Description:BACKGROUND
Field of Invention
[001] Embodiments of the present invention generally relate to a test bench for ball bearing rings and particularly to a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals.
Description of Related Art
[002] Bearing faults in industrial machinery can lead to costly downtime, production delays, and potential safety hazards. Early detection and diagnosis of these faults are critical for maintaining operational efficiency and preventing catastrophic failures. Traditional methods of detecting bearing faults often rely on manual inspection or rudimentary vibration analysis techniques, which may not be sensitive enough to identify subtle or early-stage faults.
[003] In recent years, there has been a growing interest in leveraging machine learning techniques for automated fault detection in machinery. Among these techniques, decision tree algorithms have shown promise due to their ability to handle complex datasets and extract meaningful patterns from raw sensor data. Decision trees partition the feature space into regions based on the values of input variables, making them particularly well-suited for classification tasks such as fault detection.
[004] However, implementing a robust and accurate fault detection system based on machine learning presents several challenges. One of the key challenges is the selection and extraction of relevant features from the raw vibrational data. This process requires domain expertise and careful consideration of various signal processing techniques to highlight fault signatures while minimizing noise and irrelevant information.
[005] There is thus a need for an improved and advanced leveraging decision tree machine learning framework to detect bearing faults via vibrational signals that can administer the aforementioned limitations in a more efficient manner.
SUMMARY
[006] Embodiments in accordance with the present invention provide a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals. The framework comprising: a test bearing worn on a shaft. The shaft is connected to a motor using a coupler and the motor is adapted to rotate the shaft bearing the test bearing at a standard velocity. The framework further comprising: an accelerometer adapted to capture vibration data in the test bearing while in rotation. The framework further comprising: a microphone adapted to capture noise data in the test bearing while in the rotation. The framework further comprising: a data acquisition card adapted to receive the captured vibration data and the captured noise data from the accelerometer and the microphone respectively. The framework further comprising: a computer device, connected with the data acquisition card, and adapted to receive the vibration data and the noise data from the data acquisition card, wherein the computer device comprises a LabVIEW software. The framework further comprising: a processing unit installed in the computer device. The processing unit is configured to: receive the vibration data and the noise data from the data acquisition card; conduct a time domain analysis on the received vibration data and the noise data to compute statistical parameters; feed the statistical parameters into a machine learning algorithm with fault classification, wherein the machine learning algorithm classifies fault in the test bearing as a Type-1 fault or a Type-2 fault; and plot a time-amplitude graph using the computed statistical parameters for the test bearing.
[007] Embodiments in accordance with the present invention further provide a method for detecting bearing faults via vibrational signals by leveraging a decision tree machine learning framework. The method comprising steps of: receiving a vibration data and a noise data from a data acquisition card; conducting a time domain analysis on the received vibration data and the noise data to compute statistical parameters; feeding the computed statistical parameters into a machine learning algorithm with fault classification; and plotting a time-amplitude graph using the computed statistical parameters for a test bearing.
[008] Embodiments of the present invention may provide a number of advantages depending on their particular configuration. First, embodiments of the present application may provide a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals.
[009] Next, embodiments of the present application may provide a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals that proposes a solution that lies in its enhanced accuracy and reliability in identifying bearing faults. By harnessing the power of the Decision Tree Machine Learning Framework, the solution leverages its inherent ability to effectively analyze complex datasets and extract meaningful patterns. This enables the system to discern even subtle anomalies in the vibrational signals emitted by faulty bearings, thereby improving the overall fault detection capabilities.
[0010] Next, embodiments of the present application may provide a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals that exhibit a remarkable level of versatility and adaptability. It can handle various types of bearing faults, ranging from minor defects to severe damages. This flexibility ensures that the solution is not limited to specific fault scenarios, making it highly applicable in diverse industrial settings.
[0011] Next, embodiments of the present application may provide a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals that demonstrates a notable reduction in the computational requirements and processing time. Through the intelligent utilization of the Decision Tree Machine Learning Framework, the solution streamlines the analysis process by eliminating redundant or irrelevant features, effectively focusing on the most significant characteristics of the vibrational signals. This optimization leads to expedited fault detection and diagnosis, resulting in improved operational efficiency.
[0012] Next, embodiments of the present application may provide a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals that integrate advanced feature engineering techniques, enhancing the system's ability to extract meaningful information from the raw vibrational signals. By incorporating domain-specific knowledge and utilizing innovative signal processing algorithms, the solution effectively captures and extracts distinctive fault signatures, thus enabling more accurate fault classification and diagnosis.
