Abstract: The invention relates to a process for combined ethylbenzene reforming and xylene isomerisation comprising contacting a hydrocarbon feedstock containing ethylbenzene and xylene with a catalyst comprising a catalyst carrier and one or more metal(s) supported on the catalyst carrier, wherein the catalyst carrier is an extrudate comprising (i) a ZSM-48 and/or EU-2 type zeolite and (ii) an alumina binder, the extrudate having a shape with a C/A value of at least 3, where C is the circumference of the extrudate and A is the cross-sectional area of the extrudate. The metal may be platinum and the alumina may be a wide-pore alumina. The process displays high conversion rates whilst maintaining low levels of side-product formation.

Abstract: The invention relates to a process for combined ethylbenzene reforming and xylene isomerisation comprising contacting a hydrocarbon feedstock containing ethylbenzene and xylene with a catalyst comprising a catalyst carrier and one or more metal(s) supported on the catalyst carrier, wherein the catalyst carrier is an extrudate comprising (i) a ZSM-48 and/or EU-2 type zeolite and (ii) an alumina binder, the extrudate having a shape with a C/A value of at least 3, where C is the circumference of the extrudate and A is the cross-sectional area of the extrudate. The metal may be platinum and the alumina may be a wide-pore alumina. The process displays high conversion rates whilst maintaining low levels of side-product formation.

Abstract: A connection mechanism (4) comprises first and second components (6, 8) which can be selectively coupled together, one of the first or second components comprises a locking member (25) which is moveable between a first state and a second state, the other of the first or second component comprises means (16) for initially moving the locking member from the first to the second state as the first and second components are initially drawn together and means (19) for returning the locking member to the first state once the first and second components are fully engaged together to prevent disengagement of the first and second components.

Abstract: A connection mechanism (4) comprises first and second components (6, 8) which can be selectively coupled together, one of the first or second components comprises a locking member (25) which is moveable between a first state and a second state, the other of the first or second component comprises means (16) for initially moving the locking member from the first to the second state as the first and second components are initially drawn together and means (19) for returning the locking member to the first state once the first and second components are fully engaged together to prevent disengagement of the first and second components.

Abstract: A computer-implemented method for layered training of machine-learning architectures includes receiving a plurality of data elements wherein each data element is associated with a timestamp, determining a training window for each model layer of a layered stack of model layers, determining a plurality of training data elements for each training window by identifying the data elements with timestamps corresponding to each of the training windows, identifying a previous checkpoint for each model layer wherein the previous checkpoint for each model layer is generated by a parent model layer, training each model layer with the determined training data elements for each model layer and the identified previous checkpoint for each model layer, generating a plurality of current checkpoints wherein each current checkpoint of the plurality of current checkpoints is associated with a model layer, and storing the plurality of current checkpoints at the memory.

Abstract: A computer-implemented method for layered training of machine-learning architectures includes receiving a plurality of data elements wherein each data element is associated with a timestamp, determining a training window for each model layer of a layered stack of model layers, determining a plurality of training data elements for each training window by identifying the data elements with timestamps corresponding to each of the training windows, identifying a previous checkpoint for each model layer wherein the previous checkpoint for each model layer is generated by a parent model layer, training each model layer with the determined training data elements for each model layer and the identified previous checkpoint for each model layer, generating a plurality of current checkpoints wherein each current checkpoint of the plurality of current checkpoints is associated with a model layer, and storing the plurality of current checkpoints at the memory.

Abstract: A method for inspecting a rail vehicle includes receiving signals representative of temperatures of a wheel of the rail vehicle and generating a sensed wheel temperature profile of the wheel based on the signals. At least a portion of the sensed wheel temperature profile is representative of the temperatures of the wheel. The method also includes identifying waveform attributes in the sensed wheel temperature profile and designating the sensed wheel temperature profile as at least one of a typical wheel profile and an atypical wheel profile based on the waveform attributes in the sensed wheel temperature profile.

Abstract: An apparatus and methods for identifying a defect and/or an operating characteristic of a system being monitored (and/or one or more of the system's components) are described. In an embodiment, orthogonally related data monitored by two or more detectors may be fused to determine whether a component of a system is defective and/or malfunctioning. Additionally or alternatively, data from a first detector may be determined to be accurate using non-orthogonally related data outputted by a second detector. Both types of determinations may be made with minimal or no false indications, which lowers the cost of operating the system being monitored. Embodiments of the invention may also be configured to forecast and/or prevent accidents and/or damage to the system being monitored by predicting whether a defect and/or a malfunction will occur.

Abstract: A system for detecting a moving hot bearing or wheel of a rail car is provided. The system includes a first comparator to receive input signals representative of radiation emitted by the moving hot rail car bearing or wheel, and to compare the input signals to a threshold value. The system further includes a counter for counting incidents of the input signals exceeding the threshold value and a second comparator to compare a number of incidents of the input signals exceeding the threshold value to a count threshold as an indication of detection of a hot rail car bearing or wheel.

