Hazardous chemical warehouse PCF pressure sensor design method based on MI-GA

Abstract

The invention discloses a hazardous chemical warehouse PCF pressure sensor design method based on MI-GA, relates to the technical field of sensors, and solves the technical problems that before a PCF sensor is used in the prior art, parameters of the PCF sensor need to be adjusted each time, so that the applicability of the PCF sensor is low, and the labor cost and the time cost are high. The method comprises the steps of obtaining preset structure data; initializing a photonic crystal fiber structure according to the structure data; obtaining specific requirements of the design; performing electromagnetic simulation on the device structure according to specific requirements to obtain a structure parameter matrix; individuals entering a reproduction link are screened out, and the structure of the PCF sensor is optimized; constructing a simulation file according to the structure parameter matrix, and performing iterative loop; the set number of iterations is reached; the structure of the PCF device can be adjusted at any time according to different devices so as to adapt to different measurement scenes in a warehouse management system, and the method has excellent universality.

Description

Technical Field

[0001] The present invention belongs to the field of sensor design, and relates to a PCF pressure sensor design technology for a hazardous chemical warehouse, and specifically to a PCF pressure sensor design method for a hazardous chemical warehouse based on MI-GA. Background Art

[0002] The matrix input legacy algorithm (MI-GA) is an optimization search method based on the principles of classical genetics. It continuously iterates the "population" to screen out the "dominant" individuals, that is, the local optimal solution to the problem being solved; Photonic crystal fiber (PCF) is an optical fiber with periodically arranged air holes. Various photonic crystal fiber devices are derived on its basis and are widely used in various optical communications, optical integrated systems, and optical sensors. Hardware sensors play an extremely important role in warehouse management systems. PCF sensors are widely used in warehouse management systems due to their excellent anti-electrical interference, strong acid and alkali resistance, high-precision measurement, and passive measurement. Especially in the sensing and monitoring of hazardous environments in hazardous chemical warehouses, PCF pressure sensors have incomparable safety performance compared to ordinary sensors.

[0003] The design of existing PCF sensors mainly relies on artificial a priori design, adopts artificial design configuration, repeatedly conducts simulation experiments and manually adjusts the structure parameters, and finally obtains a sensor that meets the sensing performance requirements; however, the requirements for PCF sensors are different in different measurement environments; before using the PCF sensor, the parameters of the PCF sensor need to be adjusted each time, resulting in low applicability of the PCF sensor and high labor and time costs.

[0004] The present invention provides a design method of a PCF pressure sensor for a hazardous chemical warehouse based on MI-GA to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a design method for a PCF pressure sensor for a hazardous chemical warehouse based on MI-GA, which is used to solve the technical problems in the prior art that before using the PCF sensor, the parameters of the PCF sensor need to be adjusted each time, resulting in low applicability of the PCF sensor and high labor and time costs.

[0006] To achieve the above object, the first aspect of the present invention provides a design method for a PCF pressure sensor for a hazardous chemical warehouse based on MI-GA, comprising:

[0007] Step S1: obtaining preset structural data; initializing the photonic crystal fiber structure according to the structural data;

[0008] Step S2: Obtain specific design requirements; perform electromagnetic simulation on the device structure according to the specific requirements to obtain a structural parameter matrix;

[0009] Step S3: Screening out individuals that enter the reproduction stage and optimizing the structure of the PCF sensor;

[0010] Step S4: construct a simulation file according to the structural parameter matrix and perform an iterative cycle until the set number of iterations is reached.

[0011] Preferably, the initializing the photonic crystal fiber structure according to the structural data comprises:

[0012] Retrieve structural data; wherein the structural data includes: population number NP and number of iterations NG;

[0013] The initial population lattice constant P is randomly generated according to the population number in the structural data initial ={p1, p2, ..., pn}; randomly generate the parameter matrix of the air hole according to the number of iterations in the structural data;

[0014] The structure generation function is used to write the population number into the simulation software according to the coding information in the storage matrix, and the algorithm is initialized; wherein the coding information is generated by real number coding.

