Peer Anomaly Detection for Rotational Equipment Under Varying Contexts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring a plurality of peer devices, such as gearboxes in jack-up vessels, are cumbersome, costly, and time-consuming, particularly in handling variability due to different contexts and identifying root causes in real-time or near-real-time.
Innovation Solution
A rotational equipment peer anomaly system that includes condition sensors to detect rotational equipment conditions, a processor to analyze sensor data, and a storage medium with executable instructions to determine data scope parameters, analyze data features, compare feature values, and identify anomalies satisfying an anomaly threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If known monitoring methods and systems are used to identify potential issues by comparing current sensor data with historical sensor data, then anomaly detection capability is provided, but the system cannot account for variability due to different contexts and cannot identify root causes in real-time or near-real-time
Solution Approach 1:
The patent segments the sensor data analysis by introducing context parameters that divide the data into distinct operational contexts. Instead of treating all sensor data uniformly, the system segments it according to operational conditions, allowing for more precise anomaly detection within each context while maintaining real-time processing capabilities.
Solution Approach 2:
The system dynamically adjusts the analysis approach based on identified context parameters. When contextual variability is detected, the system adapts its comparison methodology to account for the specific operational context, enabling real-time root cause identification while maintaining high anomaly detection accuracy across varying conditions.
2Reliability
If comprehensive sensor data is collected from multiple peer devices to improve monitoring accuracy, then detection reliability is improved, but the complexity of processing and analyzing the data increases
Solution Approach 1:
The patent extracts and identifies specific context parameters from the comprehensive sensor data that are most indicative of operational conditions. By focusing on these key contextual features rather than processing all raw sensor data uniformly, the system maintains high monitoring reliability while significantly reducing the complexity of data processing and analysis.
3Measurement precision
If detailed analysis of sensor data is performed to identify root causes of anomalies, then diagnostic precision is improved, but the time required for analysis and the computational resources increase
Solution Approach 1:
The system performs preliminary identification of context parameters and operational conditions before conducting detailed anomaly analysis. By pre-processing and categorizing data according to operational context, the system enables faster root cause identification while maintaining high diagnostic precision, as the detailed analysis can focus on pre-segmented contextual groups rather than raw data.
Data Source
AI summary
A rotational equipment peer anomaly system includes rotational equipment condition sensors, a rotational equipment peer anomaly processor, and a non-transitory tangible storage medium. The sensors are configured to detect rotational equipment conditions associated with the peer devices and generate rotational equipment condition sensor data corresponding to the conditions. The non-transitory tangible storage medium stores rotational equipment peer anomaly processor executable instructions that, when executed, causes the processor to receive the sensor data, determine a data scope parameter, apply the data scope parameter to the sensor data, analyze the sensor data to determine rotational equipment condition data features associated with the peer devices, compare each data feature with each of the other data features to determine rotational equipment condition data feature values associated with the peer devices, and analyze the data feature values to identify at least a first rotational equipment condition data feature value satisfying an anomaly threshold.