Abstract: Aspects of the present disclosure provide systems, methods, and computer-readable storage media that support ontology driven processes to generate digital twins having extended capabilities. To generate the digital twin, an ontology may be obtained and modified to define additional types of data, such as events and metrics, for incorporation into the digital twin. The ontology, once modified, may be instantiated as a knowledge graph having the additional types of data embedded therein. The embedded data may be used to convert the knowledge graph to a probabilistic graph model that may be queried to extract information from the digital twin in a probabilistic manner. Additionally, multiple ontologies may be utilized to create a digital twin-of-digital twins, which enables more complex digital twins to be generated (e.g., digital twins of entire ecosystems), and enables new insights and understanding of the various components and interactions between the components of the ecosystem.

Abstract: Aspects of the present disclosure provide systems, methods, and computer-readable storage media that support ontology driven processes to create digital twins that extend the capabilities of knowledge graphs. A dataset including an ontology and domain data corresponding to a domain associated with the ontology is obtained. A knowledge graph is constructed based on the ontology and the domain data is incorporated into the knowledge graph. The knowledge graph is exploited to derive random variables of a probabilistic graph model. The random variables may be associated with probability distributions, which may include unknown parameters. A learning process is executed to learn the unknown parameters and obtain a joint distribution of the probabilistic graph model, which may enable querying of the probabilistic graph model in a probabilistic and deterministic manner.

Abstract: A systems implements a gradient descent calculation, regression calculation, or other machine learning calculation on a dataset (e.g., a global dataset) using a coordination node including coordination circuitry that coordinates multiple worker nodes to create a distributed calculation architecture. In some cases, the worker nodes each hold a portion of the dataset and operate on their respective portion. In some cases, the gradient descent calculation, regression calculation, or other machine learning calculation is used to implement a targeted maximum likelihood scheme for causal inference estimation. The targeted maximum likelihood scheme may be used to conduct causal analysis of the observational data.

Abstract: The present disclosure relates to a system, a method, and a product for optimizing hyper-parameters for generation and execution of a machine-learning model under constraints. The system includes a memory storing instructions and a processor in communication with the memory. When executed by the processor, the instructions cause the processor to obtain input data and an initial hyper-parameter set; for an iteration, to build a machine learning model based on the hyper-parameter set, evaluate the machine learning model based on the target data to obtain a performance metrics set, and determine whether the performance metrics set satisfies the stopping criteria set. If yes, the instructions cause the processor to perform an exploitation process to obtain an optimal hyper-parameter set, and exit the iteration; if no, perform an exploration process to obtain a next hyper-parameter set, and perform a next iteration with using the next hyper-parameter set as the hyper-parameter set.

Abstract: In some implementations, a control system may obtain historical data associated with usage of a distillation column during a historical time period. The control system may configure a prediction model to monitor the distillation column for a hazardous condition. The prediction model may be trained based on training data that is associated with occurrences of the hazardous condition. The control system may monitor, using the prediction model, the distillation column to determine a probability that the distillation column experiences the hazardous condition within a threshold time period. The prediction model may be configured to determine the probability based on measurements from a set of sensors of the distillation column. The control system may perform, based on the probability satisfying a probability threshold, an action associated with the distillation column to reduce a likelihood that the distillation column experiences the hazardous condition within the threshold time period.

Abstract: A resource prediction system for executing machine learning models and method are provided. The system includes non-transitory memory storing instructions and a processor configured to execute the instructions to obtain input data including a targeted objective and the constraints, select a deployable machine learning model having an evaluation score that meets a predetermined criterion from among candidate machine learning models, virtually execute the deployable machine learning model on each of candidate hardware platforms according to the constraints, generate an assessment report of the virtual performance metrics set of the deployable machine learning model executed on each of the candidate hardware platforms, and select the suggested hardware platform meeting the predetermined criterion from among the candidate hardware platforms.

