BMS Supervisory Fault Detection for Conflicting HVAC Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing building management systems (BMS) face challenges in accurately detecting faults in HVAC systems, leading to inefficiencies and potential damage due to undetected issues.
Innovation Solution
A method involving multiple fault detection methods and a neural network that processes time series data to generate and analyze fault detection results, using AI and statistical inferences to determine fault conditions in HVAC systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple fault detection methods are used to improve detection accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The fault detection system is divided into multiple independent detection methods (temporal detection, peer detection, AI detection) that each process data independently and provide separate fault detection results. This segmentation allows each method to specialize in detecting different types of faults while maintaining overall system accuracy without requiring one complex monolithic system.
Solution Approach 2:
Multiple fault detection results from different detection methods are merged and combined as inputs to a neural network that synthesizes all results to make the final fault determination. This merging approach consolidates the strengths of multiple simpler detection methods into a unified decision-making process, improving overall detection accuracy while managing complexity through modular architecture.
2Reliability
If multiple fault detection methods and neural networks are implemented to reduce false alarms, then reliability is improved, but loss of time in processing increases
Solution Approach 1:
Fault detection results from multiple methods are generated and prepared in advance as inputs to the neural network before final fault determination is made. The neural network receives pre-processed fault detection results that have already been computed by various detection methods, allowing for rapid synthesis and decision-making without requiring complex real-time calculations at the moment of fault detection.
Solution Approach 2:
The manual or rule-based synthesis of multiple fault detection results is replaced with a neural network that automatically processes and integrates results from multiple detection methods. This substitution of mechanical/rule-based processing with intelligent automated processing reduces the time required to synthesize multiple detection results while improving reliability through more accurate pattern recognition.
Data Source
AI summary
A method for correcting faults in a building management system (BMS) includes receiving time series data characterizing an operating performance of one or more BMS devices, generating a first fault detection result by processing the time series data using a first fault detection technique, generating a second fault detection result that conflicts with the first fault detection result by processing the time series data using a second fault detection technique different than the first fault detection technique, resolving a conflict between the first fault detection result and the second fault detection result by applying both the first and second fault detection results as inputs to a neural network configured to output an indication of whether a fault condition is occurring in the BMS, and initiating an action to resolve the fault condition in response to the indication indicating that the fault condition is occurring in the BMS.