This technology is a set of metrics that can be used for accurately monitoring the performance of capital, operations, and maintenance investments to infrastructure.
Due to increased electricity demands and rapidly deteriorating infrastructure, the United States power grid faces enormous problems over the next several decades. Current efforts revolve around transitioning to a “smart grid” system with distributed, renewable sources of energy and connections between all components of the cyber-physical system to enhance control and optimize performance. However, these current approaches lack an empirical, unbiased method of determining the realized effectiveness of capital improvement projects. As such, there remains a need to evaluate the accuracy of such predictive models after the work has been performed, and if necessary, implement changes to these predictive models so that future predictions are more accurate.
This technology includes a set of techniques for evaluating the predicted effectiveness of maintenance of and improvements to infrastructure. The accuracy of these predictions involves collecting data, representative of at least one pre-defined metric, from the infrastructure during first and second time periods corresponding to before and after a change has been implemented. A machine learning system can then receive compiled data representative of the different time periods and generate corresponding machine learning data. As such, this technology monitors the cause-and-effect implications of operational field actions and validates whether actual performance matches what is expected from efficient energy planning. This allows for the optimization of management and financial decisions that are made to various infrastructures based on their overall effectiveness.
IR M10-068
Licensing Contact: Richard Nguyen