HVAC Predictive Control Using Simulated Data for New Buildings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building management systems often fail to provide user comfort and optimal energy usage in new buildings, as they lack historical data to generate effective control settings for energy management.
Innovation Solution
A building energy system that uses a predictive model based on a building model and historical weather data to optimize equipment settings, allowing for the simulation of energy usage data to pre-train the model, enabling immediate energy savings upon deployment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a building management system is deployed in a new building without historical data, then the system can be immediately operational, but the system fails to provide user comfort and optimal energy usage
Solution Approach 1:
The system performs preliminary actions by generating synthetic historical operation data through simulation before actual deployment. The simulation engine creates pseudo-historical data that reflects typical building operation patterns, allowing the predictive model to be pre-trained and ready for immediate effective operation without waiting for actual historical data accumulation
Solution Approach 2:
The system creates a synthetic copy of historical operation data through simulation. Instead of requiring actual historical data from the specific building, the system generates synthetic data that copies the essential patterns and characteristics of typical building operation, enabling the predictive model to learn from this synthetic replica
2Reliability
If the building management system waits for historical data to accumulate, then it can provide optimal energy management, but the system cannot operate effectively in the interim period
Solution Approach 1:
The system performs the data accumulation function in advance through simulation. Rather than passively waiting for historical data to accumulate from actual building operation, the simulation engine proactively generates synthetic historical data, eliminating the time delay associated with data accumulation
Solution Approach 2:
The system skips the time-consuming data accumulation phase by directly generating synthetic historical data through simulation. This rushes through the normally sequential process of data collection by creating equivalent information instantaneously, allowing immediate model training and deployment
3Productivity
If the building management system uses generic control settings, then it can operate immediately, but it fails to achieve optimal energy usage for the specific building
Solution Approach 1:
The system creates a synthetic copy of building-specific operation patterns through simulation. The simulation engine generates pseudo-historical data that captures the unique characteristics of the specific building, allowing the predictive model to learn building-specific patterns without requiring actual historical data or complex manual configuration
Solution Approach 2:
The system performs self-service by automatically generating synthetic historical data and training the predictive model without requiring external data sources or manual configuration. The simulation engine self-generates the necessary training data, and the system automatically adapts to building-specific patterns, eliminating the need for complex setup procedures
Data Source
AI summary
A system for controlling heating, ventilation, or air conditioning (HVAC) equipment of a building includes one or more processing circuits configured to generate simulated building data using a simulation model of the building, pre-train a reinforcement learning (RL) model using the simulated building data, operate the HVAC equipment of the building using the RL model, and retrain the RL model using actual building data generated responsive to operating the HVAC equipment using the RL model.