[0015] It should be noted that the structure generation function is function_struture.

[0016] Preferably, the expression of the parameter matrix of the air holes is as follows:

[0017]

[0018] Among them, a and b are the air hole radii in two orthogonal directions of the air hole, respectively, and n is the structural dimension of PCF, that is, the number of air holes; the size of the parameter matrix is ​​expressed as D = size (a, b) × n.

[0019] Preferably, the encoding information is generated by real number encoding, including:

[0020] Retrieve the initial structure of the initialized PCF sensor, analyze the population individuals in the initial structure, and obtain the values ​​of the optimized parameters;

[0021] The numerical value of the optimized parameter is placed into the gene sequence of the corresponding individual to obtain the genetic information of the individual; the genetic information of the individual is used as the coding information.

[0022] The present invention generates individual genetic information by means of real number coding, which can theoretically enable the optimized structure to reach the limiting birefringence coefficient of the device, thereby obtaining the optimal pressure sensing performance.

[0023] Preferably, the electromagnetic simulation of the device structure is performed according to specific requirements to obtain a structural parameter matrix, including:

[0024] Retrieve the specific requirements of the design; the specific requirements include size, drawing process, measurement range, and measurement accuracy;

[0025] Determine the initial device structure of the PCF sensor according to the specific requirements of the design; input the device structure of the PCF sensor into the electromagnetic simulation software, solve several groups of effective refractive indexes of the PCF sensor, and construct several simulation files; wherein each group of effective refractive index values ​​is the effective refractive index in two different polarization directions;

[0026] Call the structure function, use the structure function to calculate the fitness of several simulation files, and read and store the corresponding fitness matrix.

[0027] Preferably, the step of calculating the fitness of a plurality of simulation files using a structure function comprises:

[0028] Retrieve the effective refractive index of the PCF sensor in two different polarization directions, marked as n eff1 and n eff2 ;

[0029] Through the fitness function Fitness(i) = abs(n eff1 -n eff2 ) calculates the fitness of individuals in a population; where Fitness(i) represents the fitness of the i-th individual in the population.

[0030] The present invention calculates the fitness of an individual according to the effective refractive index in two different polarization directions; the performance index of the device can be directly used as the calculation result, and the convergence speed and optimization index in the optimization process can be intuitively expressed.

[0031] Preferably, the screening out of individuals entering the reproduction stage and optimizing the structure of the PCF sensor includes:

[0032] Obtaining diameter parameters of the air holes and the air holes to be optimized; wherein the diameter parameters include a major axis a and a minor axis b;

[0033] Calculate the cumulative selection probability of individuals; select the individuals to be reproduced according to the cumulative selection probability of individuals; obtain a two-dimensional matrix after encoding the air holes to be optimized according to the reproduced individuals; wherein the size of the two-dimensional matrix is ​​2×n, n is the number of air holes to be optimized; the parameter data of the corresponding air holes are stored in the two-dimensional matrix.

[0034] Preferably, the step of calculating the cumulative selection probability of an individual comprises:

[0035] Retrieve individuals from the population and use the formula Calculate the selection probability of an individual; where F i is the fitness function, and its general expression is F(X)={|f(x)|}; i=1,2,…,n; n is a positive integer, indicating the total number of individuals;

[0036] By formula Calculate the cumulative selection probability of an individual; where j = 1, 2, …, i.

[0037] Preferably, the step of selecting individuals for reproduction according to the cumulative selection probability of the individuals comprises:

[0038] Get a floating point number in a preset interval; where the preset interval is [0,1]; the floating point number is obtained by random selection;

[0039] Retrieve the cumulative selection probability of the individual; determine whether the floating point number is less than the cumulative selection probability; if yes, select the corresponding individual as the individual for reproduction; if no, proceed to the next round of cumulative selection probability q i+1 , until the individuals that reproduce in this round of the turntable are selected.

