A path planning method based on the particle swarm optimization trained fuzzy neural network algorithm

Cluster Computing - The basic fuzzy neural network algorithm has slow convergence and large amount of calculation, so this paper designed a particle swarm optimization trained fuzzy neural network...     

Culture conditions optimization of hyaluronic acid production by Streptococcus zooepidemicus based on radial basis function neural network and quantum-behaved particle swarm optimization algorithm

This study aimed to optimize the culture conditions (agitation speed, aeration rate and stirrer number) of hyaluronic acid production by Streptococcus zooepidemicus. Two optimization algorithms were used for comparison: response surface methodology (RSM) and radial basis function neural network coupling quantum-behaved particle swarm optimization algorithm (RBF-QPSO). In RBF-QPSO approach, RBF is employed to model the microbial HA production and QPSO algorithm is used to find the optimal culture conditions with the established RBF estimator as the objective function. The predicted maximum HA yield by RSM and RBF-QPSO was 5.27 and 5.62 g/l, respectively, while a maximum HA yield of 5.21 and 5.58 g/l was achieved in the validation experiments under the optimal culture conditions obtained by RSM and RBF-QPSO, respectively. It was indicated that both models provided similar quality predictions for the above three independent variables in terms of HA yield, but RBF model gives a slightly better fit to the measured data compared to RSM model. This work shows that the combination of RBF neural network with QPSO algorithm has good predictability and accuracy for bioprocess optimization and may be helpful to the other industrial bioprocesses optimization to improve productivity.     

Medium optimization for chitinase production from Trichoderma virens using central composite design

Medium development for chitinase production by Trichoderma virens was first carried out using conventional method of one-factor-at-a-time. The medium was further optimized using Central Composite Design in which response surface was generated later from the derived model. An experimental design of four variables including various initial pH values, chitin, ammonium sulphate, and methanol concentrations were created using Design Expert® Software, Version 6.0. The design consists of 30 experiments, which include 6 replicates at center points. The optimal value for each variable are 3.0 g/L, chitin; 0.1 g/L, ammonium sulphate; 0.4% (v/v), methanol; and initial pH, 4.0 with predicted chitinase activity of 0.1495 U/mL. These predicted parameters were tested in the laboratory and the final chitinase activity obtained was 0.1471 U/mL, which is almost reaching the predicted value. The optimal medium design showed an improvement of chitinase activity of 80.9% compared to activity obtained from the original Absidia medium composition.     

Application of artificial neural network coupling particle swarm optimization algorithm to biocatalytic production of GABA

The biotransformation of L-sodium glutamate (L-MSG) to gamma-aminobutyric acid (GABA) catalyzed by the cells of Lactobacillus brevis with higher glutamate decarboxylase activity was investigated. The results showed that pH, temperature, and FeSO(4) x 7H(2)O concentration had significantly positive effect on GABA yield. The individual and interactive effects of pH, temperature, and FeSO(4) x 7H(2)O concentration were further optimized in terms of GABA yield. In the present work, an artificial neural network (ANN) and response surface methodology (RSM) models were developed, which incorporated pH, temperature, and FeSO(4) x 7H(2)O concentration as input variables, and GABA yield as output variable. The optimized ANN topology included four neurons in the hidden layer and the best network architecture was 3-4-1. The trained ANN gave total root-mean square error (sigma) equal to 1.84 for GABA yield while the RSM gave sigma equal to 2.63. The results demonstrated a slightly higher prediction accuracy of ANN compared to RSM. The modeled maximum GABA yield was identified by applying particle swarm optimization algorithm to the ANN model developed. The modeled maximum GABA yield reached 91 mM under the following optimal conditions: 25 mL Na(2)HPO(4)-citric acid buffer (100 mM, pH 4.23), 120 mM L-MSG, 0.83 g/L FeSO(4) x 7H(2)O, 10 microM PLP, the resting cells obtained from a 60-h culture broth, 2.68 g dry cell weight (DCW)/L, and without agitation at 40 degrees C for 5 h. The previous high value of GABA yield that was observed was 81.8 mM. The optimized conditions allowed GABA yield to be increased from 81.8 to 90.57 mM after verification experiments test.     

