초록 ▼

A kiln thermal and combustion control. A predictive model is provided of the dynamics of selected aspects of the operation of the system for modeling the dynamics thereof. The model has at least two discrete models associated therewith that model at least two of the selected aspects, the at least tw

A kiln thermal and combustion control. A predictive model is provided of the dynamics of selected aspects of the operation of the system for modeling the dynamics thereof. The model has at least two discrete models associated therewith that model at least two of the selected aspects, the at least two discrete models having different dynamic responses. An optimizer receives desired values for the selected aspects of the operation of the system modeled by the model and optimizes the inputs to the model to minimize error between the predicted and desired values. A control input device then applies the optimized input values to the system after optimization thereof.

대표청구항 ▼

We claim: 1. A controller for controlling a multi-variable input plant having a plurality of manipulable variables (MVs) as inputs, and operable to provide a plurality of measurable outputs and at least one unmeasurable output, which unmeasurable output can not be measured in substantially real tim

We claim: 1. A controller for controlling a multi-variable input plant having a plurality of manipulable variables (MVs) as inputs, and operable to provide a plurality of measurable outputs and at least one unmeasurable output, which unmeasurable output can not be measured in substantially real time and requires periodic external analysis for the determination of periodic values of the at least one unmeasurable output, comprising: a plant predictive model that provides a model of the dynamics of selected aspects of the operation of the plant for modeling the dynamics thereof and providing at least one predicted output for at least a select one of the measurable outputs; an external predictive model, wherein, responsive to at least one input that is not an MV, the external predictive model is operable to: predict intermediate values of the at least one unmeasurable output in substantially real-time for use between use of the periodic values of the at least one unmeasurable output, and a desired unmeasurable output value for that at least one unmeasurable output; and predict the dynamics of a select one of the measurable outputs as a function of the at least one unmeasurable output and the desired value for the at least one unmeasurable output to predict the dynamics as a desired value of the select one of the measurable outputs required to achieve the desired unmeasurable output value; an optimizer for receiving desired values for the selected aspects of the operation of the plant modeled by said predictive model and said predicted outputs from said predictive model in addition to the desired value of the select one of the measurable outputs generated by said external predictive model and optimizing the inputs to the predictive model to minimize error between the predicted and desired values; and a control input device for applying the optimized input values to the plant after optimization thereof. 2. The controller of claim 1, wherein said optimizer is further operable to receive constraints on said input values such that the optimization operation changes the inputs within said constraints when minimizing the error between the predicted and desired values. 3. The controller of claim 1, wherein the plant is subject to at least one event that alters the operation of the plant and wherein said predictive model will predict the behavior of the selected aspects modeled thereby in response to the at least one event, and said optimizer will change the inputs to again minimize the error between the predicted and desired values. 4. The controller of claim 1, where the at least one input of the external predictive model is a feed composition change. 5. The controller of claim 1, where the at least one input of the external predictive model is a coating drop. 6. The controller of claim 1, wherein said external predictive model comprises a linear model. 7. The controller of claim 1, wherein the unmeasurable output comprises a product that is fabricated by the system. 8. The controller of claim 1, wherein said optimizer is not operable to utilize said external predictive model during the operation thereof for the purpose of prediction therewith. 9. A method for controlling a multi-variable input system having a plurality of manipulable variables (MVs) as inputs, and operable to provide a plurality of measurable outputs and at least one unmeasurable output, which unmeasurable output can not be measured in substantially real time and requires periodic external analysis for the determination of periodic values of the at least one unmeasurable output, comprising the steps of: providing a system predictive model that provides a model of the dynamics of selected aspects of the operation of the system for modeling the dynamics thereof and providing at least one predicted output for at least a select one of the measurable outputs; providing an external predictive model, operable to perform: responsive to at least one input that is not an MV, predicting intermediate values of the at least one unmeasurable output in substantially real-time for use between use of the periodic values of the at least one unmeasurable output, and a desired unmeasurable output value for the at least one unmeasurable output; and predicting the dynamics of a select one of the measurable outputs as a function of the at least one unmeasurable output and the desired value for the at least one unmeasurable output required to achieve the desired unmeasurable output value; receiving in an optimizer desired values for the selected aspects of the operation of the system modeled by the predictive model and the predicted outputs from the predictive model in addition to the desired value of the select one of the measurable outputs generated by the external predictive model and optimizing the inputs to the predictive model to minimize error between the predicted and desired values; and applying the optimized input values to the system after optimization thereof. 10. The method of claim 9, wherein said optimizing comprises: receiving constraints on the input values; and changing the inputs within the constraints when minimizing the error between the predicted and desired values. 11. The method of claim 9, wherein the system is subject to at least one event that alters the operation of the system, and wherein the system predictive model predicts the behavior of the selected aspects modeled thereby in response to the at least one event, and said optimizing changes the inputs to again minimize the error between the predicted and desired values. 12. The method of claim 9, wherein the external predictive model comprises a linear model. 13. The method of claim 9, wherein the unmeasurable output comprises a product that is fabricated by the system. 14. The method of claim 9, wherein said optimizing is not operable to utilize the external predictive model during the operation thereof for the purpose of prediction therewith. 15. The controller of claim 9, where the at least one input of the external predictive model is a feed composition change. 16. The controller of claim 9, where the at least one input of the external predictive model is a coating drop.