PMU Waveform Similarity Detection for Grid Anomaly Response
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power grid monitoring systems using phasor measurement units (PMUs) face challenges in processing massive data volumes, leading to delayed recognition of critical grid information and abnormal behavior due to high CPU and I/O resource requirements, and inability to detect unforeseen events caused by natural phenomena.
Innovation Solution
A system and method that quickly identifies events in PMU data by correlating incoming data with historical data without relying on pre-defined feature values or indices, using a processor to capture and compare measurement data windows, and recommending actions based on waveform correlation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If PMUs are used to monitor the power grid with high resolution, then real-time feedback regarding power system disturbances is improved, but data volume increases 100 to 1000 times larger than SCADA systems
Solution Approach 1:
The patent extracts only the most critical grid information and abnormal behaviors from the massive PMU data streams, filtering out redundant data. This allows the system to maintain high measurement precision while managing data volume by focusing on essential information only.
Solution Approach 2:
The system performs preliminary processing and analysis of PMU data before it becomes overwhelming, identifying and flagging critical events in advance. This preliminary action reduces the burden of processing massive data volumes by pre-identifying important patterns and anomalies.
2Reliability
If feature values are calculated from large databases to enable event detection, then event detection capability is improved, but CPU and I/O resources are significantly consumed
Solution Approach 1:
Instead of calculating feature values for all data points in the database, the system applies partial action by focusing computational resources only on suspected abnormal events or critical time periods. This reduces CPU and I/O consumption while maintaining reliable event detection for important incidents.
Solution Approach 2:
The system dynamically adjusts processing parameters such as analysis depth, data sampling rates, and feature extraction intensity based on system conditions. During normal operation, minimal processing is applied; during suspected events, processing intensity increases, optimizing resource usage while maintaining detection reliability.
3Measurement precision
If computational time is increased to calculate feature values accurately, then detection accuracy is improved, but response time exceeds acceptable thresholds of tens of milliseconds or one second
Solution Approach 1:
The system uses periodic action by implementing continuous monitoring at a lower computational intensity, with intensified analysis triggered periodically or event-driven. This allows the system to maintain detection accuracy for critical events while keeping average response time within acceptable thresholds through rhythmic, controlled processing cycles.
Solution Approach 2:
For critical time-sensitive events, the system skips detailed feature value calculations and rushes through with simplified detection algorithms that provide sufficiently accurate results faster. This selective skipping maintains detection accuracy for urgent cases while meeting strict response time requirements.
4Speed
If pre-defined event information indices are used to search historical events, then search speed is improved, but ability to detect unforeseen events is lost
Solution Approach 1:
The system dynamically adapts its search strategy by combining pre-defined event indices for fast searches with flexible, unsupervised anomaly detection methods. When predefined patterns are found, fast indexed search is used; when unusual patterns emerge that don't match predefined events, the system dynamically switches to more versatile detection approaches, maintaining both speed and adaptability.
Data Source
AI summary
Example implementations described herein are directed to detection of historical anomalous events that are similar to currently occurring events in a transmission power system based on phasor management unit (PMU) data to provide information to grid operators with online decision support. From the high-resolution time synchronized PMU data, the historical events can be quickly retrieved and compared to the currently occurring event so that operators can be provided with remedy actions that were attempted in response to the historical events. Utilization of PMU information for such decision support may compliment operation practices relying on supervisory control and data acquisition (SCADA) measurements by allowing a much fast response to the currently occurring event. Accurate identification of similar, historical events can advise grid operators of the cause of disturbances and provide ideas for response. Implementations of the proposed technology may improve the resilience and reliability of the transmission power systems.