Monitoring Control Using Relevance Models for Error Cause Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing monitoring systems for plants, such as water treatment plants, are limited in identifying causal relationships between monitored items, leading to inaccurate error detection and maintenance.
Innovation Solution
A monitoring control apparatus that defines relevance between monitored items, manages relevant models, calculates causal relationships using monitoring data, and identifies error causes based on chronological changes in data, incorporating various types of relevance models including equipment, procedural, and correlation relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If relevance is defined only by upstream and downstream positional relationships of piping, then the causal model can be built using simple spatial relationships, but the causal relationship identification is limited and inaccurate
Solution Approach 1:
The patent defines multiple types of relevance relationships (spatial, functional, operational) that can be applied across different monitored items. The relevance definition unit creates a universal relevance model that works for various equipment and parameters, not limited to just piping spatial relationships. This multi-functional approach expands causal relationship coverage while maintaining identification accuracy through chronological data analysis.
2Reliability
If multiple types of relevant models are managed, then more comprehensive causal relationships can be identified, but the system complexity increases
Solution Approach 1:
The patent segments the relevance definition into distinct types (spatial relevance, functional relevance, operational relevance) and manages them separately through the relevance definition unit. Each relevance type is calculated and stored independently, allowing the system to handle multiple relevance models without overwhelming complexity. The segmentation enables selective application of different relevance types based on the specific monitoring scenario.
Solution Approach 2:
The relevance definition unit acts as an intermediary between the monitored items and the causal relationship calculator. It pre-processes and structures the relevance relationships before they are used for causal analysis, simplifying the overall system architecture. This intermediary layer manages the complexity of multiple relevance models by providing a standardized interface for storing and retrieving relevance information.
3Measurement precision
If chronological change analysis is used to calculate causal relationships, then accurate error causes can be identified, but more monitoring data processing is required
Solution Approach 1:
The system performs preliminary actions by pre-defining relevance relationships between monitored items before errors occur. The relevance definition unit establishes the causal framework in advance, so when an error is detected, the causal relationship calculator only needs to apply the pre-defined relevance models to chronological data, rather than building the entire causal structure from scratch. This reduces real-time processing time while maintaining high detection accuracy.
Data Source
AI summary
A monitoring control apparatus that can increase the accuracy of maintaining a monitored target includes: a relevance defining unit to define relevance between monitored items in a plurality of monitored items; a relevant model management DB to manage relevant models, using relevant monitored items of the monitored items; a monitoring data. management DB to manage monitoring data on the monitored items, a causal relationship calculator to calculate, when an error occurrence is detected from at least one of the monitored items, with which monitored item of the relevant models the error occurrence has a causal relationship, based on the relevant models managed by the relevant model management DB and the monitoring data managed by the monitoring data management database; and a cause identifier to identify a cause of the error occurrence, based on the monitored item with the causal relationship.