Search

Thermoplastic elastomer SEBS foams were prepared by using carbon dioxide (CO2) as a blowing agent and the process is classified as physical foaming method. During the foaming process, the diffusivity of CO2 need to be controlled since it is one of the parameter that will affect the final cellular structure of the foam. Conventionally, the rate of CO2 diffusion was measured experimentally by using a highly sensitive device called magnetic suspension balance (MSB). Besides, this expensive MSB machine is not easily available and measurement of CO2 diffusivity is quite complicated as well as time consuming process. Thus, to overcome these limitations, a computational method was introduced. Particle Swarm Optimization (PSO) is a part of Swarm Intelligence system which acts as a beneficial optimization tool where it can solve most of nonlinear complications. PSO model was developed for predicting the optimum foaming temperature and CO2 diffusion rate in SEBS foam. Results obtained by PSO model are compared with experimental results for CO2 diffusivity at various foaming temperature. It is shown that predicted optimum foaming temperature at 154.6 °C was not represented the best temperature for foaming as the cellular structure of SEBS foamed at corresponding temperature consisted pores with unstable dimension and the structure was not visibly perceived due to foam shrinkage. The predictions were not agreed well with experimental result when single parameter of CO2 diffusivity is considered in PSO model because it is not the only factor that affected the controllability of foam shrinkage. The modification on the PSO model by considering CO2 solubility and rigidity of SEBS as additional parameters needs to be done for obtaining the optimum temperature for SEBS foaming. Hence stable SEBS foam could be prepared. [ABSTRACT FROM AUTHOR]

Dermatology or skin disease is one of the popular diseases among other diseases these days. The features similarities between different types of skin diseases make diagnosis of skin diseases very complex. A patient needs dermatologist that has a sound and vast good experience in skin diseases in order to give precise results at the right time. This paper elaborates a prototype with back propagation neural network (BPNN) to assist the dermatologist. This prototype improves expert diagnosis method in term of time efficiency and diagnosis accuracy. The use of two feature selection methods namely Correlation Feature Selection (CFS) and Fast Correlation-based Filter (FCBF) help by providing a smaller number of features with greater accuracy and faster response time. The adjustment of parameter in BPNN gives good performance. The findings show that FCBF method offers the shortest elapsed time and highest result compared to CFS method and the full features with an accuracy of 91.2%. [ABSTRACT FROM PUBLISHER]