Inference of genetic regulatory networks with recurrent neural network models using particle swarm optimization

Genetic regulatory network inference is critically important for revealing fundamental cellular processes, investigating gene functions, and understanding their relations. The availability of time series gene expression data makes it possible to investigate the gene activities of whole genomes, rather than those of only a pair of genes or among several genes. However, current computational methods do not sufficiently consider the temporal behavior of this type of data and lack the capability to capture the complex nonlinear system dynamics. We propose a recurrent neural network (RNN) and particle swarm optimization (PSO) approach to infer genetic regulatory networks from time series gene expression data. Under this framework, gene interaction is explained through a connection weight matrix. Based on the fact that the measured time points are limited and the assumption that the genetic networks are usually sparsely connected, we present a PSO-based search algorithm to unveil potential genetic network constructions that fit well with the time series data and explore possible gene interactions. Furthermore, PSO is used to train the RNN and determine the network parameters. Our approach has been applied to both synthetic and real data sets. The results demonstrate that the RNN/PSO can provide meaningful insights in understanding the nonlinear dynamics of the gene expression time series and revealing potential regulatory interactions between genes.     

Statistical optimization of culture conditions for bacterial cellulose production using Box-Behnken design

Culture conditions in a jar fermentor for bacterial cellulose (BC) production from A. xylinum BPR2001 were optimized by statistical analysis using Box-Behnken design. Response surface methodology was used to predict the levels of the factors, fructose (X1), corn steep liquor (CSL) (X2), dissolved oxygen (DO) (X3), and agar concentration (X4). Total 27 experimental runs by combination of each factor were carried out in a 10-L jar fermentor, and a three-dimensional response surface was generated to determine the effect of the factors and to find out the optimum concentration of each factor for maximum BC production and BC yield. The fructose and agar concentration highly influenced the BC production and BC yield. However, the optimum conditions according to changes in CSL and DO concentrations were predicted at almost central values of tested ranges. The predicted results showed that BC production was 14.3 g/L under the condition of 4.99% fructose, 2.85% CSL, 28.33% DO, and 0.38% agar concentration. On the other hand, BC yield was predicted in 0.34 g/g under the condition of 3.63% fructose, 2.90% CSL, 31.14% DO, and 0.42% agar concentration. Under optimized culture conditions, improvement of BC production and BC yield were experimentally confirmed, which increased 76% and 57%, respectively, compared to BC production and BC yield before optimizing the culture conditions.     

Statistical optimization of culture conditions for 1,3-propanediol by Klebsiella pneumoniae AC 15 via central composite design

A central composite design was used to study the effect of glycerol, rate of stirring, air aeration and pH on the synthesis of 1,3-propanediol (1,3-PD) by Klebsiella pneumoniae AC 15. Among the four variables, glycerol and rate of stirring significantly affected 1,3-PD productivity, whereas air aeration and pH were not effective. A quadratic polynomial equation was obtained for 1,3-PD productivity by multiple regression analysis using response surface methodology. The validation experimental confirmed with the predicted model. The optimum combinations for 1,3-PD productivity was glycerol, rate of stirring, air aeration, and pH of 50 g/l, 318 rpm, 0.6 vvm, 6.48, respectively. The subsequent fed batch experiments produced 1,3-PD of 70 g/l at a fermentation of 30 h.     

Optimization for high-density cultivation of heterotrophic Chlorella based on a hybrid neural network model

AIMS: The purpose of this study was to develop a reliable hybrid neural network (HNN) model for heterotrophic growth of Chlorella, based on which optimization for fed-batch (FB) cultivation of Chlorella may be successfully realized. METHODS AND RESULTS: Deterministic kinetic model was preliminarily developed for the optimization of FB cultivation of Chlorella. The highest biomass concentration and the maximum productivity were obtained as: 104.9 g l(-1) dry cell weight and 0.613 g l(-1) h(-1), respectively. After several cultivations had been performed, an HNN model was developed. The efficiency of biomass production was further increased by the optimization using this model. The highest biomass concentration and the maximum productivity attained was: 116.2 g l(-1) dry cell weight and 1.020 g l(-1) h(-1), respectively. CONCLUSION: The HNN model agreed well with experimental results in different cultivations. Comparison between the HNN model and the deterministic model showed that the former had better generalization ability, which made it a reliable tool in modelling and optimization. SIGNIFICANCE AND IMPACT OF THE STUDY: The high cell density and productivity of biomass obtained in this study is of significance for the commercial cultivation of Chlorella. The simple and efficient optimization strategy proposed in this paper may be employed in heterotrophic mass culture of Chlorella as well as other similar organisms.     

