Real-Time Automation Control With Profile Adaptation and Fallback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern, highly complex automation processes require real-time control capabilities, but existing numerical approximation methods are not real-time capable, as they cannot guarantee termination within a maximum time period, and artificial neural networks lack quality guarantees despite being real-time capable.
Innovation Solution
A method that uses a real-time-capable recognition method based on non-linear optimization to determine a change profile, which is then adapted using a numerical algorithm to ensure it meets boundary and secondary conditions, with a fallback profile used if the adapted profile does not satisfy these conditions, ensuring the automation process is controlled within a predetermined time frame.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If numerical approximation methods are used for optimization, then optimization quality can be guaranteed, but real-time capability is lost because termination within maximum time period cannot be guaranteed
Solution Approach 1:
The patent segments the control task into two distinct phases: an offline phase where numerical approximation methods are used to generate optimized control profiles without time constraints, and an online phase where pre-computed profiles are selected and applied in real-time. This segmentation allows quality-guaranteed optimization to be performed beforehand, while real-time operation uses pre-prepared solutions that guarantee termination within maximum time period.
Solution Approach 2:
The patent performs preliminary optimization actions offline to compute multiple candidate control profiles that satisfy boundary and secondary conditions. These profiles are stored and ready for immediate use during real-time operation. By performing the computationally intensive optimization work in advance, the system ensures that during real-time control, pre-computed profiles can be selected and applied without exceeding maximum time period constraints.
2Loss of time
If artificial neural networks are used for optimization, then real-time capability is achieved with deterministic runtime complexity, but quality guarantee is lost since no quality criteria are included
Solution Approach 1:
The patent introduces a hybrid approach where numerical approximation methods serve as an intermediary to bridge the gap between neural network speed and optimization quality. The system uses numerical methods offline to generate high-quality control profiles that satisfy all boundary and secondary conditions, then uses these pre-computed profiles during real-time operation. This intermediary preparation phase ensures both quality guarantees and real-time capability without requiring neural networks to provide quality criteria.
3Adaptability or versatility
If the number of process parameters is increased to represent equivalent solutions, then optimization flexibility is improved, but computational complexity increases making real-time control difficult
Solution Approach 1:
The patent handles multiple process parameters and equivalent solutions by segmenting the computational work: offline, numerical approximation methods explore the parameter space with full flexibility to find optimized profiles satisfying all conditions; online, the system selects from pre-computed profiles without re-running complex optimization. This segmentation allows high adaptability in profile generation while maintaining real-time performance during actual control operations.
Data Source
AI summary
A method for controlling an automation process in real time based on a change profile of at least one process variable, comprises determining a first change profile by a real-time-capable recognition method based on a non-linear optimization process taking into consideration at least one boundary condition of the process variable, determining a second change profile by a numerical algorithm based on the first change profile, including adapting a selected profile function to the first change profile by a numerical adaptation process and identifying the adapted profile function as a second change profile, checking whether the second change profile satisfies at least one secondary condition of the process variable, controlling the automation process based on the second change profile if the second change profile satisfies the secondary condition, and controlling the automation process based on a predetermined fallback profile if the second change profile does not satisfy the secondary condition.