Latent Neural ODE for Integrating Multi-timescale measurements in Smart Distribution Grids

TL;DR

This paper introduces a latent neural ODE method to effectively aggregate and analyze irregularly sampled multivariate measurements in smart grids, improving real-time monitoring and control capabilities.

Contribution

The paper presents a novel latent neural ODE framework specifically designed for handling multi-timescale, irregularly sampled sensor data in smart distribution grids.

Findings

Efficiently imputes missing measurements in smart grid data.

Accurately predicts system states from irregularly sampled data.

Demonstrates superior performance on IEEE 37 bus test system.

Abstract

Under a smart grid paradigm, there has been an increase in sensor installations to enhance situational awareness. The measurements from these sensors can be leveraged for real-time monitoring, control, and protection. However, these measurements are typically irregularly sampled. These measurements may also be intermittent due to communication bandwidth limitations. To tackle this problem, this paper proposes a novel latent neural ordinary differential equations (LODE) approach to aggregate the unevenly sampled multivariate time-series measurements. The proposed approach is flexible in performing both imputations and predictions while being computationally efficient. Simulation results on IEEE 37 bus test systems illustrate the efficiency of the proposed approach.