Batch Process Optimization Using Steady-State Data Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional real-time optimization techniques are challenging to implement in batch processes due to transient behavior, which is not effectively addressed by existing methods.
Innovation Solution
A method and system for detecting events in batch processes, determining whether data satisfies steady-state criteria, generating optimization targets, and applying control signals using processors to optimize subsequent cycles, specifically considering drum change and online spalling transition events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional real-time optimization techniques are applied to batch processes, then optimization can be performed using periodic sampling methods, but the transient behavior during cycle transitions causes inaccurate steady-state determination and suboptimal control signals
Solution Approach 1:
The patent segments the batch cycle into distinct portions: a transient portion (from cycle start to steady-state achievement) and a steady-state portion (from steady-state achievement to cycle end). This segmentation allows the system to selectively use data only from the steady-state portion for optimization calculations, eliminating the corrupting influence of transient behavior on steady-state determination accuracy.
Solution Approach 2:
The system performs preliminary detection of steady-state achievement within each batch cycle before proceeding with optimization calculations. By identifying when steady-state conditions are met and when they cease, the system prepares the appropriate data subsets in advance, ensuring that only valid steady-state data is used for generating control signals.
2Quantity of substance
If data from the entire batch cycle including transient periods is used for optimization, then more data points are available for analysis, but the transient behavior contaminates the steady-state data and reduces optimization accuracy
Solution Approach 1:
The patent extracts and isolates the steady-state data portion from the complete batch cycle data. By separating the steady-state portion from the transient portion and using only the extracted steady-state data for optimization calculations, the system maintains high data quality and accuracy while performing optimization.
3Productivity
If real-time optimization is implemented in batch processes with frequent cycle transitions, then continuous improvement can be achieved, but the frequent transient behavior increases the difficulty of implementing conventional periodic optimization methods
Solution Approach 1:
The patent implements a dynamic optimization approach that adapts to the varying conditions of batch processes. Instead of using fixed periodic sampling intervals, the system dynamically identifies steady-state periods within each cycle and adjusts the data selection and optimization timing accordingly. This dynamic adaptation enables effective optimization even with frequent cycle transitions.
Solution Approach 2:
The system continuously monitors process variables throughout the batch cycle to detect when steady-state conditions are achieved and when they cease. This feedback mechanism allows the system to adaptively select appropriate data portions for optimization calculations, enabling continuous improvement across multiple batches while accounting for transient behavior.
Data Source
AI summary
Systems and methods for optimizing a system operating in a first mode of operation include detecting an event associated with a cycle in a first mode of operation. When the cycle ends (i.e., when another event is detected), steady state criteria is applied to determine whether data associated with the cycle satisfies the steady state criteria. Optimization targets can be calculated based on the data associated with the cycle. Control signals including the optimization targets can be applied to a next cycle operating in the first mode of operation to improve process performance.