Abstract: This paper discusses on the use of Spline Interpolation
and Mean Square Error (MSE) as tools to process data acquired from
the developed simulator that shall replicate sea bed logging environment.
Sea bed logging (SBL) is a new technique that uses marine
controlled source electromagnetic (CSEM) sounding technique and is
proven to be very successful in detecting and characterizing hydrocarbon
reservoirs in deep water area by using resistivity contrasts. It uses
very low frequency of 0.1Hz to 10 Hz to obtain greater wavelength.
In this work the in house built simulator was used and was provided
with predefined parameters and the transmitted frequency was varied
for sediment thickness of 1000m to 4000m for environment with and
without hydrocarbon. From series of simulations, synthetics data were
generated. These data were interpolated using Spline interpolation
technique (degree of three) and mean square error (MSE) were
calculated between original data and interpolated data. Comparisons
were made by studying the trends and relationship between frequency
and sediment thickness based on the MSE calculated. It was found
that the MSE was on increasing trends in the set up that has the
presence of hydrocarbon in the setting than the one without. The MSE
was also on decreasing trends as sediment thickness was increased
and with higher transmitted frequency.
Abstract: This paper aims to develop a NOx emission model of
an acid gas incinerator using Nelder-Mead least squares support
vector regression (LS-SVR). Malaysia DOE is actively imposing the
Clean Air Regulation to mandate the installation of analytical
instrumentation known as Continuous Emission Monitoring System
(CEMS) to report emission level online to DOE . As a hardware
based analyzer, CEMS is expensive, maintenance intensive and often
unreliable. Therefore, software predictive technique is often
preferred and considered as a feasible alternative to replace the
CEMS for regulatory compliance. The LS-SVR model is built based
on the emissions from an acid gas incinerator that operates in a LNG
Complex. Simulated Annealing (SA) is first used to determine the
initial hyperparameters which are then further optimized based on the
performance of the model using Nelder-Mead simplex algorithm.
The LS-SVR model is shown to outperform a benchmark model
based on backpropagation neural networks (BPNN) in both training
and testing data.