[0040] Preferably, the construction of the simulation file according to the structural parameter matrix and the iterative loop include:

[0041] Obtain the structure of the individual with the best fitness in the previous generation and save it in the simulation file; use the structure generation function to write the new population into the simulation software for simulation solution, and count the number of optimizations; wherein the new population is obtained according to the optimized individual structure;

[0042] Determine whether the number of optimizations is less than the number of iterations; if yes, proceed to the next round of optimization; if no, the optimization is completed, all results are saved and the main program is exited.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. The present invention proposes an improved MI-GA algorithm, which innovatively introduces real number coding. Compared with the binary coding serial form of the classical genetic algorithm, which cannot characterize the device configuration of the PCF sensor, the algorithm of the present invention realizes the encoding and genetic operation of the PCF pressure sensor structure, solves the Hamming cliff problem, and ensures the ability to capture advantageous features during the algorithm optimization process; the parameters to be optimized are input in matrix form, and different structural parameters are independently placed in the input matrix, and genetic operations such as crossover and mutation are performed according to the device characteristics, thereby improving the algorithm's optimization effect on the sensor.

[0045] 2. The present invention proposes to use an improved MI-GA algorithm to optimize the design of PCF pressure sensors in warehouses. Compared with existing design methods, the algorithm is more efficient and has better performance. It can adjust the PCF device structure at any time according to different device sizes, drawing processes, measurement ranges, measurement accuracy, etc. to adapt to different measurement scenarios in warehouse management systems, and has excellent versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A schematic diagram of the overall steps of the design method of the present invention;

[0048] Figure 2 This is a schematic diagram of the initial PCF sensor results of the present invention;

[0049] Figure 3 is a graph of effective refractive index and birefringence coefficient of the present invention;

[0050] Figure 4 This is a birefringence coefficient curve diagram after optimization of the present invention;

[0051] Figure 5 This is a schematic diagram of the structure of the PCF pressure sensor optimized by the present invention;

[0052] Figure 6 This is a characteristic diagram of the PCF pressure sensor optimized by the present invention;

[0053] Figure 7 It is a box diagram of repeated optimization of the MI-GA algorithm of the present invention. DETAILED DESCRIPTION

[0054] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] See also Figure 1 The first embodiment of the present invention provides a design method of a PCF pressure sensor for a hazardous chemical warehouse based on MI-GA, comprising:

[0056] Step S1: obtaining preset structural data; initializing the photonic crystal fiber structure according to the structural data;

[0057] Step S2: Obtain specific design requirements; perform electromagnetic simulation on the device structure according to the specific requirements to obtain a structural parameter matrix;

[0058] Step S3: Screening out individuals that enter the reproduction stage and optimizing the structure of the PCF sensor;

[0059] Step S4: construct a simulation file according to the structural parameter matrix and perform an iterative cycle until the set number of iterations is reached.

[0060] See also Figure 2-Figure 3 , obtain the preset structural data; randomly generate the initial population lattice constant P according to the population number in the structural data initial ={p1, p2, ..., pn}; randomly generate the parameter matrix of the air hole according to the number of iterations in the structural data; analyze the population individuals in the initial structure to obtain the value of the optimized parameter;

[0061] The numerical value of the optimization parameter is placed in the gene sequence of the corresponding individual to obtain the genetic information of the individual; the genetic information of the individual is used as the coding information; the structure generation function is used to write the population number into the simulation software according to the coding information in the storage matrix, and the algorithm is initialized.

[0062] The expression of the parameter matrix of the air hole is as follows:

[0063]

[0064] Among them, a and b are the air hole radii in two orthogonal directions of the air hole, respectively, and n is the structural dimension of PCF, that is, the number of air holes; the size of the parameter matrix is ​​expressed as D = size (a, b) × n.

