Dynamic load curtailment system and method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing energy management systems for load curtailment often require extensive tuning and retuning due to seasonal changes, leading to either inadequate or excessive energy curtailment, resulting in occupant disruption and inefficiency.
Innovation Solution
A dynamic load curtailment algorithm that learns optimum energy consumption conditions for a building by analyzing historical data and using real-time monitoring to predict peak demand, strategically curtail loads within user-configured parameters, employing separate control schemes for HVAC and non-HVAC loads, and logging curtailment actions for transparency and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If schedule or threshold-based curtailment systems are used, then energy consumption is reduced, but extensive tuning and retuning are required to avoid occupant disruption
Solution Approach 1:
The system uses machine learning algorithms that automatically learn and adapt to building patterns and occupant behaviors over time, eliminating the need for manual tuning. The algorithm self-adjusts curtailment strategies based on historical data and real-time conditions, making the system self-sufficient in optimization.
Solution Approach 2:
The system continuously monitors energy consumption, occupancy patterns, and curtailment effects, using this feedback to dynamically adjust curtailment strategies. This closed-loop control ensures optimal performance without manual intervention while adapting to changing conditions.
2Productivity
If manual tuning is performed to optimize curtailment, then energy efficiency improves, but frequent retuning is needed due to seasonal changes
Solution Approach 1:
The system transitions from static, manually-tuned parameters to dynamic, automatically adapting curtailment strategies. The machine learning model continuously learns from seasonal patterns and environmental changes, automatically adjusting operations to maintain optimal efficiency throughout the year.
Solution Approach 2:
The algorithm automatically detects seasonal changes and adjusts curtailment parameters without human intervention, eliminating the need for seasonal retuning while maintaining peak efficiency.
3Loss of energy
If curtailment is increased to reduce energy consumption, then energy loss decreases, but occupant disruption increases
Solution Approach 1:
The system applies different curtailment strategies to different building zones and systems based on occupancy patterns and usage priorities. Critical areas maintain full operation while non-critical areas receive curtailment, optimizing energy reduction while protecting occupant comfort.
Solution Approach 2:
The system implements curtailment at the optimal level needed to reduce peak demand, avoiding both insufficient curtailment (which wastes energy) and excessive curtailment (which disrupts occupants). The machine learning algorithm precisely determines the right amount of curtailment for each condition.
Data Source
AI summary
A system and method are disclosed for dynamically learning the optimum energy consumption operating condition for a building and monitor/control energy consuming equipment to keep the peak demand interval at a minimum. The dynamic demand limiting algorithm utilized employs two separate control schemes, one for HVAC loads and one for non-HVAC loads. Separate operating parameters can be applied to the two types of loads and multiple non-HVAC (control zones) loads can be configured. The algorithm uses historical peak demand measurements in its real-time limiting strategy. The algorithm continuously attempts to reduce peak demand within the user configured parameters. When a new peak is inevitable, the algorithm strategically removes and/or introduces loads in a fashion that limits the new peak magnitude and places the operating conditions within the user configured parameters. In an embodiment, the algorithm that examines the previous seven days of metering information to identify a peak demand interval. The system then uses real-time load information to predict the demand peak of the upcoming interval, and strategically curtails assigned loads in order to limit the demand peak so as not to set a new peak.