Abstract: A computing system detects a defective object. An image is received of a manufacturing line that includes objects in a process of being manufactured. Each pixel included in the image is classified as a background pixel class, a non-defective object class, or a defective object class using a trained neural network model. The pixels included in the image that were classified as the non-defective object class or the defective object class are grouped into polygons. Each polygon is defined by a contiguous group of pixels classified as the non-defective object class or the defective object class. Each polygon is classified in the non-defective object class or in the defective object class based on a number of pixels included in a respective polygon that are classified in the non-defective object class relative to a number of pixels included in the respective polygon that are classified in the defective object class.

Abstract: Systems and methods for determining risk to operating a turbomachine are provided. According to one embodiment of the disclosure, a method may include receiving historical risk profile data associated with a fleet of turbomachines by at least one processor from a repository. The method can also include receiving ambient conditions of an environment in which a turbomachine is to be operated. Based at least in part on the historical risk profile data and in view of the ambient conditions, at least one risk threshold for at least one known operating profile can be developed. The method may continue with determining that the at least one risk threshold for the at least one known operating profile is reached. Based at least in part on a determination that the at least one risk threshold is reached, a mitigating action associated with the turbomachine can be taken.

Abstract: A system is provided comprising a memory configured to store instructions and a processor configured to execute the instructions. The processor is configured to execute the instructions to receive sensor data comprising sensed operations for a machinery, the sensed operations sensed via one or more sensors disposed in the machinery, and to derive a first model matrix based on the sensor data. The processor is further configured to derive a covariance regression model based on the first model matrix, wherein the covariance regression model is configured to be executed to derive a predictive event based on operational machinery data as input.

Abstract: Embodiments of the disclosure relate to performance testing of a gas turbine. In one embodiment, a gas turbine performance testing system can include a server, a transducer system, a signal converter, and an automated gas chromatograph. The transducer system acquires one or more functional parameters of the gas turbine and the signal converter converts the functional parameters acquired by the transducer system to gas turbine operational data. The automated gas chromatograph automatically analyzes a test sample of a natural gas that is used to operate the gas turbine. The gas turbine operational data generated by the signal converter and the analysis information obtained from the test sample are provided to the server for propagating via a communication network, to a client computer where the gas turbine operational data and the analysis information can be used to obtain a gas turbine performance test result.

Abstract: A system is provided comprising a memory configured to store instructions and a processor configured to execute the instructions. The processor is configured to execute the instructions to receive sensor data comprising sensed operations for a machinery, the sensed operations sensed via one or more sensors disposed in the machinery, and to derive a first model matrix based on the sensor data. The processor is further configured to derive a covariance regression model based on the first model matrix, wherein the covariance regression model is configured to be executed to derive a predictive event based on operational machinery data as input.

Abstract: Systems and methods for determining risk to operating a turbomachine are provided. According to one embodiment of the disclosure, a method may include receiving historical risk profile data associated with a fleet of turbomachines by at least one processor from a repository. The method can also include receiving ambient conditions of an environment in which a turbomachine is to be operated. Based at least in part on the historical risk profile data and in view of the ambient conditions, at least one risk threshold for at least one known operating profile can be developed. The method may continue with determining that the at least one risk threshold for the at least one known operating profile is reached. Based at least in part on a determination that the at least one risk threshold is reached, a mitigating action associated with the turbomachine can be taken.

Abstract: Embodiments of the disclosure relate to performance testing of a gas turbine. In one embodiment, a gas turbine performance testing system can include a server, a transducer system, a signal converter, and an automated gas chromatograph. The transducer system acquires one or more functional parameters of the gas turbine and the signal converter converts the functional parameters acquired by the transducer system to gas turbine operational data. The automated gas chromatograph automatically analyzes a test sample of a natural gas that is used to operate the gas turbine. The gas turbine operational data generated by the signal converter and the analysis information obtained from the test sample are provided to the server for propagating via a communication network, to a client computer where the gas turbine operational data and the analysis information can be used to obtain a gas turbine performance test result.

