Norizan Mohamed

Comparing forecasting performances between multilayer feedforward neural network and recurrent neural network in Malaysia’s load

Abstract This paper presents the use of two artificial neural networks models, namely the multilayer feedforward neural network (MLFF) and the recurrent neural network (RNN) are applied for Malaysia’s load forecasting. For this purpose, a half hourly load data is divided equally into three distinct sets for training, validation and testing. We use backpropagation as the learning algorithm and the sigmoid function as the transfer function for both hidden land output layers. The forecasting performances of were compared between these two models. We use the sum squared error (SSE) as the measure of performance and the correlation coefficient r , as the measure of relationship between the actual and the predicted values. Results show that, multilayer feedforward neural network (MLFF) and recurrent neural network (RNN) have comparable accuracy but the sum squared error for multilayer feedforward neural network (MLFF) is lower, thus making it better model than recurrent neural network (RNN).

Carta Kawalan EWMA dan Carta Kawalan CUSUM: Satu Perbandingan Prestasi Menggunakan ARL

Carta kawalan kualiti EWMA (Exponentially Weighted Moving Average) dan CUSUM (Cumulative Sum) digunakan dengan meluas untuk mengawal kualiti produk yang dihasilkan daripada proses pengeluaran. Program komputer ditulis dalam bahasa pengaturcaraan SAS (Statistical Analysis System) digunakan untuk mengira purata panjang larian, ARL (average run length) bagi anjakan dalam min proses bagi carta kawalan optima EWMA dan CUSUM. ARL nominal ditetapkan pada 50, 100, 250, 500, 1000, 1500 dan 2000. Analisis dijalankan terhadap prestasi carta kawalan optima EWMA dan CUSUM bagi mengesan anjakan kecil dan besar dalam min proses. Hasil daripada carta kawalan EWMA dan CUSUM bagi mengesan anjakan kecil dan besar dalam min proses dibincangkan. Katakunci: Purata Panjang Larian; Anjakan; EWMA (Exponentially Weighted Moving-Average) ; CUSUM (Cumulative Sum); kawalan optima. EWMA (Exponentially Weighted Moving-Average) control charts and CUSUM (Cumulative Sum) control charts are widely used in industry for prosess and measurement control. This paper provides a computer implimentation of a fast and accurate algorithm in programming SAS (Statistical Analysis System) to compute average run length (ARL) for CUSUM and EWMA charts to monitor the process mean. The in-control ARL is set to be at 50, 100, 250, 500, 1000, 1500 dan 2000. The sensitive for optimal control EWMA and CUSUM in monitoring shift is then be analyze. The results from EWMA and CUSUM control charts for monitoring any small and large increases in process mean be discuss. Keywords: Average Run Length; Shift; EWMA (Exponentially Weighted Moving-Average); CUSUM (Cumulative Sum) ; optimal control.