in large-scale power generation, distributed power generation and combined heat and power, transportation, peak load storage and other fields, and is the most cutting-edge fuel cell technology. It is worth noting that the accuracy of the SOFC model mainly depends on the identification of several unknown parameters in the model. However, current parameter identification technologies are too idealistic and do not take into account the impact of insufficient measurement data and interference data on parameter identification. Therefore, how to accurately and quickly identify SOFC parameters has become one of the most difficult problems hindering its development.
[0003] The preceding description is intended to provide general background information and does not necessarily constitute prior art. Summary of the invention
[0004] Based on this, it is necessary to address the above problems and propose a solid oxide fuel cell parameter identification method, computer equipment and computer-readable storage medium, which can quickly and accurately identify the parameters of the solid oxide fuel cell.
[0005] The present application solves the technical problem by adopting the following technical solutions:
[0006] The present application provides a solid oxide fuel cell parameter identification method, comprising the following steps: constructing an electrochemical model of a solid oxide fuel cell, substituting the battery information of the solid oxide fuel cell into the electrochemical model and then determining an unknown parameter set and an objective function; substituting the unknown parameter set as an optimization target into a pre-constructed mountain gazelle optimizer for iteration; after each iteration, obtaining the iteration result output by the mountain gazelle optimizer and substituting it into the objective function to calculate and obtain an identification result; repeating the iteration until the mountain gazelle optimizer meets the convergence condition, screening all identification results according to the optimal condition, taking the identification result that meets the optimal condition as the electrochemical model parameter and outputting it.
[0007] In an optional embodiment of the present application, the battery information of the solid oxide fuel cell is substituted into the electrochemical model to determine the unknown parameter set, including: obtaining the load current density in the battery information to construct a voltage loss expression; substituting the voltage loss expression into the first output voltage-current characteristic equation of the electrochemical model to obtain the second output voltage-current characteristic equation; and determining the unknown parameter set based on the second output voltage-current characteristic equation.
[0008] In an optional embodiment of the present application, the voltage loss expression includes an activation voltage loss expression, an ohmic voltage loss expression, and a concentration voltage loss expression; constructing the voltage loss expression includes: constructing the activation voltage loss expression according to the load current density:
[0009]
[0010] V act represents the activation voltage loss, A represents the slope of the Tafel line, I load Represents the load current density, I0 represents the exchange current density of the simplified battery model; the ohmic voltage loss expression is constructed based on the load current density:
[0011] V ohm =I load R ohm ;
[0012] V ohm represents the ohmic voltage loss, R ohm represents the ionic resistance; the load current density constructs the concentration voltage loss expression:
[0013]
[0014] V con represents the concentration voltage loss, B represents the first constant parameter, I L represents the limiting current density.
[0015] In an optional embodiment of the present application, the first output voltage-current characteristic equation of the electrochemical model is expressed as:
[0016] V c =N cell (E o -V act -V ohm -V con );
[0017] V c Expressed as output voltage, included in the battery information; E o Expressed as solid oxide fuel cell open circuit voltage; N cell It is represented by the number of series-connected cells in the solid oxide fuel cell stack; Substituting the activation voltage loss expression, the ohmic voltage loss expression and the concentration voltage loss expression into the first output voltage-current characteristic equation, the second output voltage-current characteristic equation can be obtained, which is expressed as:
[0018]
[0019] Determine the unknown parameter set according to the second output voltage-current characteristic equation, including: converting the solid oxide fuel cell open circuit voltage E o , the slope of the Tafel line A, the exchange current density I0 of the simplified battery model, the first constant parameter B, the limiting current density I L and ionic resistance R ohm Mark as unknown parameters and get the unknown parameter set.
[0020] In an optional embodiment of the present application, after substituting the battery information of the solid oxide fuel cell into the electrochemical model, the unknown parameter set and the objective function are determined, including: obtaining the measured voltage and the measured current in the battery information, and determining the root mean square error of the measured voltage and the measured current as the objective function, which is expressed as:
[0021]
[0022] Where x is the solution vector, which is determined based on the unknown parameter set; N is the number of measured voltages or measured currents; V m,k is the kth measured voltage; V c,k is the calculated voltage corresponding to the kth measured current.
