Multiple-input/multiple-output control blocks with non-linear predictive capabilities

Abstract

A process controller for controlling a process having a set of process outputs affected by a set of process control input signals, the process controller includes a multiple input / multiple output controller using Process output to establish a set of process control input signals and a process model, the set of process control input signals is received to generate one or more predictive signals of process output, wherein the process model may be a nonlinear process model. The multiple-input / multiple-output control unit includes another process model for establishing a predictive vector for each process output, which can be a standard linear process model, and a correction unit that uses the predictive signals for one or more process outputs to modify the predicted vectors of one or more process outputs to compensate for the nonlinearity of the process.

Description

technical field [0001] The present invention relates generally to process control systems, and more particularly to the use of advanced control blocks, such as model predictive control blocks or optimizers, in process control systems having non-linear response characteristics. Background technique [0002] Process control systems, such as those used in chemical, petroleum, or other processes, are distributed or scalable, and typically include one or more process controllers that communicate via analog, digital, or combined analog / digital bus, communicatively coupled to each other, communicatively coupled to at least one host computer or operator workstation, and communicatively coupled to one or more field devices. For example, the field devices may be valves, valve positioners, switches, and transmitters (such as temperature, pressure, and flow rate sensors) that perform functions such as opening and closing valves and measuring process parameters in a process. Process con...

AI Technical Summary

Problems solved by technology

However, the recalculation of gains is complex, it involves calculation of derived values ​​from the process model which will lead to additional modeling errors, can be computationally expensive and time consuming to implement

Therefore, methods of performing these MPC adaptations inside the MPC logic are difficult and complex, and probably unreasonable or impractical in most process plant situations

Images