K. Hans Raj

Modeling of manufacturing processes with ANNs for intelligent manufacturing

Modern manufacturing often caters to rapidly changing product specifications determined by the continuously increasing productivity, flexibility and quality demands. Metal forming and machining are two important manufacturing processes in present day manufacturing. Automatic selection of tools and accessories in these processes heavily relies on forming force/cutting force estimation. Complex relationships exist between process parameters and these forces. In the present work, the applicability and relative effectiveness of Artificial Neural Network based models has been investigated for rapid estimation of these, invoking the function approximation capabilities of the ANN models. The results obtained are found to correlate well with the finite element simulation data in cases of metal forming, and experimental data in cases of metal cutting. This work has considerable implications in selection of the tools and on-line monitoring of tool wear. The actual forming and cutting forces can be compared with predicted ones to signal the onset of tool wear, and thus prevent damage to the tool and work piece during the course of manufacturing.

Artificial bee Colony Algorithm Integrated with Differential Evolution Operators for Product Design and Manufacturing Optimization

Artificial bee colony (ABC) algorithm is a nature-inspired algorithm that mimics the intelligent foraging behavior of honey bees and it is steadily gaining popularity. It is observed that convergence of ABC algorithm in local minimum is slow. This paper presents an effort to improve the convergence rate of ABC algorithm by integrating differential evolution (DE) operators into it. The proposed ABC-DE algorithm is first tested on three product design optimization problems and the results are compared with co-evolutionary differential evolution (CDE), hybrid particle swarm optimization-differential evolution (PSO-DE) and ABC algorithms. Further, the algorithm is applied on three manufacturing optimization problems, and the results are compared with genetic algorithm (GA), real coded genetic algorithm (RCGA), and RCGA with Laplace Crossover and Power Mutation (LXPM) algorithm and ABC algorithm. Results indicate that ABC-DE algorithm is better than the state of the art algorithms for the aforesaid problems on selected performance metrics.

Finite Element Simulation of Twist Extrusion on ECAPed Al6061 Specimen

Recently, the application of Severe Plastic Deformations (SPD) processes to produce Ultra‐Fine Grained materials with improved mechanical properties are gaining prominence. A number of researchers are working on Equal Channel Angular Pressing (ECAP) and Twist Extrusion (TE) independently. In this work an effort is made to study a hybrid TE process where the Al6061 specimen is subjected to ECAP and subsequently TE in the same die setup. Finite Element (FE) modeling of the above hybrid process is attempted in FORGE 2007 environment. The simulation results clearly depict the change in equivalent strain in the entire specimen on account of this process. A comparison is made between FE results of one pass obtained using ECAP, TE and hybrid TE. Also the variation of strain with change in friction conditions and channel angle are studied using current FE model. The hybrid TE opens up new possibilities for investigating and forming UFG materials.

Evolution of Strain in Multipass Hybrid Equal Channel Angular Pressing Using 3D Finite Element Analysis

Severe Plastic Deformation (SPD) is well known process for producing nanostructured material from coarse material. Present paper is an effort to integrate the two well known SPD techniques Equal Channel Angular Pressing (ECAP) and Twist Extrusion (TE) to develop a new Hybrid ECAP (HECAP) technique that can produce nanostructured material more economically. In this technique, the specimen is subjected to both ECAP and TE in the same die setup. Finite Element (FE) modeling of metal forming processes has become an important tool for designing feasible production processes, because of its unique capability to describe the complex geometry and boundary conditions. FE Modeling of the above hybrid process is attempted in FORGE. The simulation results clearly depict the change in equivalent strain in the entire specimen on account of this process upto four passes. A comparison is made between FE results of simple ECAP and HECAP upto four passes. The study indicated that equivalent strain is much higher in case of HECAP in comparison to ECAP for same friction conditions. Also, the study is extended to analyse the effect of friction, channel angle and forging force on equivalent strain using current FE model. HECAP opens new possibilities for improving equivalent strain in same number of passes as compared to ECAP. This study is expected to contribute in forming UFG materials that are useful for automobile and aerospace industries.

Mechanical Properties of Al6061 Processed by Equal Channel Angular Pressing

The severe plastic deformation (SPD) is an effective approach for producing bulk nanostructured materials. The Equal Channel Angular Pressing (ECAP) is the most efficient SPD solution for achieving ultra-fined grained (UFG) material as billet undergoes severe and large deformation. The process parameters of ECAP (Channel Angle, angle of curvature, friction, number of passes, etc) influences major impact on the properties. In present work, the ECAP process is performed by pressing a specimen through a die consisting of two intersecting channels meeting at an angle φ and outer corner meeting at an angle ψ. Experiments with a circular specimen of Al6061 were conducted to investigate the changes in mechanical properties upto 2 passes. 3-D finite element simulations were also performed using metal forming software FORGE to study the evolution of strain in the specimen during the ECAP process. Simulation results were investigated by comparing them with experimental measured data in terms of load variations. The present work clearly shows that ECAP caused accentuated increase in Al6061 hardness and tensile strength during multi-pass processing. This study is beneficial in developing high quality, high strength products in manufacturing industry on account of its ability to change microstructure of materials.

