Evolutionary and Swarm Intelligence in Optimization of α-Amylase From Bacillus velezensis Sp.

doi:10.21203/rs.3.rs-3270983/v1

Download PDF

Research Article

Evolutionary and Swarm Intelligence in Optimization of α-Amylase From Bacillus velezensis Sp.

https://doi.org/10.21203/rs.3.rs-3270983/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

To optimize α-amylase synthesis from fermented broth, this research employed evolutionary and swarm intelligence-based techniques such as genetic algorithm (GA) and particle swarm optimisation (PSO). The nonlinear response surface model (RSM) of α-amylase from Bacillus velezensis sp.was used as the objective function.

Results

In contrast to the enzyme activity, 418.25 U/mL, achieved by the thirteen factors OFAT technique was increased by 2.76 times using GA and PSO optimization with only nine significant parameters. The optimal process parameters predicted using GA and PSO were pH (5.37), temperature (34.18 ^oC), carbon source (4.12%), nitrogen source (2.04%), K₂PO₄ (0.34%), MgSO₄ (0.2%), NaCl (0.14%), fructose (2.0%), and NaNO₃ (0.47%).

Conclusions

The results showed that α-amylase activity was significantly improved with both approaches, with similar predictions of optimized process parameters and α-amylase activity of 1157.2 U/ml. PSO, on the other hand, surpasses GA in terms of optimized α-amylase activity and convergence rate, which might be attributed to its simple structure and efficient memory capabilities. In conclusion, the suggested GA and PSO techniques are thought to be natural and may be used instead of existing gradient-based optimization strategies in downstream enzyme processing.

Applied & Industrial Microbiology

α-amylase

Bacillus velezensis sp.

Genetic algorithm

Optimization

Particle swarm optimization

α-amylase is a starch-degrading enzyme with numerous commercial and pharmacological applications in various sectors, including washing, textiles, chocolate, detergent, and biofuel manufacturing. It is crucial in the development of digestive aids, the preparation of biodegradable polymers, and the use of cross-linked starches in tablets. This research aims to optimize microbial fermentation process variables in enzyme recovery from fermented broth using global evolutionary optimization algorithms, such as artificial and swarm intelligence-based approaches. Nature-inspired optimization techniques, such as ant-colony optimization (ACO), particle swarm optimization (PSO), and firefly optimization (FO), have been developed to analyze problems in computational studies. Evolutionary optimization techniques are used in physiochemical parametric studies, based on biological evolution of reproduction, mutation, and other factors. Swarm optimization techniques, such as particle swarm, beehive, and ant-colony optimization techniques, are considered robust, simple, and easier to handle. Genetic algorithm (GA) is the best explanation for microorganism growth, as it involves factors such as crossover or recombination and mutation. The statistical association between the ultimate goal (α-amylase activity) and each of the independent variables was determined using a second-order polynomial equation. GA is an evolutionary algorithm that mimics natural evolution and involves three operators: reproduction, crossover, and mutation. Reproduction selects good strings in a population and forms a mating pool. Crossover allows for new string formation by exchanging strings with another chromosome. Mutation perturbs the child vector, achieving local search and maintaining diversity. Scale-up studies using optimization techniques are crucial for better productivity, reduced cost, and time. Advanced techniques like ANN and nature-inspired optimization are more suitable for scaling up of SmF and SSF processes, as they help understand the growth of microorganisms and their impact on the overall process. The nonlinear response surface model (RSM) of α-amylase from Bacillus velezensis sp. was used as the objective function. Response surface methodology (RSM) optimization was applied to consider only nine significant process variables: pH, temperature, agitation, inoculum size, aeration, carbon source, nitrogen source, K₂PO4, MgSO₄, NaCl, incubation period, fructose, and NaNO3. This study aimed to optimize the nonlinear RSM model of α-amylase fermentation from Bacillus velezensis sp. using artificial intelligence-based GA and swarm intelligence-based PSO approaches. GA is an evolutionary algorithm that mimics natural evolution, involving reproduction, crossover, and mutation.

Materials and microorganism

All reagents used in this study were purchased from HiMedia Labs, C40, Road No. 21Y, MIDC, Wagle Industrial Area, Thane (West) − 400 604, Maharashtra, India. Bacillus velezensis NH-1 (CCTC M2017336), was obtained from School of Pharmaceutical Sciences, Lovely Professional University, Punjab, India. The chosen strain is purified using the streak plate technique and kept at 4^oC as slant cultures.