[0023] In an optional embodiment of the present application, the unknown parameter set is substituted as an optimization target into a pre-built mountain gazelle optimizer for iteration, including: initializing the parameters and iteration number of the mountain gazelle optimizer, marking the iteration number t, and the iteration number t after initialization = 0; obtaining the output voltage and current of the t-round electrochemical model, substituting it into the objective function to obtain the t-round fitness value; screening the t-round fitness value according to the optimal condition to obtain the t-round optimal solution; generating a mountain gazelle population based on the unknown parameter set, the mountain gazelle population including male mountain gazelles and female mountain gazelles, male Mountain gazelles and female mountain gazelles are used to produce t+1 generation mountain gazelles; male mountain gazelles are determined by the optimal solution of round t, and female mountain gazelles are determined by the exploration mechanism of the mountain gazelle optimizer; the mountain gazelle population is represented by the mountain gazelle vector position, and the mountain gazelle vector position is determined according to the possible solutions of the unknown parameter set; the t generation mountain gazelles in the mountain gazelle population are processed according to the exploration mechanism to produce t+1 generation mountain gazelles; the t+1 round fitness value of the t+1 generation mountain gazelles is calculated to determine the male mountain gazelles in the t+1 generation mountain gazelles and use them for the next round of iteration until the preset conditions are met.
[0024] In an optional embodiment of the present application, the exploration mechanism includes: a solitary territorial male mechanism, a female herd mechanism, a single male herd mechanism and a migration mechanism for finding food; the young male herd coefficient vector is determined according to the total number of mountain gazelles, and the calculation equation is:
[0025]
[0026] BH is the young male herd coefficient vector; N is the total number of mountain gazelles; Xra is a random solution in the interval ra; M pr is a randomly selected search agent The average number of; r1 and r2 are random values between 0 and 1; the first coefficient is calculated according to the current male mountain gazelle vector position, and the calculation equation is:
[0027] D=(|X(t)|+|male gazelle |)×(2×r6-1);
[0028] D is the first coefficient; X(t) is the position of the mountain gazelle vector in this iteration; male gazelle is the current male mountain gazelle vector position, r6 is a random number between 0 and 1; the second coefficient is determined according to the first coefficient and the current iteration number, and the calculation equation is:
[0029]
[0030] F is the second coefficient; N1 is a random number from a standard distribution; exp is the Exponential function; MaxIter is the total number of iterations; Iter is the current number of iterations; the third coefficient is determined based on the total number of iterations and the current number of iterations, and the calculation equation is:
[0031]
[0032] The random selection coefficient vector is determined based on the third coefficient and the first coefficient, and the calculation equation is:
[0033]
[0034] Cof i is the random selection coefficient vector; a is the third coefficient; r3 and r4 are random numbers in the range of 0 to 1; N2, N3 and N4 are random numbers in the preset range and the dimension of the unknown parameter set; cos represents the cosine function; the solitary territorial male mechanism is constructed according to the current male mountain gazelle vector position, the young male herd coefficient vector, the position of the mountain gazelle vector in this iteration, the second coefficient and the random selection coefficient vector. The mathematical model of the solitary territorial male mechanism is expressed as:
[0035] TSM=male gazelle -|(ri1×BH-ri2×X(t))×F|×Cof i ;
[0036] TSM is the solitary territorial male mechanism; i , ri2 represents a number uniformly distributed in the range of [1,2]; the female herding mechanism is determined according to the young male herding coefficient vector, the random selection coefficient vector, and the current male mountain gazelle vector position. The mathematical model of the female herding is expressed as:
[0037] MH=(BH+Cof i )+(ri3×male gazelle -ri4×X rand )×Cof i ;
[0038] MH is the female herding mechanism; ri3 and ri4 are random integers 1 or 2; X rand is the vector position of a mountain gazelle randomly selected from the mountain gazelle population; the single male herding mechanism is determined according to the position of the mountain gazelle vector in this iteration, the first coefficient, the current male mountain gazelle vector position, the current male mountain gazelle vector position, the young male herd coefficient vector and the random selection coefficient vector. The mathematical model of the single male herding mechanism is expressed as:
[0039] BMH=(X(t)-D)+(ri5×male gazelle -ri6×BH)×Cof i ;
[0040] BMH is a single male herding mechanism; ri5 and ri6 are random integers 1 or 2; the upper and lower limits of the unknown parameter set are obtained to construct a migration mechanism for finding food. The mathematical model of the migration mechanism for finding food is expressed as:
[0041] MSF = (ub-lb) × r7 + lb;
[0042] ub is the upper limit of the solution of the unknown parameter set; lb is the lower limit of the solution of the unknown parameter set; r7 is a preset random number.
[0043] In an optional embodiment of the present application, the iterative result output by the mountain gazelle optimizer is obtained and substituted into the objective function to calculate the recognition result, including: the iterative result is an unknown parameter solution, and the unknown parameter solution is substituted into the electrochemical model to obtain iterative voltage and current data; the iterative voltage and current data is substituted into the objective function to calculate the recognition result including multiple fitness values, each fitness value corresponds to an unknown parameter solution; the iteration is repeated until the mountain gazelle optimizer meets the convergence condition, all the recognition results are screened according to the optimal condition, and the recognition result that meets the optimal condition is taken as the electrochemical model parameter and output, including: when the number of iterations is greater than or equal to the preset maximum number of iterations, and / or when the iterative result does not change in adjacent M rounds of iterations, it is determined that the convergence condition is met, and M is a positive integer greater than or equal to 2; all unknown parameter solutions are arranged in ascending order according to the fitness value, and the unknown parameter solution corresponding to the fitness value ranked first is marked as the electrochemical model parameter and output.