Abstract: A system for remote detection of surge in a fleet of turbine engines includes an on-site monitoring device coupled to each turbine engine of the fleet of turbine engines. The on-site monitoring device is configured to continuously receive operating parameter measurements indicative of operational and thermodynamic conditions of the turbine engine. The operational condition includes a compressor exit condition of the turbine engine compressor. The on-site monitoring device is configured to compile and transmit a snapshot of the operating parameter measurements to a remote monitoring unit. The remote monitoring unit is positioned remote from each turbine engine of the fleet of turbine engines. The remote monitoring unit is configured to receive the snapshot of operating parameter measurements from the on-site monitoring device. The remote monitoring unit is further configured to detect surge in the turbine engine based on analysis of the snapshot of operating parameter measurements.

Abstract: A system for remote detection of surge in a fleet of turbine engines includes an on-site monitoring device coupled to each turbine engine of the fleet of turbine engines. The on-site monitoring device is configured to continuously receive operating parameter measurements indicative of operational and thermodynamic conditions of the turbine engine. The operational condition includes a compressor exit condition of the turbine engine compressor. The on-site monitoring device is configured to compile and transmit a snapshot of the operating parameter measurements to a remote monitoring unit. The remote monitoring unit is positioned remote from each turbine engine of the fleet of turbine engines. The remote monitoring unit is configured to receive the snapshot of operating parameter measurements from the on-site monitoring device. The remote monitoring unit is further configured to detect surge in the turbine engine based on analysis of the snapshot of operating parameter measurements.

Abstract: A system for performance monitoring of power plants is provided. The system includes a plurality of power plants respectively owned and operated by a plurality of customers, a monitoring center, at least one of the customers and the monitoring center including a plurality of analytic algorithms configured to automate performance anomaly detection, alarming, analytics and prognosis and a secured network by which on-site monitoring (OSM) data generated at each of the plurality of the power plants is receivable by the monitoring center. The monitoring center is configured to analyze the OSM data along with site specific data for each respective power plant in accordance with the plurality of analytics algorithms to derive trends and to take action with respect to each respective power plant in accordance with changes and abnormalities indicated by the derived trends.

Abstract: An approach for detecting rotor anomalies is disclosed. An on-site monitoring unit monitors vibration measurements obtained from a rotor during a transient speed operation. In one aspect, the on-site monitoring unit classifies the vibration measurements into predetermined ranges of rotor speed during the transient speed operation, determines maximum vibration data for each of the predetermined ranges of rotor speed during the transient speed operation and compiles the maximum vibration data into a snapshot of the vibration measurements obtained during the transient speed operation. A remote monitoring unit detects a rotor anomaly from the snapshot of vibration measurements generated by the on-site monitoring unit.

Abstract: Systems and methods for protecting rotating machines are provided. Measurements data collected by a plurality of sensors may be received by a controller that includes one or more computers. The plurality of sensors may be configured to monitor vibrations associated with the rotating machine. Based at least in part upon the measurements data, the controller may determine that a respective amplitude change for at least two of the plurality of sensors exceeds a threshold condition. The controller may also determine that the threshold condition is exceeded for a predetermined period of time. Based at least in part upon determining that the threshold condition is exceeded for a predetermined period of time, the controller may identify an alarm event.

Abstract: An approach for detecting rotor anomalies is disclosed. An on-site monitoring unit monitors vibration measurements obtained from a rotor during a transient speed operation. In one aspect, the on-site monitoring unit classifies the vibration measurements into predetermined ranges of rotor speed during the transient speed operation, determines maximum vibration data for each of the predetermined ranges of rotor speed during the transient speed operation and compiles the maximum vibration data into a snapshot of the vibration measurements obtained during the transient speed operation. A remote monitoring unit detects a rotor anomaly from the snapshot of vibration measurements generated by the on-site monitoring unit.

Abstract: Systems and methods for protecting rotating machines are provided. Measurements data collected by a plurality of sensors may be received by a controller that includes one or more computers. The plurality of sensors may be configured to monitor vibrations associated with the rotating machine. Based at least in part upon the measurements data, the controller may determine that a respective amplitude change for at least two of the plurality of sensors exceeds a threshold condition. The controller may also determine that the threshold condition is exceeded for a predetermined period of time. Based at least in part upon determining that the threshold condition is exceeded for a predetermined period of time, the controller may identify an alarm event.