Plant Operating Condition Control With Virtual Change Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing plant control methods using prediction models are vulnerable to disturbance factors and prediction errors, leading to potential malfunctions when operation indices deviate from target values, especially in regions with significant changes near target values.
Innovation Solution
A plant operating condition determination device that employs a prediction model to calculate both scheduled and virtual change values of manipulation parameters, ensuring that the operation index meets the criterion by outputting a scheduled change value only if both values confirm adherence to the criterion, thereby reducing the risk of deviation due to disturbances or errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a prediction model is used to determine manipulation parameter changes, then control speed and responsiveness are improved, but reliability deteriorates due to vulnerability to disturbance factors and prediction errors
Solution Approach 1:
The patent applies preliminary action by calculating a virtual change value that is larger than the scheduled change value before actual control execution. This virtual change value is used to predict future operation indices in advance, allowing the system to anticipate potential deviations caused by disturbance factors or prediction errors. By performing this preliminary prediction with a larger change magnitude, the system ensures that the actual control action (scheduled change value) will not cause the operation index to deviate from the criterion even under uncertain conditions.
Solution Approach 2:
The patent implements preliminary anti-action by using the virtual change value calculation to counteract potential harmful effects before they occur. The virtual change value serves as a preventive measure that anticipates and compensates for the impact of disturbance factors and prediction errors. By ensuring that the operation index remains within the criterion even when the manipulation parameter changes by the larger virtual amount, the system pre-establishes a safety margin that prevents future deviations and potential malfunctions.
2Measurement precision
If the manipulation parameter is changed to reach the target operation index value, then control precision is improved, but reliability deteriorates due to dramatic changes in the operation index near the target value
Solution Approach 1:
The patent applies preliminary action by performing a preliminary prediction using the virtual change value (which is larger than the scheduled change value) to assess whether the operation index will remain within the criterion. This preliminary assessment is conducted before executing the actual control action, allowing the system to verify that even with a larger parameter change, the operation index stays acceptable. This pre-verification ensures that the subsequent smaller scheduled change value will not cause dramatic or unstable changes in the operation index near the target value.
3Measurement precision
If learning is accumulated in the actual plant under various operating conditions, then prediction accuracy is improved, but cost and time increase due to trial operations
Solution Approach 1:
The patent applies the copying principle by using a prediction model that replicates plant behavior without requiring physical trial operations. Instead of accumulating learning data through actual plant trials under various operating conditions, the system uses a prediction model (which can be trained offline or using simulated data) to copy and predict plant responses. This virtual copying allows the system to achieve prediction accuracy without incurring the time and cost penalties of extensive real-world trial operations.
Data Source
AI summary
It is judged whether a first predicted value of an operation index obtained by inputting a scheduled change value of a manipulation parameter of a plant meets an operation criterion, and whether a second predicted value of the operation index obtained by inputting a virtual change value with a greater change amount from a current value than the scheduled change value to a prediction model meets the operation criterion. If it is judged that the first predicted value and the second predicted value meet the operation criterion, the scheduled change value is output as a command value of the manipulation parameter.