AI Building Energy Modeling with Profile Clustering and Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing building energy management systems lack the necessary granularity in building information for modern real-time energy modeling, making AI-based building energy modeling time-consuming and inefficient.
Innovation Solution
A method and system for generating an AI-based building energy model by defining building parameters, simulating energy profiles, clustering, and calibrating a physical building model using a processor to create an energy profile database and selecting the closest match for a client building, followed by generating an AI-based model using known energy data and metadata.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detailed building energy data is collected and processed for AI-based modeling, then model accuracy is improved, but time consumption increases
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing building energy data in structured databases before actual modeling is needed. Energy profiles, building parameters, and weather data are organized in advance, allowing rapid retrieval and reducing real-time processing requirements while maintaining model accuracy.
Solution Approach 2:
The system creates simplified copies or representations of complex building energy data through energy profiles and standardized building parameters. These copied representations capture essential characteristics without requiring full detailed data, enabling faster processing while preserving sufficient accuracy for AI-based modeling.
2Reliability
If comprehensive building parameters are defined and simulated, then model reliability is improved, but system complexity increases
Solution Approach 1:
The system segments comprehensive building parameters into distinct categories and stores them in separate database tables (building parameters table, energy profiles table, weather parameters table). This segmentation allows the system to manage complex data through modular, organized structures that can be processed independently, reducing overall system complexity while maintaining complete parameter coverage for reliable modeling.
Solution Approach 2:
The system introduces intermediary components such as standardized building parameter definitions, energy profile templates, and structured data formats that mediate between raw building data and AI modeling requirements. These intermediaries organize and standardize comprehensive parameters, making them more manageable and reducing system complexity while preserving model reliability.
3Measurement precision
If building energy models are customized for each client building, then model precision is improved, but productivity decreases
Solution Approach 1:
The system creates universal building parameter definitions and standardized energy profile templates that can be applied across multiple client buildings. These universal structures serve multiple functions: they provide consistent data organization, enable rapid model deployment for different buildings, and maintain precision by capturing essential building characteristics. The same framework adapts to various building types without requiring complete customization for each client.
Solution Approach 2:
The system uses standardized building parameters that can be efficiently adjusted or changed for different client buildings rather than creating entirely customized models. By defining a comprehensive set of building parameters in advance, the system allows rapid modification of specific parameters for each client while maintaining the overall model structure, thus improving setup speed without sacrificing precision.
Data Source
AI summary
A method for generating an AI-based building energy model for a client building, comprising: generating an energy profile database by: defining a set of building parameters; generating energy profiles by simulating a set of physical building models; and, populating the energy profile database with the energy profiles; determining an energy profile for the client building by: splitting the energy profile database into groups and clustering each group into a set of clusters; selecting a cluster; and, selecting the energy profile in the cluster that is a closest match to that of the client building; selecting a physical building model from a building model database that corresponds to the energy profile; calibrating the physical building model to generate an adjusted building model; and, generating a set of training datasets from the adjusted building model and inputting the set of training datasets into an AI module to generate the AI-based model.