Methods

Submerged fermentation

In order to evaluate the growth pattern, Bacillus velezensis sp., a potential α-amylase-producing bacterial strain, is cultured in seed media containing Luria-Bertani broth. A 2% inoculum of the 16-hour-grown seed medium is added to basal media containing (g/L) starch 5.0, peptone 20.0, MgSO₄ 1.0, and K₂HPO₄ 3.0, and the incubation process is repeated. Every hour, turbidity at 600 nm and the 3,5-dinitro salicylic acid test (DNSA) are used to determine the rate of bacterial growth and the synthesis of the enzyme -amylase. To get the crude extract, which serves as an enzyme source for the α-amylase test, the medium is centrifuged at 5000 RPM for 15 minutes. Each test is run three times, and the average results are reported. The bacterium strain with the highest α -amylase activity is then cultured in a seed medium that has been specially designed for it.

Enzyme Assay

The 3, 5-dinitro salicylic acid test is used to measure α -amylase activity. At pH 5.5 and 55°C, one mL of 1% starch is incubated with 0.05 mL of enzyme-containing supernatant for 8 minutes before adding 0.5 mL DNSA to the reaction mixture. After 10 minutes in a boiling water bath, the reaction mixture is cooled and 3.45 mL of distilled water is added. The released reduced sugar is measured at 540 nm by using Miller's method. Blank is prepared without the use of enzyme and the concentration of protein is estimated by Lowry's method.

Optimization methodology

Statistical optimization of α-amylase production process parameters by RSM

The second-order polynomial response surface model RSM design was developed at three levels in the selection of each independent bioprocess parameter. Thirteen variables were found to be potent among the total parameters evaluated for their significance on α-amylase activity in the OFAT method (results are not shown here), namely pH, temperature, agitation, inoculum size, aeration, carbon source, nitrogen source, K₂PO₄, MgSO₄, NaCl, incubation duration, fructose, and NaNO₃. The response function was estimated by a second-degree polynomial with quadratic and interaction effects using the least squares approach (Rajulapati et al. 2011). The Definitive Screen (DSD) Design was used to assess pH (A), temperature (B), carbon source (C), nitrogen source (D), K₂PO₄ (E), MgSO₄ (F), NaCl (G), fructose (H), and NaNO₃ (I) ranges for experimental investigation. Table 1 shows the true ranges of coded factors based on the results of the OFAT technique (data is not provided). The statistical association between the ultimate goal (α-amylase activity) and each of the independent variables was determined using a second-order polynomial equation.

Y = 1085.35 + 1.15A + 0.25B + 1.0C + 1.5D + 17.5E + 6.0F + 17.5G + 1.0H + 5.0J-1.69A²-5.53C²-24.33D²-20.86E²-16.33F²-35.09G²-5.34H²-13.64J^{2 Eq. (1)}

Where, Y is the level of α-amylase activity.

Use of RSM produced the ensuing quadratic regression equation for the final response of α-amylase activity [Eq. (1)]. The ultimate objective's optimised bioprocess parameter values were identified as 5.20, 35.92, 4.22, 2.06, 0.34, 0.14, 0.23, 1.48, and 0.53, respectively, for pH, temperature, carbon source, nitrogen source, K₂PO₄, MgSO₄, fructose, and NaNO₃ .

Table 1

Variables used in experimental design
Factor Code	Name	Lower limit (-1)	Upper limit (+ 1)
A	pH	4.0	6.0
B	Temp(^oC)	32.0	36.0
C	Carbon source (%)	3.0	5.0
D	Nitrogen source (%)	1.0	3.0
E	K₂HPO₄ (%)	0.20	0.4
F	MgSO₄ (%)	0.05	0.2
G	NaCl (%)	0.1	0.3
H	Fructose (%)	1.0	2.0
J	NaNO₃ (%)	0.3	0.7

Evolutionary and swarm intelligence-based optimization

This study aims to optimize the nonlinear RSM model of α-amylase fermentation from Bacillus velezensis sp. using artificial intelligence-based GA and swarm intelligence-based PSO approaches.

