Deep Learning Control for Environmental Systems Cost Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for controlling environmental conditions in spaces like data centers do not consider changes that result in higher reductions in cost functions while maintaining critical control errors within acceptable thresholds, leading to suboptimal energy and resource consumption.
Innovation Solution
A deep learning model is trained using historic data to predict optimal changes in environmental maintenance module setpoints that minimize cost functions while keeping critical control errors within acceptable ranges, utilizing a graphical user interface for visualization and user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current control techniques are used to minimize critical control error, then control precision is improved, but cost function reduction is limited
Solution Approach 1:
The patent changes the control approach by introducing a deep learning model that predicts optimal setpoint adjustments based on historical data and environmental factors. The system transforms discrete setpoint adjustments into continuous optimization by learning from patterns in energy consumption and control errors, enabling parameter changes that reduce overall energy cost while maintaining acceptable control precision.
Solution Approach 2:
The system performs preliminary analysis by training a deep learning model on historical control data before making setpoint adjustments. The model pre-learns the relationships between environmental factors, setpoint changes, and energy consumption patterns, allowing it to predict optimal adjustments in advance while considering future energy costs rather than just immediate control precision.
2Loss of energy
If setpoint adjustments are made to reduce cost function, then energy consumption is reduced, but critical control error increases
Solution Approach 1:
The patent implements feedback by continuously monitoring both energy consumption and critical control errors, feeding this data into the deep learning model. The model learns from historical feedback patterns to predict setpoint adjustments that balance energy reduction with maintaining control errors within acceptable thresholds, creating a closed-loop optimization system.
Solution Approach 2:
The system applies partial action by making conservative setpoint adjustments that achieve moderate energy savings while staying within acceptable control error thresholds. Rather than maximizing energy reduction at any cost, the model learns to apply just enough adjustment to achieve meaningful energy savings without compromising system reliability beyond acceptable limits.
3Loss of energy
If deep learning model predicts optimal setpoint changes, then cost function reduction is improved, but system complexity increases
Solution Approach 1:
The patent introduces a deep learning model as an intermediary between historical control data and setpoint adjustment decisions. This intermediary processes complex patterns in energy consumption and control errors, translating historical data into optimized setpoint recommendations without requiring direct complex interactions between multiple control components.
Solution Approach 2:
The system uses copying by training the deep learning model on historical control data and setpoint patterns. The model creates a virtual copy of past system behavior and relationships, allowing it to predict optimal future adjustments based on learned patterns rather than requiring complex real-time calculations or additional physical sensors.
Data Source
AI summary
Embodiments provide a machine learning model to identify changes in setpoints of one or more environmental control modules that, while having a high critical error, provide greater reductions in a cost function associated with the one or more environmental modules provided in an environmentally controlled space. The high critical error associated with the identified changes may still be within an acceptable threshold range associated with the environmentally controlled space. Thus, contrary to rule-based methods, the artificial intelligence (AI) based model described herein may recommended optimal changes to the system that yield to greater savings in the cost function and may not focus on minimizing the critical control error. Rather, the AI-based technique may simply aim at keeping the critical control error within an acceptable threshold range.