Plant Event Analysis Using Process Trend Change Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional event analyzing devices in plants cannot analyze the cause-effect relationship between operator operations and process changes without a direct causal link to alarms, leading to incomplete analysis and inability to automate or rationalize operations effectively.
Innovation Solution
An event analyzing device that integrates event data and process change data to analyze cause-effect relationships, including operations with no direct causal link to alarms, by converting process data trends into event data and applying a time concept to identify relationships between events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional event analyzing devices only analyze events with direct causal links to alarms, then the analysis is simpler and more focused, but the analysis coverage is incomplete and cannot capture operations without direct alarm connections
Solution Approach 1:
The patent segments the analysis process into distinct modules: event data collection, process data collection, trend change detection, event data conversion, and cause-effect relationship analysis. This segmentation allows the system to handle complex analysis tasks in manageable steps, enabling comprehensive analysis of all events including those without direct alarm links while maintaining organizational clarity and computational efficiency.
Solution Approach 2:
The patent introduces process data and trend change information as intermediary elements between events and alarms. These intermediaries enable the system to establish indirect causal relationships by detecting trend changes in process data and converting them into event data, thereby connecting operations that do not directly trigger alarms to their underlying causes through the intermediary process state analysis.
2Loss of information
If the system integrates process data and event data for comprehensive cause-effect analysis, then the analysis completeness improves, but the data processing complexity increases
Solution Approach 1:
The patent merges process data and event data into a unified analysis framework. By integrating these two data types and applying consistent processing methods (trend detection, event conversion, cause-effect analysis), the system achieves comprehensive information coverage while managing complexity through unified processing logic rather than separate handling of each data type.
Solution Approach 2:
The patent transforms process data parameters into event data parameters through trend change detection. By monitoring changes in process parameters (such as deviations from normal ranges) and converting these changes into standardized event data format, the system enables comprehensive analysis without requiring completely separate processing pipelines for different data types.
3Reliability
If trend change detection is applied to process data, then operations without direct alarm links can be identified, but the detection complexity and computational load increase
Solution Approach 1:
The patent applies trend change detection selectively rather than continuously to all process data. By focusing detection on relevant process parameters and applying it only when necessary for identifying operational changes, the system achieves reliable operation identification while avoiding the computational burden of exhaustive continuous monitoring of all process variables.
Data Source
AI summary
An event analyzing device includes an event data collector configured to collect event data which represents an event including an alarm which has occurred in a plurality of devices in a plant and an operation performed to the devices, a process data collector configured to collect process data of the devices in the plant, a trend change detector configured to detect a trend change of the process data collected by the process data collector, an event data converter configured to convert the trend change detected by the trend change detector into a process change event represented in the same format as the event, and a cause-effect relationship analyzer configured to integrate the event data collected by the event data collector and process change event data which represents the trend change in the process change event converted by the event data converter to analyze a cause-effect relationship between the event and the process change event.