Genetic algorithm (GA)

GA is an evolutionary algorithm that mimics natural evolution. It involves three operators: reproduction, crossover, and mutation. Reproduction selects good strings in a population and forms a mating pool. Crossover allows for new string formation by exchanging strings with another chromosome. Mutation perturbs the child vector, achieving local search and maintaining diversity. The process is repeated until a termination criterion is met. In this study, a binary-coded GA was used to optimize α-amylase extraction from fermented broth for enhanced α-amylase activity. The minimization problem is converted to a maximization problem using negative sign before enzyme activity

Particle swarm optimization (PSO)

The swarm intelligence-based optimization (PSO) approach, developed by Kennedy and Eberhart, overcomes drawbacks of generalized optimization (GA) such as convergence toward local optima and difficulty in dynamic sets. PSO involves a population of particles, each with a memory for its previous best position. It has advantages over GA, such as easier implementation, fewer parameters needed for adjustment, and higher memory capability. However, PSO can sometimes suffer from premature convergence, leading to suboptimal solutions. This study uses the multiobjective PSO-crowding distance (MOPSO-CD) approach to solve the optimization problem of α-amylase extraction.MOPSO-CD incorporates the crowding distance operator, which affects global best selection criteria by deleting non-dominated solutions. The algorithm's mutation operator is adapted for exploratory capability, initially performing mutations on the entire population and rapidly decreasing coverage over time to prevent premature convergence. The MOPSO-CD algorithm's working procedure is shown in a flow chart in Fig. 2.

To compute the new velocity, V[i]:

V[i] = W * V[i] + R₁ * [Pbest (i) - P (i)] + R₂ * [A(Gbest) - P (i)]

Where, W is the inertia weight, which is equal to 0.4, R1 and R2 are the random numbers in the range of (0–1), P_best(i) is the best position reached by particle i and A(G_best) is the global best guide for each dominated solution.

To calculate the new position of P[i]:

P[i] = P[i] + V[i]

The MOPSO-CD algorithm was used to optimize lipase extraction by exploring the nonlinear RSM model and particle size parameters.

Evolutionary and swarm intelligence-based optimization

α-amylase production process optimization using artificial and swarm intelligence-based algorithms enhances recovery and activity, offering flexibility, ease of operation, and global perspective.

Optimization by GA

GAs are an alternative to traditional optimization methods that struggle to predict optimal conditions in complex optimization problems with numerous local optima. The selection of population size, number of generations, mutation probability, and crossover mechanism play a crucial role in exploring the input space of the problem of interest by GA. In this study, the RSM model of α-amylase extraction is posed as an optimization problem for maximizing amylase activity. A systematic study was conducted to determine the GA parameters responsible for an optimal value of amylase activity. The results showed that the majority of the population was in the group of 1150.0–1160.0 U/ml, with over 75% consistently above this value. The optimum values of the process parameters were found to be pH (5.37), temperature (34.18 ^oC), carbon source (4.12%), nitrogen source (2.04%), K₂PO₄ (0.34%), MgSO₄ (0.2%), NaCl (0.14%), fructose (2.0%), and NaNO₃ (0.47%). The maximum amylase activity for Bacillus velezensis sp. was found to be 1157.2 U/ml, with the maximum α-amylase activity of Bacillus velezensis sp. being 1157.2 U/ml. This optimized set of variables was chosen for experimental model validation. The experimental amylase activity under the stated conditions was found to be 1155.1 U/ml, which is in good agreement with the GA-optimized value. The OVAT approach was used to obtain amylase activity of 418.25 U/mL, indicating a significant improvement (2.8 folds) in α-amylase activity.

Optimization by PSO

PSO is a computer-based approach that uses parameters like particle size, generation number, inertia weight, and cognitive and social components. Randomly generated particles with a specific size were used to generate optimal amylase activity. The optimal amylase activity of 1157.2 U/ml was obtained with the process variables of pH (5.37), temperature (34.18 ^oC), carbon source (4.12%), nitrogen source (2.04%), K₂PO₄ (0.34%), MgSO₄ (0.2%), NaCl (0.14%), fructose (2.0%), and NaNO₃ (0.47%). Confirmation experiments were conducted in triplicate with the PSO-predicted conditions, resulting in an experimental amylase activity of 1155.1 U/ml, which is in good agreement with the predicted value.

