Abstract: Generating a synthetic dataset from road vehicle positioning data. The method involves iteratively selecting a seed record and determining a set of candidate partner records based on matching bigrams between the selected seed record and potential partner records. The method further includes determining subsequences for the candidate partner records and selected seed record based on locations of the matching bigrams. A candidate partner record is selected based on whether a length difference, in terms of time, distance, or number of road events, between the subsequences is less than a defined threshold. The subsequence of the selected seed record is replaced with the subsequence of the selected candidate partner record to produce a mutated seed record. The mutated seed records to produce a synthetic dataset.

Abstract: Relates to improving accuracy of a primary predictive model based on a residual model. A method includes determining, using a primary predictive model, a first set of training prediction residuals based on a first labeled training set and a first set of testing prediction residuals based on a first labeled testing set. A second dataset includes a second labeled training set, labeled based on the first set of training prediction residuals, and a second labeled testing set, labeled based on the first set of testing prediction residuals. The method further includes training the residual model using the second labeled training set. The primary predictive model is adjusted based on a set of training predictions and a set of testing predictions of the residual model to produce more accurate predictions.

Abstract: Generating a synthetic dataset from road vehicle positioning data. The method involves iteratively selecting a seed record and determining a set of candidate partner records based on matching bigrams between the selected seed record and potential partner records. The method further includes determining subsequences for the candidate partner records and selected seed record based on locations of the matching bigrams. A candidate partner record is selected based on whether a length difference, in terms of time, distance, or number of road events, between the subsequences is less than a defined threshold. The subsequence of the selected seed record is replaced with the subsequence of the selected candidate partner record to produce a mutated seed record. The mutated seed records to produce a synthetic dataset.

Abstract: A system utilizing internal and external online, real-time direct leak detection based on applying fiber optic sensors to detect leak generated acoustic waves or negative pressure wave inside the pipeline and utilizing the same fiber optic sensing cable or a different fiber optic sensing cable to detect leak outside of the pipeline. This approach encompasses two different fiber optic technologies to detect leaks both internally and externally and involves quasi-distributed and/or distributed sensing techniques. This system will detect leaks internally or intrinsically by sensing the leak generated acoustic signals propagated inside tubulars, pipes, pipelines, flow channels, or any pressurized system, as well as externally or extrinsically by detecting leak released fluid or leak event associated phenomena outside of the leaking systems respectively. These two technologies may detect leaks separately or jointly.

Abstract: A radiant area heater includes a columnar base with a hollow compartment and a heating column disposed on the columnar base. The heating column includes a control portion with at least one controller, a burner, and a sparker and a flame portion that includes a combustion chamber. The combustion chamber is disposed on a cylindrical member that has at least one vent allowing air to circulate to the burner and within the combustion chamber. The heating column further includes at least one heat emitting screen disposed above the flame portion that radiates heat outwards from the heater. A cover is disposed over the heating column to direct heat outwards and away from the heater.

Abstract: A radiant area heater includes a columnar base with a hollow compartment and a heating column disposed on the columnar base. The heating column includes a control portion with at least one controller, a burner, and a sparker and a flame portion that includes a combustion chamber. The combustion chamber is disposed on a cylindrical member that has at least one vent allowing air to circulate to the burner and within the combustion chamber. The heating column further includes at least one heat emitting screen disposed above the flame portion that radiates heat outwards from the heater. A cover is disposed over the heating column to direct heat outwards and away from the heater.

Abstract: Power switch device includes a wide-bandgap semiconductor switch, and a gate driver. The gate driver includes a driver circuit and a diagnostic circuit. The driver circuit is configured to provide a driver signal to control the wide-bandgap semiconductor switch. The diagnostic circuit is configured to sense an electrical characteristic of the wide-bandgap semiconductor switch, and perform a diagnostic test for the wide-bandgap semiconductor switch in response to the electrical characteristic of the wide-bandgap semiconductor switch.

Abstract: Disclosed are methods and systems for generating a media mix model. A time series data set is received specifying media delivered to recipients via a plurality of media channels at a plurality of times and one or more responses at the plurality of times. A random forest model is trained, the random forest splitting the time series data into subsets based on media channel of the plurality of media channels. Response curves are generated using the trained random forest model, each of the response curves corresponding to a media channel of the plurality of media channels, the response curves forming a media mix model adapted to predict responses based on media delivered and media channel.

Abstract: Converting an Electronic Data Capture (EDC) dataset to a Standard Data Tabulation Model (SDTM) dataset is described, including processing metadata for an EDC dataset, the metadata comprising EDC field names, to produce vectors of a particular dimensionality. The vectors are processed in an embedding model to produce embedded vectors of a lesser dimensionality. The embedded vectors are processed, in a prediction model, to produce class predictions for the EDC field names, the classes corresponding to the SDTM field names. The EDC field names are mapped with the SDTM field names based on the class predictions for the EDC field names. The embedding model is obtained from a trained Siamese neural network.

Abstract: A cable tether system includes a base member and at least one elongate member extending away from the base member. The elongate member is configured to surround at least a portion of a first cable. A second elongate member can be configured to surround at least a portion of a second cable. The base member can include one or more engagement portions that couple with one or more engagement portions on the first and second elongate members.

