Predicting the wear rate of AA6082 aluminum surface composites produced by friction stir processing via artificial neural network

2019 ◽  
Vol 16 (2) ◽  
pp. 409-423
Author(s):  
Isaac Dinaharan ◽  
Ramaswamy Palanivel ◽  
Natarajan Murugan ◽  
Rudolf Frans Laubscher
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Wear Rate ◽  
Friction Stir Processing ◽  
Aluminum Surface ◽  
Content Type ◽  
Friction Stir ◽  
Surface Composites ◽  
Input Parameters ◽  
Artificial Neural

Purpose Friction stir processing (FSP) as a solid-state process has the potential for the production of effective aluminum matrix composites (AMCs). In this investigation, various ceramic particles including B4C, TiC, SiC, Al2O3 and WC were incorporated as the dispersed phase within AA6082 aluminum alloy by FSP. The wear rate of the composite is then investigated experimentally by making use of a design of experiments technique where wear rate is evaluated as the output parameter. The input parameters considered include tool rotational speed, traverse speed, groove width and ceramic particle type. An artificial neural network (ANN) simulation was then used to describe the wear rate of the surface composites. The weights of the network were adjusted to minimize the mean squared error using a feed forward back propagation technique. The effect of the individual input parameters on wear rate was then inferred from the ANN models. Trends are presented and related to the associated microstructures observed. The TiC infused AMC displayed the lowest wear rate whereas the Al2O3 infused AMC displayed the highest, within the scope of the current investigation. The paper aims to discuss these issues. Design/methodology/approach The paper used ANN for the research study. Findings The finding of this paper is that the wear rate of AA6063 aluminum surface composites is influenced remarkably by FSP parameters. Originality/value Original work of authors.

Estimation of mechanical properties of friction stir processed Al 6061/Al2O3-Tib2 hybrid metal matrix composite layer via artificial neural network and response surface methodology

2021 ◽  
pp. 146442072110345
Author(s):  
Vahid M Khojastehnezhad ◽  
Hamed H Pourasl ◽  
Arian Bahrami
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Response Surface Methodology ◽  
Response Surface ◽  
Process Parameters ◽  
Friction Stir Processing ◽  
Wear Behavior ◽  
Friction Stir ◽  
Al 6061 ◽  
Artificial Neural

Friction stir processing is one of the solid-state processes which can be used to modify the structure and properties of alloys. In addition, it has become one of the most promising techniques for the preparation of the surface layer composites. To pursue cost savings and a time-efficient design, the mathematical model and optimization of the process can represent a valid choice for engineers. Friction stir processing was employed to generate an Al 6061/Al2O3-TiB2 hybrid composite layer, and mechanical properties such as the hardness and wear behavior were also measured. The relationship between the hardness and wear behavior, process parameters of friction stir processing were evaluated using an artificial neural network and response surface methodology. The rotational speed (1500–1800 rpm), traverse speeds (25, 50, 100 mm/min), and the number of passes (1–4) with constant axial force (2.61 kN) were used as the input, while the hardness and weight loss values were the output. Experimentally, the results showed that the process parameters have significant effect on hardness and wear behavior of Al 6061/Al2O3-TiB2. In addition, the developed artificial neural network and response surface methodology models can be employed as alternative methods to compute the hardness and weight loss for given process parameters. The results of both models showed that the estimated values for the hardness and wear behavior of the processed zone had an error less than 0.60%, which indicated reliability, and an evaluation of the estimated values of both models and the experimental values confirmed that the artificial neural network is a better model than response surface methodology.

Dynamic prediction of Indian stock market: an artificial neural network approach

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Himanshu Goel ◽  
Narinder Pal Singh
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Stock Market ◽  
Institutional Investors ◽  
Time Series Data ◽  
Gradient Algorithm ◽  
Series Data ◽  
Content Type ◽  
Artificial Neural ◽  
Indian Stock Market

Purpose Artificial neural network (ANN) is a powerful technique to forecast the time series data such as the stock market. Therefore, this study aims to predict the Indian stock market closing price using ANNs. Design/methodology/approach The input variables identified from the literature are some macroeconomic variables and a global stock market factor. The study uses an ANN with Scaled Conjugate Gradient Algorithm (SCG) to forecast the Bombay Stock Exchange (BSE) Sensex. Findings The empirical findings reveal that the ANN model is able to achieve 93% accuracy in predicting the BSE Sensex closing prices. Moreover, the results indicate that the Morgan Stanley Capital International world index is the most important variable and the index of industrial production is the least important in predicting Sensex. Research limitations/implications The findings of the study have implications for the investors of all categories such as foreign institutional investors, domestic institutional investors and investment houses. Originality/value The novelty of this study lies in the fact that there are hardly any studies that use ANN to forecast the Indian stock market using macroeconomic indicators.

Improving the electromagnetic compatibility of electronic products by using response surface methodology and artificial neural network

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ching-Hsiang Chen ◽  
Chien-Yi Huang ◽  
Yan-Ci Huang
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Response Surface Methodology ◽  
Response Surface ◽  
Electromagnetic Compatibility ◽  
Heat Dissipation ◽  
Optimal Parameter ◽  
Content Type ◽  
Early Stages ◽  
Artificial Neural

Purpose The purpose of this study is to use the Taguchi Method for parametric design in the early stages of product development. electromagnetic compatibility (EMC) issues can be considered in the early stages of product design to reduce counter-measure components, product cost and labor consumption increases due to a number of design changes in the R&D cycle and to accelerate the R&D process. Design/methodology/approach The three EMC characteristics, including radiated emission, conducted emission and fast transient impulse immunity of power, are considered response values; control factors are determined with respect to the relevant parameters for printed circuit board and mechanical design of the product and peripheral devices used in conjunction with the product are considered as noise factors. The optimal parameter set is determined by using the principal component gray relational analysis in conjunction with both response surface methodology and artificial neural network. Findings Market specifications and cost of components are considered to propose an optimal parameter design set with the number of grounded screw holes being 14, the size of the shell heat dissipation holes being 3 mm and the arrangement angle of shell heat dissipation holes being 45 degrees, to dispose of 390 O filters on the noise source. Originality/value The optimal parameter set can improve EMC effectively to accommodate the design specifications required by customers and pass test regulations.

