Taguchi Quality Method Research Articles

An Evolutionary Optimal Fuzzy System With Information Fusion of Heterogeneous Distributed Computing and Polar-Space Dynamic Model for Online Motion Control of Swedish Redundant Robots

This paper presents an evolutionary optimal fuzzy system with information fusion of heterogeneous distributed computing and polar-space dynamic model for online motion control of Swedish redundant robots. The intelligent fuzzy system incorporated with the parallel metaheuristic Bacteria Foraging Optimization (BFO)-Artificial Immune System (AIS), called FS-PBFOAIS and its field-programmable gate array (FPGA) realization to optimal polar-space online motion control of four-wheeled redundant mobile robots. This hybrid paradigm gains the benefits of Taguchi quality method, BFO, AIS, distributed processing, and FPGA technique. Experimental results are conducted to present effective optimization and high accuracy of the proposed FPGA-based FS-PBFOAIS tracking controller. Finally, the comparative works are provided to demonstrate the superiority of the FPGA-based FS-PBFOAIS polar-space redundant controller over other conventional control methods.

Backstepping Holonomic Tracking Control of Wheeled Robots Using an Evolutionary Fuzzy System with Qualified Ant Colony Optimization

This paper presents a backstepping holonomic tracking control method of embedded wheeled robots using an evolutionary fuzzy system with qualified ant colony optimization (ACO). The Taguchi method is applied to design an optimal ACO via an orthogonal array. This qualified ACO is then employed to develop an evolutionary fuzzy system, called FS-TACO. The proposed hybrid intelligent fuzzy system FS-TACO is realized in a field-programmable gate array (FPGA) to dynamic tracking control of three-wheeled holonomic mobile robots. In comparison with conventional control systems, this approach takes the advantages of Taguchi quality method, fuzzy system, ACO and FPGA technique, thereby obtaining better population diversity, avoiding premature convergence, and achieving self-adaptive holonomic control. The FPGA realization using system-on-a-programmable chip methodology of the proposed FS-TACO is more effective in practice for real-world embedded applications. Experimental results and comparative works are conducted to exhibit the merits of the proposed FPGA-based FS-TACO controller for three-wheeled holonomic mobile robots.

Wear Improvement of Tools in the Cold Forging Process for Long Hex Flange Nuts.

Cold forging has played a critical role in fasteners and has been widely used in automotive production, manufacturing, aviation and 3C (Computer, Communication, and Consumer electronics). Despite its extensive use in fastener forming and die design, operator experience and trial and error make it subjective and unreliable owing to the difficulty of controlling the development schedule. This study used finite element analysis to establish and simulate wear in automotive repair fastener manufacturing dies based on actual process conditions. The places on a die that wore most quickly were forecast, with the stress levels obtained being substituted into the Archard equation to calculate die wear. A 19.87% improvement in wear optimization occurred by applying the Taguchi quality method to the new design. Additionally, a comparison of actual manufacturing data to simulations revealed a nut forging size error within 2%, thereby demonstrating the accuracy of this theoretical analysis. Finally, SEM micrographs of the worn surfaces on the upper punch indicate that the primary wear mechanism on the cold forging die for long hex flange nuts was adhesive wear. The results can simplify the development schedule, reduce the number of trials and further enhance production quality and die life.

A Taguchi-Based Heterogeneous Parallel Metaheuristic ACO-PSO and Its FPGA Realization to Optimal Polar-Space Locomotion Control of Four-Wheeled Redundant Mobile Robots

This paper presents an efficient Taguchi-based heterogeneous parallel metaheuristic ant colony optimization (ACO)-particle swarm optimization (PSO), called THPACOPSO, and its field-programmable gate array (FPGA) realization to optimal polar-space locomotion control of four-wheeled redundant mobile robots. This THPACOPSO consists of one slave Taguchi-based ACO (TACO) and one slave Taguchi-based PSO (TPSO) along with a master communication operator in one chip using system-on-a-programmable chip (SoPC) methodology. This approach takes the advantages of the Taguchi quality method, ACO, PSO, parallel processing, and FPGA technique, thereby obtaining better population diversity, avoiding premature convergence and keeping parallelism. Experimental results and comparative works are conducted to present effective optimization and high accuracy of the proposed FPGA-based THPACOPSO locomotion controller for four-wheeled Swedish redundant mobile robots.

An Evolutionary Radial Basis Function Neural Network with Robust Genetic-Based Immunecomputing for Online Tracking Control of Autonomous Robots

This paper presents an evolutionary radial basis function neural network with genetic algorithm and artificial immune system (GAAIS-RBFNN) for tracking control of autonomous robots. Both the GAAIS-RBFNN computational intelligence and online tracking controller are implemented in one field-programmable gate array (FPGA) chip to cope with the optimal control problem of real-world mobile robotics. The hybrid GAAIS paradigm incorporated with Taguchi quality method is employed to determine the optimal structure of RBFNN. The control parameters of tracking controller are online tuned by minimizing the performance index using the proposed GAAIS-RBFNN to achieve trajectory tracking. Experimental results and comparative works are conducted to show the effectiveness and merit of the proposed FPGA-based GAAIS-RBFNN tracking controller using system-on-a-programmable-chip technology. This FPGA-based online hybrid GAAIS-RBFNN intelligent controller outperforms the existing bio-inspired RBFNN controllers using individual GA and AIS algorithms.

APPLICATION OF THE TAGUCHI QUALITY METHOD TO DESIGN NAVIGATION CONSTANTS THAT ARE ROBUST TO TARGET MANEUVERS

The object of this paper is to present a quality control approach to design the navigation constants for missile guidance law that are robust for target’s maneuverability and missile’s performance. Target’s maneuvers and missile's time constants are considered to be noise factors and Taguchi Quality Method (TQM) is used to conduct matrix experiments and determine the robust navigation constants. From the simulation results for three kinds of guidance laws, we find that robust navigation constants have a greater ability to enhance the interception of targets that are performing uncertain maneuvers with less interception time and smaller miss distances in the hitting phase.

