discovery of nonlinear inception of stall [16]. Understanding the stall behavior
from its infant stage is the vital part to initiate the control loop and also very
helpful in clarifying the flow mechanism of rotating stall. Thus the research
scenery in the period of early 90
th
could be signified as the co-development of
these two branches, active control and stall inception. The active control technique
kept developing in many ways. Good examples are the aero-mechanical scheme
of control on surge and rotating stall [17, 18], which used the structural feedback
instead of sensors and external actuators, so that the cost of implementation
could be reduced. The scope of stall inception study also increased covering the
precursor detection, characterization, signal processing, underlying mechanism
and stall warning. This was also the period when we started the research in the
field, resulted in the successful active control of surge by Nie et al. [19] and the
early study of stall inception in centrifugal compressor by Chen et al. [20].
In the following period from around mid 90
th
, as the research efforts went
on and diversified stall and its inception characteristics were evidenced [21], the
need became more apparent for going deeper into the fluid mechanics aspect of
the phenomenon rather than the description of only its dynamic characteristics.
Therefore the strengthened efforts in studying the internal flow mechanism in
blade passages can be witnessed. The question to be answered is what happened
in the blade passages while the rotating stall takes place in the compressor. This
would make our understanding of the flow instability more complete, in both blade
and system scales, and make the control strategy more pertinent and effective to
the causal factors for the loss of compressor stability, in particular if one wishes to
manipulate the blade design for stability improvement. In a long run, it should be
an important step forward towards solving one of the key problems in designing
turbomachinery, the prediction of the full range compressor performance.
It is interesting to note that as the research on stall inception and on internal
flow mechanism strengthened, the strategy of active control also updated. The
history is likely going back, via a spiral development, to its original aim of
stability enhancement. This was indeed the scenery from the end of 90
tpth
to the
present time, when the passive control strategy such as casing treatment regained
its development. The new development of passive control is in many ways
benefited by the innovative ideas gained in the course of active control study. Other
possibilities such as combined active/passive control are also under investigation.
It seems that the ten-plus years development of compressor instability research,
started from the breakthrough idea of its active control and enriched by the
numerous successes and lessons, is now entering a new period, which is viewed
not by the scale of its activities but by the realistic aim and refined means to
achieve its goal.
Pre-Stall Behavior.
Numerous evidences have been accumulated for the stall
behavior immediately prior to the established stall cells. This time period is usually
referred as the period of stall inception and the flow disturbances discovered and
believed to trigger the stall are called stall precursors. Normally, stall precursors
are identified in the dynamic pressure or velocity signals measured by sensors in
the tip region in front of rotor blades and commonly fall into two types. The first is
the mode type of long length scale wave which appears tens of rotor revolutions
time in advance of stall and builds up linearly before evolving into the finite
stall cells [11]. The second is the spiky type stall inception which initiates with
a small amplitude short length scale disturbances that can be detected only few
rotor revolutions time before straightly developing into stall cells [16].
ISSN 0236-3941. Вестник МГТУ им. Н.Э. Баумана. Сер. “Машиностроение”. 2006. № 2 115
