how to identify input and disturbance in ARX 221 model identification
조회 수: 1 (최근 30일)
이전 댓글 표시
I am doing system identification using linear parametric model.When i am simulating with one input ,I am getting four outputs with respect to four inputs.I am unable to find the original input to output.
Kindly help me in this regard. Thanks for your advanced help.
댓글 수: 0
답변 (1개)
Rajiv Singh
2011년 4월 27일
Hi Ganesh, I saw your email to me as well (I recommend using this forum rather than email for asking questions). I believe your model has two outputs and two measured inputs. To extract the transfer function for one I/O pair (i, j), you could do:
sys = arx221('m') sysij = sys(i, j)
Note that an identified model is of the general form: y=G u + H e, where G is the transfer function between the measured inputs and outputs and H is the transfer function between the unmeasured or noise inputs and outputs. When you convert a model to tf format of Control System Toolbox, by default, the noise channels are treated as measured ones and the overall transfer function has ny extra inputs, where ny is the number of model outputs (since there is one noise channel per output).
If you believe arx221 model should have had 1 input channel, please check your estimation data; is it SISO? What does size(data) return? The most likely scenario is that you are inadvertently using a 2 input, 2 output data set.
HTH. Rajiv
댓글 수: 0
참고 항목
카테고리
Help Center 및 File Exchange에서 Linear Model Identification에 대해 자세히 알아보기
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!