CIFER®-MATLAB Interfaces: Development and ... - Cal Poly

In the case of a multiple input system, the frequency responses can be conditioned to remove the
effects of correlation between different inputs. After this conditioning, the frequency responses
from each separate window are combined to form a single frequency response based on the most
accurate segments of each window. The combination is achieved by optimizing the data that has
useful content in both low and high frequency regions. This is the last step in the frequency
response generation process.
Less involved analysis, compared to full state-space model generation, can include calculations of
RMS, cutoff frequencies, bandwidth, crossover characteristics, and data consistency checks using
frequency response arithmetic. All of these tools can give useful insight to system behavior
without generating more complex math models. CIFER ® offers more complete identification
through programs that will fit transfer functions to individual responses, and full state-space
models to a series of responses. In addition to these powerful analysis tools, CIFER ® offers
utilities for plotting and organizing output, as well as managing the storage and organization of
data.
One last important tool is the ability to validate state-space models against other time history data.
These validation time histories are typically generated using a doublet maneuver as opposed to a
sweep. A doublet is a short maneuver that moves through a range of motion for a control surface,
beginning and ending in steady level flight as shown in Figure 1.3.
4