BMS Outlier Limit Retraining for Adaptive Fault Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Building management systems face challenges in determining appropriate outlier detection limits for controlled processes, such as HVAC systems, as these limits need to adapt to changes in the process to accurately detect faults and maintain sensitivity over time.

Innovation Solution

A building management system that includes sensors, equipment, and a controller capable of monitoring performance values relative to initial outlier detection limits, generating new limits in response to detected changes, and adjusting these limits based on confidence differences to maintain accurate outlier detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If fixed outlier detection limits are used for controlled processes, then the system structure is simple, but the system cannot adapt to process changes and detection accuracy deteriorates over time

Engineering Contradiction:

Improveadaptability to process changesVSAvoidcomplexity of limit adjustment mechanism

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic outlier detection limits that automatically adjust to process changes. The system monitors process data over time and dynamically recalibrates detection limits based on observed process behavior, transforming the static limit structure into an adaptive one that evolves with the process without requiring manual intervention.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs self-calibration by automatically detecting process changes and adjusting detection limits independently. The controller monitors process data, identifies when process behavior has changed, and autonomously retrains the detection algorithm to establish new baseline limits, eliminating the need for external manual recalibration.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If outlier detection limits are frequently adjusted to maintain sensitivity, then detection accuracy is maintained, but system complexity and computational resources increase

Engineering Contradiction:

Improveoutlier detection accuracyVSAvoidcomplexity of retraining mechanism

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system employs feedback mechanisms where detection performance is continuously monitored and fed back into the limit adjustment process. The controller evaluates whether process changes have occurred based on detected outliers and uses this feedback to determine when retraining is necessary, creating a closed-loop system that maintains accuracy while avoiding unnecessary adjustments.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system changes the parameters of the detection limits based on observed process data characteristics. By analyzing statistical properties of the process data and adjusting detection threshold parameters accordingly, the system maintains measurement precision while the parameter adjustments are driven by data-driven insights rather than complex manual tuning.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If manual recalibration of detection limits is performed, then detection accuracy can be maintained, but time and operational complexity increase

Engineering Contradiction:

Improvereliability of fault detectionVSAvoidtime for limit recalibration

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system automatically performs recalibration without human intervention. The controller continuously monitors process data, detects when process behavior has changed, and autonomously retrains the outlier detection algorithm to establish new limits, eliminating the time and effort required for manual recalibration while maintaining reliable detection accuracy.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system maintains continuous monitoring and automatic adaptation of detection limits rather than relying on periodic manual recalibration. This continuous automated process ensures detection reliability is maintained over time without interrupting operations or requiring manual intervention, eliminating downtime and operational disruptions.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS10120375B2Systems and methods for retraining outlier detection limits in a building management system

Publication Date: 2018.11.06 TYCO FIRE & SECURITY GMBH

AI summary

A building management system (BMS) includes a controller that monitors performance values for a controlled process during a first time period relative to initial outlier detection limits and generates new outlier detection limits for the controlled process in response to a detected change in the controlled process during the first time period. The controller monitors the performance values relative to the new outlier detection limits during a second time period to detect outliers during the second time period. The controller calculates a confidence difference for an estimated confidence parameter based on a number of outliers detected using the new outlier detection limits during the second time period. The controller adjusts the new outlier detection limits in response to the confidence difference dropping below a threshold value.