Failure Detection System Integrating Disparate Plant Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manufacturers face challenges in improving plant asset performance due to scattered and disconnected data silos across operations, maintenance, and finance divisions, with different data measurement cycles and vendor-specific codes, leading to inefficiencies and revenue loss.
Innovation Solution
A computer program and system for failure signature recognition and anomaly detection that analyzes historical and real-time sensor data to identify patterns indicative of equipment failures, using learning agents and multivariate models to generate alarms and detect anomalies, integrating with computerized maintenance management systems to provide unified performance management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data is collected from multiple divisions (operations, maintenance, finance), then information completeness is improved, but data integration complexity increases due to scattered silos and different measurement cycles
Solution Approach 1:
The patent merges data from multiple divisions (operations, maintenance, finance) into a unified performance management system. The system integrates disparate data sources including real-time control data, maintenance cycle data, and financial data into a single platform that can process and analyze all information together, eliminating the siloed approach.
Solution Approach 2:
The performance management system acts as an intermediary layer between different data sources with varying measurement cycles (real-time seconds, calendar-based maintenance, fiscal periods). The system includes translation and synchronization mechanisms that mediate between these different time scales and data formats, enabling integration without direct connection between all source systems.
2Reliability
If real-time monitoring is implemented across all assets, then failure detection capability is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The system segments monitoring by creating separate learning agents for different equipment types and failure modes. Each agent is trained on specific historical data relevant to particular assets, allowing specialized failure detection without requiring a single monolithic complex system. The segmentation enables modular deployment and management.
Solution Approach 2:
The system performs preliminary training of learning agents using historical sensor data and failure information before actual failure detection begins. This preliminary action creates pre-configured detection models that can operate autonomously in real-time, reducing the complexity of real-time decision-making while maintaining high detection capability.
Data Source
AI summary
A system for performing failure signature recognition training for at least one unit of equipment. The system includes a memory and a processor coupled to the memory. The processor is configured by computer code to receive sensor data relating to the unit of equipment and to receive failure information relating to equipment failures. The processor is further configured to analyze the sensor data in view of the failure information in order to develop at least one learning agent for performing failure signature recognition with respect to the at least one unit of equipment.