Attribute-Filtered Operation State Prediction for Target Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing prediction systems for target devices like plants fail to consider factors such as device specifications and environmental attributes, which significantly influence future behavior.
Innovation Solution
A prediction system that extracts operation state history from multiple devices based on attribute information filters, using statistical analysis or machine learning to predict future operation states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If attribute information is not considered in prediction, then the prediction system is simpler, but the prediction accuracy deteriorates
Solution Approach 1:
The system performs preliminary classification of target devices based on attribute information before prediction. By pre-grouping devices with similar attributes (e.g., device type, specifications, operating conditions) into categories, the system prepares filtered datasets in advance, which improves prediction accuracy without adding complex real-time processing requirements.
Solution Approach 2:
The prediction system segments the overall device population into distinct groups based on attribute information. Each segment is handled by a dedicated prediction model trained on specific attribute characteristics, allowing the system to maintain high accuracy for each segment while keeping individual model complexities manageable.
2Measurement precision
If attribute information filtering is applied, then prediction accuracy improves, but data processing time increases
Solution Approach 1:
Attribute-based filtering and device classification are performed in advance to pre-organize historical data into attribute-specific datasets. This preliminary action eliminates the need for time-consuming filtering operations during real-time prediction, as the system can directly access pre-filtered data corresponding to the target device's attributes.
Solution Approach 2:
The system applies different filtering and processing strategies tailored to each attribute category. Rather than uniformly processing all data, the system selectively applies attribute-specific filters and models only to relevant data subsets, reducing overall processing time while maintaining accuracy for each specific case.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A prediction system includes a storage unit that stores a history of an operation state of each of a plurality of target devices and attribute information indicating an attribute of each of the plurality of target devices; a first acquisition unit that acquires an attribute information filter condition in which at least one attribute information included in attribute information of a prediction target device is specified; a second acquisition unit that acquires an operation state filter condition in which at least one operation state included in a history of an operation state of the prediction target device is specified; an extraction unit that extracts a history of an operation state of a target device satisfying the attribute information filter condition and the operation state filter condition of the plurality of target devices, with reference to the storage unit; and an estimating unit that predicts the operation state of the prediction target device based on the extracted history of the operation state.