Modeling and optimization of microbial hyaluronic acid production by Streptococcus zooepidemicus using radial basis function neural network coupling quantum‐behaved particle swarm optimization algorithm

Hyaluronic acid (HA) is a natural biopolymer with unique physiochemical and biological properties and finds a wide range of applications in biomedical and cosmetic fields. It is important to increase HA production to meet the increasing HA market demand. This work is aimed to model and optimize the amino acids addition to enhance HA production of Streptococcus zooepidemicus with radial basis function (RBF) neural network coupling quantum‐behaved particle swarm optimization (QPSO) algorithm. In the RBF‐QPSO approach, RBF neural network is used as a bioprocess modeling tool and QPSO algorithm is applied to conduct the optimization with the established RBF neural network black model as the objective function. The predicted maximum HA yield was 6.92 g/L under the following conditions: arginine 0.062 g/L, cysteine 0.036 g/L, and lysine 0.043 g/L. The optimal amino acids addition allowed HA yield increased from 5.0 g/L of the control to 6.7 g/L in the validation experiments. Moreover, the modeling and optimization capacity of the RBF‐QPSO approach was compared with that of response surface methodology (RSM). It was indicated that the RBF‐QPSO approach gave a slightly better modeling and optimization result compared with RSM. The developed RBF‐QPSO approach in this work may be helpful for the modeling and optimization of the other multivariable, nonlinear, time‐variant bioprocesses. © 2009 American Institute of Chemical Engineers Biotechnol. Prog., 2009     

Modeling and optimization of lipase-catalyzed production of succinic acid ester using central composite design analysis

Esterification of succinic acid with oleyl alcohol catalyzed by immobilized Candida antarctica lipase B (Novozym 435) was investigated in this study. Response surface methodology (RSM) based on a five-level, four-variable central composite design (CCD) was used to model and analyze the reaction. A total of 21 experiments representing different combinations of the four parameters including temperature (35–65°C), time (30–450 min), enzyme amount (20-400 mg), and alcohol:acid molar ratio (1:1-8:1) were generated. A partial cubic equation could accurately model the response surface with a R² of 0.9853. The effect and interactions of the variables on the ester synthesis were also studied. Temperature was found to be the most significant parameter that influenced the succinate ester synthesis. At the optimal conditions of 41.1°C, 272.8 min, 20 mg enzyme amount and 7.8:1 alcohol:acid molar ratio, the esterification percentage was 85.0%. The model can present a rapid means for estimating the conversion yield of succinate ester within the selected ranges.     

Rational optimization of culture conditions for the most efficient ethanol production in Scheffersomyces stipitis using design of experiments

Optimization of culture parameters for achieving the most efficient ethanol fermentation is challenging due to multiple variables involved. Here we presented a rationalized methodology for multi‐variables optimization through the design of experiments DoE approach. Three critical parameters, pH, temperature, and agitation speed, affecting ethanol fermentation in S. stipitis was investigated. A predictive model showed that agitation speed significantly affected ethanol synthesis. Reducing pH and temperature also improved ethanol production. The model identified the optimum culture conditions for the most efficient ethanol production with the yield and productivity of 0.46 g/g and 0.28 g/l h, respectively, which is consistent with experimental observation. The results also indicated the scalability of the model from shake flask to bioreactor. Thus, DoE is a promising tool permitting the rapid establishment of culture conditions for the most efficient ethanol fermentation in S. stipitis. The approach could be useful to reduce process development time in lignocellulosic ethanol industry. © 2012 American Institute of Chemical Engineers Biotechnol. Prog., 2012     

Optimization of fermentation conditions for the production of human soluble catechol-O-methyltransferase by Escherichia coli using artificial neural network

