Abstract: The disclosed subject matter provides systems and methods for allocating resources within an infrastructure, such as an electrical grid, in response to changes to inputs and output demands on the infrastructure, such as energy sources and sinks.

Abstract: Boosting algorithms are provided for accelerated machine learning in the presence of misclassification noise. In an exemplary embodiment, a machine learning method having multiple learning stages is provided. Each learning stage may include partitioning examples into bins, choosing a base classifier for each bin, and assigning an example to a bin by counting the number of positive predictions previously made by the base classifier associated with the bin.

Abstract: The present application provides methods and systems for quantitatively predicting an effectiveness of a proposed capital improvement project based on one or more previous capital improvement projects representative of one or more physical assets and including one or more attributes that includes defining a first sample pool from the previous capital improvement project data in which said previous capital improvement project has been performed, defining a second sample in which the previous capital improvement project has not been performed, the second sample pool including one or more attribute values that are the same as, or similar to, the attribute values for the first sample pool, generating a performance metric for each of the first and second sample pools, comparing the performance metric from the first sample pool with the performance metric from the second sample pool to determine a net performance metric, and, generating a prediction of effectiveness of the proposed capital improvement project conce

Abstract: The disclosed subject matter relates to an integrated decision support “cockpit” or control center for displaying, analyzing, and/or responding to, various events and contingencies that can occur within an electrical grid.

Abstract: A machine learning system creates failure-susceptibility rankings for feeder cables in a utility's electrical distribution system. The machine learning system employs martingale boosting algorithms and Support Vector Machine (SVM) algorithms to generate a feeder failure prediction model, which is trained on static and dynamic feeder attribute data. Feeders are dynamically ranked by failure susceptibility and the rankings displayed to utility operators and engineers so that they can proactively service the distribution system to prevent local power outages. The feeder rankings may be used to redirect power flows and to prioritize repairs. A feedback loop is established to evaluate the responses of the electrical distribution system to field actions taken to optimize preventive maintenance programs.

Abstract: A machine learning system creates failure-susceptibility rankings for feeder cables in a utility's electrical distribution system. The machine learning system employs martingale boosting algorithms and Support Vector Machine (SVM) algorithms to generate a feeder failure prediction model, which is trained on static and dynamic feeder attribute data. Feeders are dynamically ranked by failure susceptibility and the rankings displayed to utility operators and engineers so that they can proactively service the distribution system to prevent local power outages. The feeder rankings may be used to redirect power flows and to prioritize repairs. A feedback loop is established to evaluate the responses of the electrical distribution system to field actions taken to optimize preventive maintenance programs.

Abstract: A computer-aided lean management (CALM) controller system recommends actions and manages production in an oil and gas reservoir/field as its properties and conditions change with time. The reservoir/field is characterized and represented as an electronic-field (“e-field”). A plurality of system applications describe dynamic and static e-field properties and conditions. The application workflows are integrated and combined in a feedback loop between actions taken in the field and metrics that score the success or failure of those actions. A controller/optimizer operates on the combination of the application workflows to compute production strategies and actions. The controller/optimizer is configured to generate a best action sequence for production, which is economically “always-in-the-money.

Abstract: Boosting algorithms are provided for accelerated machine learning in the presence of misclassification noise. In an exemplary embodiment, a machine learning method having multiple learning stages is provided. Each learning stage may include partitioning examples into bins, choosing a base classifier for each bin, and assigning an example to a bin by counting the number of positive predictions previously made by the base classifier associated with the bin.

Abstract: An Innervated Stochastic Controller optimizes business decision-making under uncertainty through time. The Innervated Stochastic Controller uses a unified reinforcement learning algorithm to treat multiple interconnected operational levels of a business process in a unified manner. The Innervated Stochastic Controller generates actions that are optimized with respect to both financial profitability and engineering efficiency at all levels of the business process. The Innervated Stochastic Controller can be configured to evaluate real options. In one embodiment of the invention, the Innervated Stochastic Controller is configured to generate actions that are martingales. In another embodiment of the invention, the Innervated Stochastic Controller is configured as a computer-based learning system for training power grid operators to respond to grid exigencies.

Abstract: An Innervated Stochastic Controller optimizes business decision-making under uncertainty through time. The Innervated Stochastic Controller uses a unified reinforcement learning algorithm to treat multiple interconnected operational levels of a business process in a unified manner. The Innervated Stochastic Controller generates actions that are optimized with respect to both financial profitability and engineering efficiency at all levels of the business process. The Innervated Stochastic Controller can be configured to evaluate real options. In one embodiment of the invention, the Innervated Stochastic Controller is configured to generate actions that are martingales. In another embodiment of the invention, the Innervated Stochastic Controller is configured as a computer-based learning system for training power grid operators to respond to grid exigencies.

Abstract: A single intranet, internet, or World Wide Web-accessible interface is provided for, initiation of, interactive adjustments to, and access to the outputs of an integrated workflow of a plurality of analytical computer applications for characterization and analysis of traits and optimal management of the extraction of oil, gas, and water from a subsurface reservoir. By combining disparate analytical application tools in a seamless and remotely accessible, package, incompatibility problems caused by the disparate nature of petroleum analysis methods is reduced. The assumptions, analytic processes, and input data used for one analysis may be readily retrieved and re-evaluated for that reservoir or for future evaluations of the same or other reservoirs. Thus a flexible database of analysis tools and data may be implemented for access, input, and output of workflow and analytical data in the field, in conjunction with standard main computer servers, software and plug-ins, and portable remote computers.

Abstract: The invention utilizes 3-D and 4-D seismic surveys as a means of deriving information useful in petroleum exploration and reservoir management. The methods use both single seismic surveys (3-D) and multiple seismic surveys separated in time (4-D) of a region of interest to determine large scale migration pathways within sedimentary basins, and fine scale drainage structure and oil-water-gas regions within individual petroleum producing reservoirs. Such structure is identified using pattern recognition tools which define the regions of interest. The 4-D seismic data sets may be used for data completion for large scale structure where time intervals between surveys do not allow for dynamic evolution. The 4-D seismic data sets also may be used to find variations over time of small scale structure within individual reservoirs which may be used to identify petroleum drainage pathways, oil-water-gas regions and, hence, attractive drilling targets.