Abstract: In various examples, a cell model that partitions a geographic region into one or more cells is used to determine clusters of cell which share similarities. Sensor data is provided to one or more machine learning models trained to classify the sensor data to one or more cells of the cell model. Based on classifying sensor data to cells of a cell model, similarities between pairings of cells of the cell model may be determined and used to form clusters of the cell which are sufficiently similar in order to aid in the curation of training data used to train machine learning models in order to aid an autonomous or semi-autonomous machine in a surrounding environment.

Abstract: Apparatuses for dynamically sensing infrared (IR) radiation in an electron beam powder bed fusion (EB-PBF) printer are provided. A radiation collector receives radiation from a surface of the powder bed. An IR-transparent material rejects one or more non-IR wavelengths, and a lens focuses the IR radiation onto an optical fiber. The IR radiation is carried from the vacuum chamber of the printer to a sensor, where IR information is determined based on the received IR radiation. The IR information may be received from the sensor and used by the print controller to modify one or more parameters, such as beam intensity or scanning rate, on the fly or during the next print cycle. An occlusion member can be used to selectively block or expose the radiation collector to protect the radiation collector from condensation of vapor from vaporization of particles at high temperatures.

Abstract: In various aspects, 3D printers and recoaters incorporate sensor systems coupled to or integrated with the 3D printers. The sensor systems may include eddy current sensors and other sensors configured to measure an electromagnetic characteristic of the build piece. A three-dimensional (3-D) printer in one aspect includes a depositor configured to deposit metal, an energy beam source configured to selectively melt the metal to form a portion of a build piece, and a sensor configured to move relative to a surface of the print area and to measure an electromagnetic characteristic of the portion of the print area. The measured data can be used to detect defects and other information about the build piece that can be used to fix the defects or enhance the build piece geometry during the printing.

Abstract: In various aspects, 3D printers, and sensor systems coupled to or integrated with the 3D printers are disclosed. The sensor systems may include image and second sensors for detecting potential defects or print artifacts. During printing, an energy beam source forms a weld pool by melting selected regions of print material, which solidifies to produce the build piece. The image sensor may image an area including the weld pool to determine a landing location of matter ejected during the heating of print material to form the weld pool. The second sensor may detect a defect in the build piece based on the determination of the landing location. Print operation may be suspended while the sensor data is used to repair the defect, after which 3D printing resumes. In this way, for example, high quality build pieces can be produced with reduced post-processing times, and hence a higher manufacturing throughput.

Abstract: A radiant area heater includes a columnar base with a hollow compartment and a heating column disposed on the columnar base. The heating column includes a control portion with at least one controller, a burner, and a sparker and a flame portion that includes a combustion chamber. The combustion chamber is disposed on a cylindrical member that has at least one vent allowing air to circulate to the burner and within the combustion chamber. The heating column further includes at least one heat emitting screen disposed above the flame portion that radiates heat outwards from the heater. A cover is disposed over the heating column to direct heat outwards and away from the heater.

Abstract: A direct AC to AC converter includes a modulation stage, a transformer and a de-modulation stage. The modulation stage is configured to convert an AC input voltage with a first frequency into a bipolar PWM voltage with a second frequency, wherein the second frequency is higher than the first frequency. The transformer has a primary winding and secondary winding, wherein the primary winding is coupled to the modulation stage to receive the bipolar PWM voltage. The de-modulation stage is coupled to the secondary winding of the transformer and configured to convert the voltage across the secondary winding of the transformer into an AC output voltage with the first frequency.

Abstract: Apparatuses for dynamically sensing infrared (IR) radiation in an electron beam powder bed fusion (EB-PBF) printer are provided. A radiation collector receives radiation from a surface of the powder bed. An IR-transparent material rejects one or more non-IR wavelengths, and a lens focuses the IR radiation onto an optical fiber. The IR radiation is carried from the vacuum chamber of the printer to a sensor, where IR information is determined based on the received IR radiation. The IR information may be received from the sensor and used by the print controller to modify one or more parameters, such as beam intensity or scanning rate, on the fly or during the next print cycle. An occlusion member can be used to selectively block or expose the radiation collector to protect the radiation collector from condensation of vapor from vaporization of particles at high temperatures.

Abstract: In various aspects, an apparatus for an electron-beam powder bed fusion (EB-PBF) printer includes a radiation collector configured to collect radiation in an ultraviolet (UV) spectrum at a powder bed surface within a vacuum chamber during an electron beam scanning cycle of EB-PBF operation, an optical fiber configured to be transparent to the radiation in the UV spectrum and configured to receive the radiation at the powder bed surface via the radiation collector, and a processor configured to receive one or more extracted wavelengths of radiation in the UV spectrum based on the radiation carried on the optical fiber.

Abstract: A cable tether system includes a base member and at least one elongate member extending away from the base member. The elongate member is configured to surround at least a portion of a first cable. A second elongate member can be configured to surround at least a portion of a second cable. The base member can include one or more engagement portions that couple with one or more engagement portions on the first and second elongate members.

Abstract: A cable tether system includes a base member and at least one elongate member extending away from the base member. The elongate member is configured to surround at least a portion of a first cable. A second elongate member can be configured to surround at least a portion of a second cable. The base member can include one or more engagement portions that couple with one or more engagement portions on the first and second elongate members.