Time Series Pattern Modeling for Multi-Sensor Condition Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge lies in effectively analyzing and predicting massive amounts of multi-dimensional time series data from large numbers of sensors, which becomes increasingly difficult due to high dimensionality and the manual approach becoming less feasible as the number of sensors increases.
Innovation Solution
A computer-implemented method that involves obtaining multi-dimensional time series data, creating matrices based on this data, determining patterns using a first numerical modeling method, and creating a single time series model using a second method to predict future conditions, accounting for sensor locations and data attenuation over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of sensors collecting data is increased, then the monitoring coverage and data completeness are improved, but the complexity of manual analysis and the difficulty of processing the data increases
Solution Approach 1:
The patent introduces an intermediary system comprising a computing device that automatically processes sensor data through matrix creation, pattern determination using numerical modeling methods, and time series model generation. This intermediary automates the analysis process, resolving the contradiction by maintaining comprehensive monitoring coverage while eliminating manual analysis complexity.
Solution Approach 2:
The patent replaces manual mechanical analysis processes with automated computational methods including matrix operations, numerical modeling algorithms, and time series prediction models. This substitution transforms the data analysis from a manual mechanical process to an automated computational system, reducing complexity while maintaining reliability.
2Loss of information
If multi-dimensional time series data from multiple sensors is collected, then the comprehensiveness of system monitoring is improved, but the difficulty of analyzing the data in one period increases
Solution Approach 1:
The patent segments multi-dimensional time series data into structured matrix formats where each matrix represents a specific time period with organized sensor readings. This segmentation approach maintains comprehensive data coverage while making the data more manageable and analyzable through systematic matrix operations and pattern recognition.
Solution Approach 2:
The patent transforms raw sensor data into different parameter representations through matrix creation and numerical modeling operations. By changing the parameters from raw multi-dimensional data to structured matrices and then to pattern-based time series models, the system maintains data comprehensiveness while reducing analysis difficulty.
3Ease of operation
If manual approach to performance monitoring is used, then the interpretability of data analysis is maintained, but the feasibility becomes less viable as the number of sensors increases
Solution Approach 1:
The patent implements feedback mechanisms where the automated system generates predictions about future system conditions and can trigger alerts or actions when certain thresholds are exceeded. This feedback loop maintains operational ease by providing actionable insights while improving productivity through automated processing of large sensor datasets.
Solution Approach 2:
The patent enables the system to perform self-service monitoring and prediction by automatically processing sensor data, generating time series models, and producing predictions without requiring continuous manual intervention. This self-service capability maintains data interpretability through automated pattern recognition while significantly improving monitoring feasibility for large sensor networks.
Data Source
AI summary
A computer-implemented method includes: obtaining, by a computing device, data from sensors that collect the data in a system during a time, wherein the data is multi-dimensional time series data; creating, by the computing device, matrices based on the data; determining, by the computing device using a first computer-based numerical modeling method, patterns based on the matrices; creating, by the computing device using a second computer-based numerical modeling method, a single time series model based on the patterns; and predicting, by the computing device, a future condition of the system using the time series model with current data of the system.