Response Surface Methodology Genetic Algorithm Research Articles

OPTIMIZATION OF DUAL TRANSESTERIFICATION OF JATROPHA SEED OIL TO BIOLUBRICANT USING HYBRIDIZED RESPONSE SURFACE METHODOLOGY (RSM) AND ADAPTIVE NEURO FUZZY INFERENCE SYSTEM (ANFIS)-GENETIC ALGORITHM (GA)

The over-reliance of the industrial and automobile sectors on petroleum-based lubricants, the feedstocks of which pose environmental challenges, has generated the need for sustainable alternatives in order to promote economic development and a sustainable green environment. The study investigated the optimization of process variables for the dual transesterification of jatropha seed oil into a biolubricant using a hybridized response surface methodology-genetic algorithm (RSM-GA) and an adaptive neuro-fuzzy inference system-genetic algorithm (ANFIS-GA). The seed oil was extracted using a Soxhlet extractor and characterized for its physicochemical properties. The catalyst for the reaction was synthesized by the acid modification of clay. The experimental design was created using Design Expert, and process parameters were optimized using RSM-GA and ANFIS-GA. The yield of oil was 56%, and its properties did not impede the catalyst from transesterification without pretreatment. The modified clay effectively converted the jatropha seed oil into a biolubricant. The ANFIS-GA model attained the highest yields (92.36%) under the optimal parameters of 3h reaction time, 120oC reaction temperature, 3% wt catalyst dosage, 5:1 TMP/JSOME molar ratio, and 300rpm agitation speed. Therefore, the incorporation of ANFIS and RSM with GA was more efficient in optimizing and predicting the biolubricant yield.

Optimizing the ECAP processing parameters of pure Cu through experimental, finite element, and response surface approaches

Abstract The main aim of the current work is to investigate the effect of equal channel angular pressing (ECAP) processing parameters, namely, number of passes, ECAP die angle, route type, and processing temperature on the mechanical and electrical properties of pure copper (Cu). The finite element method was used to simulate the homogeneity of stress and plastic strain distribution during ECAP processing. The response surface methodology (RSM) was used to identify the optimum ECAP processing parameters by analyzing the impact of ECAP conditions on responses. A second-order regression model and analysis of variance were created to analyze the ECAP condition of optimum responses. A genetic algorithm (GA) was also applied to optimize the ECAP condition. Finally, a hybrid RSM–GA was created to improve the optimization of ECAP responses and corresponding conditions evaluated using GA. The developed models were validated and compared with the experimental findings to prove that they are reliable as predictive tools. The optimization findings revealed that route Bc was more effective in improving the hardness, yield stress, ductility, and impact energy whereas route A was more effective in improving the ultimate tensile strength and the electrical conductivity of the Cu billets. Furthermore, the optimum die angle, number of passes, and processing temperature for the mechanical and electrical properties were also identified individually.

An investigation of hybrid models FEA coupled with AHP-ELECTRE, RSM-GA, and ANN-GA into the process parameter optimization of high-quality deep-drawn cylindrical copper cups

Deep drawing is one of the primary sheet metal forming processes that is used all over the world. The current study focused on using Analytical Hierarchy Process-Elimination and Choice Translating the Reality (AHP-ELECTRE I), Response Surface Methodology-Genetic Algorithm (RSM-GA), and Artificial Neural Network-Genetic Algorithm (ANN–GA) for determining deep–drawing performance parameters. A hybrid FEA–MCDA–RSM–ANN–GA was built using an experimental design obtained from RSM to develop better quality products. It is integrated with finite element-based numerical deep drawing simulation to understand the intended responses and the impact of design factors without the need for costly trial tests. To improve the quality of drawn cups characterization, three process parameters-clearance, punch radius, and coefficient of friction-have been tuned to their optimum values like resultant tool force (N), spring back (µm), max forming limit curve (%), and max thinning rate. The optimization results showed the efficacy of the technique for process design, resulting in the reduction of both cost and time. The desirability index was calculated and compared with all the predictions. The hybrid models that have been developed may be suggested for accurate prediction and optimization of various process parameters and outcomes for any industrial application issues that could arise.

Implementation of hybrid RSM-GA optimization techniques in underwater friction stir welding

Standard friction stir welding process parameters have a considerable impact on the quality of functional parts produced by underwater friction stir welding (UWFSW) with additive water. Hybrid statistical techniques may be used to optimize operating parameters in order to improve the aim function. The tensile strength (UTS)of parts fabricated with UWFSW by Al 6063 material in accordance with ASTMD638-14 tests is investigated in this study. In the construction of test specimens, three parameters were varied: rotational speed from 1000 to 1800 rpm, travel speed from 4 to 10 mm/s, and shoulder diameter from 10 to 20 mm. The response surface methodology (RSM) based central composite design (CCD) matrix for the parametric combination was constructed using a second-order polynomial fitting model. The maximum UTS of testing samples on the 201T universal testing machine (UTM) was 208.27 MPa. These process parameters are also optimized using hybrid optimization approaches such as response surface methodology- genetic algorithm (RSM-GA). RSM-GA had the highest precision of 98.99 percent, which resulted in optimal characteristics such as rotating speed 1800 rpm, travelling speed 4 mm/s, and shoulder diameter 15 mm, which resulted in a maximum tensile strength of 199.0212 MPa.

