Time-Series Sensor Synchronization for Physics-Based Alert Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring and analyzing physical systems face challenges in combining structural and sensor data effectively, leading to complex and time-consuming customization of alerting software due to issues like missing values and varying sensor reporting rates.
Innovation Solution
A method that involves storing a process graph representing the physical system, physics rules, and sensor time series data, synchronizing and interpolating sensor data streams using physics models to produce metric data streams, and applying machine learning models for alert generation, while handling missing values and varying sampling rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If custom software is used to combine structure and sensor data for alerting, then meaningful alerts can be generated, but the software becomes complicated and time-consuming to modify and deploy
Solution Approach 1:
The system segments data processing into distinct modules: graph database for structural relationships, time-series database for sensor data, and separate processing pipelines for each data type. This modular segmentation allows independent modification of each component without affecting the entire system, reducing software complexity while maintaining alert generation capability.
Solution Approach 2:
The patent introduces an intermediary processing layer that automatically combines data from the graph database and time-series database using predefined relationships and physics models. This intermediary layer handles the complexity of data integration internally, allowing users to generate alerts without directly managing the complex combination logic.
2Reliability
If sensor data is processed with missing values and different reporting rates, then comprehensive monitoring is achieved, but processing becomes complicated
Solution Approach 1:
The system changes the parameter of data representation by normalizing sensor readings to a common time basis using interpolation techniques. By transforming variable sampling rates into a standardized format, the system achieves comprehensive monitoring of all sensors while simplifying the processing logic through parameter uniformity.
Solution Approach 2:
The patent applies preliminary data preprocessing steps including handling missing values through interpolation and resampling sensor data to uniform intervals before main processing. This preliminary action prepares the data in advance, eliminating the need for complex handling during alert generation and reducing overall processing complexity.
3Stability of the object's composition
If data synchronization and interpolation are performed, then equal sampling periods are achieved, but additional processing steps are required
Solution Approach 1:
Data synchronization and interpolation are performed as preliminary steps during data ingestion and storage, rather than during query-time processing. By pre-processing the data to achieve equal sampling periods, the system eliminates time-consuming operations from the critical alert generation path, reducing overall processing time while maintaining sampling consistency.
Data Source
AI summary
A method for analyzing time series sensor data of a physical system represented by a process graph retrieves sensor data streams from stored sensor time series data. Each of the sensor data streams comprises a sequence of time-value pairs and is associated with a sensor identifier, a time offset, and a sampling period. A metric data stream is produced from the retrieved sensor data streams in accordance with a stored physics model of the physical system. Producing the metric data stream includes i) synchronizing the sensor data streams by adjusting time offsets of the sensor data streams and adding interpolated values and times to the sensor data streams to produce synchronized streams with equal sampling periods; and ii) performing a point-wise computation over values of the sensor data streams in accordance with the physics model.