[0044] The present application also provides a computer device, including a processor and a memory: the processor is used to execute a computer program stored in the memory to implement the aforementioned method.
[0045] The present application also provides a computer-readable storage medium storing a computer program, which implements the aforementioned method when the computer program is executed by a processor.
[0046] The embodiments of the present application have the following beneficial effects:
[0047] In view of the obvious defects in the accuracy and speed of the current SOFC parameter identification, this application proposes a SOFC parameter identification method based on the mountain gazelle optimizer, which can accurately identify the unknown parameters of the electrochemical model by measuring the output voltage and current data of the SOFC. The operation is simple, reliable and low in cost. Specifically, the electrochemical model of SOFC is constructed to determine the unknown parameter set and the objective function. The unknown parameter set is input into the mountain gazelle optimizer as the optimization object, and the unknown parameter set is iteratively processed using the specific mechanism of the mountain gazelle optimizer, and the iterative effect is evaluated using the objective function, and the identification result is finally output. By adopting the mountain gazelle optimizer, it is possible to find high-quality solutions and try to avoid falling into the local optimal trap. It has a stronger global search capability and can accurately and quickly complete the parameter identification task. The optimization speed is fast, the accuracy is high, and the search performance is stable, thereby effectively improving the accuracy, speed, accuracy, stability and reliability of SOFC parameter identification.
[0048] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are specifically cited and described in detail with the accompanying drawings. It should be understood that the above general description and the detailed description below are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0050] in:
[0051] Figure 1 A schematic flow chart of a solid oxide fuel cell parameter identification method provided by an embodiment;
[0052] Figure 2 A flow chart of a parameter identification flow chart based on a mountain gazelle optimizer provided in an embodiment;
[0053] Figure 3 A schematic block diagram of the structure of a computer device provided by an embodiment. DETAILED DESCRIPTION
[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0055] The current parameter identification technologies are too idealistic and do not take into account the impact of insufficient measurement data and interference data on parameter identification. Therefore, the accuracy and speed of current SOFC parameter identification have obvious defects. This application proposes a solid oxide fuel cell parameter identification method. In order to clearly describe the solid oxide fuel cell parameter identification method provided in this embodiment, please refer to Figure 1 and Figure 2 , including steps S110 to S130.
[0056] Step S110: constructing an electrochemical model of a solid oxide fuel cell, substituting the battery information of the solid oxide fuel cell into the electrochemical model to determine an unknown parameter set and an objective function.
[0057] In one embodiment, the unknown parameter set is determined after the battery information of the solid oxide fuel cell is substituted into the electrochemical model, including: obtaining the load current density in the battery information for constructing a voltage loss expression; substituting the voltage loss expression into the first output voltage-current characteristic equation of the electrochemical model to obtain the second output voltage-current characteristic equation; and determining the unknown parameter set based on the second output voltage-current characteristic equation.
[0058] In one embodiment, the battery information may be data that can be directly collected by SOFC, including but not limited to output current and output voltage, etc., wherein in this embodiment, specifically the load current density. The load current density is used to construct the voltage loss expression, and the construction process will be described in detail later. Furthermore, the electrochemical model may adopt the most widely used SOFC electrochemical model, and its output voltage-current (voltage-current, VI) characteristics may be described by the first output voltage-current characteristic equation. The specific determination process will be described in detail later.
[0059] In one embodiment, the voltage loss expression includes an activation voltage loss expression, an ohmic voltage loss expression, and a concentration voltage loss expression. Specifically constructing the voltage loss expression includes: constructing the activation voltage loss expression according to the load current density:
[0060]
[0061] V act represents the activation voltage loss, A represents the slope of the Tafel line, I load represents the load current density, and I0 represents the exchange current density of the simplified battery model.
[0062] Construct an expression for the ohmic voltage loss based on the load current density:
[0063] V ohm =I load R ohm (2)
[0064] V ohm represents the ohmic voltage loss, R ohm Represents ionic resistance.
[0065] The load current density constructs the concentration voltage loss expression:
[0066]
[0067] V con represents the concentration voltage loss, B represents the first constant parameter, I L represents the limiting current density.
[0068] In one embodiment, the first output voltage-current characteristic equation of the electrochemical model can be expressed as:
[0069] V c =N cell (E o -V act -V ohm -V con ) (4)
[0070] V c Expressed as output voltage, included in the battery information; E o Expressed as solid oxide fuel cell open circuit voltage; N cell It is expressed as the number of cells in series in the solid oxide fuel cell stack. It is worth noting that among the parameters listed above, only the load current density I load are known, and the rest are unknown states, which are the parameters of SOFC that need to be solved.
