01-12-2012, 02:08 PM
A Methodology for Evaluation of Hurricane Impact on Composite Power System Reliability
1A Methodology for Evaluation.pdf (Size: 519.75 KB / Downloads: 26)
Abstract
Adverse weather such as hurricanes can have significant
impact on power system reliability. One of the challenges of
incorporating weather effects in power system reliability evaluation
is to assess how adverse weather affects the reliability parameters
of system components. In this paper, a fuzzy inference system
(FIS) built by using fuzzy clustering method is combined with the
regional weather model to solve the preceding problem. The composite
power system is assumed to be partitioned into different regions
and the FIS maps the nonlinear functional relationship between
hurricane parameters and the increment multipliers of the
failure rates (IMFR) of the transmission lines in different regions.
The possible case that transmission lines traverse bordering regions
is investigated by using the weighted average method. Since
hurricanes last only a limited time period, the short-term reliability
indices over the duration of hurricane instead of the steady-state
ones are calculated by using the minimal cut-set method (MCSM).
The proposed method is applied to the modified IEEE Reliability
Test System (RTS). The implementation demonstrates that the proposed
method is effective and efficient and is flexible in applications.
INTRODUCTION
MANY power system components, such as transmission
and distribution lines, are exposed to weather. The
severe weather like a hurricane is an extreme weather condition
and can seriously jeopardize power system operation. More
accurately predicting the impact of hurricanes can help utilities
for better preparedness and restoration arrangements [1], [2].
In the past decades, many methods have been proposed to
evaluate power system reliability considering weather effects.
These methods include two-state weather model [3], multi-state
weather model [4], regional weather model [5], and statistical
regression method [2], [6].
One of the challenges of incorporating weather effects in
power system reliability evaluation is to assess how adverse
weather affects the reliability parameters of system components,
e.g., the failure rates of transmission lines. Regression
method and Bayesian network have been used to solve this
problem in [7] and [8].
OVERALL EVALUATION SCHEME
As shown in Fig. 1, the reliability evaluation scheme of composite
power systems affected by hurricanes consists of three
steps: hurricane forecast, assessment of hurricane impact, and
system reliability evaluation.
Hurricane Impact
Since the hurricane develops and dissipates with time, its impact
on a composite power system can be described in two aspects:
temporal and spatial. Temporal refers to the fact that the
impact of hurricanes in a given region is different at different
times; spatial refers to the fact that at a given time, the effects
of hurricanes are different in different regions. The impact of
hurricanes can be represented by using some defined parameters
that indicate their severity. When a hurricane moves on land, its
effects are mainly strong winds and heavy rainfall. Thus, in this
paper, wind speed and rainfall are selected as two hurricane parameters.
Fuzzy Inference
In [1], it was pointed out that there is rough correspondence
between hurricane severity and the number of power outages. In
this step, the impact of hurricanes is mapped to the increment
of the failure rates of transmission lines by using FIS formed by
using fuzzy clustering method.
C. System Reliability Evaluation
Finally, the output of the FIS is input to the reliability evaluation
model of composite power systems to evaluate the impact
of hurricanes. Here, MCSM is used to calculate the short-term
reliability indices.
Data Collection and Preprocessing
Since the FIS in this paper is built by using fuzzy clustering
method, the required data need to be collected and preprocessed.
This includes the regional hurricane parameters and the regional
IMFR of transmission lines.
1) Data Collection: This refers to the collection of the historical
data of hurricane parameters and the failure rates of transmission
lines in the normal weather and during hurricanes, respectively.
Usually, the former can be collected by referring to
historical meteorology records; the latter cannot be directly collected
and can be obtained by using some transformation techniques
like those in [7] and [8]. Basically, the relationship between
failure frequency and failure rate is used: .
Here, is the number of failures during a time interval, and it
can be obtained from historical records.
CONCLUSION
Hurricanes have a significant impact on composite power
system reliability. To assess how hurricane affects the reliability
parameters of transmission lines, this paper combines a fuzzy
clustering-based FIS with regional weather model to map the
relationship between hurricane parameters and the IMFR of
transmission lines. Here, two fuzzy clustering methods are used
to build two types of FIS: subtractive clustering is used to build
S-FIS and FCM is used to build M-FIS. Finally, MCSM is
used to compute the short-term reliability indices of composite
power systems.