Dynamic Load Prediction Model Retraining for Building Energy Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building energy systems face challenges in managing electricity consumption due to dynamic pricing schemes, as changes in occupancy, construction, or equipment usage can make it difficult to appropriately manage electricity consumption from the grid.
Innovation Solution
A building energy system that includes an energy storage system (ESS) and a processing circuit to collect data, retrain load prediction models based on current conditions, and optimize energy storage and discharge schedules to minimize electricity costs, using models like generalized additive models or recurrent neural networks to predict energy demand and adjust charging/discharging accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a building system uses a static load prediction model, then the model structure is simple and easy to implement, but the prediction accuracy deteriorates when occupancy, construction, or equipment usage changes
Solution Approach 1:
The patent implements dynamic model retraining by periodically updating the load prediction model with new building data. The system determines whether to retrain the model based on changes in occupancy patterns, weather conditions, or load deviations, transforming a static model into a dynamic one that adapts to changing building conditions while maintaining manageable complexity through selective retraining triggers.
2Measurement precision
If the system continuously retrain the load prediction model with all available data, then the prediction accuracy improves, but the computational time and processing resources increase
Solution Approach 1:
The patent applies partial retraining by selecting specific portions of building data relevant to recent changes rather than using all historical data. The system identifies and uses only the necessary data subset for retraining, reducing computational overhead while maintaining prediction accuracy by focusing on the most relevant information for current building conditions.
Solution Approach 2:
The system performs preliminary assessment to determine whether retraining is necessary by monitoring changes in occupancy patterns, weather conditions, and load deviations. This preliminary check prevents unnecessary retraining operations, saving computational time and resources while ensuring retraining occurs only when actually beneficial for maintaining prediction accuracy.
3Measurement precision
If the system uses detailed real-time building data for energy management, then the energy optimization accuracy improves, but the data collection and processing complexity increases
Solution Approach 1:
The patent integrates multiple data collection functions into a unified building management system that simultaneously gathers occupancy data, weather information, energy consumption data, and equipment status. This multi-functional approach consolidates what would otherwise be separate complex systems, reducing overall complexity while maintaining comprehensive data collection for accurate energy optimization.
Data Source
AI summary
A building energy system includes an energy storage system (ESS) configured to store energy received from an energy source and provide the stored energy to one or more pieces of building equipment. The system includes a local building system configured to collect building data and communicate the building data to a cloud platform and the cloud platform configured to receive the building data from the local building system via the network, determine whether to retrain a trained load prediction model based on at least some of the building data, retrain the trained load prediction model based on at least some of the building data in response to a determination to retrain the trained load prediction model, determine a load prediction for the building based on the retrained load prediction model, and cause the local building system to operate.