Production Process Control Using Real-Time Simulator State Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advanced control strategies for complex production processes, such as pulp and paper production, are hindered by time-consuming initial state estimation, which results in outdated process control settings, leading to suboptimal operation.
Innovation Solution
Implementing an on-line simulator that runs in real-time parallel to the production process to provide immediate initial condition data for optimization, allowing for real-time set point generation and validation before application to controllers, thereby ensuring current and accurate control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If initial state estimation is performed using traditional methods, then control accuracy is improved, but time consumption increases significantly
Solution Approach 1:
The on-line simulator performs preliminary action by continuously calculating and maintaining up-to-date initial state estimates in real-time parallel with the production process. This preliminary computation ensures that when optimization begins, accurate initial states are already available, eliminating the need for time-consuming estimations at the start of each optimization cycle.
Solution Approach 2:
The invention creates a virtual copy of the production process through the on-line simulator. This digital twin continuously mirrors the actual process state, providing real-time initial condition data without requiring physical measurement or traditional estimation methods. The simulator copy enables immediate access to accurate process states.
2Measurement precision
If traditional control optimization is performed, then computational accuracy is improved, but the control settings become outdated by the time optimization completes
Solution Approach 1:
The on-line simulator maintains continuous useful action by running in real-time parallel with the production process, continuously updating initial state estimates as the process evolves. This continuous computation ensures that the optimizer always receives current process data, maintaining both accuracy and currentness of control settings throughout operation.
Solution Approach 2:
The system transitions from static, batch-based optimization to dynamic real-time optimization. The on-line simulator dynamically updates process states continuously, allowing the optimization to work with live data rather than outdated measurements, thus maintaining reliability and currentness of control recommendations.
3Productivity
If real-time control optimization is implemented, then operational efficiency is improved, but computational complexity increases
Solution Approach 1:
The control system is segmented into distinct functional modules: the on-line simulator module that handles real-time state estimation, and the optimization module that computes control strategies. This segmentation allows each module to specialize in its function, improving overall efficiency while managing complexity through modular architecture.
Data Source
Figure 1~3
AI summary
The present disclosure relates to a method of controlling a real production process, wherein the method comprises: a) receiving initial condition data from an on-line simulator system (7) simulating the real production process, and b) performing an optimisation based on the initial condition data and on an objective function to obtain set points for controlling the real production process.