Hierarchical Failure Model for Industrial Alarm Root Cause Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In industrial automation systems, identifying the root cause of alarms and diagnosing failures is cumbersome and time-consuming due to the complexity of large-scale infrastructure, where a single failure can trigger numerous alarms, often leading to hidden causes and costly consequential damage.
Innovation Solution
The development of failure models combining graph-based and statistical models with sensor data and prior knowledge to construct a hierarchical failure model, enabling probabilistic reasoning for diagnostic and prognostic analytics, which includes diagnostic variables and their relationships, and uses propagation of uncertainties to derive knowledge about potential or actual system failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional alarm systems are used in large-scale industrial automation systems, then monitoring coverage is comprehensive, but diagnosing root causes becomes cumbersome and time-consuming
Solution Approach 1:
The patent segments the complex alarm system into a hierarchical structure with multiple levels (field level, control level, supervisory level). Each level handles specific diagnostic tasks, breaking down the overwhelming complexity of root cause analysis into manageable segments that can be processed systematically rather than manually.
Solution Approach 2:
The patent introduces an intermediary diagnostic system that sits between the alarm generation sources and the operators. This intermediary automatically processes alarm data, correlates symptoms, and generates diagnostic recommendations, acting as a mediator that translates raw alarm data into actionable insights without requiring direct operator intervention in the complex analysis process.
2Measurement precision
If comprehensive sensor data is collected from all system components, then diagnostic accuracy can be improved, but data complexity and processing difficulty increase
Solution Approach 1:
The patent segments sensor data processing by assigning different processing tasks to different hierarchical levels. Field-level devices perform local data filtering and preprocessing, control-level systems handle component-specific analysis, and supervisory systems perform system-wide correlation. This segmentation reduces the complexity at each processing stage while maintaining overall diagnostic accuracy.
Solution Approach 2:
The patent extracts and separates relevant diagnostic information from the comprehensive sensor data at each hierarchical level. Rather than processing all raw data centrally, the system extracts key features and symptoms at local levels, transmitting only the essential diagnostic information upward, thereby reducing processing complexity while preserving diagnostic accuracy.
3Adaptability or versatility
If manual alarm definition and programming is performed by integrators, then alarm systems can be customized to specific needs, but system complexity and implementation time increase
Solution Approach 1:
The patent implements preliminary action by providing pre-configured alarm templates, standard diagnostic rules, and hierarchical structures that can be deployed immediately. These pre-prepared elements cover common industrial scenarios, allowing rapid implementation while maintaining customization capability through parameter adjustment rather than complete system design.
Solution Approach 2:
The patent creates a universal hierarchical diagnostic framework that can handle multiple alarm types, system configurations, and diagnostic scenarios through a single standardized structure. This multi-functional approach allows the same basic architecture to serve diverse customization needs, reducing implementation complexity while maintaining adaptability.
Data Source
AI summary
A computer-implemented method for detecting faults and events related to a system includes receiving sensor data from a plurality of sensors associated with the system. A hierarchical failure model of the system is constructed using (i) the sensor data, (ii) fault detector data, (iii) prior knowledge about system variables and states, and (iii) one or more statistical descriptions of the system. The failure model comprises a plurality of diagnostic variables related to the system and their relationships. Probabilistic reasoning is performed for diagnostic or prognostic purposes on the system using the failure model to derive knowledge related to potential or actual system failures.