Khairun Azizi Mohd Azizli

Use of waste gypsum to replace natural gypsum as set retarders in portland cement

The present study is focused on clarifying the influence of waste gypsum (WG) in replacing natural gypsum (NG) in the production of ordinary Portland cement (OPC). WG taken from slip casting moulds in a ceramic factory was formed from the hydration of plaster of paris. Clinker and 3-5wt% of WG was ground in a laboratory ball mill to produce cement waste gypsum (CMWG). The same procedure was repeated with NG to substitute WG to prepare cement natural gypsum (CMNG). The properties of NG and WG were investigated via X-ray Diffraction (XRD), X-ray fluorescence (XRF) and differential scanning calorimetry (DSC)/thermogravimetric (TG) to evaluate the properties of CMNG and CMWG. The mechanical properties of cement were tested in terms of setting time, flexural and compressive strength. The XRD result of NG revealed the presence of dihydrate while WG contained dihydrate and hemihydrate. The content of dihydrate and hemihydrates were obtained via DSC/TG, and the results showed that WG and NG contained 12.45% and 1.61% of hemihydrate, respectively. Furthermore, CMWG was found to set faster than CMNG, an average of 15.29% and 13.67% faster for the initial and final setting times, respectively. This was due to the presence of hemihydrate in WG. However, the values obtained for flexural and compressive strength were relatively the same for CMNG and CMWG. Therefore, this result provides evidence that WG can be used as an alternative material to NG in the production of OPC.

Modified Recursive Least Squares algorithm to train the Hybrid Multilayered Perceptron (HMLP) network

In this paper, a new learning algorithm, called the Modified Recursive Least Square (MRLS), is introduced for the Hybrid Multilayered Perceptron (HMLP) network. Adopting the Recursive Least Square (RLS) algorithm as its basis, the MRLS algorithm differs from RLS in the way that the weight of the linear connections for the HMLP network is estimated. The convergence rate of the MRLS algorithm is further improved by varying the forgetting factor, optimizing the way the momentum and learning rate are assigned. To investigate its applicability, the MRLS algorithm is demonstrated on the HMLP network using six benchmark data sets obtained from the UCI repository. The classification performance of the HMLP network trained with the MRLS algorithm is compared with those of the HMLP network trained with the Modified Recursive Prediction Error (MRPE) algorithm and the MLP trained with the standard RLS algorithm as well as with other commonly adopted machine learning classifiers. The comparison results indicated that the proposed MRLS trained HMLP network provides significant improvement over RLS trained MLP network, MRPE trained HMLP network, and other machine learning classifiers in terms of accuracy, convergence rate and mean square error (MSE).

Analysis of Mineralogical Component of Palm Oil Fuel Ash with or without Unburned Carbon

The purpose of this paper is to investigate the mineralogical components of palm oil fuel ash (POFA) with or without unburned carbon by using semi-quantitative X-ray diffraction (XRD). Original POFA taken from palm oil mill was ground in ball mill, producing ground POFA (GPOFA). Unburned carbon was removed by heating the GPOFA at 500 °C for 1 h, producing treated POFA (TPOFA). Neither glassy phase crystallization nor agglomeration of GPOFA particles occurred during the heat treatment. The content of crystalline phase and glassy phase in GPOFA or TPOFA was determined using semi-quantitative XRD analysis internal standard method. The results showed that GPOFA and TPOFA contained 67.22% and 73.89% of glassy phase, respectively, which could be used as pozzolanic materials. The presence of P2O5, K2O and Na2O in POFA was the reason of reducing the melting point of POFA in boiler.