Parametric analysis and optimization of fused deposition modeling technique for dynamic mechanical properties of acrylic butadiene styrene parts

Fused deposition modeling (FDM) technology is catching the fast global market in the real-time production of polymeric parts. Process variables highly influence the performance characteristics of FDM-generated parts, so mechanical performance is not perfect for all applications. In actual conditions, parts produced by FDM are constantly subjected to loading at different temperatures. The former studies mainly concentrated on the properties of FDM products to static loading environments. There is a scope of effective investigation on the influence of FDM processing conditions on dynamic mechanical properties using artificial intelligence (AI) based techniques. The present study focused on investigation and optimization the manufacturing process parameters to evaluate the dynamic mechanical performance of FDM-produced part. The experimental runs were obtained through central composite design in Minitab software. A DMA8000 instrument was used to test the specimens for dynamic mechanical performance. The mathematical models were developed and optimized through different approaches like response surface methodology-genetic algorithm (RSM-GA) and artificial neural network-genetic algorithm (ANN-GA). The techniques for order preference by similarity to an ideal solution (TOPSIS) is employed to obtain the best parameter settings from sets of optimized solutions. The sequential use of ANN-GA and TOPSIS methods predicted the highest values of storage modulus 1619.61 MPa and loss modulus 257.38 MPa corresponding to 68.94° raster angle, 81.48% infill density, 0.10 mm layer thickness, 237.73°C nozzle temperature and 38.97 mm/s print head speed. The confirmation tests were conducted to validate the predicted result that upscale the desired properties. The RSM-GA-TOPSIS occurred with a prediction error of 2.40% and −3.31%, corresponding to storage and loss modulus. Similarly, ANN-GA-TOPSIS shows 2.17% and 2.89% prediction error corresponding to storage and loss modulus. The experimental and analytical outcome of present study will be helpful for the designers of intricate functional parts which come under thermo-mechanical loading conditions.

Combustion exhaust release impact, diesel engine performance, and optimization studies of green diesel-petrodiesel blend in a high compression ratio direct-injection compression-ignition engine

Reports on the optimum brake-power, surrogate fuel, engine emissions, and efficiency using hybrid model on high compression ratio diesel engines are very imperative for effective application of biodiesel in power and renewable energy generation. This study presents Dyacrodes edulis biodiesel engine performance and combustion release optimization using response surface methodology-genetic algorithm (RSM-GA) as well as the variation of key engine efficiency and exhaust release indices with brake power and fuel blend in a high compression ratio (CR) diesel engine. Combustion emission impacts of the blends with respect to petro-diesel decreased in values except for NOX. Brake thermal efficiency (BTE), brake specific fuel consumption (BSFC), brake mean effective pressure (BMEP), volumetric efficiency, and exhaust temperature increased with brake power while specific energies decreased with load. Optimum conditions obtained using integrated RSM-GA were 40.03%, 0.05 kg/kW-h, 0.03%, 132.30 ppm, and 18.84 ppm for BTE, BSFC, CO, NOx, and HC respectively at low factor (engine load, engine speed, and fuel blend) conditions. At the optimum conditions, the experimental validation results were 44.01%, 0.05 kg/kW-h, 0.04%, 130.05 ppm, and 20.33 ppm for BTE, BSFC, CO, NOx, and HC respectively. The application of the feedstock in compression ignition engine is viable.

Experimental Investigation for the Multi-objective Optimization of Machining Parameters on AISI D2 Steel Using Particle Swarm Optimization Coupled with Artificial Neural Network

High Carbon High Chromium (or AISI D2) Steels, owing to the fine surface finish they produce upon grinding, find lot of applications in die casting. Machining parameters affect the surface finish significantly during the grinding operation. In this context, this work puts an effort to arrive at the optimum machining parameters relating to fine surface finish with minimum cutting force. The material removal caused by the abrasive grinding wheel makes the process a very complex and nonlinear machining operation. In many situations, traditional optimization techniques fail to provide realistic optimum conditions because of the associated complexity. In order to overcome this issue, particle swarm optimization (PSO) coupled with artificial neural network (ANN) is applied in this research work for parameter optimization with the objective of achieving minimum surface roughness and cutting force. The machining parameters selected for the investigation were table speed, cross feed and depth of cut and the responses were surface roughness and cutting force. ANNs, inspired from biological neural networks, are well capable of providing patterns, which are too complex in behavior. The ANN model developed was used as the fitness function for PSO to complete the optimization. Optimization was also carried out using conventional response surface methodology-genetic algorithm (RSM-GA) approach in which regression equation developed with RSM was considered as the fitness function for GA. Confirmatory experiments were conducted and the comparison showed that PSO coupled with ANN is a reliable tool for complex optimization problems.

