Scoring entity names of devices in a building management system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated Fault Detection & Diagnostics (FDD) analytics in building management systems face challenges due to the unavailability of contextual information in machine-readable format, as text labels require experience and knowledge to interpret, and naming conventions vary across regions and customers, making it difficult to classify devices like HVAC systems effectively.
Innovation Solution
A controller is configured to extract core names from entity names, compare them to candidate core names, determine scores, and classify devices based on the highest score, using techniques such as constructing bigrams and summing results to facilitate accurate classification of devices into classes like dedicated or common space devices within HVAC systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If text labels are used for device identification, then device classification can be performed, but the contextual information is not available in machine-readable format and requires expert interpretation
Solution Approach 1:
The patent introduces an intermediary processing system that translates human-readable device names into structured contextual information. The system uses natural language processing and classification algorithms as mediators to convert unstructured text labels into machine-readable formats that preserve contextual meaning, enabling automated FDD analytics without losing informational content.
Solution Approach 2:
The patent replaces the manual mechanical process of expert interpretation with an automated computational system. Instead of requiring human experts to manually interpret device names, the system uses algorithmic processing, pattern recognition, and classification to automatically extract contextual information from text labels, substituting human cognitive work with computational mechanisms.
2Adaptability or versatility
If naming conventions vary across regions and customers, then device classification becomes difficult, but maintaining flexibility in naming is important
Solution Approach 1:
The patent creates a universal classification framework that can handle multiple naming conventions simultaneously. The system is designed to accommodate diverse regional and customer-specific naming styles while applying consistent classification logic, making it multi-functional in handling different linguistic and cultural variations in device naming without requiring separate systems for each convention.
Solution Approach 2:
The patent dynamically adjusts classification parameters and thresholds based on the specific naming convention being processed. The system modifies its processing parameters adaptively to match different regional styles, allowing the same core algorithm to effectively classify devices across various naming conventions by changing its operational parameters rather than requiring different algorithms for each convention.
3Productivity
If automated FDD analytics is implemented without proper device classification, then processing speed increases, but accuracy of fault detection decreases
Solution Approach 1:
The patent performs preliminary device classification and contextual information extraction before the actual FDD analytics processing. By pre-processing and organizing device data into structured formats with extracted contextual attributes, the system prepares the data in advance, enabling faster and more accurate fault detection during the main analytics phase without requiring complex processing during critical detection moments.
Data Source
AI summary
A controller for classifying devices of a building management system (BMS). The controller may be configured to obtain an entity name for a device, extract a core name from the entity name, compare the core name to candidate core names, determine scores for each comparison, identify a highest score, identify a class of a candidate core name, and classify the device in the class.