27 research outputs found
An Integrated Neural Network Structure for Recognizing Autocorrelated and Trending Processes
No full textData sets collected from industrial processes may have both a particular type of trend and correlation among adjacent observations (autocorrelation). In the present paper, an integrated neural network structure is used to recognize trend stationary first order autoregressive (trend AR(1)) process. The proposed integrated structure operates as follows. (i) First a combined neural network structure (CNN), that is composed of appropriate number of linear vector quantization (LVQ) and multi layer perceptron (MLP) neural networks, is used to recognize the trended data, (ii) then, the Elman’s recurrent neural network (ENN) is used to diagnose the autocorrelation through the data. Correct classification rate is used as performance criteria. Results indicate that proposed structure is effective and competitive with other combined neural network structures
Design Optimization of a 4-Poled 1500 rpm 25 kVA SG to Obtain the Desired Magnetic Flux Density Distributions by using RSM
Get PDF84-93In this study design optimization for 4-poled 1500 rpm 25 kVA synchronous generator (SG) is performed. The aim is to
determine the optimum factor levels for the design parameters namely slot opening width (Bs0), height, and width to keep
the responses namely ‘pole-body flux density’ and ‘air-gap flux density’ distributions in a desired range. The target values
are determined as 1.75 Tesla and 0.9 Tesla for the ‘pole-body flux density’ and ‘air-gap flux density’ respectively. For this
purpose, Response Surface Methodology (RSM) is used for optimization. Numerical simulations are performed in the
Maxwell environment and the optimization by RSM is performed by Minitab statistical package. Desired goals were
achieved and optimum factor levels were determined with RSM. Then the results of RSM are compared by Genetic
Algorithm (GA), Particle Swarm Optimization algorithm (PSO), and Modified Social Group Optimization (MSGO)
algorithm. These methods are evaluated together in terms of advantages and disadvantages. The comparisons indicate that
using RSM provides acceptable results without performing coding effort and also provides users to understand the relations
visually between the factors and the responses by the aid of ‘Minitab Response Optimizer Module’
Optimizing Plastic Extrusion Process via Grey Wolf Optimizer Algorithm and Regression Analysis
Get PDF34-41One of the most widely used methods in the production of plastic products is the extrusion process. There are many factors that affect the product quality throughout the extrusion process. Examining the effects of these factors and determining the optimum process parameters which will provide the desired product characteristics; is important for reducing costs and increasing competitiveness. This study is performed in a manufacturer that produces plastic cups.
The aim is to optimize extrusion process parameters of this company in order to achieve 1.15 mm thickness at the produced plastic sheets. For this reason, in order to be able to model the problem as an optimization problem through regression modelling, the thicknesses of the sheet generated with different process parameters were observed during the production processes. Then, considering the desired 1.15 mm sheet thickness, the established model is optimized by running the grey wolf optimizer (GWO) algorithm through the model
Artificial Bee Colony Algorithm for Labor Intensive Project Type Job Shop Scheduling Problem: A Case Study
No full textOptimizing Plastic Injection Process Using Whale Optimization Algorithm in Automotive Lighting Parts Manufacturing
Get PDF360-368In this study, using the whale optimization algorithm (WOA), one of the recent optimization algorithms inspired by nature, the plastic injection process parameters of an automotive sub-industry company were tried to be optimized. For this purpose, we tried to provide the maximum weight criterion for the “356 MCA Plastic Housing” (which is an automotive lighting part) produced by plastic injection method. The decrease in the weight of the product indicates that the material injected into the mold is missing and naturally indicates that there will be quality problems. In order to achieve this aim, the best factor levels were tried to be determined for the mold temperature (°C), injection speed (m/s), injection pressure (bar), holding time (s), and injection time (s), which are the controllable parameters of injection process. Factors and factor levels addressed using WOA have not been studied for this type of problem before and this is the novelty aspect of this research. Experiments performed to confirm the findings for optimum process parameters proved that the WOA method can be successfully applied to improve plastic injection process parameters. This study contains information for practicing researchers in terms of showing how the nature-inspired algorithm WOA can be applied in practical field studies
A regression control chart for autocorrelated processes
No full textIn this study, we present a new regression control chart which is able to detect the mean shift in a production process. This chart is designed for autocorrelated process observations having a linearly increasing trend. Existing approaches may individually cope with autocorrelated and trending data. The proposed chart requires the identification of trend stationary first order autoregressive (trend AR(1)) model as a suitable time series model for process observations. For a wide range of possible shifts and autocorrelation coefficients, performance of the proposed chart is evaluated by simulation experiments. Average correct signal rate and average run length are used as performance criteria. © 2014 Inderscience Enterprises Ltd
Performance comparison of residual control charts for trend stationary first order autoregressive processes
No full textData sets collected from industrial processes may have both a particular type of trend and correlation among adjacent observations (autocorrelation). Existing statistical control charts may individually cope with autocorrelated or trending data. Applying the Shewhart, EWMA, CUSUM, or GMA charts to the uncorrelated residuals of an appropriate time series model for a process is a primary method to deal with autocorrelated process data. In the relevant literature, there exists no study that shows how these charts' performances change by the addition of a particular type of trend in autocorrelated data. In the present paper, average run lengths of these charts are computed; first, for autocorrelated data which does not include an increasing linear trend, and second, for autocorrelated data which includes an increasing linear trend. It is assumed that stationary AR(1) model and trend stationary first order autoregressive (trend AR(1) for short) model, respectively, are suitable models for the test data. ARL performances are compared within the charts and among the charts. Comparisons are made for different magnitudes of the process mean shift and various levels of autocorrelation
A new painting process for vessel radiators of transformer: wet-on-wet
No full text<div><p>The painting process of corrugated wall radiators of a distribution transformer is performed by a flow-down painting technique in the industrial field. This study has been prepared in accordance with ISO 12944-5. Correspondingly, this work is motivated by Epoxy 2-pack paints (4.3.4.2) to obtain minimum requirements for C3 atmospheric corrosivity categories (5.1.1). This standard requires from the vertical surface of the vessel of the transformer to be painted with epoxy paints that contain anti-corrosive pigments with a minimum of 100 µm dry film thickness. In the present study, a new production methodology called wet-on-wet (WOW) painting is developed which has never been used in industry. In addition, a modified response surface methodology (RSM) is proposed for designing, modeling, and optimizing the proposed process under unsteady environmental effects. The results indicate that the WOW painting can be applied to real industrial systems successfully by the aid of the proposed new RSM algorithm and provide remarkable time and cost savings.</p></div
Using response surface design to determine the optimal parameters of genetic algorithm and a case study
No full textCopyright © 2013 Taylor & Francis. This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Production Research on 09 June 2013, available online: http://www.tandfonline.com/10.1080/00207543.2013.784411Genetic algorithms are efficient stochastic search techniques for approximating optimal solutions within complex search spaces and used widely to solve NP hard problems. This algorithm includes a number of parameters whose different levels affect the performance of the algorithm strictly. The general approach to determine the appropriate parameter combination of genetic algorithm depends on too many trials of different combinations and the best one of the combinations that produces good results is selected for the program that would be used for problem solving. A few researchers studied on parameter optimisation of genetic algorithm. In this paper, response surface depended parameter optimisation is proposed to determine the optimal parameters of genetic algorithm. Results are tested for benchmark problems that is most common in mixed-model assembly line balancing problems of type-I (MMALBP-I)
Response surface methodology based prediction of engine performance and exhaust emissions of a diesel engine fuelled with canola oil methyl ester
No full text
