Caterpillar Inc.

Xuefei Hu

Yanchai Zhang

Zhaoxu Dong

Yong Sun

Zhijun Cai

Wenming Zhao

Nathan Stephen Pauli

Maurice J Dust

Chunhui Pan

Yifeng Lu

Eric James Johannsen

Perry Dwain Converse

Suresh Baddam Reddy

James David Siegle

Venkata Dandibhotla

Daniel Wilson Hoyt

Eric W Ohlson

Li Zhang

John Gates

Purna Chander Nalla

Arun Rajendran

Rodney D Harms

Shiferaw Damtie Beyene

Nathan J Wieland

Mingang Wang

Lidi Chen

Brian W Young

Daojie Zhang

Nimesh Akalanka Jayakody

An electronic control module (ECM) on-board a machine executes a reduced-order model configured to generate virtual sensor data associated with a machine component based on sensor data measured by actual sensors. The ECM can also execute a machine learning model configured to generate offset data for the sensor data and/or the virtual sensor data, based on operating conditions, an age of the machine component, and/or other factors that the reduced-order model may not consider. The sensor data and/or the virtual sensor data, adjusted based on the offset data, can indicate whether the machine component has failed or is predicted to fail at a future point in time, and/or a remaining usable life of the machine component.

On-board machine component failure detection

A system may comprise a device. The device may be configured to receive, from one or more sensor devices of the machine, sensor data associated with wear of one or more components of an undercarriage of the machine; and predict, using a machine learning model and the sensor data, an amount wear of the one or more components based on a wear rate of the one or more components. The machine learning model is trained, using training data, to predict the wear rate of the one or more components. The training data includes two or more of: historical sensor data, historical inspection data, or simulation data, of a simulation model, from one or more third devices. The device may perform an action based on the amount of wear.

Undercarriage wear prediction using machine learning model

A detection system for detecting winding faults, such as inter-turn winding faults in the stator and/or rotor of an electrical generator utilizes one or more vibration sensors that can be located on a generator housing. The vibration sensors make mechanical vibration measurements and transmit them to a fault analyzer. The fault analyzer can compare the measured vibrations with a threshold to determine if a winding is occurring. In an embodiment, the fault analyzer can convert the mechanical vibration measurements from the time domain to a frequency domain to facilitate analysis.

System and method for detecting winding faults in a generator

A hydraulic fracturing machine includes a pump failure detection system. The hydraulic fracturing machine includes a hydraulic fracturing pump with a power end and a fluid end. The power end includes a plurality of roller bearings, and the fluid end includes a flow of fluid. A particle sensor coupled to the power end is configured to transmit particle information regarding a quantity of particles in the fluid. A temperature sensor, also coupled to the power end, is configured to transmit temperature information regarding a temperature of the fluid. A vibration sensor coupled to the power end is configured to transmit vibration information regarding a vibration of each of the plurality of roller bearings. An electronic control module analyzes the particle information, the temperature information and the vibration information, and calculates a failure warning level based on the analysis.

Hydraulic fracturing pump health monitor

A system and method for protecting a genset from pole slip is disclosed. The system may comprise a generator, a prime mover and a controller. The generator includes a stator and a rotor. The prime mover is configured to drive rotation of the rotor. The controller may be configured to: determine mechanical status of the generator based on data associated with a translational displacement of the rotor; determine electrical status of the generator based on (a) a load angle or (b) the load angle and a rate of change of the load angle associated with rotation of the rotor in the stator; determine an operating condition of the generator based on fusion of the mechanical status and the electrical status; if the operating condition is a pole-slip-warning, activate an output member to display or emit a warning; and, if the operating condition is a pole-slip, activate a protective action.

Generator pole slip detection

A method for adjusting a valve lash in an internal combustion engine includes receiving a first signal generated by a sensor secured to the internal combustion engine, the first signal being indicative of a closing of a valve, receiving a second signal indicative of at least one of an engine speed of the internal combustion engine or a position of a camshaft of the internal combustion engine, and automatically determining an adjusted amount of lash associated with the valve based on the received first signal and the received second signal. The method also includes comparing the adjusted amount of lash to at least one predetermined threshold, and providing, in response to determining that the adjusted amount of lash is greater than the at least one predetermined threshold, a valve lash re-adjustment notification.