[0065] Obtain specific design requirements; determine the initial PCF sensor device structure according to the specific design requirements; input the PCF sensor device structure into electromagnetic simulation software, solve several groups of effective refractive indexes of the PCF sensor, and construct several simulation files; wherein each group of effective refractive index values ​​is the effective refractive index in two different polarization directions; call the structure function, use the structure function to calculate the fitness of several simulation files, and read and store the corresponding fitness matrix.

[0066] It should be noted that the fitness function directly uses the performance indicators of the device as the calculation result, which can intuitively show the convergence speed and optimization index in the optimization process; the settings of the PCF sensor can be easily changed according to actual needs, which has great practical value and commercial potential, and is conducive to the performance analysis of the PCF sensor.

[0067] For example: Based on the warehouse management system requirements of a chemical enterprise, the initial configuration of a PCF pressure sensor is determined. Its overall structure is a solid core with three layers of air holes. The optical fiber base material is selected as silicon dioxide. The PCF structure is as follows: Figure 2 As shown:

[0068] In the initial structure, the lattice constant P is set to 3μm, and the air hole is set to a normalized circular hole with a diameter of 2μm; the digital simulation of the initial structure is analyzed, and at a wavelength of 1550nm, the effective refractive index of the X polarization mode is 1.4418708590, and the effective refractive index of the Y polarization mode is 1.4418711330; the birefringence coefficient B at this time is calculated to be 0.274×10-7.

[0069] See also Figure 4-Figure 7 , retrieve the effective refractive index of the PCF sensor in two different polarization directions, marked as n eff1 and n eff2 ; Through the fitness function Fitness(i) = abs(n eff1 -n eff2 ) calculates the fitness of individuals in the population; where Fitenss(i) represents the fitness of the i-th individual in the population.

[0070] Get the diameter parameters of the air holes and the air holes to be optimized; the diameter parameters include the major axis a and the minor axis b; call the individuals in the population, and use the formula Calculate the selection probability of an individual; where F i is the fitness function, and its general expression is F(X)={|f(x)|}; i=1,2,…,n; n is a positive integer, indicating the total number of individuals; through the formula Calculate the cumulative selection probability of the individual; where j = 1, 2, ..., i; obtain a floating point number in a preset interval; where the preset interval is [0, 1]; the floating point number is obtained by random selection; retrieve the cumulative selection probability of the individual; determine whether the floating point number is less than the cumulative selection probability; if yes, select the corresponding individual as the individual for reproduction; if no, proceed to the next round of cumulative selection probability q i+1 until the reproduced individuals of this round of turntables are selected; a two-dimensional matrix is ​​obtained after encoding the air holes to be optimized according to the reproduced individuals; wherein the size of the two-dimensional matrix is ​​2×n, n is the number of air holes to be optimized; the parameter data of the corresponding air holes are stored in the two-dimensional matrix.

[0071] It is worth noting that the hybridization of the parents is also divided into two parts: for the lattice constant P, because the real number encoding is only one digit, there is no hybridization operation in this part, and the mutation operation is directly performed. The mutation method is to randomly select the vector P = {p1, p2, ..., pn}, and the value of the selected digit is randomly increased or decreased by ±5%; for the air hole input matrix, a "breakpoint" position is randomly selected according to the length of the matrix, and the gene sequence of the parent is exchanged at this position, and the hybridization process is completed at this time; there are two mutations in the air hole gene sequence matrix Two methods are performed simultaneously. One method is to randomly select a position in the matrix and randomly generate a floating-point number within the parameter value range to overwrite the original data; the other is "digit reversal", that is, to swap the major and minor axes of the air hole; the mutation operation on the air hole gene sequence matrix is ​​performed simultaneously, where the probability pv1 of overwriting the matrix value and the probability pv2 of digit reversal are both set manually. The probability of mutation is the same as roulette selection, which is determined by comparing the size of the random number generated in the [0,1] interval. In this algorithm, both mutation probabilities are set to 0.01.