The aim of this work was to optimize the temperature, pH and stirring rate of the production of human soluble catechol-O-methyltransferase (hSCOMT) in a batch Escherichia coli culture process. A central composite design (CCD) was firstly employed to design the experimental assays used in the evaluation of these operational parameters on the hSCOMT activity for a semi-defined and complex medium. Predictive artificial neural network (ANN) models of the hSCOMT activity as function of the combined effects of these variables was proposed based on this exploratory experiments performed for the two culture media. The regression coefficients (R(2)) for the final models were 0.980 and 0.983 for the semi-defined and complex medium, respectively. The ANN models predicted a maximum hSCOMT activity of 183.73 nmol/h, at 40 °C, pH 6.5 and stirring rate of 351 rpm, and 132.90 nmol/h, at 35 °C, pH 6.2 and stirring rate of 351 rpm, for semi-defined and complex medium, respectively. These results represent a 4-fold increase in total hSCOMT activity by comparison to the standard operational conditions used for this bioprocess at slight scale.     

Wind power prediction based on neural network with optimization of adaptive multi-group salp swarm algorithm

Salp swarm algorithm (SSA) is a swarm intelligence algorithm inspired by the swarm behavior of salps in oceans. In this paper, a adaptive multi-group salp swarm algorithm (AMSSA) with three new communication strategies is presented. Adaptive multi-group mechanism is to evenly divide the initial population into several subgroups, and then exchange information among subgroups after each adaptive iteration. Communication strategy is also an important part of adaptive multi-group mechanism. This paper proposes three new communication strategies and focuses on promoting the performance of SSA. These measures significantly improve the cooperative ability of SSA, accelerate convergence speed, and avoid easily falling into local optimum. And the benchmark functions confirm that AMSSA is better than the original SSA in exploration and exploitation. In addition, AMSSA is combined with prediction of wind power based on back propagation (AMSSA-BP) neural network. The simulation results show that the AMSSA-BP neural network prediction model can achieve a better prediction effect of wind power.

     

Determining the optimum model parameters for oligosaccharide production efficiency using response surface integrated particle swarm optimization method: an experimental validation study

Abstract

Glucansucrases (GTFs) catalyzes the synthesis of α-glucans from sucrose and oligosaccharides in the presence of an acceptor sugar by transferring glucosyl units to the acceptor molecule with different linkages. The acceptor reactions can be affected by several parameters and this study aimed to determine the optimal reaction parameters for the production of glucansucrase-based oligosaccharides using sucrose and maltose as the donor and acceptor sugars, respectively via a hybrid technique of Response Surface Method (RSM) and Particle Swarm Optimization (PSO). The experimental design was performed using Central Composite Design and the tested parameters were enzyme concentration, acceptor:donor ratio and the reaction period. The optimization studies showed that enzyme concentration was the most effective parameter for the final oligosaccharides yields. The optimal values of the significant parameters determined for enzyme concentration and acceptor:donor ratio were 3.45?U and 0.62, respectively. Even the response surface plots for input parameters verified the PSO results, an experimental validation study was performed for the reverification. The experimental verification results obtained were also consistent with the PSO results. These findings will help our understanding in the role of different parameters for the production of oligosaccharides in the acceptor reactions of GTFs.     

Optimization of a capillary zone electrophoresis method by using a central composite factorial design for the determination of codeine and paracetamol in pharmaceuticals

Capillary zone electrophoresis was optimized to quantitatively determine codeine and paracetamol via central composite factorial design. Critical parameters (concentration, buffer, pH, voltage) assessed effects on resolution, analysis time and efficiencies. Optimum separation conditions were achieved using phosphate buffer 20 mM (pH 6.8) and voltage (15 kV). The optimized procedure easily determined codeine and paracetamol with separation in less than 3 min. Calibration curves (R > 0.999) were prepared, with LODs of 13.5 and 340 ng mL(-1) for codeine and paracetamol, respectively, and a good R.S.D.% (<3%). This method was applied to determine codeine and paracetamol in pharmaceutical formulations; recoveries coincided with stated contents.     