Comparison between GA and PSO

The study analyzed amylase extraction optimization using GA and PSO approaches in fermented broth. Results showed that PSO converges faster than GA, with an experimental amylase activity of 1157.57 U/mL, which is in agreement with the predicted value. The PSO approach's optimal amylase activity and parameters are slightly better than GA due to easier implementation, fewer parameters, and higher memory capability.

Table 2

Experimental and predicted lipase activity (U/ml) with optimal variable conditions of OVAT, GA and PSO approaches.
	α-amylase production process parameters									α-amylase acitivity (U/ml)
Approach	pH	Temp.	Carbon	Nitrogen	K₂HPO₄	MgSO₄	NaCl	Fructose	NaNO3	predicted	Experimental
OVAT	5	34	4	2	0.3	0.1	0.2	1.5	0.5	---	418.25
GA	5.38	34.18	4.12	2.04	0.34	0.2	0.14	2.0	0.47	1157.24	1155.13
PSO	5.38	34.18	4.12	2.04	0.34	0.2	0.14	2.0	0.47	1157.24	1155.15

The optimization of α-amylase production using evolutionary and swarm intelligence-based algorithms improves recovery and activity, offering flexibility, ease of operation, and global perspective. Evolutionary (GA) and swarm intelligence-based (PSO) optimization is an alternative to traditional optimization methods that struggle to predict optimal conditions in complex optimization problems with numerous local optima. In this study, the RSM model of α-amylase extraction was posed as an optimization problem for maximizing amylase activity. A systematic study was conducted to determine the GA parameters responsible for an optimal value of amylase activity. The majority of the population was in the group of 1150.0–1160.0 U/ml, with over 75% consistently above this value. The optimum values of process parameters were pH (5.37), temperature (34.18^oC), carbon source (4.12%), nitrogen source (2.04%), K₂PO₄ (0.34%), MgSO₄ (0.2%), NaCl (0.14%), fructose (2.0%), and NaNO₃ (0.47%). The maximum amylase activity for Bacillus velezensis sp. was found to be 1157.2 U/ml. Optimization by PSO is a computer-based approach that uses parameters like particle size, generation number, inertia weight, and cognitive and social components. Randomly generated particles with a specific size were used to generate optimal amylase activity. The optimal amylase activity of 1157.2 U/ml was obtained with the process variables of pH (5.37), temperature (34.18^oC), carbon source (4.12%), nitrogen source (2.04%), K₂PO₄ (0.34%), MgSO₄ (0.2%), NaCl (0.14%), fructose (2.0%), and NaNO₃ (0.47%). The PSO approach's optimal amylase activity and parameters are slightly better than GA due to easier implementation, fewer parameters, and higher memory capability.

Competing interests: The authors declare no competing interests.

