Smart Sensor Alarm Signatures for Rotary Equipment Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for monitoring rotary equipment require large sets of data, leading to bandwidth issues and have a 50/50 success rate in diagnosing equipment alarms, necessitating the need for a more effective method to monitor and diagnose alarm conditions.
Innovation Solution
A system utilizing smart sensors to perform data analytics, featuring an alarm signature and diagnostic detector with a signal processor that receives and processes time-stamped data to determine alarm diagnoses, which can self-learn and update diagnostic signatures, enabling efficient monitoring and maintenance of rotary equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large sets of data are collected from sensor modules for equipment diagnosis, then diagnostic accuracy can be improved, but bandwidth issues and data transmission problems occur
Solution Approach 1:
The patent segments the diagnostic process by deploying edge computing devices at distributed locations (equipment sites, regional centers) that perform local data processing and filtering. Only essential diagnostic data and alarm conditions are transmitted to the central server, reducing overall bandwidth consumption while maintaining diagnostic accuracy through distributed intelligence.
Solution Approach 2:
The system performs preliminary data processing, filtering, and feature extraction at the edge devices before transmission. Diagnostic algorithms pre-process sensor data locally to identify only relevant alarm conditions and diagnostic information, eliminating the need to transmit entire raw datasets and thereby reducing bandwidth requirements.
2Ease of operation
If physics-based models are used to diagnose equipment alarms, then diagnostic capability is provided, but success rate is only 50/50 and resident experts are required
Solution Approach 1:
The patent replaces traditional physics-based diagnostic models with machine learning and AI-based diagnostic algorithms. These intelligent systems learn from historical equipment data and alarm patterns to provide more accurate diagnoses without requiring resident experts, thereby improving both ease of operation and diagnostic success rate.
Solution Approach 2:
The diagnostic system performs self-learning and self-improvement through continuous analysis of equipment data and alarm conditions. The AI algorithms automatically update their diagnostic models based on accumulated experience, enabling the system to diagnose equipment alarms independently without requiring external expert intervention while maintaining high success rates.
3Measurement precision
If smart sensors and data analytics are implemented for equipment diagnosis, then diagnostic accuracy is improved, but system complexity increases
Solution Approach 1:
The patent implements universal diagnostic algorithms and AI models that can diagnose multiple types of equipment across different industries. The same core diagnostic engine and machine learning frameworks are applied to various rotary equipment, pumps, motors, and machinery, reducing system complexity through standardized multi-functional diagnostic capabilities rather than requiring separate systems for each equipment type.
Data Source
AI summary
A system for performing sensor data analytics for equipment diagnostics featuring an alarm signature and diagnostic detector having a signal processor configured to receive signaling containing information about an alarm signature for sensed data that is time-stamped for captured alarm signature parameters in order to monitor rotary equipment, and also about diagnostic detector signatures for diagnostic detectors related to alarm conditions for the rotary equipment; and determine corresponding signaling containing information about an alarm diagnosis based upon a match between the alarm signature and one of the diagnostic detector signatures contained in the signaling received.