Methods and systems of diagnosing machine components using analog sensor data and neural network
Systems and methods for data collection in an industrial environment are disclosed. A system can include a plurality of analog sensors, wherein each of the plurality of analog sensors is operationally coupled to a respective data collection point of a machine component, and generates a respective stream of detection values. A data acquisition and analysis circuit can receive the respective stream of detection values and analyze the respective stream of detection values using an expert system analysis circuit, wherein the expert system analysis circuit determines an occurrence of an anomalous condition based on an analysis of the respective stream of detection values, wherein the expert system analysis circuit utilizes a neural network including one of a probabilistic, a time delay, and a convolutional neural network.
1. A system comprising:
a plurality of sensors, wherein each of the plurality of sensors is operationally coupled to a respective data collection point of a machine component in an industrial environment and generates a respective stream of detection values relating to the respective data collection point of the machine component, wherein at least one of the plurality of sensors monitors a rotating machine component, and wherein the respective stream of detection values is digitally sampled and filtered waveform data from a respective one of the plurality of sensors; and
a data acquisition and analysis circuit for receiving the respective stream of detection values and structured to analyze the respective stream of detection values using an expert system analysis circuit, wherein the expert system analysis circuit determines an occurrence of an anomalous condition for the rotating machine component based on an analysis of the respective stream of detection values, wherein the expert system analysis circuit utilizes a neural network including at least one of a probabilistic, a time delay, or a convolutional neural network,
wherein the expert system analysis circuit controls a plurality of data collection bands for determining collection schedules of different groupings of the plurality of sensors, and
wherein the data collection bands each include one or more frequencies to be measured by a respective grouping of the plurality of sensors.
2. The system of claim 1 , wherein the neural network comprises a probabilistic neural network that determines the occurrence of the anomalous condition based on pattern recognition.
3. The system of claim 1 , wherein the neural network comprises a time delay neural network that determines the occurrence of the anomalous condition based on pattern recognition.
4. The system of claim 3 , wherein the time delay neural network is trained with machine learning.
5. The system of claim 1 , further comprising an analyzed stream of detection values that represents sound.
6. The system of claim 1 , wherein the neural network is a convolutional neural network that determines the occurrence of the anomalous condition based on pattern recognition.
7. The system of claim 1 , further comprising an analyzed stream of detection values that comprises image data.
8. The system of claim 1 , further comprising an analyzed stream of detection values data comprises video data.
9. The system of claim 1 , wherein one of the plurality of sensors comprises a tri-axial sensor structured to monitor three orthogonal directions of the rotating machine component.
10. The system of claim 1 , wherein the expert system analysis circuit is configured to analyze respective streams of detection values from a first and a second of the plurality of sensors to determine a relative phase at one or more times, and wherein the expert system analysis circuit is further configured to determine a failure state in response to the determined relative phase.
11. The system of claim 1 , wherein the respective stream of detection values includes at least one of an interpolated waveform or a decimated waveform.
12. The system of claim 1 , wherein the one or more frequencies includes at least one of a group of spectral peaks, a true-peak level, a crest factor derived from a time waveform, or an overall waveform derived from a vibration envelope.
13. A computer-implemented method for data collection in an industrial environment, the method comprising:
collecting streams of detection values relating to a plurality of machine components by a plurality of sensors, wherein each of the plurality of sensors is operationally coupled to a respective data collection point of a machine component in the industrial environment and generates a respective stream of detection values relating to the respective data collection point of the machine component, wherein at least one of the plurality of sensors monitors a rotating machine component, and wherein the detection values of at least one of the respective streams of detection values are digitally sampled and filtered waveform data from a respective one of the plurality of sensors;
analyzing the streams of detection values using an expert system analysis circuit, wherein the expert system analysis circuit determines an occurrence of an anomalous condition for the rotating machine component based on an analysis of the streams of detection values, wherein the expert system analysis circuit utilizes a neural network including at least one of: a probabilistic, a time delay, or a convolutional neural network; and
controlling a plurality of data collection bands for determining collection schedules of different groupings of the plurality of sensors, wherein the data collection bands each include one or more frequencies to be measured by a respective grouping of the plurality of sensors.
14. The method of claim 13 , further comprising determining a failure state for the rotating machine component based on the analysis and providing the failure state to a data storage.
15. The method of claim 14 , further comprising analyzing a first stream of detection values corresponding to a first sensor and a second stream of detection values corresponding to a second sensor for a relative phase determination and detecting the failure state for the rotating machine component in response to the relative phase determination.
16. The method of claim 13 , further comprising operating the expert system analysis circuit to control data collection bands of a plurality of input channels.
17. The method of claim 13 , wherein the neural network comprises a probabilistic neural network that determines the occurrence of the anomalous condition based on pattern recognition.
18. The method of claim 13 , wherein the neural network comprises a time delay neural network and an analyzed stream of detection values represents sound.
19. The method of claim 13 , wherein the neural network comprises a convolutional neural network that determines the occurrence of the anomalous condition based on pattern recognition of an analyzed stream of detection values which represents image data.
20. The method of claim 13 , wherein the respective stream of detection values includes at least one of: an interpolated waveform or a decimated waveform.
21. A system comprising:
a plurality of sensors, wherein each sensor is operationally coupled to a respective data collection point of a machine component in an industrial environment and generates a respective stream of detection values relating to the respective data collection point of the machine component, wherein at least one of the sensors monitors a rotating machine component, and wherein the respective stream of detection values is digitally sampled and filtered waveform data from a respective one of the plurality of sensors,
wherein the respective stream of detection values is digitally sampled at an effective sampling rate that is higher at frequency bands proximal to an operating speed of the rotating machine component;
a means for receiving the respective stream of detection values and structured to analyze the respective stream of detection values using an expert system analysis circuit; and
a means for determining an occurrence of an anomalous condition for the rotating machine component based on an analysis of the respective stream of detection values, wherein the means utilize a neural network including at least one of: a probabilistic, a time delay, or a convolutional neural network.
22. The system of claim 21 , further comprising a means for analyzing respective streams of detection values from a first and a second of the plurality of sensors to determine a relative phase at one or more times, and further comprising a means for determining a failure state in response to the determined relative phase.
23. The system of claim 21 , wherein the effective sampling rate is realized by means for interpolating and decimating the respective stream of detection values.