Estimation of Uncertainty Bounds for the Future Performance of a Power Plant

Prediction of the future performance of large-scale power plants can be very relevant for the operators of these plants, as the predictions can indicate possible problems or failures due to current operating conditions and/or future possible operating conditions. A problem in predicting the future performance of these plants is that available models of the plants are uncertain. In this brief, three schemes for predicting uncertain dynamical systems are presented. The schemes estimate upper and lower bounds on the system performance. Two of the schemes are statistically based, one only based on recent data and the other is based on operating points as well. The third proposed scheme uses dynamical models of the prediction uncertainties, like in H infin-control. The proposed schemes are subsequently applied to experimental data from a coal-fired power plant. Two sets of data from an actual power plant are used, one containing normal plant operation and in the second set, coal is accumulating in the coal mill due to an unbalance in the operating conditions. These tests showed that Schemes II and III did bound the real system performance, while Scheme I failed doing so. In addition, the plant was simulated operating under the same conditions with additional large disturbances. These simulations were used to investigate the robustness and conservatism of the proposed schemes. In this test, Schemes I and II failed, while Scheme III succeeded.

[1]  Arvid Naess.  Prediction of extreme response of nonlinear structures by extended stochastic linearization , 1995 .

[2]  J. Doyle,et al.  Robust and optimal control , 1995, Proceedings of 35th IEEE Conference on Decision and Control.

[3]  Jie Chen,et al.  Robust Model-Based Fault Diagnosis for Dynamic Systems , 1998, The International Series on Asian Studies in Computer and Information Science.

[4]  Evgeny Loskutov,et al.  Prognosis of Qualitative Behavior of a Dynamic System by the Observed Chaotic Time Series , 2001 .

[5]  Jan Dimon Bendtsen,et al.  Bumpless transfer between advanced controllers with applications to power plant control , 2003, 42nd IEEE International Conference on Decision and Control (IEEE Cat. No.03CH37475).

[6]  Cécile Viboud,et al.  Prediction of the spread of influenza epidemics by the method of analogues. , 2003, American journal of epidemiology.

[7]  Jan Dimon Bendtsen,et al.  Multi-dimensional gain scheduling with application to power plant control , 2003, 42nd IEEE International Conference on Decision and Control (IEEE Cat. No.03CH37475).

[8]  L. Ya. Bukharbaeva,et al.  Dynamic System Simulation for Prediction of the Number of Patients with Malignant Tumors in Various Clinical Groups and Assessment of the Demand for Financial Resources , 2004 .

[9]  Gabriele Pannocchia,et al.  Robust model predictive control with guaranteed setpoint tracking , 2004 .

[10]  Moritz Diehl,et al.  Robust dynamic programming for min-max model predictive control of constrained uncertain systems , 2004, IEEE Transactions on Automatic Control.

[11]  Biao Huang,et al.  Robust digital model predictive control for linear uncertain systems with saturations , 2004, IEEE Trans. Autom. Control..

[12]  James B. Rawlings,et al.  A new robust model predictive control method. II: examples , 2004 .

[13]  M. Tippett,et al.  Using a general circulation model to forecast regional wheat yields in northeast Australia , 2004 .

[14]  J. Rawlings,et al.  A new robust model predictive control method I: theory and computation , 2004 .

[15]  Carlo Novara,et al.  Set membership prediction of nonlinear time series , 2005, IEEE Transactions on Automatic Control.

[16]  Peter Fogh Odgaard,et al.  Estimation of Moisture Content in Coal in Coal Mills , 2006 .

[17]  B. Mataji,et al.  Preventing performance drops of coal mills due to high moisture content , 2007, 2007 European Control Conference (ECC).

[18]  B. Mataji,et al.  Prediction of lacking control power in power plants using statistical methods , 2007, 2007 European Control Conference (ECC).

[19]  Peter Fogh Odgaard,et al.  Using reference trajectories to predicted uncertain systems: exemplified on a power plant , 2007 .

[20]  B. Mataji,et al.  Using reference trajectories for predicting uncertain systems exemplified for a power plant , 2007, 2007 American Control Conference.

[21]  B. Mataji,et al.  Using dynamical uncertainty models for estimation of uncertainty bounds on power plant performance prediction , 2007, 2007 European Control Conference (ECC).