Ensemble Fault Detection Models for Building Management Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing building management systems (BMS) face challenges in accurately detecting and diagnosing faults, often resulting in false alarms and inefficient resource allocation.
Innovation Solution
A method and system for generating fault determinations in a BMS using multiple fault detection models and a weighting function to produce a fault score, which is then compared to a fault value to determine the existence of a fault.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple fault detection models are used to improve detection accuracy, then the reliability of fault determination is improved, but the device complexity increases
Solution Approach 1:
The patent combines multiple fault detection models into an ensemble system that shares common components such as the fault indicator generation mechanism, weighting function, and fault score calculation. This merging approach allows the system to achieve higher reliability through multiple models while avoiding the full complexity of implementing entirely separate detection systems for each model.
Solution Approach 2:
The ensemble of fault detection models uses a universal framework where different models can detect various types of faults using the same underlying architecture and processing steps. The system generates fault indicators, applies weighting functions, and calculates fault scores consistently across all models, making the system multi-functional and adaptable to different fault detection needs without proportionally increasing complexity.
2Reliability
If fault detection sensitivity is increased to detect more faults, then the reliability of fault detection is improved, but the number of false alarms increases
Solution Approach 1:
The system incorporates feedback mechanisms where the fault score, derived from weighted fault indicators across multiple models, provides a confidence metric that feeds back into the fault determination process. This feedback allows the system to distinguish between true faults and false alarms by comparing the aggregated signal strength against thresholds, reducing false alarms while maintaining high sensitivity.
Solution Approach 2:
The patent changes the parameter of fault determination from binary (fault/not fault) to a continuous fault score that reflects confidence level. By transforming the output into a scaled parameter that can be thresholded at different levels, the system can adjust its sensitivity dynamically, maintaining high detection capability while filtering out false alarms through configurable thresholding based on the fault score magnitude.
Data Source
AI summary
A method of generating a fault determination in a building management system (BMS), the method including receiving signal data, generating, using a number of fault detection models, a number of fault indications based on the signal data, generating, using a weighting function, based on the number of fault indications, a fault score, comparing the fault score to a fault value, and determining, based on the comparison, an existence of a fault.