Steady-state bumpless transfer under controller uncertainty using the state/output feedback topology
- University of Illinois, Urbana, IL (United States)
Linear quadratic (LQ) bumpless transfer design introduced recently by Turner and Walker gives a very convenient and straightforward computational procedure for the steady-state bumpless transfer operator synthesis. It is, however, found to be incapable of providing convergence of the output of the offline controller to that of the online controller in several industrial applications, producing bumps in the plant output in the wake of controller transfer. An examination of this phenomenon reveals that the applications in question are characterized by a significant mismatch, further referred to as controller uncertainty, between the dynamics of the implemented controllers and their models used in the transfer operator computation. To address this problem, while retaining the convenience of the Turner and Walker design, a novel state/output feedback bumpless transfer topology is introduced that employs the nominal state of the offline controller and, through the use of an additional controller/model mismatch compensator, also the offline controller output. A corresponding steady-state bumpless transfer design procedure along with the supporting theory is developed for a large class of systems. Due to these features, it is demonstrated to solve a long-standing problem of high-quality steady-state bumpless transfer from the industry standard low-order nonlinear multiloop PID-based controllers to the modern multiinput-multioutput (MIMO) robust controllers in the megawatt/throttle pressure control of a typical coal-fired boiler/turbine unit.
- OSTI ID:
- 20712271
- Journal Information:
- IEEE Transactions on Control Systems Technology, Vol. 14, Issue 1; ISSN 1063-6536
- Country of Publication:
- United States
- Language:
- English
Similar Records
Design of operator interfaces for ``bumpless`` transfers between operator behaviors
Experimental comparison of conventional and nonlinear model-based control of a mixing tank