Robust Optimum Design of PIFA for RFID Mobile Dongle Applications

A novel planar inverted F-antenna (PIFA) structure for a dongle-type RFID reader module with a resonant frequency of 920 MHz is proposed using a robust optimum design process. In order to take into account the robustness against noise factors, the Taguchi quality method is complemented to the evolutionary strategy optimization method. The validity of the proposed design process is verified through a noise test and comparisons of the computation and the experiment results.

A Practical Approach to Robust Design of a RFID Triple-Band PIFA Structure

This paper presents a practical methodology of obtaining a robust optimal solution for a multi U-slot PIFA (Planar Inverted F-Antenna) structure with triple bands of 433 MHz, 912 MHz and 2.45 GHz. Using evolutionary strategy, a global optimum is first sought out in terms of the dimensions of the slots and shorting strip. Then, starting with the optimized values, Taguchi quality method is carried out in order to obtain a robust design of the antenna against the changes of uncontrollable factors such as material property and feed position. To prove the validity of the proposed method, the performance of the antenna is predicted by general-purpose electromagnetic software and also compared to experimental results.

Construction and Analysis in Combining the Taguchi Method and the Back Propagation Neural Network in the PEEK Injection Molding Process

In order to promote the accuracy and reduce the dimension deviation of injection molding, the Taguchi quality method and the Back Propagation Neural Network (BPNN) were adopted in the injection molding process. The aim of the research is to enhance the optimization of the injection molding process using the most precursory material, Polyether Ether Ketone (PEEK), in the current plastic industry. First, the Taguchi quality method is applied to establish the design of experiment. An analysis of variance is done to arrive at the significant factors that influence the injection molding quality the most. Also, the significant factors are used to construct the BPNN prediction system. Next, the BPNN is employed to fine-tune the optimum conditions obtained from the Taguchi method for achieving optimum quality. The experimental results are within the 95% confidence interval, confirming that the factor's effects are reproducible and that the experimental results are reliable.

Optimization of the processing conditions and prediction of the quality for dyeing nylon and lycra blended fabrics

This paper is intended to determine the optimal processing parameters applied to the dyeing procedure so that the desired color strength of a raw fabric can be achieved. Moreover, the processing parameters are also used for constructing a system to predict the fabric quality. The fabric selected is the nylon and Lycra blend. The dyestuff used for dyeing is acid dyestuff and the dyeing method is one-bath-two-section. The Taguchi quality method is applied for parameter design. The analysis of variance (ANOVA) is applied to arrange the optimal condition, significant factors and the percentage contributions. In the experiment, according to the target value, a confirmation experiment is conducted to evaluate the reliability. Furthermore, the genetic algorithm (GA) is combined with the back propagation neural network (BPNN) in order to establish the forecasting system for searching the best connecting weights of BPNN. It can be shown that this combination not only enhances the efficiency of the learning algorithm, but also decreases the dependency of the initial condition during the network training. Most of all, the robustness of the learning algorithm will be increased and the quality characteristic of fabric will be precisely predicted.

Optimization of the electron‐beam‐lithography parameters for the moth‐eye effects of an antireflection matrix structure

AbstractIn this study, we mainly used the characteristics of electron‐beam lithography in measurement control and direct‐write technology to improve the physical restrictions and production processes of optical lithography and other nanopattern production methods. We did this by using a silicon wafer as a substrate, coating a negative‐tone photoresist, and using scattering and the reflection produced by the collision of an electron beam with the wafer lattice and the proximity effect of a secondary electron inside the electron‐beam photoresist to produce an antireflection matrix structure with a moth‐eye effect. In addition, we used the Taguchi quality method with an orthogonal array to plan the experiment and the signal‐to‐noise ratio to analyze the experimental data, and in the experimental process, we produced a full factorial equivalent experiment, using very few experiment repetitions and deriving optimum conditions. Also, we used back‐propagation neural networks to fine‐tune significant factors, allowing the production of the deepest process control parameters and thereby imparting to the antireflection matrix structure the best effect. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 102: 5303–5313, 2006

The Construction and Analysis of a Prediction Model for Combining the Taguchi Method and General Regression Neural Network for Injection Moulding

Plastics always show mould deformation after the injection moulding process and this can cause dimensional errors. The extent of a dimensional error can be controlled effectively by choosing adequate processing parameters. The target of this work was to reduce the dimensional error in the hexagonal screw diameter when processing polyether ether ketone (PEEK), and to improve the polymer's quality. First, we selected the factor levels according to the chosen quality characteristics. We decided that a prediction model for the injection process could be constructed using the Taguchi quality method and a general regression neural network (GRNN). The optimum conditions determined by the Taguchi method can be modified by the neural network. The injection moulding product can achieve a reduced dimensional error by adjusting the factor levels using this combined approach.

Application of a Taguchi-based neural network prediction design of the film coating process for polymer blends

In this paper, an approach for developing the prediction model for polymer blends using a back-propagation neural network (BPNN) combined with the Taguchi quality method is presented in an attempt to improve the deficiencies in current neural networks associated with the design of network architecture, including the selection of one optimal set of learning parameters to accomplish faster convergence during training and the desired accuracy during the recall step. The objective of the prediction model is to explore the relationships between the control factor levels and surface roughness in the film coating process. In addition, the feasibility of adopting this approach is demonstrated in the study optimizing the learning parameters of the BPNN structure to forecast the target characteristics of the product or process with various control conditions in the manufacturing system.