Showing 7 results for Analysis of Variance
Volume 14, Issue 1 (6-2014)
Abstract
Dyes are being extensively used in textile, dyeing and leather industries and usually have a synthetic origin and complex chemical structure that may be mutagenic and carcinogenic. Among various processes of dye removal from wastewater, adsorption is considered to be the most efficient process. Activated carbon is the most widely used adsorbent with great success because of its high adsorption capacity, but its use is limited due to its high-cost, has led to a search for cheaper substitutes. Optimization methods, such as the Taguchi method has been generally adopted to optimize the design parameters, because this systematic approach can significantly minimize the overall testing and the experimental costs. In this study; the optimum condition of Reactive Blue 19(RB19) removal by activated carbon based on pomegranate residual is determined using Taguchi method. For this purpose, 4 main factors such as pH, adsorbent dose, initial dye concentration and contact time were considered in 5 levels. Therefore, an L25 orthogonal array was chosen, and the experimental conditions were obtained. In addition, after design of experiments, an analysis of the signal-to-noise (S/N) ratio was needed to evaluate the experimental results. The analysis of mean (ANOM) statistical approach was adopted herein to construct the optimal conditions. In addition to ANOM, the analysis of variance (ANOVA) statistical method was also used to analyze the influence of each controllable factor on the dye removal efficiency. To prepare the To prepare the activated carbon (AC), pomegranate residual was collected. Collected sample has been dried in an oven for 2 h at 100 0C. It was then ground in a ball mill and the material passed through the No.30 mesh was collected and tested. They were soaked for 24 h in a 1:1 wt. ratio with 50 wt. % phosphoric acid as the carbonized at room temperature. The sample is then decanted and dried in a muffle furnace for 1 h at 500 0C. They were then washed sequentially several times with hot distilled water, until pH of the washing solution became neutral. AC was finally powdered and sieved by the No. 100 mesh. The experiments were done 3 times according to the tests conditions determined by the Taguchi method and the dye removal efficiency was measured. Due to the results of S/N and ANOM, it can be inferred that the optimum mixture proportions to obtain the highest RB19 removal efficiency are as follows. (1) Initial dye concentration of 100 mg/L; (2) pH of 11; (3) adsorbent dose of 1.75 gr/L; and (4) contact time of 7 minutes. The experiment with aforementioned condition was done and result was shown the highest efficiency (98.94%). According to the result of ANOVA, the rank order of the contribution percentage of each factor on RB19 removal efficiency is as follows: (1) initial dye concentration (52.67%), (2) adsorbent dose (33.32%), (3) pH (13.61%), and (4) contact time (2.72%).
Sina Ariaee, Abolfazl Tutunchi, Abbas Kianvash,
Volume 14, Issue 9 (12-2014)
Abstract
In the present study, central composite algorithm was used in order to model and optimize the mechanical behavior of “glass fiber reinforced epoxy composite - structural steel “connections. Initial tests showed that the polymer curing variables play a significant role as key process parameters in producing strong and reliable connections. After conducting Thermal Gravimeteric Analysis on polymer, by selecting curing time and curing temperature as input variables, the parameters were coded and each of them was studied in five levels. In order to estimate the desirable response and provide appropriate models, thirteen tests were conducted systematically. In order to assess the accuracy and to validate the proposed model, analysis of variance was performed successfully. The effect of curing time and curing temperature on the connection’s strength quality was studied utilizing two-dimensional graphs. Utilizing this approach the optimal bonding process variables was achieved at 40°C and 180 min for curing temperature and curing time respectively. Finally, the results obtained from micro structural characterization and fractography analyses of joints by Optical and Scanning Electron Microscope were in good agreement with the results achieved by the developed model.
