Dynamic Network Island Event Assignment for Component Fault Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predictive maintenance applications for complex networks, such as electrical grids, struggle to accurately identify the specific components causing interruptions, leading to inefficient response and control, especially during dynamic changes like islanding and remerging events.
Innovation Solution
A machine learning latent variable model is employed to disaggregate network events, allowing for the precise assignment of interruptions to individual components and predicting future behavior by using probabilistic modeling and iterative probability updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If simple logic is used for predictive maintenance in complex networks, then the system is easier to operate, but the measurement precision of event localization deteriorates
Solution Approach 1:
The patent segments the complex network into multiple islands based on topological connectivity. Each island is independently analyzed for events, allowing precise localization within specific islands while keeping the overall system manageable. The disaggregation model further segments events into component-level contributions, enabling accurate event attribution without requiring complex global analysis.
Solution Approach 2:
The patent introduces an intermediary disaggregation model that acts as a bridge between observed island-level events and unobserved component-level events. This intermediary model infers component-level contributions from island-level observations, enabling precise event localization without direct component monitoring, thus maintaining ease of operation while improving measurement precision.
2Measurement precision
If component-level event detection is implemented in dynamically changing networks, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The patent implements dynamic island detection that automatically adapts to network topology changes. Islands are reidentified whenever network connections change, allowing the system to maintain accurate component-level event detection precision without requiring manual reconfiguration. The disaggregation model dynamically updates its inference based on current island configurations, handling network dynamics without increasing operational complexity.
Solution Approach 2:
The patent employs feedback mechanisms where the disaggregation model continuously refines its event assignments based on observed patterns. The model learns from historical event data and improves its component-level event detection accuracy over time. This feedback-driven approach enables high measurement precision while keeping the system structure relatively simple, as the complexity is managed through adaptive learning rather than rigid complex architecture.
3Reliability
If probabilistic modeling is used for event assignment, then the reliability of predictions improves, but the loss of computational time increases
Solution Approach 1:
The patent applies probabilistic modeling selectively at the island level rather than exhaustively at the component level for all events. The disaggregation model uses probability distributions to assign events to components based on island-level observations, providing reliable predictions without performing full probabilistic analysis on every component. This partial application of probabilistic methods maintains prediction reliability while reducing computational time compared to exhaustive component-level analysis.
Data Source
AI summary
The present disclosure provides systems and methods that may advantageously apply machine learning to detect and ascribe network interruptions to specific components or nodes within the network. In an aspect, the present disclosure provides a computer-implemented method comprising: mapping a network comprising a plurality of islands that are capable of dynamically changing by splitting and/or merging of one or more islands, wherein the plurality of islands comprises a plurality of individual components; and detecting and localizing one or more local events at an individual component level as well as at an island level using a disaggregation model.