Predictive Maintenance Data Fusion for Multi-Equipment Vibration Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional predictive maintenance methods in the petrochemical industry rely solely on single equipment signals, lacking integration with production line schedules and process parameters, leading to misjudgments and failures.
Innovation Solution
An electronic system that combines data from multiple vibration sensing devices on associated equipment and process parameters to train a predictive model, performing data fusion, feature extraction, and time-series analysis for accurate health status diagnosis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single equipment signal monitoring is used, then device complexity is reduced, but predictive maintenance accuracy deteriorates
Solution Approach 1:
The patent combines vibration signals from multiple equipment pieces into a unified analysis system. The signal integration module aggregates vibration data from different equipment sources, and the predictive model analyzes them collectively to improve diagnostic accuracy while managing system complexity through modular architecture.
Solution Approach 2:
The patent introduces intermediate processing modules including signal integration, feature extraction, and predictive modeling layers that mediate between raw multi-equipment signals and final diagnostic results. These intermediaries transform complex multi-source data into actionable maintenance insights.
2Measurement precision
If production line schedule and process parameter integration is added, then predictive maintenance accuracy is improved, but data processing complexity increases
Solution Approach 1:
The predictive model is designed to handle multiple data types universally - vibration signals, production schedules, and process parameters - through a single integrated analysis framework. This multi-functional approach improves diagnostic accuracy without requiring separate processing systems for each data type.
Solution Approach 2:
The patent transforms diverse data types (time-series vibration signals, discrete production schedules, continuous process parameters) into unified feature representations suitable for predictive modeling. This parameter transformation enables accurate health status diagnosis while managing data processing complexity through standardized feature extraction.
3Reliability
If multi-equipment vibration data fusion is implemented, then diagnostic reliability is improved, but information processing requirements increase
Solution Approach 1:
The patent extracts key diagnostic features from multi-equipment vibration data through feature extraction modules that identify and isolate relevant signal characteristics. This extraction process reduces information processing load by focusing on discriminative features while maintaining diagnostic reliability.
Solution Approach 2:
The system performs preliminary signal processing including filtering, feature extraction, and data fusion before final predictive analysis. This preliminary action prepares the data in advance, reducing the computational burden on the predictive model while ensuring diagnostic reliability through thorough preprocessing.
Data Source
AI summary
An electronic system includes a first vibration sensing device, a second vibration sensing device, a database, and a processor. The first vibration sensing device is disposed on a surface of first equipment, and detects a first time-domain vibration signal when the first equipment is operating. The second vibration sensing device is disposed on a surface of second equipment, and detects a second time-domain vibration signal when the second equipment is operating. The database stores process parameters associated with at least one of the first equipment and the second equipment. The processor receives the first time-domain vibration signal, the second time-domain vibration signal, and the process parameters, and trains a predictive model based on the first time-domain vibration signal, the second time-domain vibration signal, and the process parameters. The predictive model diagnoses a health status of the first equipment.