Optimization of the culture conditions for the production of a bacteriocin from halophilic archaeon Sech7a

An extremely halophilic haloarchaeon Sech7a, isolated from a solar saltern, was found to excrete halocin, a bacteriocin like substance. Optimal antimicrobial activity was obtained at 45 degrees C using 0.5% (w/v) glycerol and 0.5% (w/v) yeast extract as nutrients in SW media containing 3.4 M NaCl with pH value 7.5. Halocin Sech7a is a 10.7-kDa polypeptide, which is stable in a wide range of pH and is thermolabile at temperatures above 80 degrees C. As many other halophilic proteins, halocin Sech7a loses part of its activity upon exposure to low salt conditions, yet its activity can be restored after dialysis against initial saline conditions. Microscopic inspection revealed swelling and lysis of sensitive cells upon exposure to halocin Sech7a. These results indicate that haloarchaeon Sech7a excretes a novel bacteriocin.     

Optimisation of a microwave-assisted method for extracting withaferin A from Withania somnifera Dunal. using central composite design

Introduction – Recently, there have been growing attention on the modification and optimisation of new extraction and quantification methods, caused by the lack of environmentally friendly methodologies for the extraction of phytochemicals from complex matrices. In the case of pharmaceutical compounds, not only the extraction procedure but also the analysis method should be efficient, precise, fast and easy. Objectives – The essential pharmaceutical characteristics and trace concentration of withanolides led us to modify and optimise the previously reported extraction and quantification procedure for withaferin A (WA) as a candidate for withanolides. Matrial and methods – The WA from the air‐dried aerial part of Withania somnifera Dunal. was extracted using a microwave‐assisted extraction (MAE) technique. Four variables affecting the extraction procedure were optimised using the central composite design approach. The method of high‐performance thin‐layer chromatography assay was validated and applied for the quantification of each experiment. Results – The optimum values of factors were: extraction time (150 s), extraction temperature (68°C) and 17 mL of methanol : water in the ratio 25 : 75 as extracting solvent. The solvent system consisted of ethyl acetate : toluene : formic acid : 2‐propanol (7.0 : 2.0 : 0.5 : 0.5, v/v/v/v), and densitometric scanning at 220 nm was applied for the analysis. The dynamic linear range, LOD, LOQ and recovery with the inter‐day, and intra‐day RSDs of the developed method indicated the validity of the method. Conclusion – A pressurised MAE method for extracting WA from the plant's aerial part was optimised using factorial‐based design. The net effect of time, temperature, solvent volume and its ratio suggests that the yield of WA increases until each factor reaches its optimum value, and decreases with further increase in temperature or solvent ratio. Copyright © 2010 John Wiley & Sons, Ltd.     

Optimization of culture parameters for extracellular protease production from a newly isolated Pseudomonas sp. using response surface and artificial neural network models

Radial basis function (RBF) artificial neural network (ANN) and response surface methodology (RSM) were used to build a predictive model of the combined effects of independent variables (pH, temperature, inoculum volume) for extracellular protease production from a newly isolated Pseudomonas sp. The optimum operating conditions obtained from the quadratic form of the RSM and ANN models were pH 7.6, temperature 38 °C, and inoculum volume of 1.5 with 58.5 U/ml of predicted protease activity within 24 h of incubation. The normalized percentage mean squared error obtained from ANN and RSM models were 0.05 and 0.1%, respectively. The results demonstrated an higher prediction accuracy of ANN compared to RSM. This superiority of ANN over other multi factorial approaches could make this estimation technique a very helpful tool for fermentation monitoring and control.     

Optimization of a single HPLC-PDA method for quantifying Metformin,Gliclazide, Pioglitazone,Dapagliflozin, Empagliflozin,Saxagliptin, Linagliptin and Teneligliptin using central composite design

1. Download : Download high-res image (127KB)
2. Download : Download full-size image

     

Optimization of the hydrolysis of lignocellulosic residues by using radial basis functions modeling and particle swarm optimization

The concentrations of glucose and total reducing sugars obtained by chemical hydrolysis of three different lignocellulosic feedstocks were maximized. Two response surface methodologies were applied to model the amount of sugars produced: (1) classical quadratic least-squares fit (QLS), and (2) artificial neural networks based on radial basis functions (RBF). The results obtained by applying RBF were more reliable and better statistical parameters were obtained. Depending on the type of biomass, different results were obtained. Improvements in fit between 35% and 55% were obtained when comparing the coefficients of determination (R²) computed for both QLS and RBF methods. Coupling the obtained RBF models with particle swarm optimization to calculate the global desirability function, allowed to perform multiple response optimization. The predicted optimal conditions were confirmed by carrying out independent experiments.