Papers by Keyword: Genetic Algorithm (GA)

Abstract: The bi-layer materials have been used widely during past decades due to their specific characteristics like lighter weight, more corrosion resistance, and insulation features in comparison with mono-layers which consisting them. In this research the aim is achieving to best combination of bi-layer material (Al3105-St14) to satisfy two objectives of weight and formability while it has a constant total thickness. The represent the formability objective and is derived from M-K model associated with Barlat-Lian yield criteria. Another objective is weight of per unit area. The data of experiments are designed based on full factorial method and the surfaces are best polynomial which can fit the variables and objectives. The MATLAB software and the genetic algorithm (GA) are used to generate feasible combination of thickness to provide to minimize the weight and maximize the formability. The Pareto frontier is utilized to satisfy two objective functions simultaneously. The best answer is selected with norm approaching and minimum distance method.

Abstract: Optimization of energy-related systems with by-products that involve environmental degradation has never been so crucial today with depleting resources and global concerns over negative impacts on our environment. This paper reports the results of an optimization scheme on the coefficient of performance (COP) of a standing wave thermoacoustic refrigerator based on genetic algorithm. The environmentally friendly refrigerator operates without any CFCs, which has been associated with the depletion of ozone, a substance that prevents uv light from reaching the earth’s atmosphere. A single-objective optimization to maximize the COP of a thermoacoustic refrigerator has been completed. The variables investigated include the length of the stack, Lsn, center position of the stack, xsn, blockage ratio, B and drive ratio, DR. The results show that a COP of up to 1.64 is achievable which provides promise for future improvements in the present systems.

Abstract: During the last three decades the concept of the traditional cooling systems was modified to include single, double, and multi-layer micro channels. The new studies, applications, fabrication, and research focus on four main areas: the geometrical shape of the micro channels, the number of stacked layers, the type of the coolants, and the heat performance optimization. The previous studies have shown a significant reduction in the power consumption as the optimization is accomplished. In this paper, a semi-review for the previous works is provided, an attempt to interpret the nature of the work done, and show another trial for optimization. In this study, water was used as coolant, stacked multi-channel was adopted, and thermal resistance network was calculated. The heat sink under consideration is a rectangle of width W and length L. The thickness Hsub of the base of the micro-channel is 100 [μ m] while the depth Hc of the micro-channel is 500[μ m], both kept constant for all future optimization cases.

Abstract: Research in two-phase flow in heat exchanging devices plays an important part in today’s applications in miniaturization of engineering systems. The phase change process factors in the flow conditions and heat transfer in evaporators and condensers. Numerous studies in the past have looked at the predicted and measured frictional pressure drop of coolants as the vapor quality increases. This paper reports a preliminary attempt at modeling of the relationship between the frictional pressure drop and vapor quality in an ammonia-cooled and R22-cooled mini-channel of 1.5 mm diameter under optimized conditions using multi-objective genetic algorithm. R22 is a being phased-out due to its ozone-depleting characteristic and ammonia is being considered as its potential replacement. The properties of ammonia and R22 used have been obtained experimentally at the saturation temperature of 5°C and 10°C respectively. Modeling of the minimized pressure drop per unit tube length together with the Lockhart-Martinelli parameter was completed under optimized flow rate and vapor quality.The outcomes obtained are similar to those that have been reported experimentally with other coolants, increasing pressure drop with increasing vapor quality.

Abstract: In order to improve the working efficiency of a manufacturing system, tool life estimation is very essential. In this paper, the dominant factors affecting tool life are analyzed by theoretical analysis. According to the nonlinear relationship between affecting factors and tool life, a tool life prediction model based on BP neural network, which is optimized by genetic algorithm (GA), is built up. 15 network patterns are trained to get the best network structure. The accuracy of GA-BP model is verified through computing and compared with the standard BP model. The results show that GA-BP model prediction value is exactly closed to the expected value of tool life and the prediction accuracy can be improved more than 5% compared than the standard BP model. The model is proved to be accuracy and it can be used as an effective method of tool selection decision.

