Building Controller DOE Optimization for Equipment Upgrades
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems face challenges in determining the optimal combinations of building equipment upgrades due to the exponentially large number of possible selections, making it inefficient and time-consuming to evaluate and compare various combinations for optimizing building performance.
Innovation Solution
A building management system that uses a design of experiments analysis to identify facility improvement measures, generate objective functions, and optimize economic values by selecting appropriate levels for each measure based on output responses, including binary, linear, and quadratic relationships, and models operating and capital expenses to determine optimal upgrades.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional brute force evaluation of all potential equipment upgrade combinations is performed, then complete optimization can be achieved, but the computation time becomes excessively long and the process becomes impractical
Solution Approach 1:
The patent segments the exhaustive evaluation process into two phases: (1) Design of Experiments phase that evaluates a representative subset of combinations to build statistical models, and (2) Optimization phase that uses these models to predict optimal combinations without exhaustive evaluation. This segmentation reduces computation time while maintaining optimization accuracy through statistical sampling and modeling.
Solution Approach 2:
The patent performs preliminary action by conducting Design of Experiments analysis on a selected subset of equipment combinations before full optimization. This preliminary evaluation builds statistical models (regression, ANOVA) that capture the relationships between equipment selections and building performance, enabling subsequent optimization without re-evaluating all combinations.
2Measurement precision
If the number of equipment upgrade selections is increased to achieve better building performance optimization, then the quality of optimization improves, but the complexity of the system increases exponentially
Solution Approach 1:
The patent changes the parameter representation by using statistical models (regression coefficients, ANOVA results) to describe equipment performance relationships instead of evaluating each individual combination. This transforms the complex discrete optimization problem into a continuous parameter optimization problem that is computationally manageable while maintaining optimization quality.
Solution Approach 2:
The patent introduces statistical models as intermediaries between equipment selections and building performance outcomes. These models (regression equations, ANOVA models) act as mediators that predict performance without requiring direct evaluation of all equipment combinations, thereby reducing system complexity while preserving optimization capability.
3Productivity
If Design of Experiments analysis is used to reduce the number of combinations evaluated, then computation efficiency improves, but the complexity of the analysis method increases
Solution Approach 1:
The patent applies Design of Experiments methodology that serves multiple functions simultaneously: (1) screening important factors, (2) building predictive models, (3) quantifying interactions, and (4) guiding optimization. This multi-functional approach achieves computation efficiency through a unified framework rather than separate analysis steps, managing complexity through methodological integration.
Solution Approach 2:
The patent performs preliminary Design of Experiments analysis to establish statistical models before the optimization phase. This preliminary action captures the essential relationships between equipment selections and building performance, enabling subsequent efficient optimization without repeating complex analysis. The factorial designs and statistical modeling are performed once to enable multiple optimization queries.
Data Source
AI summary
A building management includes building equipment operable to affect a variable state or condition of a building and a controller. The controller is configured to identify one or more facility improvement measures (FIMS), each of the FIMS representing a potential upgrade or addition to the building equipment. The controller is further configured to perform a design of experiments analysis to determine a plurality of combinations of the FIMS, each combination including one or more of the FIMS and a level for each FIM in the combination. The controller is configured to generate an objective function based on the combinations of the FIMS. The objective function indicates an economic value as a function of the FIMS and optimize the objective function to determine an optimal combination of the FIMS.