Building Energy Model Calibration via Sensitivity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional building energy model calibration methods are time-consuming, labor-intensive, and computationally expensive, often requiring manual adjustments and insufficient data, leading to discrepancies between projected and actual energy consumption.
Innovation Solution
A system and method that uses machine learning to generate sensitivity values for building energy model variables, creating a more computationally efficient second model by modifying these variables based on historical energy consumption data, thereby reducing the need for extensive simulation iterations and integrating separate datasets for enhanced calibration quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional manual calibration approaches are used to adjust model inputs, then calibration accuracy can be achieved, but the process becomes time-consuming and labor-intensive
Solution Approach 1:
The system performs self-calibration by automatically comparing simulated energy consumption with actual building energy data and adjusting model inputs without requiring manual analyst intervention. The calibration engine autonomously iterates through parameter adjustments to minimize discrepancies between modeled and measured data.
Solution Approach 2:
The patent replaces the manual mechanical adjustment process with an automated computational system using machine learning algorithms and optimization techniques to adjust model parameters, substituting human analysts with algorithmic processes that can rapidly evaluate multiple scenarios.
2Measurement precision
If hundreds or thousands of parametric simulations are run for calibration, then model accuracy improves, but computational resources and time requirements increase significantly
Solution Approach 1:
The system performs preliminary sensitivity analysis to identify which model inputs have the greatest impact on energy consumption outputs. By pre-screening parameters to determine their relative importance, the system can focus computational resources on adjusting only the most influential parameters rather than exhaustively searching all possible inputs.
Solution Approach 2:
The patent applies partial action by running simulations with a selective subset of parameters identified as most critical through sensitivity analysis, rather than performing exhaustive simulations on all model inputs. This approach achieves sufficient calibration accuracy with significantly reduced computational effort.
3Measurement precision
If comprehensive building data is collected for calibration, then model predictive accuracy improves, but data processing complexity and time requirements increase
Solution Approach 1:
The system extracts and isolates the most critical building parameters and operational data that have the greatest influence on energy consumption. By selectively extracting only the essential data elements needed for calibration rather than processing all available building information, the system reduces data processing complexity while maintaining model accuracy.
4Reliability
If manual trial-and-error adjustment of model inputs is performed, then calibration can be achieved, but the process becomes expensive and labor-intensive
Solution Approach 1:
The system implements automated feedback loops where the calibration engine continuously compares simulated energy consumption results with actual measured building data, uses the discrepancy information to guide parameter adjustments, and iterates until convergence is achieved. This feedback-driven approach replaces manual trial-and-error with systematic automated optimization.
Solution Approach 2:
The calibration process serves itself by automatically performing data validation, parameter adjustment, and accuracy verification without requiring manual analyst intervention at each step, thereby improving productivity while maintaining reliable calibration results through systematic automated procedures.
Data Source
AI summary
In some embodiments, a system for automated calibration of building energy models may be configured to obtain a first building energy model for a building, generate plurality of sensitivity values by determining, for each of the plurality of variables, a respective sensitivity value that indicates a sensitivity of an output of the first building energy model to variations in the respective variable. The system may be further configured to, based on the plurality of variables and the plurality of sensitivity values, generate a second building energy model for the building. The second building energy model may include the plurality of variables and may be more computationally efficient than the first building energy model. The system may be configured to tune the second building energy model by modifying one or more of the plurality of variables based on historical energy consumption data for the building.