Automated control-schedule acquisition within an intelligent controller
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional control systems lack the flexibility and intelligence to adapt and produce desired operational behaviors specified after controller design and implementation, particularly in dynamic environments where traditional design techniques fail to accommodate changing conditions.
Innovation Solution
Intelligent controllers that learn and modify control schedules over time through schedule-creation and schedule-modification interfaces, using immediate-control inputs and sensor feedback to encode desired operational behaviors into control schedules, enabling adaptive control in dynamic environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional control design techniques are used, then the controller can ensure predetermined system behavior under normal conditions, but the controller lacks flexibility to adapt to changing conditions specified after design and implementation
Solution Approach 1:
The controller transitions from a static control model to a dynamic learning system that continuously adapts control schedules based on observed system behavior and user inputs. The control schedules are no longer fixed but evolve over time through machine learning processes, allowing the controller to dynamically adjust to changing conditions while maintaining manageable complexity through automated learning.
Solution Approach 2:
The controller performs self-learning and self-adjustment of control schedules without requiring manual reprogramming or complex external intervention. The machine learning algorithms enable the system to automatically observe, learn, and modify its own control behavior based on operational data and user feedback, reducing the need for complex external configuration while improving adaptability.
2Adaptability or versatility
If the controller continuously learns and modifies control schedules, then the flexibility and adaptability improve, but the computational resources and processing time increase
Solution Approach 1:
The controller performs preliminary learning during periods of low system activity or when predictive models can be updated without affecting real-time control performance. By separating the learning process from critical control paths and performing updates during idle periods or using historical data, the system reduces computational energy consumption during active control while maintaining adaptability improvements.
3Speed
If the controller uses frequent immediate-control inputs for learning, then the learning speed and adaptability improve, but the system responsiveness and control stability may deteriorate
Solution Approach 1:
The controller applies learning updates at partial frequency rather than continuously, selecting specific moments or conditions for learning interventions. By using partial action (intermittent learning updates rather than continuous) and excessive action (aggressive learning during initial phases followed by conservative refinement), the system achieves adequate learning speed while maintaining control stability through selective update timing and magnitude.
Data Source
AI summary
A method of using a server to update stored control schedules for environmental controllers includes communicating with an environmental controller that controls, during a monitoring period, an environmental system in an enclosure according to a stored control schedule; receiving a first immediate-control input provided through the environmental controller during the monitoring period; receiving a first control-schedule change provided through the environmental controller during the monitoring period; receiving a second immediate-control input provided through a user device during the monitoring period; receiving a second control-schedule change provided through the user device during the monitoring period; processing at least the first immediate-control input, the first control-schedule change, the second immediate-control input, and the second control-schedule change together to generate an updated control schedule; and causing the environmental controller to control the environmental system according to the updated control schedule.