Sas Institute Inc.

Arnulfo D De Castro

Xinmin Wu

Richa Chauhan

Tae Yoon Lee

Greg Link

Kanchan Kamat

Hemil Shah

Glenn Lampley

Harish Yadav

In one example, a system can receive an input from a user indicating a target variable to be forecasted over a future time window. The system can then determine independent variables that influence the target variable and generate a set of candidate variables, including combinations of the independent variables. The system can then execute a random forest classifier to identify a subset of candidate variables having a threshold level of influence on the target variable. The system can then construct a machine-learning model configured to receive the identified subset of candidate variables as inputs and generate a forecast of the target variable. After constructing the machine-learning model, the system can train the machine-learning model using historical data and then execute the machine-learning model to generate the forecast.

Deep learning model for energy forecasting

Techniques to determine settings for an electrical distribution network are described. Some embodiments are particularly directed to techniques to determine settings for an electrical distribution network using power flow heuristics. In one embodiment, for example, an apparatus may comprise a model reception component, a forecast component, and an optimization component. The model reception component may be operative to receive a model of an electrical distribution network having multiple capacitor banks and multiple voltage regulators, each of the multiple capacitor banks represented in the model by a model capacitor bank, each of the multiple voltage regulators represented in the model by a model voltage regulator, the electrical distribution network having a radial layout in which power flows from a source to multiple nodes in which each node is associated with one voltage regulator. The forecast reception component may be operative to receive a forecast for demand on the electrical distribution network. The optimization component may be operative to receive the model capacitor banks and model voltage regulators and determine one or more settings for the multiple capacitor banks and multiple voltage regulators that allow for providing power within predetermined limits while reducing power loss as compared to a power loss of the existing settings or reducing power usage as compared to a power usage of the existing settings, the one or more settings for the multiple voltage regulators determined according to a heuristic in which potential settings are iteratively determined for each of the model voltage regulators based on a least squares model of load flow analysis. Other embodiments are described and claimed.

Techniques to determine settings for an electrical distribution network

Growing community of inventors

Durham, NC, United States of America

Arnulfo D De Castro