Identifying Key Reliability Factors in Micro- Grid Systems Using Principal Component Analysis.
No Thumbnail Available
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Description
One way to solve the shortage in power supply and the rapid load
growth is by operating power systems that could improve power supply
reliability. The study aims to carry out a holistic evaluation by identifying the
several reliability variables that could influence the micro-grid power system's
reliability which is vital in electricity generation. Thirty-three reliability
variable factors that are commonly observed to influence power systems
reliability were chosen for the micro-grid power systems study and examined
using the principal component analysis (PCA). The system reliability key
variables were evaluated using the StatistiXL software. A structured
questionnaire was crafted considering thirty-three reliability variables,
harvested from literature, and administered to respondents in the micro-grid
power system industry. The respondent size was determined at a level of
confidence of 95% and an error margin of 5% was deployed to corroborate an
adequate population size representation which validated the study data.
StatistiXL software was deployed to analyze the (mxn) data matrix, collated
from the respondents' scores. The matrix was used as the input variable for the
model deployed for the factor analysis. Nine factors with eigenvalues (λ>1)
were mined and labeled for the analysis, but all the trivial variables were
discarded. The PCA result holistically pinpointed the key reliability variables
that influence the micro-grid reliability, revealing that system availability
represented by factor 1 (F1) loaded 24% of the total variables studied, with
reliability cluster including Mean Time Between Failures (MTBF) = - 0.844,
Mean Time to Repair (MTTR) = - 0.737, Demand Response (DR) technique =
0.752, Failure Rate = 0.647 among others. The failure rate and the frequency of
outages in F1, were an indication that system availability would be influenced,
thereby affecting the micro-grid performance. The study also extracted some
weak factor loading, F8 and F9 indicting them as reliability variables whose influences do not impact negatively on the micro-grid reliability but should not
be discarded in the study of the reliability of micro-grid power systems. Hence
an attempt to improve the system's reliability, concentrating on the key
variables factors, the weak variables should not be neglected.
Keywords
TK Electrical engineering. Electronics Nuclear engineering