Plant Inspection Trend Graphs Using Event-Based Measurement Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional trend graphs with a horizontal axis representing measurement time suffer from reduced visibility due to large amounts of data during plant shutdown periods and blank periods during operation, making it difficult to detect equipment failures.
Innovation Solution
A data collection device and method that assigns the same event number to measured values within a specified time difference and stores them accordingly, allowing for a trend graph display with event numbers as the horizontal axis, enhancing visibility and facilitating failure detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If measured values are displayed in a trend graph with measurement time on the horizontal axis, then the complete time series data is preserved, but the visibility is lowered due to large amounts of data during plant shutdown periods
Solution Approach 1:
The patent segments the time series data by dividing it into multiple periods based on plant operation status (operation period and shutdown period). Each period is assigned a unique event number, allowing the graph to display multiple segments along the horizontal axis instead of continuous time. This segmentation reduces visual clutter while preserving all measurement data points within their respective periods.
Solution Approach 2:
The patent introduces a new dimension (event number) to the traditional time-series graph. Instead of plotting only time vs. measured value, the graph now plots event number vs. measured value, where event numbers represent discrete operational periods. This dimensional transformation allows compact representation of data from different time periods while maintaining data completeness.
2Stability of the object's composition
If the horizontal axis is set to measurement time, then chronological order is maintained, but blank periods during plant operation reduce visibility of actual measurement data
Solution Approach 1:
The patent extracts and removes the blank operation periods from the graph display. By using event numbers that represent only periods with measurement data (shutdown periods for insulation resistance measurements), the graph excludes unnecessary blank time intervals. This extraction maintains the chronological sequence of measurements while eliminating visual waste space.
3Quantity of substance
If all measured values are displayed continuously, then data completeness is maintained, but the graph becomes difficult to read during long operation periods
Solution Approach 1:
The patent segments the continuous data stream into discrete periods with unique event numbers. Each event number represents a specific shutdown period where measurements were taken. This segmentation organizes large quantities of data into manageable, readable groups while maintaining complete data representation. Users can easily distinguish between different measurement periods without being overwhelmed by continuous data.
Solution Approach 2:
By transforming the single-dimensional time axis into a two-dimensional structure (event number vs. measured value), the patent improves graph readability. The event number dimension provides categorical organization that makes it easier to interpret data from different periods, while the complete measurement data is preserved within each event category.
Data Source
AI summary
A plant monitoring system comprises: a data collection device configured to collect multiple measured values measured during a plant equipment inspection; and a trend graph display part configured to generate a trend graph based on the multiple measured values. The data collection device comprises: a measured value collection part configured to collect multiple measured values; an event number assignment part configured to assign a same event number or a different event number to the multiple measured values based on a time difference between times at which the multiple measured values were acquired; and a measured value storage part configured to store the multiple measured values. The trend graph display part is configured to render the multiple measured values stored in the measured value storage part in a graph having measured value as a first axis and event number as a second axis.