Abstract: A system receives sensor data from sensing parameters of a piece of factory equipment. The system includes a first model to generate predicted degradation states of the piece of factory equipment by being trained to generate a stochastic degradation model for classification of the predicted degradation states of a particular asset. The system includes a second model to which the predicted degradation states are provided. The second model trained to generate a covariate indicative of a failure condition of the piece of factory equipment. The system may supply the covariate to the first model to generate predicted degradation states compensated with the covariate. From the predicted degradation states compensated with the covariate a policy of a maintenance action may be generated with the system to optimize life expectancy of the piece of factory equipment. The system may adjust operation of the piece of factory equipment based on the maintenance action.

Abstract: In some implementations, a control system may obtain historical data associated with usage of a distillation column during a historical time period. The control system may configure a prediction model to monitor the distillation column for a hazardous condition. The prediction model may be trained based on training data that is associated with occurrences of the hazardous condition. The control system may monitor, using the prediction model, the distillation column to determine a probability that the distillation column experiences the hazardous condition within a threshold time period. The prediction model may be configured to determine the probability based on measurements from a set of sensors of the distillation column. The control system may perform, based on the probability satisfying a probability threshold, an action associated with the distillation column to reduce a likelihood that the distillation column experiences the hazardous condition within the threshold time period.

Abstract: The present disclosure relates to a system, a method, and a product for optimizing hyper-parameters for generation and execution of a machine-learning model under constraints. The system includes a memory storing instructions and a processor in communication with the memory. When executed by the processor, the instructions cause the processor to obtain input data and an initial hyper-parameter set; for an iteration, to build a machine learning model based on the hyper-parameter set, evaluate the machine learning model based on the target data to obtain a performance metrics set, and determine whether the performance metrics set satisfies the stopping criteria set. If yes, the instructions cause the processor to perform an exploitation process to obtain an optimal hyper-parameter set, and exit the iteration; if no, perform an exploration process to obtain a next hyper-parameter set, and perform a next iteration with using the next hyper-parameter set as the hyper-parameter set.

Abstract: A systems implements a gradient descent calculation, regression calculation, or other machine learning calculation on a dataset (e.g., a global dataset) using a coordination node including coordination circuitry that coordinates multiple worker nodes to create a distributed calculation architecture. In some cases, the worker nodes each hold a portion of the dataset and operate on their respective portion. In some cases, the gradient descent calculation, regression calculation, or other machine learning calculation is used to implement a targeted maximum likelihood scheme for causal inference estimation. The targeted maximum likelihood scheme may be used to conduct causal analysis of the observational data.

Abstract: A system receives sensor data from sensing parameters of a piece of factory equipment. The system includes a first model to generate predicted degradation states of the piece of factory equipment by being trained to generate a stochastic degradation model for classification of the predicted degradation states of a particular asset. The system includes a second model to which the predicted degradation states are provided. The second model trained to generate a covariate indicative of a failure condition of the piece of factory equipment. The system may supply the covariate to the first model to generate predicted degradation states compensated with the covariate. From the predicted degradation states compensated with the covariate a policy of a maintenance action may be generated with the system to optimize life expectancy of the piece of factory equipment. The system may adjust operation of the piece of factory equipment based on the maintenance action.

Abstract: A plurality of measurement signals may be evaluated to detect a poorly damped oscillation mode in an electric power delivery system. An oscillation mode of interest may be detected, and the oscillation mode of interest may be analyzed using a frequency transform. A plurality of amplitudes of the oscillation mode of interest in each measurement signal may be determined using a sliding window. The plurality of amplitudes may be used to calculate a damping of the oscillation mode of interest. The damping may be calculated solving a linearized system of equations. The linearized system of equations may be a least square estimate of the damping based on the logarithm of each amplitude. If the damping indicates that the oscillation mode of interest is poorly damped, a control action may be taken.

Abstract: A plurality of measurement signals may be evaluated to detect a poorly damped oscillation mode in an electric power delivery system. An oscillation mode of interest may be detected, and the oscillation mode of interest may be analyzed using a frequency transform. A plurality of amplitudes of the oscillation mode of interest in each measurement signal may be determined using a sliding window. The plurality of amplitudes may be used to calculate a damping of the oscillation mode of interest. The damping may be calculated solving a linearized system of equations. The linearized system of equations may be a least square estimate of the damping based on the logarithm of each amplitude. If the damping indicates that the oscillation mode of interest is poorly damped, a control action may be taken.