Content deleted Content added
JanHRichter (talk | contribs) m Capitalization |
JanHRichter (talk | contribs) Changed the way references are defined |
||
Line 1:
'''Control reconfiguration''' is an active approach in [[control theory]] to achieve [[Fault-Tolerant Control|fault-tolerant control]] for [[dynamic systems]] <ref>{{Harv|Blanke|Kinnaert|Lunze|Staroswiecki|2006}}</ref>. It is used when severe [[Fault (technology)|faults]], such as actuator or sensor outages, cause a break-up of the [[control loop]], which must be restructured to prevent [[failure]] at the system level. In addition to loop restructuring, the [[Controller (control theory)|controller]] parameters must be adjusted to accommodate changed plant dynamics. Control reconfiguration is a building block toward increasing the [[dependability]] of systems under [[feedback]] control <ref>{{Harv|Patton|1997}}</ref>.
== Reconfiguration problem ==
Line 52:
=== Fault hiding ===
[[Image:FaultHiding with Goals.png|frame|Fault hiding principle. A reconfiguration block is placed between faulty plant and nominal controller. The reconfuigured plant behaviour must match the nominal behaviour. Furthermore, the reconfiguration goals are pointed out.]]
This paradigm aims at keeping the nominal controller in the loop. To this end, a reconfiguration block is placed between the faulty plant and the nominal controller. Together with the faulty plant, it forms the reconfigured plant. The reconfiguration block has to fulfill the requirement that the behaviour of the reconfigured plant matches the behaviour of the nominal, that is fault-free plant <ref>{{Harv|Steffen|2005}}</ref>.
=== Linear model following ===
In linear model following, a formal feature of the nominal closed loop is attempted to be recovered. In the classical pseudo-inverse method, the closed loop system matrix <math>\bar{\mathbf{A}} = \mathbf{A}-\mathbf{B}\mathbf{K}</math> of a state-feedback control structure is used. The new controller <math>\mathbf{K}_f</math> is found to approximate <math>\bar{\mathbf{A}}</math> in the sense of an induced matrix norm <ref>{{Harv|Gao|Antsaklis|1991}} {{Harv|Staroswiecki|2005}}</ref>.
In perfect model following, a dynamic compensator is introduced to allow for the exact recovery of the complete loop behaviour under certain conditions.
Line 62:
=== Optimisation-based control schemes ===
Linear-quadratic regulator design (LQR), model predictive control (MPC) <ref>{{Harv|Looze|Weiss|Eterno|Barrett|1985}},{{Harv|Lunze|Rowe-Serrano|Steffen|2003}},{{Harv|Maciejowski|Jones|2003}}</ref>
=== Probabilistic approaches ===
<ref>{{Harv|Mahmoud|Zhang|Jiang|2003}}</ref>
=== Learning control ===
Learning automata, neural networks etc. <ref>{{Harv|Rauch|1994}}</ref>.
== Mathematical tools and frameworks ==
The methods by which reconfiguration is achieved differ considerably. The following list gives an overview of mathematical approaches that are commonly used <ref>{{Harv|Zhang|Jiang|2003}}</ref>.
* [[Adaptive control]] (AC)
Line 89:
Prior to control reconfiguration, it must be at least determined whether a fault has occurred ([[Fault Detection|fault detection]]) and if so, which components are affected ([[Fault Isolation|fault isolation]]). Preferably, a model of the faulty plant should be provided ([[Fault Identification|fault identification]]). These questions are addressed by [[fault diagnosis]] methods.
[[Fault Accommodation|Fault accommodation]] is another common approach to achieve [[Fault Tolerance|fault tolerance]]. In contrast to control reconfiguration, accommodation is limited to internal controller changes. The sets of signals manipulated and measured by the controller are fixed, which means that the loop cannot be restructured <ref>{{Harv|Blanke|Kinnaert|Lunze|Staroswiecki|2006}}</ref>.
== References ==
<references/>
== Further reading ==
| |||