Abstract: Stringent legislative, social concerns & clean carbon emissions are constraining companies to take a fresh look at the impact of supply chain operations on environment, society and individuals when designing reverse supply chain networks. A challenging task in today’s globalised environment where companies mandatorily have to collect back goods after its reliable life is making companies integrate supply chain decisions objective here is to minimize transportation cost and distance. In this paper problem of designing a TSP (Travelling Sales man problem) is addressed which is one of the NP-hard problem in combinatorial optimization. Computational experiments conducted with GA (Genetic algorithm) on large and small size TSP cases where compared with NNA (Nearest neighboring algorithm) and have proved that the GA provides optimal tour every time in reasonable time by outperforming the NNA solution when number of cities are increased.

Abstract: In metal cutting surface finish is a crucial output parameter in determining the quality of the product. Good surface finish not only assures quality, but also reduces manufacturing cost. Surface finish is an important parameter in terms of tolerances, it reduces assembly time and avoids the need for secondary operation, thus reduces operation time and leads to overall cost reduction. It is very important to select optimum parameters in metal operations. Traditionally, the experience of the operator plays a major role in the selection of optimum metal cutting conditions. However, attaining optimum values each time by even a skilled operator is difficult. The non-linear nature of the machining process has compelled engineers to search for more effective methods to attain optimization. The main aim of the present work is to build a model to solve real world optimization problems in manufacturing processes.The selection of optimal cutting parameters are speed, feed and depth of cut. are important for all machining process. Experiments have been designed using Taguchi technique, dry and single pass turning of AISI No. 1042 (EN-41B) steel with cermet insert tool performed on PSG A141 lathe. By using signal to noise (S/N) ratio and Analysis of variance (ANOVA) are performed to find the optimum level and percentage of contribution of each parameter. A mathematical model is developed using regression analysis for surface roughness and the model is validated.Moreover, the proposed algorithm, namely GA and PSO were utilized to optimize the output parameter R_a in terms of cutting speed, feed and depth of cut by using MATLAB.

Abstract: In Cellular manufacturing industries are producing similar products using cells, or groups of team members, workstations, or equipment, to make easy operations by eliminating setup and unnecessary costs between operations, Cells might be designed for a specific process, part, or a complete product. There is a strong propensity towards the effectiveness of manufacturing system, proper scheduling (determining the sequence of operations is to be performed) of jobs is essential for the flourishing operation of a shop. Group technology has become a more and more popular concept in manufacturing, which is designed to take advantage of mass production layout and techniques in smaller batch production system. Since the conventional scheduling methods need more computation time. In this paper, an effort has been made in two parts from the first part of this work is to optimize scheduling in different types of products in the job-shop environment are identified and grouping of cells is performed using Rank Order Clustering Method. In the second part, optimization procedure has been developed for the scheduling problem for processing in the machine cells. Particle Swarm Optimization and Genetic Algorithm are used in this paper for explore the optimum schedule by minimizing the total penalty cost due to the delay in meeting the due date. Better scheduling is obtained by comparing the two methods.

Abstract: A mathematical prediction model has been developed in order to detect particles with a diameter of 10 micrometers or less (PM-10) that are responsible for adverse health effects because of their ability to cause serious respiratory conditions in areas of high pollution such as Chiang Mai City moat area. The prediction model is based on 3 types of Artificial Neural Networks (ANNs), including Multi-layer perceptron (MLP-NN), Radial basis function (RBF-NN), and hybrid of RBF and Genetic algorithm (RBF-NN-GA). The model uses 8 input variables to predict PM-10, consisting of 4 air pollution substances ( CO, O₃, NO₂ and SO₂) and 4 meteorological variables related PM-10 (wind speed, temperature, atmospheric pressure and relative humidity). These 3 types of ANN have proved efficient instrument in predicting the PM-10. However, the performance of RBF-NN was superior in comparison with MLP-NN and RBF-NN-GA respectively.

Abstract: To maximize the fundamental frequency of composite laminates, a hybrid optimization algorithm which combines the respective merits of the genetic algorithm and the simulated annealing algorithm is adopted. This hybrid algorithm also incorporates adaptive mechanisms designed to adjust the probabilities of the cross-over and mutation operators. Then, this algorithm is applied to optimize the fiber angle of each layer of a composite laminate such that its fundamental natural frequency is maximized. The results indicate that this hybrid optimization algorithm could quickly find the optimal fiber angles and maximize the fundamental frequency, even under complicated choices of fiber angle and boundary conditions.