Geometric balancing algorithm for multi-metric building design optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated building design software lacks the ability to efficiently optimize multiple performance metrics such as cost, energy, daylight, and thermal comfort simultaneously, requiring specialized expertise and being computationally intensive, limiting its use in architecture, engineering, and construction management.
Innovation Solution
A computer-implemented method using a geometric balancing algorithm and machine learning to automate simulations, check input quality, and provide automated reports, allowing for rapid analysis and optimization of building components to balance competing objectives, reducing the need for specialized consultants.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If current automated building design software is used, then building performance simulations can be run, but the process requires specialized technicians with years of training and is computationally intensive
Solution Approach 1:
The system performs self-service by automatically setting up simulations, checking input data quality, running multiple simulations, and generating interpreted results without requiring specialized technician intervention. The automated system serves itself through machine learning algorithms that handle both the technical setup and interpretation phases.
Solution Approach 2:
The patent replaces the mechanical system of manual simulation setup and interpretation with an automated digital system using machine learning algorithms. The machine learning model automatically processes input data, configures simulations, and generates interpreted results, eliminating the need for human expertise in these tasks.
2Adaptability or versatility
If multiple building performance metrics are analyzed separately, then each metric can be optimized, but the process requires hours of spreadsheet manipulation and multiple tools
Solution Approach 1:
The system merges multiple separate building performance analysis tools and processes into a single integrated platform. It combines energy analysis, daylight simulation, glare assessment, and thermal comfort evaluation into one unified system that can process all metrics simultaneously through a single interface, eliminating the need for separate spreadsheet manipulations.
Solution Approach 2:
The patent creates a universal software system that can handle multiple building performance metrics through a single interface and processing framework. The machine learning model is designed to accommodate various performance criteria (energy, daylight, glare, thermal comfort, cost) and automatically selects and processes the relevant metrics based on user inputs, providing multi-functional capability.
3Ease of operation
If building design decisions are made based on past projects or team familiarity, then the decision-making process is simple, but the optimization of competing objectives such as daylight, glare, and thermal comfort is limited
Solution Approach 1:
The system incorporates feedback mechanisms where machine learning algorithms continuously analyze building performance data and provide automated recommendations. The system learns from input data patterns and generates optimized design recommendations that balance competing objectives, providing both ease of use through automated guidance and precision through data-driven optimization.
Solution Approach 2:
The patent utilizes parameter changes by automatically adjusting multiple building performance parameters simultaneously based on machine learning optimization. The system modifies design parameters (daylighting, insulation, orientation, etc.) through automated algorithms that find optimal balances between competing objectives, achieving precise optimization without manual intervention.
4Measurement precision
If specialized consultants are hired for building performance optimization, then accurate analysis can be provided, but the cost increases significantly
Solution Approach 1:
The system creates a digital copy of specialized consultant expertise through machine learning models trained on building performance data and design optimization principles. The AI model replicates the analytical capabilities and optimization techniques of experienced consultants, providing accurate analysis and recommendations without requiring human consultants, thus reducing costs while maintaining precision.
Data Source
AI summary
A method of optimizing computer-implemented building design, includes the following: defining one or more options for each building component; providing an energy use intensity versus cost optimization value for each option for a plurality of metrics; selecting a subset of the plurality of metrics applicable to each option; defining a metric vector for each metric through connecting the energy use intensity versus cost optimization value for each option; arranging each metric vector on a coordinate grid with an equal angle between each metric vector; constructing a two-dimensional polygon on an XY-plane by interconnecting for all the metric vectors the energy use intensity versus cost optimization value for each option; providing a performance value for each option for each metric vector based on a percentage the metric vector the associated option represents; and representing a fitness factor for each option as a function of each of the plurality of metrics.