Adaptive Model Predictive Control for Dead Time Dominant Loops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing process control systems face challenges in creating an adaptive model predictive controller (MPC) that can be updated online without disturbing the process, as traditional methods require process upsets and are computationally expensive, leading to model mismatch and suboptimal control, especially in dead time dominant processes.
Innovation Solution
A method using a model switching technique to periodically determine a parameterized process model online, generating an MPC control model and algorithm without artificial excitation, allowing for adaptive MPC control during normal operation, which is computationally efficient and can be implemented in distributed controllers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional model determination methods are used, then accurate process models can be obtained, but the process must be disturbed and significant computational resources are required
Solution Approach 1:
The system performs model determination in advance during commissioning or maintenance periods when process disturbance is acceptable, storing the model for later use during normal operation. This allows accurate models to be obtained without disturbing the process during critical production periods.
Solution Approach 2:
The invention creates a mathematical copy (model) of the process behavior that can be used for control calculations without physically disturbing the actual process. The model captures the essential dynamics and can be updated periodically without requiring continuous process upsets.
2Reliability
If complex MPC algorithms are implemented, then control performance improves, but computational cost and implementation complexity increase
Solution Approach 1:
The control algorithm is divided into segments: computationally intensive model determination is performed offline during commissioning, while the online control phase uses simplified calculations with pre-determined model parameters. This segmentation allows high performance without continuous heavy computational burden.
Solution Approach 2:
The system adapts its complexity based on operating conditions - using full MPC algorithms when computational resources are available and process conditions warrant it, while switching to simpler control strategies during periods of high computational load or when process dynamics are relatively stable.
3Measurement precision
If adaptive control is implemented, then control accuracy improves, but computational resources and processing time increase
Solution Approach 1:
The system performs adaptive model updates periodically rather than continuously, balancing the need for accurate adaptive control with computational resource constraints. Model determination is triggered by significant process changes or at scheduled intervals, reducing processing time while maintaining control accuracy.
Solution Approach 2:
The control system automatically determines when model updates are needed based on process behavior analysis, eliminating the need for continuous computational resources. The system self-manages the balance between adaptivity and computational load by monitoring process changes and triggering updates only when necessary.
Data Source
AI summary
A method of creating and using an adaptive DMC type or other MPC controller includes using a model switching technique to periodically determine a process model, such as a parameterized process model, for a process loop on-line during operation of the process. The method then uses the process model to generate an MPC control model and creates and downloads an MPC controller algorithm to an MPC controller based on the new control model while the MPC controller is operating on-line. This technique, which is generally applicable to single-loop MPC controllers and is particularly useful in MPC controllers with a control horizon of one or two, enables an MPC controller to be adapted during the normal operation of the process, so as to change the process model on which the MPC controller is based to thereby account for process changes. The adaptive MPC controller is not computationally expensive and can therefore be easily implemented within a distributed controller of a process control system, while providing the same or in some cases better control than a PID controller, especially in dead time dominant process loops, and in process loops that are subject to process model mismatch within the process time to steady state.