| 摘要 |
An optimizing device is provided with a target value calculating portion 105 that includes a variable constraint setting portion, a slope constraint setting portion, and a solution calculating portion. The variable constraint setting portion sets constraints for the manipulated variables and control variables. The slope constraint setting portion sets constraints for the slopes for the control variables that include integrating systems. The constraint settings are changed in accordance with the possibility of control variables that include integrating systems going outside of upper and lower limits. The various types of constraints, models for control, and applied optimization evaluating functions that are set/calculated are passed to the solution calculating portion, and optimal target values for minimizing the evaluating function wherein all of the variables are within the constraints are calculated. |
| 主权项 |
1. An optimizing device comprising:
a data collecting portion that collects data on a control object including a manipulated variable used in controlling the control object and a control variable for outputting the control object; a model storing portion that stores a mathematical model of the control object; a future predicting portion that predicts a value, after a prescribed time, for at least a control variable that does not include an output of an integrating element, of optimization object variables that are optimization objects from among the data collected by the data collecting portion; an integrating system slope predicting portion that predicts a magnitude of change per unit time, after a prescribed time, for those control variables that include an output of an integrating element; a variable constraint setting portion that sets a constraint for a value of a variable that is an optimization object, after a prescribed time; a slope constraint setting portion that sets a constraint on the magnitude of change after a prescribed time for a control variable that includes an output of an integrating element; and a solution calculating portion that calculates an optimal solution for the variables that are subject to optimization, satisfies the constraints set by the variable constraint setting portion and the slope constraint setting portion, and optimizes an optimization evaluating function that is applied, wherein: the slope constraint setting portion changes a slope constraint depending on a collected value or a predicted value for a control variable for which a slope constraint is set, to set, when a possibility of the control variable going outside of the constraints that are set by the variable constraint setting portion is low, to a slope constraint that is more relaxed than a slope constraint when the possibility is high. |
