Artificial neural networks which are inspired by the concept of the biological neurons are commonly used in many applications including in the field of weather forecasting. The neural networks approaches have provided an educated process for weather forecasting as well as a viable means of the prediction of raindrop. This paper attempts to determine the suitability and the applicability of artificial neural networks for rain prediction based on temperature, pressure and humidity. Those conditions have been used as input data and solution was classified as percentage of raining. Multilayer perceptron network with two different learning algorithms have been studied. The multilayered perceptron trained using Lavenberg Marquardt algorithm has been proven to produce better results with accuracy percentage (99.75%) as compared to back propagation (94.57%). pressure; humidity.

No abstract

No abstract

This paper is aimed to detect the heart abnormality by using multilayer perceptron (MLP) network. Several data from the electrocardiogram (ECG) signal is extracted to be set as the input parameters. In order to get the best result unipolar sigmoid, bipolar sigmoid, tangent hyperbolic and conic section function. The result obtained is then compared among the activation function techniques.

A new class of colloidal insulating transformer oil is formulated by dispersing CNT nanomaterials with mineral oil in purpose to enhance the breakdown strength performanc existing transformer oil. This paper represents the experimental studied dealing with the influence of CNT nanomaterials mix with transformer mineral oil in term of AC breakdown voltage at various gap distances. The concentrations of CNT nanomaterials were measured in range 0.01g/L to 0.2g/L. The impact of three gap distances (1.5mm, 2.5mm and 3.5mm) for mushroom-mushroom electrode configuration is studied in order to observe and trend between conventional mineral oil and CNT nanofluids. To employ CNTs as effective reinforcement in mineral oil, proper dispersion methods need to be applied.

No abstract

No abstract

This paper presenting a new electrocardiogram (ECG) noise removal based on wavelet transform technique. The system involves two stages filtration, which denoising with ECG thresholds with several threshold methods. The residual signal will feed to the second stage filtration using the high/low wavelet based filter. Wavelet transform methods capable to counter the unpredictability of motion artifact n movement. Wavelets may reduce the motion artifact effects since it has the capability to filter both high and low frequency signals and produce good results for both stationary and with ECG thresholds with several threshold methods. The residual signal will feed to the second stage filtration using the high/low wavelet based filter. Wavelet transform methods oise which produce by physical movement. Wavelets may reduce the motion artifact effects since it has the capability to filter both high and low frequency signals and produce good results for both stationary and he wavelet arrangement is shown by using ECG