[0071] Substituting the activation voltage loss expression, ohmic voltage loss expression and concentration voltage loss expression of formulas (1) to (3) into the first output voltage-current characteristic equation (4) can obtain the second output voltage-current characteristic equation, which is expressed as:
[0072]
[0073] According to formula (5), only the load current density I load is known, and the rest will be the solid oxide fuel cell open circuit voltage E o , the slope of the Tafel line A, the exchange current density I0 of the simplified battery model, the first constant parameter B, the limiting current density I L and ionic resistance R ohm is unknown. That is, 6 unknown parameters need to be identified, namely E o ,A,I0,B,I L and R ohm Among them, I L >I load , the above 6 parameters can be summarized to get the unknown parameter set.
[0074] In one embodiment, the battery information of the solid oxide fuel cell is substituted into the electrochemical model to determine the unknown parameter set and the objective function, including: obtaining the measured voltage and measured current in the battery information, and determining the root mean square error of the measured voltage and measured current as the objective function.
[0075] In one embodiment, the main purpose of parameter identification is to extract a set of appropriate parameters for the above model so that the model is equivalent to the actual SOFC with the minimum fitting error. Therefore, the objective function can be selected as the root mean square error (RMSE). The objective function can be specifically expressed as:
[0076]
[0077] Among them, x is the solution vector, which is determined according to the unknown parameter set and can be specifically expressed as x = [E o ,A,R,B,I0,I L ]; N is the number of measured voltage or measured current, that is, the number of data points; V m,k is the kth measured voltage; V c,k is the calculated voltage corresponding to the kth measured current. The measured voltage and the measured current are included in the battery information and can be directly measured or calculated.
[0078] Step S120: Substitute the unknown parameter set as the optimization target into the pre-built Gazelle optimizer for iteration; after each iteration, obtain the iteration result output by the Gazelle optimizer and substitute it into the objective function to calculate and obtain the recognition result.
[0079] In one embodiment, the unknown parameter set is substituted as the optimization target into a pre-built mountain gazelle optimizer for iteration, including: initializing the parameters and iteration number of the mountain gazelle optimizer, marking the iteration number t, and the iteration number t after initialization is 0; obtaining the output voltage and current of the electrochemical model in round t, substituting it into the objective function to obtain the fitness value of round t; screening the fitness value of round t according to the optimal condition to obtain the optimal solution of round t; generating a mountain gazelle population based on the unknown parameter set, the mountain gazelle population including male mountain gazelles and female mountain gazelles, the male mountain gazelles The antelope and female mountain gazelles are used to produce t+1 generation mountain gazelles; the male mountain gazelles are determined by the optimal solution of round t, and the female mountain gazelles are determined by the exploration mechanism of the mountain gazelle optimizer; the mountain gazelle population is represented by the mountain gazelle vector position, and the mountain gazelle vector position is determined according to the possible solutions of the unknown parameter set; the t generation mountain gazelles in the mountain gazelle population are processed according to the exploration mechanism to produce t+1 generation mountain gazelles; the t+1 round fitness value of the t+1 generation mountain gazelles is calculated to determine the male mountain gazelles in the t+1 generation mountain gazelles and use them for the next round of iteration until the preset conditions are met.
[0080] Step S130: Repeat the iteration until the Mountain Gazelle Optimizer meets the convergence condition, filter all the recognition results according to the optimal condition, take the recognition results that meet the optimal condition as the electrochemical model parameters and output them.
[0081] In one embodiment, the iterative result output by the mountain gazelle optimizer is obtained and substituted into the objective function to calculate the recognition result, including: the iterative result is an unknown parameter solution, and the unknown parameter solution is substituted into the electrochemical model to obtain iterative voltage and current data; the iterative voltage and current data is substituted into the objective function to calculate the recognition result including multiple fitness values, each fitness value corresponds to an unknown parameter solution; the iteration is repeated until the mountain gazelle optimizer meets the convergence condition, all the recognition results are screened according to the optimal condition, and the recognition results that meet the optimal condition are taken as the electrochemical model parameters and output, including: when the number of iterations is greater than or equal to the preset maximum number of iterations, and / or when the iterative result does not change in adjacent M rounds of iterations, it is determined that the convergence condition is met, and M is a positive integer greater than or equal to 2; all unknown parameter solutions are arranged in ascending order according to the fitness value, and the unknown parameter solution corresponding to the fitness value ranked first is marked as the electrochemical model parameter and output.