Mohammad Ali Ansari, Emadoddin Sadeqzadeh Naeini, Mohammad Kazem Besharati Givi, Ghader Faragi,
Volume 15, Issue 6 (8-2015)
Abstract
Friction Stir Extrusion (FSE) is a modern one step process with high efficiency for conversion and recycling of materials which capable of producing Nano-engineered material via production with good deformability, mechanical and physical properties. Novelty of this production technique is utilization of frictional heat and severe plastic deformation for material flowing, mechanical alloying and finally amendment of powder, chips and other recyclable scraps directly to useful products. Sample’s microstructure was revealed and average grain size was gained for 18 samples. Experimental parameters by use of design of experiments for two factors and analysis of variance were investigated and by the use of experimental results were validated. In this study, the effect of rotational speed in 3 levels and plunge rate in 2 levels were examined on microstructure of produced wires via FSE process. Based on process parameters, there is an equation for grain size prediction was presented by using full factorial design of experiment. Furthermore, normal possibility diagram and residual versus order based on residual theorem were evaluated for systematic error entry and reliability to experimental results. The efficient region on contour diagram reveals that suitable condition of minimum grain size and maximum strength occurred at 250 rpm for rotational speed and 14 mm/min for feed rate. It should be noted that analysis of variance showed that rotational speed, feed rate and interaction of rotational speed and feed rate respectively have a meaningful effect on the grain size of produced wire.
Ali Reza Nikravan, Farhad Kolahan,
Volume 15, Issue 9 (11-2015)
Abstract
The use of TI-6AL-4V alloy is increasing in various industries, due to its characteristics such as high strength to weight ratio and excellent creep and corrosion resistance. However, because of titanium alloys poor machinability, non-traditional machining is usually used in shaping them. In this research, the wire cut machining of Ti-6Al-4V alloy has been investigated. In this regard, the effects of 8 process parameters and their optimal values are determined, for two process characteristics, cutting rate and surface roughness. One of the innovative aspects of the current study is the consideration of cutting height as one of the main parameters affecting the process output quality. Therefore, for three different cutting heights (each with two replicates), the L18 Taguchi matrix is employed to gather the required experimental data. Then, using the 108 data, statistical analysis, including signal to noise and analysis of variance were carried out in order to determine optimal process parameters and their effects on the two process outputs. Optimal process parameters values for maximum cutting rate and minimum surface roughness were verified by validation experiments. The results illustrate that the cutting height would alter the effects of process parameters and their optimal values. Finally, multi-objective optimization was performed to determine optimal parameters settings based on relative importance of process outputs for cutting heights of 30, 60 and 90 mm, in the form of technology table. The contribution of uncontrolled parameters and errors were less than 10% in all cases, which indicate the accuracy of the proposed technique.
Volume 17, Issue 4 (11-2017)
Abstract
The pollution of soil with 2-methylpropane-2-thiol as an odorant hydrocarbon is an environmental problem. It also causes secondary impacts such as social dissatisfaction and economic problems due to tourist revenue reduction. 2-methylpropane-2-thiol is a hazardous material and remediation of soil polluted by this material with a fast method is important to study.
In this study, modified Fenton treatment is investigated for oxidation of 2-methylpropane-2-thiol. Central Composite Design (CCD) based on Response Surface Methodology (RSM) was used to obtain appropriate effects of the main factors (initial H2O2 concentration, FeSO4 to soil ratio and stirring time interval percentage) and their interactions on the removal efficiency. Treatments were set up to monitor 2-methylpropane-2-thiol removal efficiency for initial contaminant concentration of 64690 part per million by weight. Samples were analyzed by gas chromatograph equipped with FID and TCD detector and HP-Plot Q column. Design of experiment in the three-factor with five-level matrix include 20 experiment. Randomization technique is used to guard against unknown and uncontrolled factors as lurking nuisance factors. Moreover, blocking technique is used for investigation of probable effect of initial soil temperature on results.