Abstract: A system for detecting a moving hot bearing or wheel is provided. The system includes a summer for combining an input signal representative of radiation emitted by the moving hot rail car bearing with a feedback signal. The system also includes an integrator to accumulate an error resulting from the combination of the input signal and the feedback signal. The system further includes a feedback loop to feedback output of the integrator to the summer.

Abstract: A method for inspecting a rail vehicle includes receiving signals representative of temperatures of a wheel of the rail vehicle and generating a sensed wheel temperature profile of the wheel based on the signals. At least a portion of the sensed wheel temperature profile is representative of the temperatures of the wheel. The method also includes identifying waveform attributes in the sensed wheel temperature profile and designating the sensed wheel temperature profile as at least one of a typical wheel profile and an atypical wheel profile based on the waveform attributes in the sensed wheel temperature profile.

Abstract: In a railroad track system that provides for communications through a track rail without insulated joints between a specific transmitter and a specific receiver when a plurality of transmitters and a plurality of receivers are communicating using the track rail, a method including emitting a unique signal from the specific transmitter during a specific time. The unique signal is transmitted through a railway rail, which is without an insulated joint between successive rails and is the medium through which the unique signal travels, wherein the unique signal is detectable but not readable by the plurality of receiver. The specific receiver is activated to read the unique signal during the specific time.

Abstract: A system to facilitate reducing false train stops includes an infrared sensor that is coupled to a processing unit. The processing unit is programmed to receive inputs from the sensor, generate a bearing profile using the received inputs, classify the bearing profile as at least one of a typical bearing profile and an atypical bearing profile, and if the bearing profile is classified as a typical bearing profile, analyze the typical bearing profile to determine if at least a portion of the typical bearing profile meets the alarm criteria. A method of reducing train stops and generating maintenance alerts are also described herein.

Abstract: A system to facilitate reducing false train stops includes an infrared bearing scanner that is coupled to a processing unit. The processing unit is programmed to receive inputs from the infrared bearing scanner, generate a bearing profile for each bearing passing the Hot Box Detection System, assign a classification to each bearing profile, determine a quantity of bearing profiles assigned to each classification, and generate a maintenance alert if the quantity of bearings assigned to at least one classification exceeds a predetermined threshold. A method of reducing false trains stops is also described herein.

Abstract: An apparatus and methods for identifying a defect and/or an operating characteristic of a system being monitored (and/or one or more of the system's components) are described. In an embodiment, orthogonally related data monitored by two or more detectors may be fused to determine whether a component of a system is defective and/or malfunctioning. Additionally or alternatively, data from a first detector may be determined to be accurate using non-orthogonally related data outputted by a second detector. Both types of determinations may be made with minimal or no false indications, which lowers the cost of operating the system being monitored.

Abstract: An apparatus and methods for identifying a defect and/or an operating characteristic of a system being monitored (and/or one or more of the system's components) are described. In an embodiment, orthogonally related data monitored by two or more detectors may be fused to determine whether a component of a system is defective and/or malfunctioning. Additionally or alternatively, data from a first detector may be determined to be accurate using non-orthogonally related data outputted by a second detector. Both types of determinations may be made with minimal or no false indications, which lowers the cost of operating the system being monitored. Embodiments of the invention may also be configured to forecast and/or prevent accidents and/or damage to the system being monitored by predicting whether a defect and/or a malfunction will occur.

Abstract: A method and system to detect misalignment of sensors used in train inspection systems are provided. The method includes receiving inputs from a train inspection sensor that is coupled to a railroad track, determining a position of the train inspection sensor using a motion detector, and identifying the inputs received from the train inspection sensor as at least one of acceptable or not acceptable based on the determined position of the train inspection sensor.

Abstract: A method and system to detect misalignment of sensors used in train inspection systems are provided. The method includes receiving inputs from a train inspection sensor that is coupled to a railroad track, determining a position of the train inspection sensor using a motion detector, and identifying the inputs received from the train inspection sensor as at least one of acceptable or not acceptable based on the determined position of the train inspection sensor.

Abstract: An apparatus and methods for identifying a defect and/or an operating characteristic of a system being monitored (and/or one or more of the system's components) are described. In an embodiment, orthogonally related data monitored by two or more detectors may be fused to determine whether a component of a system is defective and/or malfunctioning. Additionally or alternatively, data from a first detector may be determined to be accurate using non-orthogonally related data outputted by a second detector. Both types of determinations may be made with minimal or no false indications, which lowers the cost of operating the system being monitored.

Abstract: A system for detecting a moving hot bearing or wheel of a rail car is provided. The system includes a first comparator to receive input signals representative of radiation emitted by the moving hot rail car bearing or wheel, and to compare the input signals to a threshold value. The system further includes a counter for counting incidents of the input signals exceeding the threshold value and a second comparator to compare a number of incidents of the input signals exceeding the threshold value to a count threshold as an indication of detection of a hot rail car bearing or wheel.