[0072] Obtain the structure of the individual with the best fitness in the previous generation and save it in the simulation file; use the structure generation function to write the new population into the simulation software for simulation solution, and count the number of optimizations; the new population is obtained based on the optimized individual structure; determine whether the number of optimizations is less than the number of iterations; if yes, proceed to the next round of optimization; if not, the optimization is completed, save all results and exit the main program.

[0073] It should be noted that the MI-GA algorithm performs crossover and mutation operations on the optimization parameters independently, and adopts a crossover and mutation method with matrix input to independently encode and store different structural parameters. The crossover and mutation operations of different coding segments are performed separately. Compared with the binary coding serial form of the classical genetic algorithm, which cannot represent the device configuration of the PCF sensor, it solves the Hamming cliff problem and ensures the ability to capture advantageous features during the algorithm optimization process.

[0074] For example, the parameters of the MI-GA algorithm are set as follows:

[0075]

[0076] The computer configuration used for optimization simulation is as follows: Inter(R)Xeon(R)Gold 5218CPU@2.3GHz; 256GB RAM; Matrox G200eW3 (Nuvoton) WDDM 1.2 graphics card. After 150 optimization iterations, the curve of the change of birefringence coefficient with the number of iterations is shown in the figure. Figure 4 As shown; the final PCF pressure sensor structure given by the MI-GA algorithm is as follows Figure 5 As shown;

[0077] The optimized PCF pressure sensor was subjected to numerical pressure simulation analysis, with the pressure increased from 0MPa to 720MPa, the pressure increase step was set to 40MPa, and the incident wavelength was 1550nm; the phase mode birefringence change rate PCF Pressure Sensor Characteristics Figure 6 As shown, it has excellent detection sensitivity; according to the phase mode birefringence sensitivity formula Calculation shows that the pressure sensing sensitivity of the sensor reaches 162.14rad / (MPa·m).

[0078] Finally, the robustness analysis of the MI-GA algorithm proposed in the present invention is carried out; the consistency of the optimization effect of the improved genetic algorithm used in the present invention is tested. When the initial population is 100 and the number of iterations is 150, 20 optimization simulations are independently performed. The box plot of the optimization results is as follows: Figure 7 As shown in the figure, the results show that the historical best birefringence coefficients obtained by 20 separate calculations are all within the confidence interval of 25% to 75%, and there are no outliers in the optimization data. In addition, the data obtained by multiple simulations are relatively concentrated, and most of the data are distributed near the mean. It is proved that the algorithm proposed in the present invention is insensitive to the initial structure in the optimization of the PCF pressure sensor, and the algorithm has excellent robustness.

[0079] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0080] The working principle of the present invention is as follows: the present invention obtains preset structural data; initializes the photonic crystal fiber structure according to the structural data; obtains specific design requirements; performs electromagnetic simulation on the device structure according to the specific requirements to obtain a structural parameter matrix; screens out individuals entering the reproduction link and optimizes the structure of the PCF sensor; constructs a simulation file according to the structural parameter matrix, and performs an iterative cycle; until the set number of iterations is reached.

[0081] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA is characterized by: include: Step S1: Obtaining preset structural data; Initialize the photonic crystal fiber structure according to the structure data; Step S2: Obtain specific design requirements; perform electromagnetic simulation on the device structure according to the specific requirements to obtain a structural parameter matrix; Step S3: Screening out individuals that enter the reproduction stage and optimizing the structure of the PCF sensor; Step S4: construct a simulation file according to the structural parameter matrix and perform an iterative cycle until the set number of iterations is reached; The sensitivity of the optimized PCF pressure sensor was evaluated.

2. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 1 is characterized in that: Initializing the photonic crystal fiber structure according to the structural data includes: Retrieve structural data; wherein the structural data includes: population number NP and number of iterations NG; The initial population lattice constant P is randomly generated according to the population number in the structural data initial ={p1, p2, ..., pn}; randomly generate the parameter matrix of the air hole according to the number of iterations in the structural data; The structure generation function is used to write the population number into the simulation software according to the coding information in the storage matrix, and the algorithm is initialized; wherein the coding information is generated by real number coding.

3. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 2 is characterized in that: The expression of the parameter matrix of the air hole is as follows: Among them, a and b are the air hole radii in two orthogonal directions of the air hole, respectively, and n is the structural dimension of PCF, that is, the number of air holes; the size of the parameter matrix is ​​expressed as D = size (a, b) × n.

4. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 2 is characterized in that: The encoding information is generated by real number encoding, including: Retrieve the initial structure of the initialized PCF sensor, analyze the population individuals in the initial structure, and obtain the values ​​of the optimized parameters; The numerical value of the optimized parameter is placed into the gene sequence of the corresponding individual to obtain the genetic information of the individual; the genetic information of the individual is used as the coding information.

5. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 1 is characterized in that: The device structure is subjected to electromagnetic simulation according to specific requirements to obtain a structural parameter matrix, including: Retrieve the specific requirements of the design; the specific requirements include size, drawing process, measurement range, and measurement accuracy; Determine the initial device structure of the PCF sensor according to the specific requirements of the design; input the device structure of the PCF sensor into the electromagnetic simulation software, solve several groups of effective refractive indexes of the PCF sensor, and construct several simulation files; wherein each group of effective refractive index values ​​is the effective refractive index in two different polarization directions; Call the structure function, use the structure function to calculate the fitness of several simulation files, and read and store the corresponding fitness matrix.

6. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 5 is characterized in that: The method of calculating the fitness of a plurality of simulation files by using a structure function includes: Retrieve the effective refractive index of the PCF sensor in two different polarization directions, marked as n eff1 and m eff2 ; Through the fitness function Fitness(i) = abs(n eff1 -n eff2 ) calculates the fitness of individuals in a population; where Fitness(i) represents the fitness of the i-th individual in the population.

7. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 1 is characterized in that: The step of selecting the individuals that have entered the reproduction stage and optimizing the structure of the PCF sensor includes: Obtaining diameter parameters of the air holes and the air holes to be optimized; wherein the diameter parameters include a major axis a and a minor axis b; Calculate the cumulative selection probability of individuals; select the individuals to be reproduced according to the cumulative selection probability of individuals; obtain a two-dimensional matrix after encoding the air holes to be optimized according to the reproduced individuals; wherein the size of the two-dimensional matrix is ​​2×n, n is the number of air holes to be optimized; the parameter data of the corresponding air holes are stored in the two-dimensional matrix.

8. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 7 is characterized in that: The step of calculating the cumulative selection probability of an individual comprises: Retrieve individuals from the population and use the formula Calculate the selection probability of an individual; where F i is the fitness function, and its general expression is F(X)={|f(x)|}; i=1,2,…,n; n is a positive integer, indicating the total number of individuals; By formula Calculate the cumulative selection probability of an individual; where j = 1, 2, …, i.

9. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 7 is characterized in that: The method of selecting individuals for reproduction according to the cumulative selection probability of the individuals includes: Get a floating point number in a preset interval; where the preset interval is [0,1]; the floating point number is obtained by random selection; Retrieve the cumulative selection probability of the individual; determine whether the floating point number is less than the cumulative selection probability; if yes, select the corresponding individual as the individual for reproduction; if no, proceed to the next round of cumulative selection probability q i+1 , until the individuals that reproduce in this round of the turntable are selected.

10. The design method of PCF pressure sensor for hazardous chemical warehouse based on MI-GA according to claim 1 is characterized in that: The construction of the simulation file according to the structural parameter matrix and the iterative cycle include: Obtain the structure of the individual with the best fitness in the previous generation and save it in the simulation file; use the structure generation function to write the new population into the simulation software for simulation solution, and count the number of optimizations; wherein the new population is obtained according to the optimized individual structure; Determine whether the number of optimizations is less than the number of iterations; if yes, proceed to the next round of optimization; if no, the optimization is completed, all results are saved and the main program is exited.

Images