Rules-Based Chemical Dosing Control for Stable Papermaking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pulp, paper, and board manufacturing processes face challenges in accurately and adaptively controlling chemical dosing due to unpredictable process changes and the difficulty in obtaining real-time measurements, leading to quality disturbances and excessive chemical use.
Innovation Solution
Implementing a rules engine with a communication interface and sensors to adjust chemical dosing based on a set of rules, allowing for simulation and adaptation of dosing strategies using historical data to optimize chemical use and improve process control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If adaptive control of chemical dosing is implemented, then chemical use is optimized and performance is improved, but device complexity increases due to the need for rules engine and multiple measurements
Solution Approach 1:
The control system is segmented into modular rule sets that can be independently developed, tested, and activated. Each rule set addresses specific process conditions or chemical dosing scenarios, allowing the complex adaptive control function to be built from manageable, interchangeable components through the rules engine.
Solution Approach 2:
Multiple rule sets are prepared and validated in advance for different process scenarios. The rules engine selects and activates the appropriate pre-prepared rule set based on current process conditions, eliminating the need for real-time complex decision-making and reducing operational complexity while maintaining adaptive performance.
2Measurement precision
If online measurements of desired properties are obtained, then dosing control accuracy is improved, but measurement cost and system complexity increase
Solution Approach 1:
The system uses intermediary measurements of process parameters (such as flow rates, temperatures, or readily measurable chemical properties) that correlate with the desired end-product properties. These intermediary measurements serve as proxies that are easier and cheaper to obtain while still providing sufficient information for accurate dosing control through the rules engine.
Solution Approach 2:
Instead of directly measuring difficult-to-obtain product properties, the system uses available process measurements as copies or proxies that represent the state of the process. The rules engine processes these copied measurements to infer the required dosing adjustments, avoiding the need for complex direct measurement systems.
3Manufacturing precision
If multiple concurrent changes are taken into account in dosing control, then dosing accuracy is improved, but control difficulty increases due to opposing effects
Solution Approach 1:
The rules engine dynamically selects and switches between different rule sets based on current process conditions. Each rule set is optimized for specific scenarios and contains pre-determined relationships between process parameters and dosing rates, allowing the system to adapt to multiple concurrent changes without requiring complex real-time calculations or resolving opposing effects manually.
Solution Approach 2:
The system changes operational parameters by switching between different pre-configured rule sets, each representing a different set of parameter relationships. This allows the control system to handle multiple concurrent changes and opposing effects by transitioning to a rule set that is already optimized for the current combination of conditions, rather than attempting to calculate all interactions simultaneously.
Data Source
AI summary
A method, apparatus, and computer program control a process in making pulp, paper, or board automatically controlling by a rules engine a dosing of a chemical agent in the process into a fibrous suspension. The rules engine receives a plurality of measurements relating to the process and maintains a plurality of rules sets, including an active rules set. Each rules set defines how the dosing depends on the measurements. The rules engine controls the dosing of the chemical agent based on the measurements and the active rules set, logs at least some of the measurements and actualised dosing of the chemical agent, and receives at least one rules set candidate. The dosing of the chemical agent is simulated with the received at least one rules set candidate based on the earlier measurements; and simulation results are output for optimising performance of the chemical agent in the process.