Simulation-Guided Turbine Control for Safe State Exploration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for controlling technical systems, such as gas turbines and wind turbines, face challenges in exploring new system states without disrupting operations or causing malfunctions, particularly when generating training data for machine-learning-based control methods.
Innovation Solution
A method that uses a predefined control method to determine initial actions and an exploration rule to predict subsequent states via a simulation model, ensuring that new states meet predetermined threshold criteria to avoid operational disruptions, allowing for the generation of new training data while maintaining system stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the technical system is operated in as yet unknown states to generate new training data, then the data set for machine learning is expanded, but the operation of the technical system may be disrupted or severely deteriorated
Solution Approach 1:
The simulation model predicts subsequent states before the technical system actually executes actions, allowing exploration of new states to be planned and validated in advance. This preliminary prediction step ensures that only safe exploratory actions are executed on the real system, resolving the contradiction between generating new training data and maintaining system reliability
Solution Approach 2:
A simulation model is introduced as an intermediary between the control method and the technical system. This intermediary predicts the outcomes of exploratory actions, allowing the system to evaluate potential new states without actually executing them on the real technical system, thus expanding training data while preserving system operation reliability
2Adaptability or versatility
If exploration rules define second actions to approach new states, then the state space is explored, but system stability may be compromised
Solution Approach 1:
The simulation model provides feedback on the predicted outcomes of exploratory actions by comparing them against threshold criteria for system stability. This feedback mechanism allows the exploration rule to adjust second actions dynamically, ensuring that state space exploration proceeds only when system stability requirements are satisfied
Solution Approach 2:
The control approach dynamically adjusts between executing first actions from the predefined control method and second actions from the exploration rule, based on whether predicted subsequent states meet stability criteria. This dynamic adjustment enables flexible state space exploration while maintaining system stability
Data Source
AI summary
The invention concerns a method for the computerized control and/or regulation of a technical system (T). Within the context of the method according to the invention, there is implemented in a preset regulating process (CO1, CO2) an exploration rule (EP) by means of which new, as yet unknown, states (x) of the technical system (T) are started, a simulation model (SM) of the technical system (T) checking whether the actions (a2) of the exploration rule (EP) lead to sequential states (x') lying within predetermined thresholds. Only in that case is the corresponding action (a2) performed according to the exploration rule (EP) on the technical system. The method according to the invention enables new states to be explored within the framework of the operation of a technical system, it being ensured through checking of appropriate thresholds that the exploration is carried out imperceptibly and does not lead to incorrect operation of the technical system. The method according to the invention is suitable in particular for controlling or regulating gas turbines or wind turbines.