Asset prioritization algorithm for optimizing electrical network upgrade and replacement strategies

This technology is a machine learning algorithm that estimates the mean time between failures (MTBF) for components in electrical networks.

Unmet Need: Planning system for selecting assets for improvement within an infrastructure

With rising energy costs, efficiently using and maintaining the current energy grid has become a priority. Because of the complexity and breadth of assets in many locations across the energy grid, the management and decisions on replacing equipment can be difficult. As the assets in the system deteriorate, the probability of interruptions to the power supply increases. And so currently, many assets are often replaced before their true lifecycle, increasing costs. Current enterprise asset management systems cannot identify the most probable lifecycle of each individual asset, thus failing to maximize efficiencies while maintaining low failure rates in systems.

The Technology: Capital asset prioritization algorithm that suggests improvements to electrical networks

This technology is a machine-learning algorithm can be used to estimate the MTBF of components in electrical networks. Through utilization of historical and current inputs, the training of the algorithm allows for reliable predictions of areas where improvements in MTBF can be made. These suggestions can be applied to all categories of capital assets, allowing optimization of the replacement and upgrade of components. This technology can enable cost-effective solutions to be applied to the highly complex process of managing assets across energy grids. Through better management of equipment replacement in electrical grids, this technology can provide higher operating efficiency and reduce costs associated with repairs and upgrades.

Applications:

• Management of replacements and upgrades in capital assets
• System to provide data that will reduce the mean time between failures of equipment
• System to be implemented by utility companies to increase efficiency
• Maximizing the use of assets at high-intensity operations, such as power plants, gas, oil refining, mass transit, communication networks

Advantages:

• Finds optimal MTBF for electrical network components
• Can be sold as an all-inclusive software/hardware package
• Custom tailored to each application and can improve as it acquires more data
• Software can be upgraded to deal with radical changes in energy consumption

Lead Inventor:

Roger N. Anderson, Ph.D.

Patent Information:

Patent Status

Related Publications:

• Rudin C, Waltz D, Anderson RN, Boulanger A, Salleb-Aouissi A, Chow M, Dutta H, Gross PN, Huang B, Ierome S, Isaac DF, Kressner A, Passonneau RJ, Radeva A, Wu L. “Machine Learning for the New York City Power Grid” IEEE Trans Pattern Anal Mach Intell. 2012 Feb; 34(2): 328-345.

• Anderson RN, Boulanger A. “Knowledge management and computational learning for lean energy management” Oil & Gas Journal. 2004 Nov; 102(44): 36-40.

Tech Ventures Reference:

• IR M09-094

• Licensing Contact: Richard Nguyen