[0082] In one embodiment, for how to use the mountain gazelle optimizer (MGO) to iterate in steps S120 to S130, please refer to Figure 2 , Figure 2 It is a parameter identification flow chart based on the Mountain Gazelle Optimizer, which specifically includes steps S210 to S270.
[0083] Step S210: Initialize the parameters and iteration times of the Mountain Gazelle optimizer, mark the iteration times t, and the iteration times after initialization are t=0.
[0084] In one embodiment, the initialization not only includes initializing the MGO parameters, but also needs to initialize the number of iterations and set it to 0.
[0085] Step S220: Obtain the output voltage and current of the t-round electrochemical model, and substitute them into the objective function to obtain the t-round fitness value.
[0086] In one embodiment, the output voltage and current of the t-round electrochemical model are obtained, that is, the battery information collected by the SOFC in the initial state, including the specific numerical values of the output voltage and current, are obtained, and are substituted into the objective function represented by formula (6), and the fitness value of each individual is calculated by formula (6).
[0087] Step S230: Filter the t-round fitness values according to the optimal conditions to obtain the t-round optimal solution.
[0088] Step S240: Determine whether the convergence condition is met.
[0089] In one embodiment, the optimal condition in this embodiment can be specifically to arrange all fitness values in descending order, and take the parameter solution with the smallest fitness value as the solution of the unknown parameter set. And further determine whether the convergence condition is met in the current iteration round. The convergence condition can be specifically that the number of iterations is greater than or equal to the preset maximum number of iterations, and / or that the iteration result does not change in adjacent M rounds of iterations, and M is a positive integer greater than or equal to 2. The iteration result is also the solution of the unknown parameter set determined after each round of iteration, and the unknown parameter solution is substituted into the electrochemical model to obtain iterative voltage and current data. The iterative voltage and current data are substituted into the objective function to calculate the recognition result including multiple fitness values, and each fitness value corresponds to an unknown parameter solution.
[0090] If the convergence condition is not met, step S250 is executed: generating a t-generation mountain gazelle population based on the unknown parameter set; and processing the t-generation mountain gazelles in the mountain gazelle population according to the exploration mechanism to generate t+1-generation mountain gazelles.
[0091] Step S260: Let t=t+1, calculate the fitness value of the t-th generation of mountain gazelles in round t, so as to determine the male mountain gazelles in the t-th generation of mountain gazelles. After step S260, return to step S230, repeat the execution, and start iteration.
[0092] In one embodiment, when the convergence condition is not met, MGO is used for iteration to update the current optimal solution. Specifically, the unknown parameter set is used to generate a mountain gazelle population, wherein the mountain gazelle population includes male mountain gazelles and female mountain gazelles, and the male mountain gazelles and female mountain gazelles are used to generate t+1 generation mountain gazelles. Specifically, the male mountain gazelles are determined by the optimal solution of the t round, and the female mountain gazelles are determined by the exploration mechanism of the mountain gazelle optimizer. The determination method of the female mountain gazelle will be described in detail later. It can be understood that the mountain gazelle population is represented by the mountain gazelle vector position, and the mountain gazelle vector position is determined according to the possible solution of the unknown parameter set.
[0093] Iteration in MGO is to process the t generation of mountain gazelles in the mountain gazelle population according to the exploration mechanism to produce the t+1 generation of mountain gazelles; calculate the t+1 round fitness value of the t+1 generation of mountain gazelles to determine the male mountain gazelles in the t+1 generation of mountain gazelles and use them for the next round of iteration.
[0094] In one embodiment, the exploration mechanism includes: a solitary territorial male mechanism, a female herd mechanism, a single male herd mechanism, and a migration mechanism for finding food. The four mechanisms are applied to all mountain gazelles in the mountain gazelle population, and will have an impact on the mountain gazelle vector position of all mountain gazelles during the iteration process to update the position of the mountain gazelle.
[0095] The exploration mechanism requires several basic parameters for construction. First, the young male herd coefficient vector is determined based on the total number of mountain gazelles. The calculation equation is:
[0096]
[0097] BH is the young male herd coefficient vector; N is the total number of mountain gazelles; X ra is a random solution in the interval ra; M pr is a randomly selected search agent ; r1 and r2 are random values between 0 and 1.
[0098] Then calculate the first coefficient based on the current male mountain gazelle vector position, and the calculation equation is:
[0099] D=(|X(t)|+|male gazelle |)×(2×r6-1) (8)
[0100] D is the first coefficient; X(t) is the position of the mountain gazelle vector in this iteration; male gazelle is the current male mountain gazelle vector position, and r6 is a random number between 0 and 1.
[0101] And the second coefficient is determined according to the first coefficient and the current iteration number, and the calculation equation is:
[0102]
[0103] F is the second coefficient; N1 is a random number from a standard distribution; exp is the Exponential function; MaxIter is the total number of iterations; Iter is the current number of iterations.
