Time-Based Power Analysis Using Pattern Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current time-based power consumption analysis in electrical systems is constrained to short time windows due to high computational resource requirements, limiting the ability to simulate realistic extended usage patterns, which affects the accuracy of power consumption profiles and other analyses like transient thermal analysis and side channel attack analysis.
Innovation Solution
A computer-implemented method that generates an estimation of power consumption over time, identifies state signatures and patterns, computes pattern output responses, and constructs a weighted sum of these responses to improve accuracy and efficiency in power consumption analysis, using techniques like low pass filtering and quantization to enhance pattern recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If traditional time-based power consumption simulations are performed over long time windows, then realistic extended usage patterns can be encompassed, but computational resources are excessively consumed
Solution Approach 1:
The patent segments the power consumption analysis into distinct operational modes or states, where each mode represents a specific usage pattern. By dividing the extended time window into multiple mode segments, the system can analyze each segment separately using traditional simulation methods, then combine results to achieve long-time-window analysis without proportionally increasing computational resource consumption.
Solution Approach 2:
The patent transforms the power consumption representation by changing parameters from continuous time-based data to discrete mode-based characteristics. By extracting key parameters that define each operational mode and using these to represent power consumption patterns, the system reduces computational complexity while maintaining accuracy for extended time windows.
2Measurement precision
If traditional time-based power consumption simulations are performed, then power consumption profiles can be computed, but the time windows are too short to encompass realistic extended usage patterns
Solution Approach 1:
The patent performs preliminary identification of operational modes and their characteristics before conducting detailed power consumption analysis. By pre-characterizing the different usage patterns and their temporal relationships, the system can then apply traditional simulation methods to each mode with appropriate time durations, ultimately constructing accurate extended-time power consumption profiles that encompass realistic usage patterns.
Solution Approach 2:
The patent creates simplified representations or copies of operational modes that capture essential power consumption characteristics. These mode models serve as surrogates that can be instantiated and combined multiple times to represent extended usage patterns without requiring equally extended simulation times, thereby achieving both accuracy and extended time window coverage.
Data Source
AI summary
Systems, machine readable media and methods are described for analyzing one or more physical systems using techniques that recognize patterns in underlying data and use the patterns to efficiently compute outputs using the patterns to reduce computations. The physical systems can be simulated with an estimation (e.g., an estimated power versus time waveform) that can be efficiently computed and then the estimation can be analyzed to detect patterns in the data. The detected patterns can each be analyzed with, in one embodiment, higher accuracy than the estimation to provide data that can be combined across multiple instances of each pattern to provide a higher accuracy evaluation of the system with a lower computational overhead.