Integrating artificial neural network and imperialist competitive algorithm (ICA), to predict the energy consumption for land leveling

2017 ◽  
Vol 11 (4) ◽  
pp. 522-540 ◽  
Author(s):  
Isham Alzoubi ◽  
Mahmoud Delavar ◽  
Farhad Mirzaei ◽  
Babak Nadjar Arrabi
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Energy Consumption ◽  
Compressibility Factor ◽  
Imperialist Competitive Algorithm ◽  
Content Type ◽  
Competitive Algorithm ◽  
Land Leveling ◽  
Soil Compressibility ◽  
Artificial Neural

Purpose This work aims to determine the best linear model using an artificial neural network (ANN) with the imperialist competitive algorithm (ICA-ANN) and ANN to predict the energy consumption for land leveling. Design/methodology/approach Using ANN, integrating artificial neural network and imperialist competitive algorithm (ICA-ANN) and sensitivity analysis (SA) can lead to a noticeable improvement in the environment. In this research, effects of various soil properties such as embankment volume, soil compressibility factor, specific gravity, moisture content, slope, sand per cent and soil swelling index on energy consumption were investigated. Findings According to the results, 10-8-3-1, 10-8-2-5-1, 10-5-8-10-1 and 10-6-4-1 multilayer perceptron network structures were chosen as the best arrangements and were trained using the Levenberg–Marquardt method as the network training function. Sensitivity analysis revealed that only three variables, namely, density, soil compressibility factor and cut-fill volume (V), had the highest sensitivity on the output parameters, including labor energy, fuel energy, total machinery cost and total machinery energy. Based on the results, ICA-ANN had a better performance in the prediction of output parameters in comparison with conventional methods such as ANN or particle swarm optimization (PSO)-ANN. Statistical factors of root mean square error (RMSE) and correlation coefficient (R2) illustrate the superiority of ICA-ANN over other methods by values of about 0.02 and 0.99, respectively. Originality/value A limited number of research studies related to energy consumption in land leveling have been done on energy as a function of volume of excavation and embankment. However, in this research, energy and cost of land leveling are shown to be functions of all the properties of the land, including the slope, coefficient of swelling, density of the soil, soil moisture and special weight dirt. Therefore, the authors believe that this paper contains new and significant information adequate for justifying publication in an international journal.

scholarly journals An artificial neural network model for predicting characteristic input parameters for physics based modelling of drying process

2019 ◽  
Author(s):  
Ankita Sinha ◽  
Atul Bhargav
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Artificial Neural Network Model ◽  
Estimation Methods ◽  
Drying Process ◽  
Many Instruments ◽  
Input Parameters ◽  
Artificial Neural

Drying is crucial in the quality preservation of food materials. Physics-based models are effective tools to optimally control the drying process. However, these models require accurate thermo-physical properties; unavailability or uncertainty in the values of these properties increases the possibility of error. Property estimation methods are not standardized, and usually involve the use of many instruments and are time-consuming. In this work, we have developed an experimentally validated deep learning-based artificial neural network model that estimates sensitive input parameters of food materials using temperature and moisture data from a set of simple experiments. This model predicts input parameters with error less than 1%. Further, using input parameters, physics-based model predicts temperature and moisture to within 5% accuracy of experiments. The proposed work when interfaced with food machinery could play a significant role in process optimization in food processing industries.

Predicting the ground vibration induced by mine blasting using imperialist competitive algorithm

2018 ◽  
Vol 35 (4) ◽  
pp. 1774-1787 ◽  
Author(s):  
Katayoun Behzadafshar ◽  
Fahimeh Mohebbi ◽  
Mehran Soltani Tehrani ◽  
Mahdi Hasanipanah ◽  
Omid Tabrizi
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Ground Vibration ◽  
Imperialist Competitive Algorithm ◽  
Empirical Models ◽  
Coefficient Of Determination ◽  
Content Type ◽  
Competitive Algorithm ◽  
Spacing Ratio ◽  
Artificial Neural

PurposeThe purpose of this paper is to propose three imperialist competitive algorithm (ICA)-based models for predicting the blast-induced ground vibrations in Shur River dam region, Iran.Design/methodology/approachFor this aim, 76 data sets were used to establish the ICA-linear, ICA-power and ICA-quadratic models. For comparison aims, artificial neural network and empirical models were also developed. Burden to spacing ratio, distance between shot points and installed seismograph, stemming, powder factor and max charge per delay were used as the models’ input, and the peak particle velocity (PPV) parameter was used as the models’ output.FindingsAfter modeling, the various statistical evaluation criteria such as coefficient of determination (R2) were applied to choose the most precise model in predicting the PPV. The results indicate the ICA-based models proposed in the present study were more acceptable and reliable than the artificial neural network and empirical models. Moreover, ICA linear model with theR2 of 0.939 was the most precise model for predicting the PPV in the present study.Originality/valueIn the present paper, the authors have proposed three novel prediction methods based on ICA to predict the PPV. In the next step, we compared the performance of the proposed ICA-based models with the artificial neural network and empirical models. The results indicated that the ICA-based models proposed in the present paper were superior in terms of high accuracy and have the capacity to generalize.

Sign in / Sign up

Export Citation Format

Share Document