Maman A. Djauhari

Improved Monitoring of Multivariate Process Variability

The standard way of using the determinant of the average of sample covariance matrices for multivariate process variability control charting leads to a biased estimate of the control limits. In this paper, we present an improved control chart having unbiased control limits. It has smaller ARL than the standard chart and, hence, it is more sensitive to shifts in the process variability. An example is given to illustrate these advantages.

Multivariate Process Variability Monitoring

Understanding multivariate variability is a difficult task because there is no single measure that can be properly used. This article presents a new measure that features good properties. If this measure is simultaneously used with generalized variance, it will give a better understanding of multivariate variability. It can also efficiently be used for large data sets with high dimensions. Furthermore, when it is used for constructing a Shewhart-type chart to monitor multivariate variability, the resulting chart has a much better out-of-control ARL than the generalized variance chart. An example illustrates its advantage.

Leakage currents and pollutants properties of porcelain insulators from geothermal area

This paper presents the experimental results of one new-clean and four different geothermal naturally polluted outdoor porcelain insulators. The conducted experiments were leakage current and applied high voltage measurements, in the hermetically sealed chamber, on various temperature, humidity, pressure and applied voltage. Both quantities were measured by a storage digital oscilloscope. The data could be recorded and transferred to a computer, saved in softcopy form. The leakage current waveforms were analyzed using FFT and the relation to environmental parameters were analyzed using correlation coefficient matrix and principal component analysis. The results indicated that power factors and leakage current amplitudes closed to humidity, the phase angle closed to THD, and temperature opposited to humidity. On the polluted insulator, the comparison of impedance between that on wet and dry conditions reached 0.134, as the lowest. The reached lowest phase angle and THD were 24 degree and 7,5% respectively. These phenomena occurred due to contamination on the porcelain insulator surface that make more fairly conductive. The most chemical element on EDAX test was silicon.

STOCHASTIC PATTERN OF TRAFFIC ACCIDENTS IN BANDUNG CITY

This paper deals with a study to gain a better understanding about traffic accident phenomena in Bandung City, the capital of West Java. The principal concern, which is very important for traffic management, is on the following traffic accident parameters; daily accident rate, daily accident-hazard-level, probability of daily fatal accidents and daily fatal victim rates. This study includes data collection and summarization, which will be used to estimate those parameters by using the following three stochastic models; Generalized Poisson (GP) distribution to explain the first two parameters, Generalized Poisson-Quasi Binomial (GPQB) distribution and Generalized Poisson-Generalized Poisson (GPGP) distribution to explain respectively the third and the fourth parameters. A goodness of fit test will show the significance of those models and some important results will be discussed.

A robust canonical correlation analysis in the study of leakage current behavior of geothermal polluted porcelain insulators

This study presents the experimental results of a new-clean and three naturally geothermal polluted porcelain insulators. The experimentals were LC and applied high voltage measurements used a two-channel storage digital oscilloscope, in the hermetically sealed chamber, where temperature, humidity, pressure and applied voltage could be adjusted and measured simultaneously. The LC waveforms were analyzed using FFT and the statistics concerns were analyzed using FMCD of robust multivariate statistical tools. Furthermore, the data were made correlations using CCA. The LCs were significantly influenced by RH. Otherwise, very small data were suspected as outliers, and the dependent variables were linearly function to independent variables, based on CCA.

Impedances and surface conditions of naturally coastal-polluted epoxy resin insulators

This paper presents the experimental results of one new-clean and six different coastal area naturally polluted outdoor epoxy resin insulators. The conducted experiments were leakage current and applied high voltage measurements, in the hermetically sealed chamber on various RH. Both quantities were measured by a storage digital oscilloscope. The data could be recorded and transferred to a computer, saved in softcopy form. In this paper, it is revealed two values of humidity, low and high. Furthermore, from each measurement, it was obtained impedance, phase angle and THD, and impedance ratio between low and high RH. The results indicated that on the new-clean, on low and high RH, the insulator was no change. Whereas on first withdrawal, the insulator impedance was significantly different on high and low RH, as ratio 65.45%. This case was caused by pollutant adhered on the insulator surface, due to dry season. Finally, on the sixth withdrawal, the ratio of impedance on high and low RH reached 67.97%, the phase angle and THD changed from 77o to 51.43o and from 18.7% to 8.8% respectively for low and high RH. It is supposed, besides pollutant, it was caused by the degradation on the surface, that indicated by SEM test results on the epoxy resin specimens. The degradations of epoxy resin surface were also shown in this paper gradually using SEM.

Investigation of comparison on porcelain and epoxy resin insulator conditions from a coastal area due to various parameters

This paper presents the experimental results of eight months naturally coastal polluted porcelain and epoxy resin insulators. The experiments were leakage current and applied voltage measurements, in a hermetically sealed chamber, where temperature, humidity, pressure and applied voltage could be adjusted and measured simultaneously. The leakage current waveforms were analyzed using FFT and the relation to environmental parameters were analyzed using correlation matrix. It was also conducted SEM and EDAX tests. The humidity contributed to leakage current and phase angle on the porcelain insulator were 0.60 and -0.66 respectively. Whereas, on polluted epoxy resin one, the humidity did not influence to leakage current and phase angle considerably. The surface of polluted specimens became rougher than the new one. The most three chemical elements in pollutants were chlorine, silicon and ferrum.

Multivariate analysis for investigation of outdoor insulator conditions

The works reported in this paper deals with experimental study on a porcelain outdoor insulator, which has been installed at a coastal switchyard, during eight months. The measurements were also carried-out on new-clean one, in order to have a comparison with the polluted one. The insulators have been subjected under high voltage, and were measured the leakage currents, in a hermetically sealed chamber, wherein the temperature, humidity and pressure could be controlled, for simulating the environmental condition of site. The leakage currents were analyzed with FFT method, and correlation coefficient and PCA of multivariate statistical tools. The correlation coefficient shows that the humidity influenced the leakage current magnitude, phase angle and the third harmonics as 56%, -88% and 59%, respectively, on the new-clean one. In case of polluted one, the humidity influenced the leakage current magnitude, phase angle and third harmonics as 60%, -68% and 53% respectively.