[0104] The third coefficient is determined according to the total number of iterations and the current number of iterations. The calculation equation is:
[0105]
[0106] The random selection coefficient vector is determined based on the third coefficient and the first coefficient, and the calculation equation is:
[0107]
[0108] Cof i is a randomly selected coefficient vector; a is the third coefficient; r3 and r4 are random numbers in the range of 0 to 1; N2, N3 and N4 are random numbers in the preset range and the dimensions of the unknown parameter set; cos represents the cosine function.
[0109] After determining the above parameters, the exploration mechanism can be constructed. The solitary territorial male mechanism is constructed according to the current male mountain gazelle vector position, the young male herd coefficient vector, the position of the mountain gazelle vector in this iteration, the second coefficient and the random selection coefficient vector. The mathematical model of the solitary territorial male mechanism is expressed as:
[0110] TSM=male gazelle -|(ri1×BH-ri2×X(t))×F|×Cofi i (12)
[0111] TSM is the solitary territorial male mechanism; ri1, ri2 represent numbers uniformly distributed in the range [1,2].
[0112] The female herding mechanism is determined according to the young male herd coefficient vector, the random selection coefficient vector, and the current male mountain gazelle vector position. The mathematical model of the female herding is expressed as:
[0113] MH=(BH+Cofi i )+(ri3×male gazelle -ri4×X rand )×Cof i (13)
[0114] MH is the female herding mechanism; ri3 and ri4 are random integers 1 or 2; X rand is the vector position of a randomly selected mountain gazelle from the mountain gazelle population.
[0115] The single male herding mechanism is determined according to the position of the mountain gazelle vector in this iteration, the first coefficient, the current male mountain gazelle vector position, the current male mountain gazelle vector position, the young male herd coefficient vector and the random selection coefficient vector. The mathematical model of the single male herding mechanism is expressed as:
[0116] BMH=(X(t)-D)+(ri5×male gazelle -ri6×BH)×Cofi i (14)
[0117] BMH is a single male herding mechanism; ri5 and ri6 are random integers 1 or 2.
[0118] Obtain the upper and lower limits of the unknown parameter set to construct a migration mechanism for finding food. The mathematical model of the migration mechanism for finding food is expressed as:
[0119] MSF=(ub-lb)×r7+lb (15)
[0120] ub is the upper limit of the solution of the unknown parameter set; lb is the lower limit of the solution of the unknown parameter set; r7 is a preset random number, which can be an integer between 0 and 1.
[0121] In one embodiment, the fitness of MSF, BMH, MH, and TSM is calculated and added to the mountain gazelle population for iteration to update the vector position of each mountain gazelle. After each iteration, all mountain gazelles are sorted in ascending order of fitness. The top-ranked mountain gazelle, with high-quality and promising solutions, is regarded as an adult male mountain gazelle that owns the territory and is protected in the total population; other individuals are regarded as old or weak mountain gazelles and are removed from the entire population. The best N mountain gazelles in the population are finally retained for the next round of iterations.
[0122] If the convergence condition is met, step S270 is executed: all recognition results are screened according to the optimal condition, and the recognition results that meet the optimal condition are taken as electrochemical model parameters and output.
[0123] In one embodiment, after completing the iteration, several recognition results can be obtained, which can be screened using the optimal conditions. Specifically, all unknown parameter solutions can be arranged in ascending order according to fitness values, and the unknown parameter solution corresponding to the first-ranked fitness value can be marked as an electrochemical model parameter and output.
[0124] Therefore, the present application aims at the obvious defects in the accuracy and speed of the current SOFC parameter identification, and proposes a SOFC parameter identification method based on the mountain gazelle optimizer, which can accurately identify the unknown parameters of the electrochemical model by measuring the output voltage and current data of the SOFC, and is simple to operate, reliable, and low in cost. Specifically, the electrochemical model of SOFC is constructed, and the unknown parameter set and the objective function are determined. The unknown parameter set is input into the mountain gazelle optimizer as the optimization object, and the unknown parameter set is iteratively processed using the specific mechanism of the mountain gazelle optimizer, and the iterative effect is evaluated using the objective function, and the identification result is finally output. By adopting the mountain gazelle optimizer, it is possible to find high-quality solutions and try to avoid falling into the local optimal trap. It has a stronger global search capability and can accurately and quickly complete the parameter identification task. The optimization speed is fast, the accuracy is high, and the search performance is stable, thereby effectively improving the accuracy, speed, accuracy, stability and reliability of SOFC parameter identification.
[0125] Figure 3 FIG. 1 shows an internal structure diagram of a computer device in an embodiment. The computer device may be a terminal or a server. Figure 3 As shown, the computer device includes a processor, a memory and a network interface connected via a system bus. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor may implement the solid oxide fuel cell parameter identification method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor may implement the solid oxide fuel cell parameter identification method. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0126] In one embodiment, the present application further proposes a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the method described in any of the aforementioned embodiments.
