逗号简历

Led a research team solving major bulk US power grid problems using real, large-scale Synchrophasor data.

Created a model to generate realistic synthetic anomaly data increasing data availability by at least 200% and a model to classify anomaly signatures achieving an F1 score of 97%.

Created a system that automatically identified and categorized unprecedented anomaly labels with 96% accuracy.

Created an online method to replace missing values during volatile anomalies while keeping errors 10%.

Created two methods (for two use cases) to detect the start of an anomaly. Both methods achieve F1 scores of at least 95%.

Created a model for topology classification and monitoring achieving 97% Accuracy from training on just 5% of the data.

Developed a method to detect electricity fraud which increases detection rates by a factor of 20.

Developed a model to predict the deaths of power equipment increasing the proportion of correct identifications by 700%.

Performed detailed financial optimization of battery storage systems via their participation in Wholesale Power Markets.

Performed battery storage degradation Modelling and optimization in power markets under degradation and managed a database of smart meter data on a Hadoop cluster.

Performed simulations and optimizations to design multijunction solar cells, increasing efficiencies by 46%.

Experimentally tested the power curves of 20 fabricated solar cells and performed wet-etching of multijunction solar cells in a cleanroom environment.