Semi-Intrusive Multivariable Model Invalidation 1
Abstract:
This paper introduces a mechanism for testing multivariable models employed by model-based controllers. Although external excitation is not necessary, the data collection includes a stage where the controller is switched to open-loop operation (manual mode). The main idea is to measure a certain “distance ” between the closed-loop and the open-loop signals, and then trigger a flag if this “distance ” is larger than a threshold level. Moreover, a provision is made for accommodating model uncertainty. Since no hard bounds are assumed with respect to the noise amplitude, the model invalidation mechanism works in a probabilistic framework. Key words: Model invalidation; performance monitoring; model-based control; multivariable control systems; stochastic noise; model uncertainty. 1
Citations
| 696 | Time Series Analysis: Forecasting and Control – Box, Jenkins, et al. - 1994 |
| 37 | System Identification: Theory for the User, 2nd Ed – Ljung - 1999 |
| 21 | Spectral analysis and its applications. Holden-Day – JENKINS, WATTS - 1968 |
| 7 | Tests for comparing two estimated spectral densities – Coates, Diggle - 1986 |
| 6 | Nonparametric comparison of cumulative periodograms – Diggle, Fisher - 1991 |
| 5 | A control-loop performance monitor – Hagglund - 1995 |
| 1 | Automatic detection of sluggish control loops – Hägglund - 1999 |
| 1 | Assessment of control loop performance. The Canadian – Harris - 1989 |
| 1 | A simple method for detection of stiction in process control loops. Control Engineering Practice 7(10 – Horch - 1999 |
| 1 | A simple nonparametric test for the equality of autocorrelation structures of two stochastic processes, with application to experimental data – Pudney, Deadman, et al. - 1999 |
