Scalable Building Asset Models for Variable-Speed HVAC Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting the energy consumption and performance of building assets, such as HVAC equipment, are inaccurate due to the lack of historical data and the inefficiency of scaling asset model coefficients, particularly when transitioning from constant speed to variable speed systems.
Innovation Solution
A system that includes scalable building asset models, which use a planning system to predict energy consumption and performance by scaling resource production and consumption data sets based on new design parameters, generating predictive models to control equipment operations, and adjusting coefficient of performance (COP) data to achieve accurate predictions without requiring extensive historical data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If asset model coefficients are scaled to predict energy consumption for new equipment, then predictions can be made without historical data, but the predictions become inaccurate particularly for variable speed systems
Solution Approach 1:
The patent transforms the approach by changing from scaling model coefficients to scaling performance data points. Instead of modifying the mathematical coefficients of the asset model, the system scales the actual performance data (energy consumption, resource production) to match the new equipment's design capacity. This parameter change resolves the contradiction by providing accurate predictions for variable speed systems without increasing model complexity.
Solution Approach 2:
The patent creates a scaled copy of the performance data from a reference equipment rather than directly scaling model coefficients. By copying and scaling actual performance data points (energy consumption at various operating points) to match the design capacity of new equipment, the system achieves accurate predictions without requiring historical data from the new equipment itself.
2Ease of manufacture
If base asset models are used for new equipment, then implementation is quick and easy, but the models do not accurately represent new or upgraded equipment performance
Solution Approach 1:
The patent changes the scaling approach from coefficient-based to data-point-based. By scaling performance data points (energy consumption, resource production) rather than model coefficients, the system maintains ease of implementation while significantly improving prediction precision for new and upgraded equipment including variable speed systems.
Solution Approach 2:
The patent introduces dynamic scaling factors based on the ratio between new equipment design capacity and reference equipment design capacity. This dynamic approach allows the model to adapt to different equipment sizes and types while maintaining ease of implementation. The scaling factors are calculated automatically based on equipment specifications, making the process both easy and accurate.
Data Source
AI summary
A system includes processors configured to perform operations including obtaining a base design resource production of a building asset and a base resource production data set comprising a base resource production of the building asset at a plurality of operating points, calculating a scaled resource production data set comprising a scaled resource production of the building asset at the plurality of operating points by scaling the base resource production data set based on a new design resource production of the building asset relative to the base design resource production of the building asset, generating a resource consumption data set comprising a resource consumption of the building asset at the plurality of operating points based on the scaled resource production data set, and initiating an automated action based on the scaled resource production data set and the resource consumption data set.