13.07.2015 Views

Assessment and Future Directions of Nonlinear Model Predictive ...

Assessment and Future Directions of Nonlinear Model Predictive ...

Assessment and Future Directions of Nonlinear Model Predictive ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

NLMPC: A Platform for Optimal Control 369Equations 2 <strong>and</strong> 3 define the process model in differential algebraic equation(DAE) form. x are the states, u are the manipulated inputs, v are the measureddisturbances or disturbance model variables, p are parameters, <strong>and</strong> y are theoutputs. In this problem statement x, u, v, <strong>and</strong>y are in absolute, not deviation,terms.Equations 4 <strong>and</strong> 5 are examples <strong>of</strong> specifications for the reference trajectoriesfor the desired closed loop behavior <strong>of</strong> the outputs. In this example, a first-orderresponse is requested with a time constant <strong>of</strong> τ c . SP h <strong>and</strong> SP l are the high<strong>and</strong> low setpoint (target) values for the outputs that defines the allowed settlingzone. P h <strong>and</strong> P l are 1-norm penalty variables on deviations <strong>of</strong> the output fromthe reference trajectory. S h <strong>and</strong> S l are the corresponding slack variables (i.e. notcosted in the objective function).Output feedback is incorporated in additive form in Eqn. 7. For scaling reasons,we sometimes incorporate output feedback in multiplicative form.Equations 8 <strong>and</strong> 9 impose absolute <strong>and</strong> incremental bounds on the manipulatedvariables. The subscripts k <strong>and</strong> k − 1 refer to adjacent zero-order holdvalues <strong>of</strong> the manipulated variables across the entire control horizon.Objective FunctionIn the objective function, the J 1 term is a weighted 1-norm <strong>of</strong> errors from thedesired closed loop output trajectory over the prediction horizon.n yn pJ 1 = 1 ∑ ∑(w hi P hi,k + w li P li,k ) (10)n pi=1 k=1Conceptually, this can be depicted as a conic section <strong>of</strong> unpenalized trajectorieswith linear penalties assigned to trajectories outside <strong>of</strong> the section. n p is thelength <strong>of</strong> the prediction horizon. n y is the number <strong>of</strong> outputs. The significance<strong>of</strong> the 1-norm is a distinct relaxation <strong>of</strong> s<strong>of</strong>t constraints from lowest ranked tohighest ranked. This one-at-a-time relaxation is more aligned with industrial expectationsas opposed to 2-norm behavior which spreads error across multipleoutputs when constraints become active [10].The J 2 term is an economic cost whose mean value over the prediction horizonis minimized to specify where within the allowed zone the system will settle. Auseful alternative explanation is that −J 2 is a net income (product value - cost<strong>of</strong> feed <strong>and</strong> control) to be maximized.n yn pJ 2 = 1 ∑ ∑( c yi y i,k +n pi=1 k=1∑n un pm=1 k=1n p∑∑n v∑c um u m,k + c vi v j,k ) (11)j=1 k=1n u is the number <strong>of</strong> manipulated inputs. n v is the number <strong>of</strong> measured disturbancevariables.The J 3 term is the cost <strong>of</strong> incremental moves <strong>of</strong> the manipulated variables.J 3 =∑n u∑n cm=1 l=1c ∆um |∆u m,l | (12)

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!