Andrab SG, Hekmat A, Yusop ZB. A review: evolutionary computations (GA and PSO) in geotechnical engineering. Computational Water, Energy, and Environmental Engineering. 2017; 6(2):154–79. https://doi.org/10.4236/cweee.2017.62012
Bagheri M, Akbari A, Mirbagheri SA. Advanced control of membrane fouling in filtration systems using artificial intelligence and machine learning techniques: A critical review. Process Safety and Environmental Protection. 2019; 123:229–52. https://doi.org/10.1016/j.psep.2019.01.013
Papazoglou G, Biskas P. Review and Comparison of Genetic Algorithm and Particle Swarm Optimization in the Optimal Power Flow Problem. Energies. 2023; 16(3):1152. https://doi.org/10.3390/en16031152
Almanaa TN, Vijayaraghavan P, Alharbi NS, Kadaikunnan S, Khaled JM, Alyahya SA. Solid state fermentation of amylase production from Bacillus subtilis D19 using agro-residues. Journal of King Saud University-Science. 2020; 32(2):1555–61. https://doi.org/10.1016/j.jksus.2019.12.011
Iram N, Shakir HA, Irfan M, Khan M, Ali S, Anwer A, Saeed S, Qazi JI. Statistical optimization of amylase production using grape fruit peels in submerged fermentation. Acta Scientiarum. Technology. 2021; 43(1):1–1. https://doi.org/10.4025/actascitechnol.v43i1.50538
Majid MH, Arshad AM. An analysis of PSO inertia weight effect on swarm robot source searching efficiency. In2017 IEEE 2nd international conference on automatic control and intelligent systems (I2CACIS) 2017 Oct 21 (pp. 173–178). IEEE. https://doi.org/10.1109/I2CACIS.2017.8239053
Bozorg-Haddad O, Azarnivand A, Hosseini-Moghari SM, Loáiciga HA. Optimal operation of reservoir systems with the symbiotic organisms search (SOS) algorithm. Journal of Hydroinformatics. 2017; 19(4):507–21. https://doi.org/10.2166/hydro.2017.085
Abdel-Salam M, El-Mohandes MT, Yousef AM, Abdel-Hakim AE, Ramadan R. A GA-Based Method for Computing MOF-Weighting Coefficients Along with Optimal Location and Sizing of DG Sources in Distribution Systems. In2019 21st International Middle East Power Systems Conference (MEPCON) 2019 Dec 17 (pp. 796–802). IEEE. https://doi.org/10.1109/MEPCON47431.2019.9007946
MirRokni MK. Applying genetic algorithm in architecture and neural network training. International Journal of Computer Science and Network Security IJCSN. 2017; 17(6):118.
Shirodkar PV, Muraleedharan UD. Enhanced α-amylase production by a marine protist, Ulkenia sp. using response surface methodology and genetic algorithm. Preparative Biochemistry and Biotechnology. 2017; 47(10):1043–9. https://doi.org/10.1080/10826068.2017.1373293
García-Bautista I, García-Cruz U, Pacheco N, García-Maldonado JQ, Aguirre-Macedo ML. Optimization of the biodegradation of aliphatic, aromatic, and UCM hydrocarbons from light crude oil in marine sediment using response surface methodology (RSM). Bulletin of Environmental Contamination and Toxicology. 2022; 108(1):107–13. https://doi.org/10.1007/s00128-021-03281-w
Sholikin MM, Alifian MD, Jayanegara A. Optimization of the Hermetia illucens larvae extraction process with response surface modelling and its amino acid profile and antibacterial activity. InIOP Conference Series: Materials Science and Engineering 2019 Jun 1 (Vol. 546, No. 6, p. 062030). IOP Publishing. 10.1088/1757-899X/546/6/062030
Saad WF, Othman AM, Abdel-Fattah M, Ahmad MS. Response surface methodology as an approach for optimization of α-amylase production by the new isolated thermotolerant Bacillus licheniformis WF67 strain in submerged fermentation. Biocatalysis and Agricultural Biotechnology. 2021; 32:101944. https://doi.org/10.1016/j.bcab.2021.101944
Alahmad Aljammas H, Yazji S, Azizieh A. Optimization of protease production from Rhizomucor miehei Rm4 isolate under solid-state fermentation. Journal of Genetic Engineering and Biotechnology. 2022; 20(1):1–3. https://doi.org/10.1186/s43141-022-00358-9
Sahu R, Meghavarnam AK, Janakiraman S. Response surface methodology: An effective optimization strategy for enhanced production of nitrile hydratase (NHase) by Rhodococcus rhodochrous (RS-6). Heliyon. 2020; 6(10). https://doi.org/10.1016/j.heliyon.2020.e05111
Kangah J, Otoo H, Acquah J. A Hybrid Optimization Model for Vehicle Routing Problem, a Case Study at Zoomlion Ghana Limited, Shama District. Asian Research Journal of Mathematics. 2022; 18(11):148–61. https://doi.org/10.9734/arjom/2022/v18i1130432
Tong L, Li X, Hu J, Ren L. A PSO optimization scale-transformation stochastic-resonance algorithm with stability mutation operator. IEEE Access. 2017; 6:1167–76. https://doi.org/10.1109/ACCESS.2017.2778022
