Methods and apparatuses for defect diagnosis in a mechanical system

A method of detecting a defect in a drilling rig, the method comprising:

• measuring a first signal, during operation of the drilling rig, to produce a time sampled first representation of motion associated with a moving component of the drilling rig;
• measuring a second signal, during operation of the drilling rig, to produce a time sampled second representation of motion associated with the moving component of the drilling rig;
• calculating a set of wavelet envelopes of the time sampled first representation;
• converting the set of wavelet envelopes from a time domain to a spatial domain based on measured samples of the time sampled second representation of motion associated with the moving component;
• determining one or more spectral energy concentrations of the converted wavelet envelopes associated with at least one expected defect frequency of the moving component, the one or more spectral energy concentrations having increased amplitudes relative to amplitudes of spectral energy concentrations, produced using a conventional enveloping order spectrum method, at the at least one expected defect frequency;
• and
• determining whether a defect in the moving component exists based on the determined one or more spectral energy concentrations with improved recognition accuracy of the defect relative to a determination based on the conventional enveloping order spectrum method and reporting whether the defect exists in the moving component of the drilling rig to a human-machine interface, enabling real-time condition monitoring and health diagnosis of the drilling rig.

A method of detecting a defect in a drilling rig, the method comprising:

• producing electric current signal measurements by measuring an electric current signal of a motor of the drilling rig, the electric current signal measured during operation of the drilling rig by a current probe, the current probe electrically coupled to the motor;
• calculating a motor current envelope from the electric current signal measurements;
• calculating one or more motor current envelope spectral energy concentrations of the motor current envelope at one or more expected defect frequencies of the motor, the motor current envelope being effective in producing the one or more motor current envelope spectral energy concentrations at the one or more expected defect frequencies with increased amplitude relative to motor current spectral energy concentrations at the one or more expected defect frequencies in a motor current spectrum of the motor current signal;
• and
• determining whether a defect in the motor exists based on the calculated one or more motor current envelope spectral energy concentrations, with improved recognition accuracy of motor defects relative to using the motor current spectral energy concentrations, and reporting whether the defect exists in the motor of the drilling rig to a human-machine interface, enabling real-time condition monitoring and health diagnosis of the drilling rig.

A method of detecting a defect in a drilling rig, the method comprising:

• measuring a first signal, during operation of the drilling rig, to produce a time sampled first representation of motion associated with a moving component of the drilling rig;
• measuring a second signal, during operation of the drilling rig, to produce a time sampled second representation of motion associated with the moving component of the drilling rig;
• converting a set of measured samples of the time sampled first representation of motion associated with a moving component of the drilling rig from time domain to spatial domain based on measured samples of the time sampled second representation of motion associated with the moving component;
• calculating a set of wavelet envelopes of the converted set of measured samples of the time sampled first representation of motion associated with the moving component during operation of the drilling rig;
• determining one or more spectral energy concentrations of the calculated wavelet envelopes associated with at least one expected defect frequency of the moving component, the one or more spectral energy concentrations having increased amplitudes relative to amplitudes of spectral energy concentrations, produced using a conventional enveloping order spectrum method, at the at least one expected defect frequency;
• and
• determining whether a defect in the moving component exists based on the calculated one or more energy concentrations with improved recognition accuracy of the defect relative to a determination based on the conventional enveloping order spectrum method and reporting whether the defect exists in the moving component of the drilling rig to a human-machine interface, enabling real-time condition monitoring and health diagnosis of the drilling rig.

An apparatus for detecting a defect in drilling rig, the apparatus comprising:

• means for measuring multi-sensor data during operation of the drilling rig, the at least one component including a motor component;
• means for measuring a motor current signal of the motor component during operation of the drilling rig;
• means for calculating a plurality of features including frequency domain features using, at least in part, the multi-sensor data, the multi-sensor data representing different physical parameters of the at least one component, the frequency domain features including motor current envelope spectral features of a motor current envelope computed from the motor current signal, the motor current envelope being effective in producing the motor current envelope spectral features at defect-related frequencies with increased amplitude relative to motor current spectral features at the defect-related frequencies in a motor current spectrum of the motor current signal;
• means for classifying a subset of the calculated plurality of features, the subset including the motor current envelope spectral frequencies, based on classification patterns constructed from multi-sensor training data corresponding to different operational conditions of the at least one component;
• and
• means for determining whether a defect exists in the at least one component based on the classified subset of features including the motor current envelope spectral frequencies from the motor current envelope, with improved recognition accuracy of motor defects relative to using the motor current spectral features from the motor current spectrum, and reporting whether the defect exists in the at least one component of the drilling rig to a human-machine interface, enabling real-time condition monitoring and health diagnosis of the drilling rig.

An apparatus comprising a memory with computer code instructions stored thereon and a processor, the memory and computer code instructions, with the processor, being configured to cause the apparatus to:

• measure, during operation of a drilling rig, multi-sensor data from multiple sensors of different sensor types associated with at least one component of the drilling rig, the at least one component including a motor component, the multiple sensors including a current probe electrically coupled to the motor component;
• measure a motor current signal of the motor component via the current probe during operation of the drilling rig;
• calculate a plurality of features including frequency domain features using, at least in part, the multi-sensor data, the multi-sensor data representing different physical parameters of the at least one component, the frequency domain features including motor current envelope spectral features of a motor current envelope computed from the motor current signal, the motor current envelope being effective in producing the motor current envelope spectral features at defect-related frequencies with increased amplitude relative to motor current spectral features at the defect-related frequencies in a motor current spectrum of the motor current signal;
• classify a subset of the calculated plurality of features, the subset including the motor current envelope spectral features, based on classification patterns constructed from multi-sensor training data corresponding to different operational conditions of the at least one component;
• and
• determine whether a defect exists in the at least one component based on the classified subset of features including the motor current envelope spectral features from the motor current envelope, with improved recognition accuracy of motor defects relative to using the motor current spectral features from the motor current spectrum, and reporting whether the defect exists in the at least one component of the drilling rig to a human-machine interface, enabling real-time condition monitoring and health diagnosis of the drilling rig.

A computer-readable medium comprising computer code instructions stored thereon, the computer code instructions when executed by a processor cause an apparatus to:

• measure, during operation of a drilling rig, multi-sensor data from multiple sensors of different sensor types associated with at least one component of the drilling rig;
• calculate a plurality of features including frequency domain features using, at least in part, the multi-sensor, the at least one component including a motor component, the multi-sensor data representing different physical parameters of the at least one component, the multiple sensors including a current probe, the current probe electrically coupled to the motor component of the drilling rig and configured to measure a motor current signal of the motor component during operation of the drilling rig, the frequency domain features including motor current envelope spectral features of a motor current envelope computed from the motor current signal, the motor current envelope being effective in producing the motor current envelope spectral features at defect-related frequencies with increased amplitude relative to motor current spectral features at the defect-related frequencies in a motor current spectrum of the motor current signal;
• classify a subset of the calculated plurality of features, the subset including the motor current envelope spectral features, based on classification patterns constructed from multi-sensor training data corresponding to different operational conditions of the at least one component;
• and
• determine whether a defect exists in the at least one component based on the classified subset of features including the motor current envelope spectral features from the motor current envelope, with improved recognition accuracy of motor defects relative to using the motor current spectral features from the motor current spectrum, and reporting whether the defect exists in the at least one component of the drilling rig to a human-machine interface, enabling real-time condition monitoring and health diagnosis of the drilling rig.