Power Switching Device Operating Time Prediction via Vicinity Data Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional apparatuses for predicting operating times in power switching devices face challenges in stably estimating operating characteristic parameters without increasing calculation complexity, especially under conditions where high accuracy is difficult to achieve.
Innovation Solution
An apparatus that includes an operating time measuring unit, an environmental condition measuring unit, an operation history retaining unit, an operating characteristic parameter estimating unit, a predicting unit, a feedback correcting unit, and a learning completion determining unit, which extracts vicinity historical data for functional approximation using lower-order functions to stabilize parameter estimation and improve prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional functional approximation methods are used to estimate operating characteristic parameters, then the prediction apparatus can operate with simple calculation, but the estimation stability deteriorates under challenging conditions
Solution Approach 1:
The patent segments the historical data by extracting only 'vicinity historical data' that are placed in the vicinity of the current operation environmental condition. This segmentation allows the functional approximation to focus on relevant data points, improving estimation stability without requiring complex global approximation methods. The segmentation is achieved by comparing current environmental conditions with historical conditions and selecting only those within a predetermined range.
Solution Approach 2:
The patent applies local quality by using different approximation approaches for different regions of the data space. Instead of using a single global approximation function, the system selectively applies functional approximation only to vicinity historical data that are locally relevant to current conditions. This local approach ensures higher estimation stability in challenging conditions while maintaining simple calculation for the approximation process itself.
2Measurement precision
If all historical data are used for functional approximation, then more data are available for estimation, but the amount of calculation increases
Solution Approach 1:
The patent extracts only the necessary subset of historical data ('vicinity historical data') that are relevant to current operation conditions. By taking out only the data placed in the vicinity of current environmental conditions and excluding unrelated historical data, the system maintains high estimation accuracy while significantly reducing the calculation amount required for functional approximation.
Solution Approach 2:
The patent applies partial action by using only a portion of the available historical data rather than all historical data. Specifically, it uses only the vicinity historical data that are sufficiently relevant to current conditions, which is a partial subset of the complete historical dataset. This partial usage achieves sufficient estimation accuracy without the excessive calculation burden of processing all historical data.
Data Source
AI summary
An apparatus for predicting an operating time in a power device includes an operating characteristic parameter estimating unit, and only the operation historical data in the vicinity of a current operation environmental condition is extracted and input by a vicinity operation historical data extracting unit.