Modeling and parametric optimization of FDM 3D printing process using hybrid techniques for enhancing dimensional preciseness

In this study, significant process parameters (layer thickness, build orientation, infill density and number of contours) are optimized for enhancing the magnitude/dimensional preciseness of fused deposition modeling (FDM) devise units. Hybrid statistical tools such as response surface methodology–genetic algorithm (RSM–GA), artificial neural network (ANN) and artificial neural network-genetic algorithm (ANN-GA) in MAT LAB 16.0 are utilized for training and optimization. An attempt has been made to build up a mathematical model in order to set up an indirect correlation between various FDM process parameters and magnitude preciseness. Sequentially to verify the different developed models and the optimum process parameters setting validation tests were also performed. The results showed that various hybrid statistical tools such as RSM-GA, ANN and ANN-GA are very adequate tools for FDM process parameter optimization. The minimum percentage variation in length = 0.06409%, width = 0.03961% and thickness = 0.85689% can be obtained by using ANN-GA.

Improved biomass and lipid production in Synechocystis sp. NN using industrial wastes and nano-catalyst coupled transesterification for biodiesel production

In this study, the improved biomass (1.6 folds) and lipid (1.3 folds) productivities in Synechocystis sp. NN using agro-industrial wastes supplementation through hybrid response surface methodology-genetic algorithm (RSM-GA) for cost-effective methodologies for biodiesel production was achieved. Besides, efficient harvesting in Synechocystis sp. NN was achieved by electroflocculation (flocculation efficiency 97.8±1.2%) in 10min when compared to other methods. Furthermore, different pretreatment methods were employed for lipid extraction and maximum lipid content of 19.3±0.2% by Synechocystis sp. NN was attained by ultrasonication than microwave and liquid nitrogen assisted pretreatment methods. The highest FAME (fatty acid methyl ester) conversion of 36.5±8.3mg FAME/g biomass was obtained using titanium oxide as heterogeneous nano-catalyst coupled whole-cell transesterification based method. Conclusively, Synechocystis sp. NN may be used as a biodiesel feedstock and its fuel production can be enriched by hybrid RSM-GA and nano-catalyst technologies.

Simulation and optimization of continuous laser transmission welding between PET and titanium through FEM, RSM, GA and experiments

An intelligent method for simulation and optimization of continuous laser transmission welding (LTW) and validated with experiments (EX) is investigated in this paper. Thermal model using finite element method (FEM) has been combined with response surface methodology (RSM) and genetic algorithm (GA) techniques to improve veracity of the model prediction with less time spending on the experiments. A three-dimensional axi-symmetric thermal model has been developed to simulate the continuous LTW process with a moving Super-Gaussian heat source. The model is confirmed with a series of experiments. A statistical technique RSM based mathematical model is proposed to establish relations between input variables (power, welding speed, stand-off-distance) and output variables (maximum temperature at the weld interface-Tmax, maximum temperature at the top surface of the transparent PET-Ttop, weld width-WW, and weld depth in the transparent PET-DT). The RSM models are trained and tested by using the data from the numerical (FEM) models. It turns out that the models are proposed to accurately predict the output variables with the corresponding input variables. Finally, the desirability function (DF) integrated with the developed non-dominating sorting genetic algorithm-II (NSGA-II) is used to find out the optimal variables that enhance the quality and efficiency of the welding. Experiments using the optimum parameters are carried out to verify the FEM and RSM models. Results show that the proposed integrated (FEM–RSM–GA–EX) approach performed very well in optimum performance of the continuous LTW process. In addition, this approach also presents the feasibility of the use of the FE simulation to guide the experiments.

STATISTICAL MODELING AND OPTIMIZATION OF ALKALINE PROTEASE PRODUCTION FROM A NEWLY ISOLATED ALKALOPHILIC Bacillus SPECIES BGS USING RESPONSE SURFACE METHODOLOGY AND GENETIC ALGORITHM

A new hyperactive alkalophilic bacterial strain (Bacillus sp. BGS) was isolated from samples collected from soil that received the effluent of a milk processing industry located in Madurai, Tamilnadu, India, and this bacterial strain was used for the production of alkaline protease. Four out of eight variables, such as molasses, peptone, pH, and inoculum size, have been identified through Plackett–Burman (PB) design and used for the alkaline protease production. These significant variables were further optimized through a hybrid system of response surface methodology (RSM) followed by genetic algorithm (GA). The optimal combination of media components and culture conditions for maximal protease production was found to be 16.827 g/L of peptone, 1.128% (v/v) of molasses, pH value of 11, and 2% (v/v) of inoculum size. A 6.36-fold increase in protease production was achieved through the RSM–GA hybrid system. The protease activity increased significantly with an optimized medium (2,992.75 U/mL) as opposed to an unoptimized basal medium (470.35 U/mL).