Distributed Power Monitoring for Adaptive Anomaly Fault Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power network anomaly detection methods face challenges in accurately identifying and isolating faults in hierarchical structures, leading to increased energy bills, reduced productivity, and safety incidents due to low detection rates, high false alarms, and limited fault isolation capabilities.
Innovation Solution
A method and device for adaptive anomaly detection in power networks that determine baseline power usage, detect anomalies by comparing active usage to baseline data, isolate faults through hierarchical analysis, and provide feedback for modifying baseline usage, utilizing predictive machine learning and historical data to identify anomalies and isolate faults at lower hierarchical levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional power network anomaly detection methods are used, then the system structure remains simple, but the anomaly detection accuracy is low and false alarms are high
Solution Approach 1:
The power network is divided into hierarchical levels (facility level, building level, section level, machine level, component level), with anomaly detection performed at each level independently. This segmentation enables precise localization of anomalies to specific hierarchical levels while maintaining manageable system complexity through modular detection units at each level.
Solution Approach 2:
The patent introduces a hierarchical dimension to the anomaly detection system, transforming a flat detection approach into a multi-level hierarchical structure. This adds a new dimension (hierarchical level) to the detection process, enabling more accurate anomaly identification by analyzing power usage patterns at multiple granularities simultaneously.
2Reliability
If comprehensive monitoring of all power network elements is implemented, then fault isolation capability improves, but the device complexity and computational requirements increase
Solution Approach 1:
The monitoring system is segmented into distributed detection units at each hierarchical level, each responsible for monitoring specific elements at that level. This segmentation enables comprehensive fault isolation capability by localizing monitoring functions to appropriate hierarchical levels, reducing the complexity any single monitoring component must handle while maintaining overall system reliability.
Solution Approach 2:
Each hierarchical level implements monitoring with quality and detail appropriate to its level - facility level monitors aggregate patterns, while component level monitors detailed electrical parameters. This local quality approach enables comprehensive fault isolation by matching monitoring granularity to the scale of potential faults at each level, avoiding unnecessary complexity in the overall system.
3Productivity
If real-time anomaly detection and fault isolation are performed, then productivity is improved through quick response, but the energy consumption and computational resources increase
Solution Approach 1:
Computational workload is segmented across multiple hierarchical levels, with each level performing detection and analysis appropriate to its scale. This segmentation enables real-time response by distributing computational tasks - simple patterns are detected at lower levels quickly, while more complex analysis occurs at higher levels, optimizing the balance between response speed and energy consumption.
Solution Approach 2:
Baseline power usage patterns are pre-computed and stored for each hierarchical level, enabling rapid real-time comparison without extensive computational analysis during actual detection. This preliminary action of pre-establishing expected patterns allows quick anomaly identification while minimizing computational energy consumption during operational monitoring.
4Adaptability or versatility
If the baseline power usage is frequently updated to adapt to changing patterns, then the system adaptability improves, but the stability of the baseline data decreases
Solution Approach 1:
The baseline power usage data is made dynamic through automatic updates at each hierarchical level based on detected patterns and anomalies. This dynamic adaptation enables the system to respond to changing power consumption patterns while maintaining stability through structured update mechanisms that prevent erratic baseline changes, balancing adaptability with data stability.
Solution Approach 2:
A feedback mechanism is implemented where detected anomalies and actual power usage patterns feed back into baseline updates at each hierarchical level. This feedback loop enables the system to adapt to changing patterns over time while maintaining stability through controlled, data-driven updates rather than arbitrary changes, ensuring baselines reflect actual operational patterns.
Data Source
AI summary
Methods, devices and systems for detecting an anomaly in a power network are described. A method for detecting an anomaly in a power network includes determining a baseline power usage in the power network, receiving data indicative of an active power usage in the power network, detecting an anomaly based on a difference between the baseline power usage and the active power usage, isolating a fault for an element in the power network, responsive to detecting the anomaly, and transmitting fault isolation information indicating the fault to a user device. Related devices and systems may perform operations of the method described herein.