Asset State Analysis Using Preselected Sensors and ML Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches for asset performance management (APM) are inadequate as they fail to determine or predict the operating state of physical assets, leading to inefficient maintenance and operation.
Innovation Solution
A computer-implemented method using machine learning (ML) methodologies to analyze the operating state of physical assets by acquiring data from sensors, correlating it with operating state templates, and generating metrics to measure deviations from optimized states, thereby providing a single index for asset performance evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning methodologies are used to determine asset operating state, then measurement precision of asset state is improved, but device complexity increases
Solution Approach 1:
The patent segments the asset performance management system into multiple specialized models: a first model for determining operating state from sensor data, a second model for analyzing the determined state, and a third model for generating actionable insights. This segmentation allows each model to focus on a specific aspect of the complex analysis task, improving overall measurement precision while managing system complexity through modular design.
Solution Approach 2:
The patent introduces an intermediary processing layer that transforms raw sensor data into meaningful operating state information through the first model, then further processes this information through the second model before generating final insights. This intermediary structure acts as a bridge between complex raw data and actionable conclusions, enhancing measurement precision at each transformation stage.
2Reliability
If multiple sensor data measurements are acquired and analyzed, then reliability of asset state determination is improved, but loss of time in data processing increases
Solution Approach 1:
The patent implements preliminary action by pre-processing sensor data and establishing the first model's operating state determination logic before actual asset monitoring begins. The system pre-defines the relationships between sensor measurements and operating states, allowing rapid real-time analysis without extensive computational processing during critical monitoring periods, thus reducing time loss while maintaining high reliability.
3Manufacturing precision
If domain-specific information is integrated into training data, then manufacturing precision of operating state model is improved, but loss of information from multiple data sources increases
Solution Approach 1:
The patent applies local quality by integrating domain-specific information selectively into the training data for the first model, focusing on the most relevant features and parameters for operating state determination. Rather than uniformly processing all available data, the system identifies and prioritizes critical domain-specific characteristics, improving model manufacturing precision while managing information loss through targeted data selection.
Data Source
AI summary
Embodiments analyze an operating state of a physical asset. An embodiment first acquires, based on one or more predetermined criteria, data measurements from one or more preselected sensors configured to sense one or more respective aspects of the physical asset. The data measurements correspond to one or more time periods, the one or more preselected sensors are preselected by correlating data measurements from a plurality of sensors of the physical asset to one or more operating states of the physical asset, and the one or more predetermined criteria are predetermined by identifying one or more data output patterns of the one or more preselected sensors. Then, via a first model, one or more operating states of the physical asset are determined based on the acquired data measurements.