[0127] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0128] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0129] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A solid oxide fuel cell parameter identification method, characterized in that: The steps include: Constructing an electrochemical model of a solid oxide fuel cell, and substituting the battery information of the solid oxide fuel cell into the electrochemical model to determine an unknown parameter set and an objective function; Substituting the unknown parameter set as the optimization target into the pre-built Gazelle optimizer for iteration; After each iteration, obtaining the iteration result output by the mountain gazelle optimizer and substituting it into the objective function to calculate and obtain the recognition result; Repeat the iteration until the mountain gazelle optimizer meets the convergence condition, screen all the recognition results according to the optimal condition, take the recognition results that meet the optimal condition as the electrochemical model parameters and output them.
2. The solid oxide fuel cell parameter identification method according to claim 1, characterized in that: Substituting the battery information of the solid oxide fuel cell into the electrochemical model to determine the unknown parameter set includes: Obtaining the load current density in the battery information for constructing a voltage loss expression; Substituting the voltage loss expression into the first output voltage-current characteristic equation of the electrochemical model to obtain a second output voltage-current characteristic equation; The unknown parameter set is determined according to the second output voltage-current characteristic equation.
3. The solid oxide fuel cell parameter identification method according to claim 2, characterized in that: The voltage loss expressions include an activation voltage loss expression, an ohmic voltage loss expression, and a concentration voltage loss expression; The constructing of the voltage loss expression includes: The activation voltage loss expression is constructed according to the load current density: V act represents the activation voltage loss, A represents the slope of the Tafel line, I load represents the load current density, I0 represents the exchange current density of the simplified battery model; The ohmic voltage loss expression is constructed according to the load current density: V ohm =I load R ohm ; V ohm represents the ohmic voltage loss, R ohm represents ionic resistance; The load current density constructs the concentration voltage loss expression: V con represents the concentration voltage loss, B represents the first constant parameter, I L represents the limiting current density.
4. The solid oxide fuel cell parameter identification method according to claim 3, characterized in that: The first output voltage-current characteristic equation of the electrochemical model is expressed as: V c =N cell (E o -V act -V ohm -V con ); V c It is expressed as output voltage and is included in the battery information; E o Expressed as solid oxide fuel cell open circuit voltage; N cell Expressed as the number of series-connected cells in a solid oxide fuel cell stack; Substituting the activation voltage loss expression, the ohmic voltage loss expression and the concentration voltage loss expression into the first output voltage-current characteristic equation can obtain a second output voltage-current characteristic equation, which is expressed as: The step of determining the unknown parameter set according to the second output voltage and current characteristic equation comprises: The open circuit voltage E of the solid oxide fuel cell o , the slope A of the Tafel line, the exchange current density I0 of the simplified battery model, the first constant parameter B, the limiting current density I L and the ionic resistance R ohm Mark as unknown parameters to obtain the unknown parameter set.
5. The solid oxide fuel cell parameter identification method according to claim 1, characterized in that: Substituting the battery information of the solid oxide fuel cell into the electrochemical model to determine the unknown parameter set and the objective function includes: The measured voltage and the measured current in the battery information are obtained, and the root mean square error of the measured voltage and the measured current is determined as the objective function, which is expressed as: Wherein, x is a solution vector determined according to the unknown parameter set; N is the number of the measured voltages or the measured currents; V m,k is the kth measured voltage; V c,k is the calculated voltage corresponding to the kth measured current.
6. The solid oxide fuel cell parameter identification method according to claim 1, characterized in that: Substituting the unknown parameter set as the optimization target into the pre-built Gazelle optimizer for iteration includes: Initialize the parameters and iteration number of the mountain gazelle optimizer, mark the iteration number t, and the iteration number t after initialization=0; Obtaining the output voltage and current of the electrochemical model in round t, and substituting them into the objective function to obtain the fitness value in round t; Filtering the t-round fitness values according to the optimal condition to obtain the t-round optimal solution; Generate a mountain gazelle population based on the unknown parameter set, the mountain gazelle population includes male mountain gazelles and female mountain gazelles, and the male mountain gazelles and female mountain gazelles are used to generate t+1 generation mountain gazelles; the male mountain gazelles are determined by the t-round optimal solution, and the female mountain gazelles are determined by the exploration mechanism of the mountain gazelle optimizer; the mountain gazelle population is represented by the mountain gazelle vector position, and the mountain gazelle vector position is determined according to the possible solution of the unknown parameter set; The t generation of mountain gazelles in the mountain gazelle population is processed according to the exploration mechanism to produce the t+1 generation of mountain gazelles; the t+1 round fitness value of the t+1 generation of mountain gazelles is calculated to determine the male mountain gazelles in the t+1 generation of mountain gazelles and use them for the next round of iteration until the preset conditions are met.
