Power Grid Cognitive Disambiguation for RTU Data Overload
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-throughput, low-latency sensor data from remote terminal units (RTUs) such as phasor measurement units (PMUs) overwhelm grid operators and engineers with considerable amounts of irrelevant data, and existing analysis approaches fail to remedy such inefficiencies.
Innovation Solution
Implementing cognitive disambiguation techniques that incorporate relevance feedback and automated inferencing to curate remote terminal unit measurements by capturing user interactions, inferring rules through machine learning, and updating a knowledge base with user-validated rules to resolve ambiguities and filter irrelevant data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-throughput sensor data from RTUs is collected for grid monitoring, then measurement precision and data availability are improved, but information overload and difficulty in identifying relevant data worsen
Solution Approach 1:
The system implements feedback loops where user interactions with the visual display (highlighting, selecting, or ignoring specific measurements) are captured and used to refine the inference rules. This feedback mechanism allows the system to learn from user behavior patterns and progressively improve its ability to filter relevant from irrelevant data, resolving the contradiction between providing comprehensive data and eliminating information overload
Solution Approach 2:
The system performs self-service by automatically inferring and updating its own inference rules based on captured user feedback. The machine learning component enables the system to autonomously improve its data curation capabilities without requiring manual rule configuration, allowing it to adapt to specific user needs and grid conditions while maintaining high measurement precision
2Quantity of substance
If comprehensive RTU measurements are displayed for analysis, then data completeness is improved, but operator workload and analysis time increase
Solution Approach 1:
The system extracts and displays only the most relevant measurements by applying inferred rules to filter the comprehensive RTU data stream. The visual display selectively highlights measurements that are likely to be relevant to the current grid event based on the inferred rules, allowing operators to quickly identify critical information without being overwhelmed by complete data sets
Solution Approach 2:
The system performs preliminary filtering and ranking of measurements before they reach the operator. By pre-processing the comprehensive RTU data through the inference engine and organizing it by relevance in the visual display, the system reduces the operator's analysis time while maintaining data completeness in the background
3Productivity
If automated rule inference is implemented through machine learning, then productivity and data curation efficiency are improved, but system complexity increases
Solution Approach 1:
The system introduces an intermediary inference engine that sits between the raw RTU measurements and the operator interface. This intermediary component handles the complex machine learning operations and rule inference, translating complex data patterns into simplified visual representations and relevance rankings that the operator can easily interpret, thus managing system complexity while maintaining high productivity
4Adaptability or versatility
If user feedback is captured and used to update inference rules, then adaptability and relevance accuracy are improved, but processing overhead and computational requirements increase
Solution Approach 1:
The system applies partial action by updating inference rules selectively based on the significance and frequency of user feedback. Not every user interaction triggers a full rule retraining process; instead, the system accumulates feedback and performs incremental updates, reducing computational overhead while maintaining adaptability to changing grid conditions and user preferences
Data Source
AI summary
Methods, systems, and computer program products for cognitive disambiguation of problem-solving tasks involving a power grid are provided herein. A computer-implemented method includes capturing user feedback pertaining to relevance of remote terminal unit measurements related to a grid event through user interface interactions carried out by the user, wherein the user interface is communicatively linked to at least one computing device; automatically inferring rules related to the grid event to curate remote terminal unit measurements across iterations of analysis by recognizing irrelevant data and/or distractions in a visual display associated with the user interface, wherein said automatically inferring comprises implementing machine learning via the at least one computing device based on the user feedback; and outputting candidate solutions to a problem-solving task involving the grid based on the inferred rules, wherein said outputting is carried out by the at least one computing device.