Equipment Behavior Profiling for Low-False-Positive Maintenance Alerts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for managing equipment in buildings are inadequate in predicting failures and optimizing operations due to misidentification of equipment behavior, leading to late or non-comprehensive repairs, increased maintenance costs, and reactive maintenance approaches with high false positives.
Innovation Solution
A system comprising a classification module, profiling module, and insights and service actionable generation module that behaviorally classifies equipment, generates time-granular behavior patterns, and provides proactive insights and recommendations for maintenance, enabling early detection of anomalies and reducing redundant processing through reusable classifications and autonomous editing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional alarm-based FDD solutions are used to identify equipment problems, then fault detection capability is provided, but false positives increase and real failure reasons are not identified
Solution Approach 1:
The system segments equipment behavior analysis into multiple independent behavior models, each focusing on specific aspects such as temperature behavior, pressure behavior, vibration behavior, etc. This segmentation allows precise analysis of individual behavior patterns without interference from other factors, reducing false positives while maintaining comprehensive fault detection capability
Solution Approach 2:
The system transforms traditional threshold-based alarm parameters into behavior-based parameters that capture temporal patterns and relationships. By changing from static threshold comparisons to dynamic behavior pattern recognition, the system achieves more accurate fault detection with reduced false positives
2Measurement precision
If comprehensive equipment behavior analysis is performed to identify all possible failure reasons, then prediction accuracy improves, but system complexity increases
Solution Approach 1:
The system divides complex equipment behavior analysis into multiple independent behavior models, each analyzing specific aspects of equipment operation. This segmentation reduces the computational complexity of each individual model while collectively achieving comprehensive and accurate fault prediction
Solution Approach 2:
The system creates universal behavior models that can be applied across different equipment types and failure modes. These multi-functional models reduce overall system complexity by avoiding the need to build separate specialized models for each specific failure scenario
3Ease of repair
If traditional reactive maintenance is performed after equipment failure, then maintenance cost is reduced, but equipment downtime increases
Solution Approach 1:
The system performs preliminary actions by detecting and analyzing equipment behavior patterns that precede actual failures. By identifying degradation trends and predicting failures before they occur, the system enables proactive maintenance scheduling that prevents unexpected downtime while optimizing maintenance cost
Data Source
AI summary
Aspects of the invention are directed towards a system and a method for generating service actionable for a plurality of equipment. Embodiments of the invention describe the method comprises steps of behaviorally classifying an equipment into normalizing classification and behavior classification. The method further comprises steps of processing the normalizing and behavior classifications to generate one or more profiles corresponding to the equipment. The one or more profiles represent time-granular behavior patterns of the equipment. The method comprises steps of generating time-granular normalized characteristics for the equipment and normalizing variances of the time-granular normalized characteristics and the time-granular behavior patterns to generate possible service actionable (SACT) recommendations that are integrated into workflows to drive action and receive prediction confirmation.