Smart Trending for Industrial Control Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial control systems face challenges in analyzing and providing timely trend information from industrial machines, as existing methods require manual analysis of extensive data sets, hindering efficient decision-making.
Innovation Solution
A system and method that acquire data from sensors associated with industrial machines, log it as a function of time, and parse it to identify subsets based on predefined conditions, allowing for real-time analysis and forecasting without human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis of extensive time series data is performed to identify specific events, then complete data coverage is achieved, but time consumption and analysis efficiency deteriorate
Solution Approach 1:
The system performs preliminary action by logging not only the raw time series data but also pre-computed trend information and event markers at the time of data acquisition. This preliminary processing enables rapid querying and analysis later without requiring manual examination of the entire data set, thus reducing analysis time while maintaining data completeness.
Solution Approach 2:
The patent introduces an intermediary layer of trend information and event markers that mediate between the raw data and the user. Instead of directly analyzing extensive time series data, users query the pre-processed trend information which acts as an intermediary, significantly reducing the time required to identify specific events while preserving access to complete data when needed.
2Reliability
If detailed time series data from industrial machines is logged continuously, then comprehensive monitoring is achieved, but data complexity and difficulty of analysis increase
Solution Approach 1:
The patent applies segmentation by dividing the continuous time series data into meaningful segments marked by events and trends. Instead of presenting raw continuous data, the system segments the data based on significant changes and events, making the information more manageable and easier to analyze while maintaining comprehensive monitoring coverage.
Solution Approach 2:
The system extracts key trend information and event markers from the extensive time series data. By taking out only the most relevant and significant information points while maintaining access to the complete data set, the system reduces data complexity and makes analysis more manageable without sacrificing monitoring comprehensiveness.
3Measurement precision
If trend information is obtained by examining entire time series data sets, then accurate event identification is achieved, but productivity and decision-making speed deteriorate
Solution Approach 1:
The system performs preliminary action by pre-processing the time series data to identify and mark events and trends before user querying. This preliminary event identification is stored as metadata, enabling rapid retrieval of accurate event information without requiring users to examine entire data sets, thus maintaining accuracy while improving productivity and decision-making speed.
Solution Approach 2:
The patent introduces an intermediary layer of pre-identified event markers and trend information that mediates between the raw time series data and user queries. This intermediary enables accurate event identification by providing pre-processed information, significantly improving decision-making speed without sacrificing identification accuracy.
Data Source
AI summary
There are provided methods and devices for use with industrial control systems. For example, there is provided a method that can include acquiring data from a sensor associated with an industrial machine. The method can include determining, while acquiring the data, whether at least one condition is satisfied in the data. Furthermore, the method can include logging the data as a function of time and logging information associated with the at least one condition as a function of time, in response to the at least one condition being satisfied. The method can also querying the logged information, subsequent to logging the data, to identify subsets of the data for which the at least one condition is satisfied.