It should be noted that MILP-based approach is becoming a
mainstream method for solving SCUC problems based on commercial
MILP solvers [8], [17]–[19]. However, the LR-based
approaches are still useful in many cases. This is because to
apply MILP-based methods efficiently, one of the most important
issues is to convert the problem objective and constraints
into a good linear formulation. Therefore, many ancillary variables
and constraints must be introduced to handle minimum
up/down time constraints, variable start-up costs, nonlinear fuel
costs, etc. In some cases, the computation efficiency would be
a serious issue. The approximation error is another important
issue seldom discussed. On the other hand, one can use a more
natural and concise formulation for the LR-based methods and
mainstream method for solving SCUC problems based on commercial
MILP solvers [8], [17]–[19]. However, the LR-based
approaches are still useful in many cases. This is because to
apply MILP-based methods efficiently, one of the most important
issues is to convert the problem objective and constraints
into a good linear formulation. Therefore, many ancillary variables
and constraints must be introduced to handle minimum
up/down time constraints, variable start-up costs, nonlinear fuel
costs, etc. In some cases, the computation efficiency would be
a serious issue. The approximation error is another important
issue seldom discussed. On the other hand, one can use a more
natural and concise formulation for the LR-based methods and