Sun W, Lin A, Yu H, Liang Q, Wu G. All-dimension neighborhood based particle swarm optimization with randomly selected neighbors. Information Sciences. 2017; 405:141–56. https://doi.org/10.1016/j.ins.2017.04.007
Wu Q, Zhou L, Sun X, Yan Z, Hu C, Wu J, Xu L, Li X, Liu H, Yin P, Li K. Altered lipid metabolism in recovered SARS patients twelve years after infection. Scientific reports. 2017; 7(1):9110.
Kangah JK, Appati JK, Darkwah KF, Soli MA. Implementation of an H-PSOGA optimization model for vehicle routing problem. International Journal of Applied Metaheuristic Computing (IJAMC). 2021; 12(3):148–62. 10.4018/IJAMC.2021070106
Shankar R, Ganesh N, Čep R, Narayanan RC, Pal S, Kalita K. Hybridized particle swarm—Gravitational search algorithm for process optimization. Processes. 2022; 10(3):616. https://doi.org/10.3390/pr10030616
Singh N, Singh SB, Singh S. Solution of Bio-Medical Problem by Genetic Algorithm. J Biomedical Sci. 2018; 7(1):2. DOI: 10.4172/2254-609X.100081
Sun T, Xu MH. A swarm optimization genetic algorithm based on quantum-behaved particle swarm optimization. Computational intelligence and neuroscience. 2017; 2017.https://doi.org/10.1155/2017/2782679
Xiaohui D, Huapeng L, Yong L, Ji Y, Shuqing Z. Comparison of swarm intelligence algorithms for optimized band selection of hyperspectral remote sensing image. Open Geosciences. 2020; 12(1):425–42. https://doi.org/10.1515/geo-2020-0155
Zhang, F. and Zhang, Y., 2021. A multi-objective optimization prediction approach for water resources based on swarm intelligence. Earth Science Informatics, 14, pp.457–468. https://doi.org/10.1007/s12145-020-00521-1
AL-Allaf ON. Particle Swarm Optimization Algorithm vs. Genetic Algorithm for Image Watermarking Based Discrete Wavelet Transform. International Journal of Computer and Information Engineering. 2017; 11(9):1106–12.
Reddy BC, Reddy PV, Rajeshwari M, Kavya YS. Correlation of GA and PSO for Analysis of Efficient Optimization. International Journal for Advance Research and Development. 2017; 2(4).
Study on Clarification of Apple Juice using Enzymes Sonam Lhamo, Sonam Tobgay, Dhan Maya & Sonam Dek https://doi.org/10.55925/btagr.22.5115
Abdin M, Hamed YS, Akhtar HM, Chen D, Mukhtar S, Wan P, Riaz A, Zeng X. Extraction optimisation, antioxidant activity and inhibition on α-amylase and pancreatic lipase of polyphenols from the seeds of Syzygium cumini. International journal of food science & technology. 2019; 54(6):2084–93. https://doi.org/10.1111/ijfs.14112
Geng W, Jia L. Hybrid Genetic Particle Swarm Optimization Based Economical Operation of Energy Hub. In2020 5th International Conference on Power and Renewable Energy (ICPRE) 2020 Sep 12 (pp. 184–188). IEEE. https://doi.org/10.1109/ICPRE51194.2020.9233191
Abushawish A, Jarndal A. Hybrid PSO-GWO optimization based parameter extraction method applied to GaN devices. In2021 4th International Conference on Circuits, Systems and Simulation (ICCSS) 2021 May 26 (pp. 38–44). IEEE. https://doi.org/10.1109/ICCSS51193.2021.9464208
Priyanka BS, Rastogi NK. Selective extraction of lipase and amylase from enzyme mixture by employing liquid emulsion membrane. Biotechnology Progress. 2018; 34(3):721–9. https://doi.org/10.1002/btpr.2624
Niu C, Zheng F, Li Y, Liu C, Li Q. Process optimization of the extraction condition of β-amylase from brewer's malt and its application in the maltose syrup production. Biotechnology and Applied Biochemistry. 2018; 65(4):639–47. https://doi.org/10.1002/bab.1650
Ousaadi MI, Merouane F, Berkani M, Almomani F, Vasseghian Y, Kitouni M. Valorization and optimization of agro-industrial orange waste for the production of enzyme by halophilic Streptomyces sp. Environmental Research. 2021; 201:111494. https://doi.org/10.1016/j.envres.2021.111494

Download PDF

Version 1

posted

You are reading this latest preprint version

Evolutionary and Swarm Intelligence in Optimization of α-Amylase From Bacillus velezensis Sp.

Status:

Version 1

Abstract

Background

Results

Conclusions

Figures

Background

Materials And Methods

Materials and microorganism

Methods

Submerged fermentation

Enzyme Assay

Optimization methodology

Statistical optimization of α-amylase production process parameters by RSM

Evolutionary and swarm intelligence-based optimization

Results

Evolutionary and swarm intelligence-based optimization

Optimization by GA

Optimization by PSO

Discussion

Comparison between GA and PSO

Conclusion

Declarations

References

Status:

Version 1