Adaptive Energy Use Model Parameter Order Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining an appropriate parameter order for building energy use models is challenging due to the difficulty in accurately determining break-even temperatures in extreme climates, which affects the accuracy of energy consumption predictions.
Innovation Solution
A method and system that calculate regression statistics under null and alternative hypotheses to determine the appropriate parameter order for energy use models by comparing test statistics to a threshold, using weather-related predictor variables like cooling and heating degree days, and updating the models accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a simple energy use model with fewer parameters is used, then the model complexity is reduced and ease of operation is improved, but the measurement precision and accuracy of energy consumption predictions deteriorate
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the number of parameters in the energy use model based on climate conditions. The system transitions between a two-parameter model (for extreme climates) and a three-parameter model (for moderate climates) by changing the balance point parameter from a fixed value to a variable that can be determined through hypothesis testing. This allows the model to adapt its complexity to match the actual climate conditions, improving measurement precision when needed while maintaining ease of operation under extreme conditions.
2Measurement precision
If a complex energy use model with more parameters is used, then the measurement precision and accuracy of energy consumption predictions is improved, but the device complexity and difficulty of determining appropriate parameter order increases
Solution Approach 1:
The patent implements dynamics by making the model structure adaptive rather than static. The system dynamically selects between different model configurations (two-parameter vs. three-parameter) based on real-time hypothesis testing of climate conditions. The balance point parameter transitions from a fixed value to a variable that is determined through statistical testing, allowing the model to automatically adjust its complexity to match the actual climate conditions without requiring manual intervention or complex determination processes.
Solution Approach 2:
The patent replaces the mechanical/manual process of determining appropriate model parameters with a statistical hypothesis testing mechanism. Instead of relying on expert judgment or complex engineering analysis to determine the balance point temperature, the system uses automated F-tests and regression analysis to objectively determine whether a two-parameter or three-parameter model is appropriate. This substitution of statistical mechanics for manual mechanical determination reduces device complexity while improving measurement precision.
3Ease of manufacture
If the balance point parameter is fixed rather than variable, then the device complexity is reduced and ease of manufacture is improved, but the adaptability to different climate conditions and measurement precision deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-establishing the hypothesis testing framework and statistical criteria that will automatically determine the appropriate model configuration. The system prepares the analytical structure in advance with predefined F-tests and significance levels, so that when actual climate data becomes available, the balance point parameter can be quickly and easily determined without complex manufacturing or configuration processes. This preliminary setup maintains ease of manufacture while enabling adaptability to different climate conditions.
Data Source
AI summary
A building analysis system includes a communications interface that receives energy consumption data for a building site including energy-consuming building equipment. A processing circuit of the building analysis system calculates first and second regression statistics indicating a fit of an energy use model to the energy consumption data under a null hypothesis that the energy use model has a first parameter order and an alternative hypothesis that the energy use model has a second parameter order different from the first parameter order. The processing circuit generates a test statistic indicating an improvement between the first regression statistic and the second regression statistic, compares the test statistic to a threshold value to determine whether the improvement warrants rejecting the null hypothesis, and determines an appropriate parameter order for the energy use model based on a result of the comparison.