Prioritized Economic Objective Functions for MPC Steady-State Targets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tuning Model Predictive Control (MPC) controllers in industrial processes is a time-consuming and challenging task due to the need for trial and error adjustments of cost factors to achieve optimal behavior, especially when dealing with multiple manipulated variables and changing process conditions.
Innovation Solution
A computer-implemented method and system that configures optimization preferences and priorities for key manipulated variables, translating them into prioritized economic objective functions to optimize industrial process behavior without requiring trial and error tuning, using a user interface to specify optimization directions and priorities for each variable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single economic objective function is used with multiple manipulated variables, then optimization of process behavior is achieved, but the tuning becomes time-consuming and sensitive to model changes
Solution Approach 1:
The patent segments the single economic objective function into multiple prioritized objective functions, each corresponding to a specific manipulated variable or process goal. This segmentation allows independent configuration of optimization preferences for each variable without requiring global trial-and-error tuning, directly reducing tuning time while maintaining optimization stability.
Solution Approach 2:
The patent introduces new parameters (optimization preferences and priorities) that change the structure of the objective function from a single weighted sum to a hierarchical multi-objective framework. This parameter change eliminates sensitivity to model variations by providing a more robust optimization structure that doesn't require frequent retuning.
2Manufacturing precision
If trial and error tuning is used to adjust cost factors, then preferred optimization behavior is achieved, but the process becomes challenging and time-consuming
Solution Approach 1:
The system performs self-service by automatically determining cost factors based on the configured optimization preferences and priorities, eliminating the need for manual trial-and-error tuning. The controller self-adjusts the optimization parameters based on the hierarchical objective functions, achieving precise optimization without requiring operator expertise in tuning.
Solution Approach 2:
The patent implements preliminary action by pre-configuring optimization preferences and priorities for each manipulated variable before operation. This preliminary configuration establishes the optimization framework in advance, eliminating the need for subsequent trial-and-error adjustments and making the system easy to operate.
3Reliability
If cost factors are adjusted to balance multiple manipulated variables, then optimal process behavior is achieved, but the tuning becomes sensitive to model and operating condition changes
Solution Approach 1:
The patent implements dynamics by making the objective function structure adaptive to different operating conditions through the hierarchical prioritization framework. The system can dynamically adjust which objectives are active and their priorities based on current process conditions, maintaining optimization stability while being adaptable to model changes without requiring retuning.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Computer system and methods for optimally controlling the behavior of an industrial process, in accordance with plant operating goals, without requiring a complicated trial and error process. The system and methods enable configuring optimization preference and optimization priority for key manipulated variables (MVs) of the industrial process. The system and methods translate the configured optimization preference and optimization priority for each key MV into prioritized economic objective functions. The system and methods calculate a set of normalized cost factors for use in a given prioritized economic functions based on a model gain matrix of manipulated variables and controlled variables of the industrial process. The system and methods automatically determine best achievable targets for the MVs by solving each prioritized economic objective functions in sequence of priority within the constraints of: (1) the determined CV best achievable steady-state targets, and (2) the determined MV best achievable steady-state targets from higher prioritized economic objective functions.