Sensor aware fusion improves state estimates in power grids

Sensor-Aware Joint Data Fusion and State Estimation under Sparse Multi-rate Distribution Grid Sensing

Machine Learning

Summary

Power grids are getting more complicated as more local devices like batteries and solar panels connect to them, making it harder for operators to keep track of everything. Traditional methods for monitoring the grid need synchronized sensors, but real-world sensors often send data at different speeds and from different places, making it tricky to combine their information. The authors created a new method that intelligently combines data from various sensors placed throughout the grid, even when the data comes in at different rates, and then estimates the overall state of the grid more accurately. They tested their approach on standard power grid models and found it works well, especially as the number of sensors and their coverage decreases.

What this means in practice

  • For power grid operators: Improve grid monitoring by better combining data from different sensor types and rates for more accurate system state estimates.
  • For energy management teams: Manage distributed energy resources more effectively using improved state estimation from sparse and varied sensor data.

Authors

Pooja Algikar, Stefano Fenu, Kumarsinh Jhala, Siby Plathottam

Abstract

Distribution-system state estimation is a growing necessity for the modern power grid as the addition of local storage and generation add system complexity that an operator or aggregator is not able to directly monitor. Classical methods work well when all system elements are time-synchronized and at least partly observable, but multi-rate monitoring of sparse heterogeneous sensors typically relies on machine learning approaches. It is a common practice in the literature to structure this task as a two-stage pipeline, first fusing data from disparate sensors and then feeding the fused pseudo-measurements to a separately trained state estimator. We propose a heterogeneous latent graph ODE in which measurement functions and sampling processes are typed and localized on specific buses, paired with a graph- transformer state estimator solving DSSE specifically for multi-rate heterogeneous sensor placements, and demonstrate its effectiveness across multiple IEEE standard topologies as observability degrades.
View PDFOpen arXiv