Building Automation Point Typing with Probabilistic String Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current building automation systems face challenges in efficiently classifying data points due to non-standard and semantically rich descriptions, requiring manual intervention and lacking expressive power to assign computable semantic types and relationships among objects, which increases the complexity of commissioning and management processes.
Innovation Solution
A computerized method and system that utilize a processing circuit to receive unclassified data points and attributes, generate a term set of substrings, calculate probability indicators for potential point types, and automatically assign point types by finding the most probable substring and type pair, using techniques like naive Bayes classification and latent semantic indexing to reduce manual effort and improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual point classification methods are used, then accurate functional classification can be achieved, but the commissioning process becomes extremely lengthy and time-consuming
Solution Approach 1:
The system enables automated self-service classification by having the computer automatically analyze point descriptions, generate classifications, and assign point types without human intervention. The computer uses algorithms to process unclassified points and assign them to appropriate point types based on the classification schema, eliminating the need for manual commissioning while maintaining accuracy.
Solution Approach 2:
The patent replaces the mechanical manual classification process with an automated computational system. The computer executes algorithms that analyze point descriptions, calculate classification probabilities, and automatically assign point types, substituting human manual evaluation with machine-based automated classification.
2Extent of automation
If standardized naming conventions are enforced, then machine processing becomes easier, but the system loses the ability to handle diverse and semantically rich descriptions
Solution Approach 1:
The system changes the parameter of classification from requiring standardized naming to accepting diverse descriptions by introducing probabilistic classification. Instead of demanding exact naming conventions, the system calculates the probability that a point belongs to each possible point type based on its description, allowing flexible handling of varied naming styles while enabling automated processing.
Solution Approach 2:
The patent introduces an intermediary classification schema that acts as a bridge between diverse point descriptions and standardized point types. The computer analyzes various description formats and maps them to the classification schema, which then assigns appropriate point types, serving as a mediator that handles both diversity and standardization.
3Reliability
If comprehensive manual evaluation of each point is performed, then relevant points can be accurately identified, but the complexity of the commissioning process increases significantly
Solution Approach 1:
The computer performs automated self-service evaluation by independently analyzing each point's description, comparing it against the classification schema, and determining relevance without requiring human commissioning personnel to manually evaluate each point. This maintains identification accuracy while eliminating process complexity.
Solution Approach 2:
The system applies partial action by focusing computational resources only on analyzing the description text for classification purposes, rather than requiring comprehensive manual evaluation of all point attributes. The computer performs sufficient analysis to achieve accurate identification without the excessive complexity of detailed manual commissioning.
Data Source
AI summary
A computerized method of assigning a building automation system point type to a plurality of unclassified data points is provided. The method includes receiving unclassified data points and attributes for each data point. The method includes receiving classifications for a first subset of the unclassified data points. Each classification associates a data point with a building automation system point type. The method includes generating a term set containing substrings that appear in the attributes. The method includes generating a first matrix describing a frequency that the substrings appear in the attributes. The method includes calculating an indicator of a probability that the presence of the selected substring results in the data point belonging to the selected point type. The method includes assigning a point type to a second subset by finding the substring and potential point type pair having the greatest indication of probability.