Uncertainty in renewable energy generation, energy consumption, and electricity prices, as well as transmission congestion, pose a number of problems in modern power grids, necessitating stability on the supply, grid, and demand sides. Grid-side stability can be achieved by dynamic line rating (DLR) forecasting, which reliably predicts the overall current carrying potential of overhead transmission lines. Long short-term memory proved beneficiary in this field, owing to its ability to learn highly variable and uncertain data. To empower this network to tackle the non-stationary nature of meteorological parameters, a novel machine learning (ML) architecture based on Dagging technique is proposed and tested on the data collected from a 400 kV overhead transmission line. Simulation results corroborate that the proposed Dagging-based stacked LSTM can successfully handle the non-stationary issue and outperform the decomposition-based technique, as the state-ofthe-art algorithm, for various forecasting horizons. The results confirm the generalizability of models with an application in forecasting DLR over the line without utilizing additional sensors and communication networks. Moreover, the proposed model is compared to several ML architectures, including support vector machines (SVM), random forest (RF), and multi-layer perceptron (MLP) in a comprehensive benchmark study. The introduced algorithm outperforms MLP by 3.4%, RF by 9.4%, and SVM by 6.7% in terms of average prediction accuracy.

No abstract

No abstract

Established methods to track the dynamics of neural representations focus at the level of individual neurons for spiking data, and individual or pair of channels for local field potentials. However, our understanding of neural function and computation has moved toward an integrative view, based upon coordinated activity of multiple neural populations across brain areas. To draw network-level inferences of brain function, we propose a new modeling framework that combines the state-space model and cross-spectral matrix estimates – this is called state-space coherence (SSCoh). We define elements of the SSCoh and derive system identification and approximate filter solution for multivariate space processes. We expand SCoh for mixed observation processes, where the observation includes different modalities of neural data including local filed potential and spiking activity. Finally, we show an application of the framework to study neural synchrony across different brain nodes of a task participant performing Stroop task under different distraction levels.

No abstract