Optimization of hot extrusion using single objective neuro stochastic search technique

This paper presents a new single-objective neuro-stochastic search technique (SONSST) for the economic load estimation problem in hot extrusion which is often used to produce long straight metal products of constant cross-sections such as bars, solid and hollow sections, tubes, wires and strips from materials that cannot be formed by cold extrusion. The shape of the dies and the temperature developed during extrusion and the velocity of the dies significantly influence forging force at which the process is to be carried out. In order to understand the complex relationship between the material and process variables, a few finite element models are developed and simulated in the FORGE2 environment. These finite element simulations are used to train a neural network (NN) model. Later the same model is incorporated along with a genetic algorithm (GA) and simulated annealing (SA) to form SONSST. It incorporates a genetic crossover operator BLX-/spl alpha/ and a problem specific mutation operator incorporating a local search heuristic: to provide it a better search capability. Extensive simulations have been carried out considering various aspects and the results are validated with those of the existing finite element method in the literature. These results indicate that the new SONSST heuristic converges to better solutions rapidly. SONSST is a truly single-objective technique as it provides the values of various process parameters for optimizing single objective (extrusion load), in a single run and thus assists in achieving energy and material saving, quality improvement and in the development of sound extruded parts.

Determination of Optimal Process Parameters of Friction Stir Welding to Join Dissimilar Aluminum Alloys Using Artificial Bee Colony Algorithm

This paper presents the efforts of joining dissimilar aluminum alloys (AA6351-T6 and AA6061-T6) by friction stir welding (FSW) process. FSW experiments are conducted according to the three factors five level central composite rotatable design method, and the response surface methodology was used to establish the empirical relationship between FSW process parameters such as tool rotational speed (N), tool traverse speed (S) and axial force (F), and the response variables such as ultimate tensile strength, yield strength, and percentage of elongation. The developed empirical models’ adequacies are estimated using the analysis of variance technique. This paper also presents the application of the artificial bee colony algorithm to estimate the optimal process parameters to achieve good mechanical properties of FS weld joints. Results suggest that the estimations of the algorithm are in good agreement with the experimental findings.

Quantum evolutionary computational technique for constrained engineering optimization

A Quantum Evolutionary Computational Technique (QECT) is proposed in this paper. The approach is based on the integration of quantum computing concepts such as superposition of states, application of quantum gate with the concept of genetic algorithm and simulated annealing. To demonstrate effectiveness and applicability of QECT, simulations are carried out on five Benchmark Test functions, which are well-known combinatorial optimization problems. These exemplify that the proposed algorithm has a capability to obtain near global optimum, without premature convergence unlike other variants. QECT has the strong capability to explore the nonlinear search regions and it is a step forward in the area of hybrid stochastic search. It is applied on a problem of mechanical engineering design of a pressure vessel. The application of proposed heuristic technique in mechanical engineering design is a step towards agility in design and manufacturing.

Hybrid evolutionary computational algorithm for dairy cattle feed cost optimization

A Hybrid Evolutionary Computational Algorithm (HECA) is proposed in this paper. It incorporates Genetic algorithm and Simulated Annealing together in Evolutionary Computational part and quantum concepts that are quantum gate, superposition of states in the improvement of initial population. In this paper HECA has been tested on 6 popular benchmark functions and compared with other algorithms reported in literature. HECA has the strong capability to explore the nonlinear search regions and it is a step forward in the area of hybrid stochastic search. HECA is applied on a real world application of cattle feed optimization for local Dairy. The objective of the present algorithm is to find minimum cost diet from the set of available ingredients while improving the quality of feed. The feed mix obtained from the present algorithm is compared with that of a local Dairy. It is found that the results obtained from the present algorithm are favourable and useful for dairy cattle feed planning and cost optimization.

Fuzzy quality function deployment (FQFD) to assess student requirement in engineering institutions: An Indian prospective

Quality is a complex and multifaceted concept. It is being increasingly recognized that high quality of products and services and their associated customer satisfaction are the key elements for survival of any organization. It is the basis that differentiates between a mediocre and an excellent organization. This concept is equally important for an engineering education system. The present paper hybridizes QFD with fuzzy dominance order to assess the student requirements in engineering education system. It uses content analysis and nominal group technique (NGT) to identify requirements. Fuzzy dominance order has been used to establish democratic wisdom to identify customer requirements. Relation matrix used in traditional QFD has been replaced with aggregated fuzzy relation matrix to justify subjective responses. The results obtained by this approach and traditional QFD approach are compared using Interpretive Structural Model (ISM) for deployable elements of technical requirements as yardstick.