Distributed Control Strategy Using Operator Interaction Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated control systems in industrial plants often require manual interventions due to unforeseen situations, leading to inefficiencies and safety concerns, as they cannot anticipate all operational scenarios during plant operation.
Innovation Solution
The implementation of two computer-implemented methods to amend and augment engineering tools for distributed control systems, using past interaction events and machine learning models to predict operator interactions, thereby reducing the need for manual interventions by automating routine tasks and adapting control strategies in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated control systems are implemented in industrial plants, then productivity and operational efficiency are improved, but the system cannot foresee all situations leading to frequent manual interventions by operators
Solution Approach 1:
The system implements feedback by capturing operator interaction events and using them to train machine learning models. The models predict future operator interactions and automatically execute predicted actions, creating a closed-loop system where past operator behavior informs future automated responses, continuously improving the system's adaptability to unforeseen situations
Solution Approach 2:
The system performs preliminary action by training machine learning models on historical operator interaction data before actual operational needs arise. The models are pre-trained to recognize patterns and predict operator actions, enabling the system to automatically respond to unforeseen situations before they require manual intervention
2Adaptability or versatility
If manual interventions are allowed to handle unforeseen situations, then system adaptability is maintained, but operator time is consumed and safety concerns arise
Solution Approach 1:
The system implements self-service by enabling automated control systems to handle their own unforeseen situations without requiring operator intervention. Machine learning models predict operator actions and automatically execute predicted interventions, allowing the system to serve itself in responding to异常情况, thereby freeing operator time and reducing safety concerns
3Reliability
If engineered control strategies are designed to cover all situations, then system reliability is improved, but device complexity and engineering effort increase significantly
Solution Approach 1:
The system applies dynamics by transitioning from static engineered control strategies to dynamic machine learning models that adapt to new situations. The models are trained on historical data and continuously improve their predictive capabilities, allowing the control system to evolve and handle unforeseen situations without requiring complex pre-programmed rules for every possible scenario
Data Source
AI summary
A method includes acquiring state variables that characterize an operational state of an industrial plant; acquiring interaction events of a plant operator interacting with the distributed control system via a human-machine interface; determining based on the interaction events, and with state variables as input data, whether one or more interaction events are indicative of the plant operator executing a task that is not sufficiently covered by engineering of the distributed control system. When this determination is positive, mapping the input data to an amendment and/or augmentation for the engineering tool that has generated the application code.