Building Point Naming With Generative AI PES Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building devices are often installed without standard naming conventions, making configuration challenging for building configuration systems due to unstructured object names.
Innovation Solution
A method using generative artificial intelligence to perform data augmentation and fine-tuning of large language models for PES classification, enabling automatic generation of standardized tags for building devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If contractors install building devices without standard naming conventions, then installation flexibility is improved, but configuration difficulty increases
Solution Approach 1:
The patent introduces an intermediary system comprising discovery mechanisms and machine learning models that translate non-standard device names into standardized PES classifications. This intermediary layer enables contractors to install devices with flexible naming while the system automatically bridges the gap to standardized configuration, resolving the contradiction between installation flexibility and configuration difficulty.
Solution Approach 2:
The system implements self-service through automated discovery and classification processes. The machine learning models autonomously analyze device data, perform PES classification, and generate standardized configurations without requiring manual intervention from contractors, thereby maintaining installation flexibility while eliminating configuration difficulty.
2Device complexity
If manual configuration methods are used for building devices, then system complexity is reduced, but productivity decreases
Solution Approach 1:
The patent replaces manual mechanical configuration processes with automated machine learning-based classification systems. The discovery mechanisms and trained models automatically perform PES classification, substituting human effort with intelligent algorithms that maintain manageable system complexity while dramatically improving configuration productivity.
Solution Approach 2:
The system performs preliminary actions by pre-training machine learning models with extensive PES classification data before deployment. This preliminary training enables the models to automatically and accurately classify devices during installation, eliminating the need for manual configuration while maintaining system simplicity and enhancing productivity.
3Measurement precision
If standardized naming conventions are enforced during installation, then configuration accuracy is improved, but ease of operation worsens
Solution Approach 1:
The patent inverts the traditional approach by not requiring standardized names during installation. Instead, it uses discovery mechanisms to capture device information and applies machine learning models to derive standardized PES classifications automatically. This inversion maintains ease of operation during installation while achieving configuration accuracy through automated classification.
Data Source
AI summary
A method for a building automation system includes performing data augmentation of a point, equipment, and subtype (PES) dataset using generative artificial intelligence to generate an augmented PES dataset. The method includes fine-tuning at least one large language model (LLM) using the augmented PES dataset, discovering points on a building network for the building automation system, performing PES classification of the points using the at least one fine-tuned LLM, and operating equipment of the building automation system using the PES classification to affect a physical condition of a building.