7. The solid oxide fuel cell parameter identification method according to claim 6, characterized in that: Said exploration mechanisms include: solitary territorial male mechanism, female herding mechanism, single male herding mechanism and migration mechanism for finding food; The young male herd coefficient vector is determined based on the total number of mountain gazelles, and the calculation equation is: BH is the young male herd coefficient vector; N is the total number of mountain gazelles; X ra is a random solution in the interval ra; M pr is a randomly selected search agent The average number of; while r1 and r2 are random values between 0 and 1; The first coefficient is calculated based on the current male mountain gazelle vector position, and the calculation equation is: D=(|X(t)|+|male gazelle |)×(2×r6-1); D is the first coefficient; X(t) is the position of the mountain gazelle vector in this iteration; male gazelle is the current male mountain gazelle vector position, r6 is a random number between 0 and 1; The second coefficient is determined according to the first coefficient and the current iteration number, and the calculation equation is: F is the second coefficient; N1 is a random number of standard distribution; exp is the Exponential function; MaxIter is the total number of iterations; Iter is the current number of iterations; The third coefficient is determined according to the total number of iterations and the current number of iterations, and the calculation equation is: The random selection coefficient vector is determined according to the third coefficient and the first coefficient, and the calculation equation is: Cof i is the randomly selected coefficient vector; a is the third coefficient; r3 and r4 are random numbers in the range of 0 to 1; N2, N3 and N4 are random numbers in a preset range and the dimension of the unknown parameter set; cos represents the cosine function; The solitary territorial male mechanism is constructed according to the current male mountain gazelle vector position, the young male herd coefficient vector, the position of the mountain gazelle vector in this iteration, the second coefficient and the random selection coefficient vector. The mathematical model of the solitary territorial male mechanism is expressed as: TSM=male gazelle -|(ri1×BH-ri2×X(t))×F|×Cofi i ; TSM is the solitary territorial male mechanism; ri1, ri2 represent numbers uniformly distributed in the range [1,2]; The female herding mechanism is determined according to the young male herding coefficient vector, the random selection coefficient vector, and the current male mountain gazelle vector position. The mathematical model of the female herding is expressed as: MH=(BH+Cof i )+(ri3×male gazelle -to t ×X rand )×Coffee i ; MH is the female herd mechanism; ri3 and ri4 are random integers 1 or 2; X rand is the vector position of a randomly selected mountain gazelle from the population of mountain gazelles; The single male herding mechanism is determined according to the position of the mountain gazelle vector in this iteration, the first coefficient, the current male mountain gazelle vector position, the current male mountain gazelle vector position, the young male herd coefficient vector and the random selection coefficient vector. The mathematical model of the single male herding mechanism is expressed as: BMH=(X(t)-D)+(ri5×male gazelle -ri6×BH)×Cofi i ; BMH is the single male herding mechanism; ri5 and ri6 are random integers 1 or 2; The upper and lower limits of the unknown parameter set are obtained to construct a migration mechanism for finding food. The mathematical model of the migration mechanism for finding food is expressed as: MSF = (ub-lb) × r7 + lb; ub is the upper limit of the solution of the unknown parameter set; lb is the lower limit of the solution of the unknown parameter set; r7 is a preset random number.
8. The solid oxide fuel cell parameter identification method according to claim 1, characterized in that: The step of obtaining the iterative result output by the mountain gazelle optimizer and substituting it into the objective function to calculate and obtain the recognition result includes: The iterative result is an unknown parameter solution, and the unknown parameter solution is substituted into the electrochemical model to obtain iterative voltage and current data; Substituting the iterative voltage and current data into the objective function to calculate and obtain the identification result including a plurality of fitness values, each fitness value corresponding to an unknown parameter solution; The iterative steps are repeated until the Gazelle optimizer meets the convergence condition, all the recognition results are screened according to the optimal condition, and the recognition results meeting the optimal condition are taken as electrochemical model parameters and outputted, including: When the number of iterations is greater than or equal to a preset maximum number of iterations, and / or when the iteration result does not change in M consecutive iterations, it is determined that the convergence condition is met, where M is a positive integer greater than or equal to 2; All the unknown parameter solutions are arranged in ascending order according to the fitness values, and the unknown parameter solution corresponding to the fitness value ranked first is marked as an electrochemical model parameter and outputted.
9. A computer device, characterized in that: including a processor and a memory; The processor is configured to execute the computer program stored in the memory to implement the method according to any one of claims 1 to 8.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.
Citation Information
Cited By
Electrochemical experimental data adaptive processing system and method for solid oxide cell
CN122862925A