Analysis of variance and Pareto analysis show that all main factor are effective. Also, stirring time interval percentage was the most influential factors on 2-methylpropane-2-thiol removal efficiency. Results of the experiments shows that at low concentration levels removal efficiency increases with hydrogen peroxide concentration up to the certain level. For higher concentration of hydrogen peroxide concentration, the removal efficiencies decreased which could be due to scavenging. Also, increasing in FeSO4:Soil ratio increases removal efficiency up to the certain level because Fe2+ is an alternative to enable more extensive and greater contaminant oxidation; however a greater ratio (greater than 0.0040) causes decrease in the removal efficiency. This phenomena could be due to side reactions which affect reactive radicals such as OH• radicals. Furthermore, investigation of the results demonstrates that 2-methylpropane-2-thiol removal efficiency rises with increasing stirring time interval percentage. This phenomena could be due to uniform distribution of oxidation agent and Fe2+ and better desorption of contaminant from soil to liquid phase.
Moreover, based on analysis of variance, the interaction between hydrogen peroxide and FeSO4: Soil ratio was significant with positive effect on the removal efficiency. This interaction could be the result of reaction between H2O2 and Fe2+. By considering main and interaction effects, with the raising H2O2 and Fe2+ concentration up to a certain level, the removal efficiency increase and with further concentration increasing the removal efficiency will be dropped. Analysis of variance indicate that initial soil temperature (21 and 25 0C) were not effective factors during the time interval of the experiments which could be due to the exothermic reaction between hydrogen peroxide, FeSO4 and contaminant. P-value of lack-of-fit (0.064) indicates that suggested model adequately fits the data with good correlation coefficient (R2=95.12%). Optimum condition suggested for maximum 2-methylpropane-2-thiol removal efficiency (94.412%) shows that concentration of H2O2 and Fe2+ ion must be at the certain level and maximum stirring time for remediation in the studied intervals.
CCD model predict 94.084% for the removal efficiency at optimum condition which is good agreement with the predicted value.
Mohammad Amin Shahrokhian Dehkordi, Majid Karimian,
Volume 17, Issue 5 (7-2017)
Abstract
Lost foam casting is a new method for casting complex parts. This method, in addition to its technical and economic advantages over the traditional methods, has environmental benefits and therefore has been of special interest. In this study, the effects of foam density, pouring temperature, and coating viscosity were studied, which are the most important factors affecting porosity and hardness in the lost foam casting method. The Taguchi method, signal to noise ratio and analysis of variance were used to design experiments and determine the optimal levels of each variable. All the considered variables were evaluated in three levels using L9 orthogonal array Taguchi analysis. Results showed that foam density of 20 kg/m3, pouring temperature of 740° C and coating viscosity of 20 sec were the optimal values for the variables due to creating appropriate condition between thermal decomposition and foam evaporation with speed of melting advancement and exhaust gas through the pores in the coating and creating lowest porosity (2.6%) and highest hardness value (27.7 HRA). Foam density and pouring temperature were the most influential parameters on the porosity and hardness with the impact factors of 64.58% and 56.35%, respectively.
Ali Reza Nikravan, Farhad Kolahan, Mahmoud Shariati,
Volume 22, Issue 8 (8-2022)
Abstract
Due to the importance of the joints in the pressurized instruments and the abilities of the laser welding in this study, the welding of AISI316L tubes have been studied and analyzed. In this regard a fixture has been drawn in order to positioning of the tubes and fixed on the welding desk. The welding input parameters includes the laser welding adjusting variables, includes (welding current, welding pulse width, and welding frequency). Moreover, the effect of the two other variables (rotating speed and the force applied to the welding seam) has also been studied. The welding output characteristics comprises the welding width, depth of penetration and welding strength. The experimental data has been collected using L27 Taguchi design. The relation between the process input variables and output characteristics has been established using different regression models. Based on the analysis of variance (ANOVA) results, pulse width and welding current with 70% contribution have an influential effect on all the three response characteristics. Moreover, the seam force has only the influential effect on the depth of penetration and strength. Next, in optimization step based on the importance of the process characteristics (strength, depth of penetration, welding width), the optimized levels have been determined. At the end, the optimized condition has been conducted using laser welding, in comparison of which the samples in the design matrix, the welding depth has a close relation with the thickness of the wall, the welding boundaries smother and the strength has a close value to the base metal.