Time-Series Transition Feature Extraction with Rate-of-Change Envelopes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying transition events in time-series signals are ineffective for non-stationary data, leading to inaccurate detection of transition start points, particularly in applications involving sensing devices exposed to unknown fluids, which can degrade due to bio-fouling.
Innovation Solution
Transform the time-series signal to a rate of change signal, calculate an envelope based on a stationary portion, and identify the transition start point where the transformed signal crosses the envelope.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing deviation point detection approaches (supervised or unsupervised) are used on non-stationary time-series data, then the method can be applied to real-world sensing data, but the detection accuracy deteriorates due to non-stationarity
Solution Approach 1:
The patent transforms the time-series signal from the original domain to a rate-of-change domain by computing the derivative. This parameter transformation converts non-stationary characteristics into stationary characteristics, making the data suitable for standard deviation point detection algorithms. The envelope is then calculated from the transformed signal, and transition points are identified where the signal crosses the envelope, providing accurate detection in non-stationary conditions.
2Reliability
If transition start points are identified accurately to reduce exposure time to unknown fluids, then device reliability improves, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary transformation of the signal into the rate-of-change domain and calculates the envelope before actual transition detection. This preparation step simplifies the subsequent detection process, allowing for rapid identification of transition points without requiring complex real-time computations during the actual transition event.
Solution Approach 2:
The patent replaces complex supervised machine learning models with a simpler mathematical approach using signal transformation and envelope detection. This substitution reduces computational complexity and processing time while maintaining high detection accuracy, making the system more efficient for real-time applications.
3Productivity
If the sensing apparatus continues to take measurements from unknown fluids, then data collection is maintained, but the sensing apparatus degrades due to bio-fouling
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors the time-series signal for transition events. When a transition start point is detected, the system can trigger alerts or automated responses to reduce or stop exposure to unknown fluids, preventing bio-fouling degradation while maintaining productivity through continuous monitoring.
Data Source
AI summary
A first signal is obtained. the first signal comprising a transition event having a transition start point. A first transformed signal is determined such that the first transformed signal is indicative of a rate of change of the first signal. A first envelope is calculated based on a substantially stationary portion of the first transformed signal. The transition start point is identified within the first signal based on a point in time where the first transformed signal crosses the first envelope. A command can be issued to an external device based on the transition starts point.