Process Controller Data Conversion for Versatile Model-Based Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Trained models used in process control lack versatility and practicality as they are not capable of performing effective control when conditions change, being specific to the learning conditions and target set values.
Innovation Solution
A controller that acquires process data for an actual controlled object, converts it using a trained model, and calculates a manipulated variable, allowing the model to approximate process data to a target set value, even when the controlled object or target set value changes, by using parameters that specify the relation between manipulated variables and process data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a trained model is used for process control under specific conditions, then control precision is improved, but adaptability to different controlled objects and target set values deteriorates
Solution Approach 1:
The patent transforms the trained model into a versatile control system by changing parameters through data conversion. The conversion unit converts process data based on target set values and controlled object parameters, allowing the same trained model to adapt to different controlled objects and target values without retraining, thus resolving the contradiction between control precision and adaptability
Solution Approach 2:
The patent makes the trained model universal by introducing a conversion unit that adjusts process data according to different controlled objects and target set values. This allows a single trained model to perform multiple control tasks across different conditions, achieving both high control precision and broad adaptability
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A controller (1) includes: an acquisition unit (31) configured to acquire process data (X) for an actually used controlled object (9); and a calculation unit (22) configured to, based on at least one of a target set value (N) and parameters (P) for the actually used controlled object (9), convert the process data (X) acquired by the acquisition unit (31), and calculate a manipulated variable (MV) for the actually used controlled object (9) by use of the converted process data (S) and a trained model (2). Upon receiving input of process data for a specific controlled object (9a), the trained model (2) outputs a manipulated variable for approximating process data for the specific controlled object (9a) to a specific target set value (Na). The parameters (P) include parameters for specifying the relation between manipulated variables for the actually used controlled object (9) and process data obtained by use of the manipulated variables.