Kreangsak Tamee

Towards clustering with XCS

This paper presents a novel approach to clustering using an accuracy-based Learning Classifier System. Our approach achieves this by exploiting the generalization mechanisms inherent to such systems. The purpose of the work is to develop an approach to learning rules which accurately describe clusters without prior assumptions as to their number within a given dataset. Favourable comparisons to the commonly used k-means algorithm are demonstrated on a number of synthetic datasets.

Distributed Sensors Using a PANDA Ring Resonator Type in Multiwavelength Router

We propose a new system of microring sensing transducer using a PANDA ring resonator type, in which the sensing unit is consisted of an optical add/drop filter and two nanoring resonators, where one ring is placed as a transducer (sensing unit), the other ring is set as a reference ring. In operation, the external force is assumed to exert on the sensing ring resonator. The method of finite difference time domain (FDTD) via the computer programming called Optiwave is used to simulate the sensing behaviors. The obtained results have shown that the change in wavelength due to the change in sensing ring radii is seen, in which the wavelength shift of 1 nm resolution is achieved. The distributed sensing system is designed, which is available for network sensing applications. The behavior of light within a PANDA ring resonator is also analyzed and reviewed.

A Learning Classifier System Approach to Clustering

This paper presents a novel approach to clustering using a simple accuracy-based learning classifier system. Our approach achieves this by exploiting the evolutionary computing and reinforcement learning techniques inherent to such systems. The purpose of the work is to develop an approach to learning rules which accurately describe clusters without prior assumptions as to their number within a given dataset. Favourable comparisons to the commonly used k-means algorithm are demonstrated on a number of datasets

Using Self-Organizing Maps with Learning Classifier System for Intrusion Detection

Learning Classifier Systems (LCS) have previously been shown to have application in Intrusion Detection. This paper extends work in the area by applying the Self-Organizing Map (SOM) for creating the new input string by 2-bit encoding rely on degree of deviation of normal behaviour. The performance of systems is investigated under an FTP-only dataset. It is shown that the proposed system is able to perform significantly better than the conventional XCS, modified XCS and twelve ML algorithms.

Resistance Spot Welding Optimization Based on Artificial Neural Network

Resistance Spot Welding (RSW) is processed by using aluminum alloy used in the automotive industry. The difficulty of RSW parameter setting leads to inconsistent quality between welds. The important RSW parameters are the welding current, electrode force, and welding time. An additional RSW parameter, that is, the electrical resistance of the aluminum alloy, which varies depending on the thickness of the material, is considered to be a necessary parameter. The parameters applied to the RSW process, with aluminum alloy, are sensitive to exact measurement. Parameter prediction by the use of an artificial neural network (ANN) as a tool in finding the parameter optimization was investigated. The ANN was designed and tested for predictive weld quality by using the input and output data in parameters and tensile shear strength of the aluminum alloy, respectively. The results of the tensile shear strength testing and the estimated parameter optimization are applied to the RSW process. The achieved results of the tensile shear strength output were mean squared error (MSE) and accuracy equal to 0.054 and 95%, respectively. This indicates that that the application of the ANN in welding machine control is highly successful in setting the welding parameters.

Muscle Sensor Model Using Small Scale Optical Device for Pattern Recognitions

A new sensor system for measuring contraction and relaxation of muscles by using a PANDA ring resonator is proposed. The small scale optical device is designed and configured to perform the coupling effects between the changes in optical device phase shift and human facial muscle movement, which can be used to form the relationship between optical phase shift and muscle movement. By using the Optiwave and MATLAB programs, the results obtained have shown that the measurement of the contraction and relaxation of muscles can be obtained after the muscle movements, in which the unique pattern of individual muscle movement from facial expression can be established. The obtained simulation results, that is, interference signal patterns, can be used to form the various pattern recognitions, which are useful for the human machine interface and the human computer interface application and discussed in detail.

PSYCHIATRIC INVESTIGATION USING WGMs IN MICRORING CIRCUITS

The use of an electrical probe is formed by whispering gallery modes (WGMs) of light within the coated microring circuits, in which the electrical signal is generated by trapped electron tunneling along the circular path of the coated microring circuit. The collection of electrons is formed within the WGMs, where in this study, a modified nonlinear microring resonator known as a PANDA ring resonator is coated by gold material and forms the mirroring circuit. The induced current (magnetic field) within the circuit occurs by the coupling effects between trapped electrons and coated ring, which can penetrate into the brain cells and transform to the required signals via the terahertz carrier for psychiatric investigations. The use of WGMs for 3D image construction using a PANDA conjugate mirror is also discussed, which is useful for thermal and imaging sensors.

Planning a sports training program using Adaptive Particle Swarm Optimization with emphasis on physiological constraints

ObjectiveAn effective training plan is an important factor in sports training to enhance athletic performance. A poorly considered training plan may result in injury to the athlete, and overtraining. Good training plans normally require expert input, which may have a cost too great for many athletes, particularly amateur athletes. The objectives of this research were to create a practical cycling training plan that substantially improves athletic performance while satisfying essential physiological constraints. Adaptive Particle Swarm Optimization using ɛ-constraint methods were used to formulate such a plan and simulate the likely performance outcomes. The physiological constraints considered in this study were monotony, chronic training load ramp rate and daily training impulse.ResultsA comparison of results from our simulations against a training plan from British Cycling, which we used as our standard, showed that our training plan outperformed the benchmark in terms of both athletic performance and satisfying all physiological constraints.

Intelligence Planning for Aerobic Training Using a Genetic Algorithm

A training plan is an important part of aerobic training. A training plan with a good sequence of high intensive training sessions and low intensive training sessions will substantially raise athletic performance. A creation of training plan require a sport scientist or a sports coach to do. An athlete who trains with limit in sports science knowledge may get injury. In this study, we propose a systemic implementation using a genetic algorithm (GA) to find optimal training plan. Comparison of this study result and an independently created, apparently reliable training plan, it reveal that GA is obtain capability to find optimal training plan.

Distributed photon network sensors via a wavelength router

This study proposes a novel system of multi-photons generation and networking using an optical add/drop filter incorporating two nanoring resonators and the wavelength routers. By using some suitable parameters of the input and the control signals via an input and add ports, the intense optical fields can form and propagate within the system. Simulation results obtained have shown that multi-photons/atoms trapping pulses can be generated by using dark solitons and Gaussian pulses into the system, which can be used to form the multi-variable network for photons/atoms trapping and transportation via the wavelength routers. In application, the proposed system can be fabricated on a chip, in which the use for network sensors (communications) can be realised.