AI Mapping of Building Management Data to Information Models

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The process of generating an information model (IM) from building management system (MS) data is cumbersome, time-consuming, and prone to errors due to the vast differences among MS providers, leading to incorrect connections and increased energy consumption when using erroneous IMs.

Innovation Solution

A method utilizing an AI processing component, particularly a Large Language Model (LLM), generates a program code executable to transform MS data into accurate IM data, allowing manual refinement to ensure correctness and completeness.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If manual analysis of MS data is used to generate IM, then flexibility in handling different MS providers is maintained, but the process becomes cumbersome and time-consuming

Engineering Contradiction:
Improvecompatibility with different MS providersVSAvoidtime to generate IM
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent introduces an AI processing component as an intermediary between the diverse MS data and the target IM format. This AI component learns mapping patterns from training data and automatically transforms MS data into IM data, eliminating the need for manual analysis while maintaining adaptability to different MS providers through its learning capability.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the manual mechanical process of analyzing and transforming MS data with an automated AI-based system. The AI processing component uses machine learning algorithms to automatically map MS data to IM data, substituting the time-consuming manual workflow with an efficient automated process that maintains high adaptability.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Loss of time

If automated methods are used to generate IM from MS data, then time consumption is reduced, but errors in device connections and logical relationships increase

Engineering Contradiction:
Improvetime to generate IMVSAvoidaccuracy of device connections
Core Design Contradiction:
Loss of timeVSReliability

Solution Approach 1:

The patent applies preliminary action by training the AI processing component on a large dataset of MS data and corresponding correct IM data before actual transformation. This training phase allows the AI to learn the correct mapping relationships and logical connections between devices, ensuring high accuracy when generating IM data in production without manual intervention.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent incorporates feedback mechanisms where the AI processing component's transformations are validated against expected IM data structures and logical relationships. Errors or inconsistencies are fed back into the system for correction and retraining, continuously improving the reliability of device connections and logical relationships in the generated IM data.

Inventive Principle:
Principle #23Feedback

3Manufacturing precision

If custom analysis processes are developed for each building's MS data, then accuracy of IM data is improved, but device complexity and costs increase

Engineering Contradiction:
Improvequality of IM dataVSAvoidcomplexity of analysis process
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent creates a universal AI processing component that can handle multiple types of MS data from different providers through a single system. This multi-functional AI component learns various mapping patterns during training and can automatically adapt to different data formats and structures, eliminating the need for separate custom analysis processes for each building while maintaining high IM data quality.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent uses parameter changes in the form of adjusting AI model parameters and transformation rules based on the specific characteristics of different MS data sources. Rather than creating entirely custom analysis processes, the system modifies existing AI parameters and mapping rules to accommodate different data formats, reducing complexity while maintaining accuracy.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4657183A1Method for obtaining information model data for one or more
Publication Date: 2025.12.03 ABB (SCHWEIZ) AG
  • EP4657183A1 patent drawingFigure 1~2
  • EP4657183A1 patent drawingFigure 3~4
  • EP4657183A1 patent drawingFigure 5

AI summary

The disclosure relates to a method (100) for obtaining information model, IM, data for one or more buildings (10), the method (100) comprising: - obtaining building management system, MS, data (22) of a building management system of the one or more buildings (10); - obtaining instruction data comprising one or more instructions for processing the MS data (22) by an Artificial Intelligence, Al, processing component (44) for generating a program code (42) executable to obtain the IM data; - providing the MS data (22) and the instruction data for processing by the Al processing component (44); and - receiving the program code (42) executable to obtain the IM data by the Al processing component (44).