Environmental Engineering Reference
In-Depth Information
Fig. 14 Variation of
γ
with
Φ
/
Φ
LBO
for different L
fuel
,
when methane is used as fuel.
Result shows that method is
capable to detect LBO
(Chaudhari et al.
2013
)
produce the robust measure of the proximity to LBO. Consequently, the compu-
tation time required for producing the desired LBO metric is higher, which hampers
the time response characteristic of sensing technique. This major shortcoming of
those discussed methods is considerably reduced in the proposed
flame color-based
LBO detecting technique as the computation time required for simply reading the
images is very low. Apart from this, the proposed method is cost-effective and
easily implementable and shows consistency at different degrees of reactant
premixing.
fl
3.2 Symbolic Time Series Analysis
In this investigation, the use of STSA of captured heat release
fluctuations was
explored as a tool for early detection of imminent LBO. This approach has been
successfully implemented by Bukkapatnam et al. (
1995
,
2000
), Chin et al. (
2005
),
Gupta et al. (
2006
,
2007
), Chakraborty et al. (
2008
) for different systems. However,
STSA for experimentally obtained time series data in combustion systems has not
been previously reported. A major advantage of working with symbols is that the
ef
fl
cantly enhanced relative to what can
be achieved by direct analysis of the original time series data.
In STSA, the range of data is divided into a number of non-overlapping regions.
A symbol is assigned to each region. The collection of symbols is referred to as
ciency of numerical computation is signi
Search WWH ::
Custom Search