System and method for engine valve lash calibration

A pump monitoring and notification system for a hydraulic pump includes an accelerometer and a controller. The accelerometer is associated with the hydraulic pump and is disposed relative to the hydraulic pump to generate acceleration data indicative of acceleration of the hydraulic pump. The controller is configured to access a fault threshold, access a time threshold, and determine an acceleration of the accelerometer based upon the acceleration data from the accelerometer. The controller is further configured to determine an RMS average of the acceleration of the accelerometer based upon the acceleration of the hydraulic pump, compare the RMS average of the acceleration of the accelerometer to the fault threshold, and generate an alert signal when the RMS average of the acceleration of the accelerometer exceeds the fault threshold for a time period exceeding the time threshold.

Pump failure differentiation system

A rig management system is disclosed. The rig management system may be configured to receive vibration data from a set of sensors installed on one or more elements of a hydraulic fracturing rig. The rig management system may be configured to perform a first processing of the vibration data utilizing a moving window technique to identify a possible failure of a set of bearings. The rig management system may be configured to perform a second processing utilizing a set of processing techniques after identifying the possible failure of the set of bearings. The rig management system may be configured to determine that the possible failure is an actual failure based on a result of performing the second processing. The rig management system may be configured to perform an action after determining that the possible failure is the actual failure.

Bearing failure detection in a hydraulic fracturing rig

A system for detecting wear or failure of a genset coupling of a genset power system is provided. The system includes a vibration sensor for measuring vibrations, and an additional sensor for measuring an operating condition. A controller is configured to process operating condition data from the additional sensor using a modeling software to generate simulated data. The controller applies time domain information of at least one of the simulated data and vibration sensor data to the modeling software using a machine learning algorithm, and perform a comparison to identify wear or failure of the coupling, wherein, when performing the comparison, the controller compares at least one of: time domain information of the vibration sensor data to time domain information of the simulated data or frequency domain information of the vibration sensor data to frequency domain information of the simulated data, to identify wear or failure of the coupling.

System and method for detecting wear or failure of genset power system coupling

Some examples described herein may involve determining an advance timing window between the valve opening or closing and a designated time that the valve is scheduled to open or close; determining a closing velocity of the valve; monitoring an engine speed of the engine; determining valve lash information based on the advance timing window, the closing velocity, and the engine speed, wherein the valve lash information identifies a magnitude of the valve lash or whether the magnitude of the valve lash associated with the valve satisfies a threshold; and performing an action based on the valve lash information.

Valve lash detection and analysis

A method for detecting a leak in a pump comprises: monitoring discharge pressure time domain signals of the pump; monitoring piston position time domain signals for each piston of a plurality of pistons; filtering the monitored discharge pressure time domain signals and the monitored piston position time domain signals of each piston of the plurality of pistons via a band pass filter; determining, at each pump revolution for each piston, a sum of product of the filtered discharge pressure time domain signals and the filtered piston position time domain signals; determine, at each pump revolution for each piston, an absolute value of each sum; determining, at each pump revolution, a largest absolute value of the determined absolute values; determining a corresponding piston of the plurality of pistons associated with the largest absolute value; and determining a cylinder of the pump associated with the determined corresponding piston as leaking.

System for detecting failure location in a pump

A pumping system for use in a well stimulation application includes a pump controlled by a controller, the controller operating to monitor operation of the pump and determine when cavitation is present or imminent in the pump. When cavitation is present, the controller automatically adjusts an operating condition of the pump reduce pump speed or increase a fluid pressure at the pump inlet.

Well stimulation pump control and method

Operating a pumping system includes moving a pumping element to transition liquid through the pump, and determining a value based at least in part upon inlet pressure and pumping speed that is indicative of a pressure of the liquid within a bore susceptible to cavitation. Pumping speed and/or inlet pressure can be varied responsive to the determined value to limit cavitation.

Cavitation limiting strategies for pumping system

A method, system, and apparatus to monitor health of a compressor are provided. The method, system, and apparatus include detecting whether the compressor has a fault associated using one or more of performance-based modeling and structure-based modeling of physical aspects associated with operation of the compressor; diagnosing any detected fault to determine the cause or causes of the fault; and evaluating each diagnosed fault to assess the significance of the fault. Data pertaining to the cause or causes of the fault and the significance of the fault is used to output health information corresponding to a health state or status of the compressor. The method, system, and apparatus also predict future health-related conditions of the compressor, and data pertaining to the predicted future health-related conditions is used to output health information corresponding to a future health state or status of the compressor.

Growing community of inventors

Dunlap, IL, United States of America

Xuefei Hu