[0013] Next, embodiments of the present application may provide a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals that offer a significant advantage over previous approaches. By leveraging the inherent interpretability of decision trees, the system provides clear and transparent insights into the underlying decision-making process. This enables technicians and domain experts to comprehend and validate the reasoning behind the fault detection outcomes, ultimately fostering trust in the system's capabilities.
[0014] These and other advantages will be apparent from the present application of the embodiments described herein.
[0015] The preceding is a simplified summary to provide an understanding of some embodiments of the present invention. This summary is neither an extensive nor exhaustive overview of the present invention and its various embodiments. The summary presents selected concepts of the embodiments of the present invention in a simplified form as an introduction to the more detailed description presented below. As will be appreciated, other embodiments of the present invention are possible utilizing, alone or in combination, one or more of the features set forth above or described in detail below.
BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The above and still further features and advantages of embodiments of the present invention will become apparent upon consideration of the following detailed description of embodiments thereof, especially when taken in conjunction with the accompanying drawings, and wherein:
[0017] FIG. 1A illustrates a leveraging decision tree machine learning framework to detect bearing faults via vibrational signals, according to an embodiment of the present invention;
[0018] FIG. 1B illustrates a healthy bearing, according to an embodiment of the present invention;
[0019] FIG. 1C illustrates a type-1 fault in the bearing, according to an embodiment of the present invention;
[0020] FIG. 1D illustrates a type-2 fault in the bearing, according to an embodiment of the present invention;
[0021] FIG. 1E illustrates a first time-amplitude graph of the healthy bearing, according to an embodiment of the present invention;
[0022] FIG. 1F illustrates a second time-amplitude graph of the bearing with the type-1 fault, according to an embodiment of the present invention;
[0023] FIG. 1G illustrates a third time-amplitude graph of the bearing with the type-2 fault, according to an embodiment of the present invention;
[0024] FIG. 1H illustrates a table comparing the healthy bearing, the bearing with the type-1 fault, and the bearing with the type-2 fault, according to an embodiment of the present invention;
[0025] FIG. 1I illustrates a prediction class, according to an embodiment of the present invention;
[0026] FIG. 2 illustrates a block diagram of a processing unit of the leveraging decision tree machine learning framework to detect bearing faults via vibrational signals, according to an embodiment of the present invention; and
[0027] FIG. 3 depicts a flowchart of a method for detecting bearing faults via vibrational signals by leveraging the decision tree machine learning framework, according to an embodiment of the present invention.
[0028] The headings used herein are for organizational purposes only and are not meant to be used to limit the scope of the description or the claims. As used throughout this application, the word "may" is used in a permissive sense (i.e., meaning having the potential to), rather than the mandatory sense (i.e., meaning must). Similarly, the words “include”, “including”, and “includes” mean including but not limited to. To facilitate understanding, like reference numerals have been used, where possible, to designate like elements common to the figures. Optional portions of the figures may be illustrated using dashed or dotted lines, unless the context of usage indicates otherwise.
DETAILED DESCRIPTION
[0029] The following description includes the preferred best mode of one embodiment of the present invention. It will be clear from this description of the invention that the invention is not limited to these illustrated embodiments but that the invention also includes a variety of modifications and embodiments thereto. Therefore, the present description should be seen as illustrative and not limiting. While the invention is susceptible to various modifications and alternative constructions, it should be understood, that there is no intention to limit the invention to the specific form disclosed, but, on the contrary, the invention is to cover all modifications, alternative constructions, and equivalents falling within the scope of the invention as defined in the claims.
[0030] In any embodiment described herein, the open-ended terms "comprising", "comprises”, and the like (which are synonymous with "including", "having” and "characterized by") may be replaced by the respective partially closed phrases "consisting essentially of", “consists essentially of", and the like or the respective closed phrases "consisting of", "consists of”, the like.
[0031] As used herein, the singular forms “a”, “an”, and “the” designate both the singular and the plural, unless expressly stated to designate the singular only.
[0032] FIG. 1A illustrates a leveraging decision tree machine learning framework 100 (hereinafter referred to as the framework 100) to detect bearing faults via vibrational signals, according to an embodiment of the present invention. In an embodiment of the present invention, the framework 100 may exam a test bearing 102, and may further classify the test bearing 102 as a healthy, the test bearing 102 with a type-1 fault, or the test bearing 102 with a type-2 fault.
[0033] According to embodiments of the present invention, the framework 100 may comprise the test bearing 102, a shaft 104, a motor 106, a coupler 108, an accelerometer 110, a microphone 112, a data acquisition card 114, a computer device 116, a LabVIEW software 118, and a processing unit 120, and a power source 122.
[0034] In an embodiment of the present invention, the test bearing 102 may be worn on the shaft 104. The shaft 104 may be connected to the motor 106 using the coupler 108 and the motor 106 may be adapted to rotate the shaft 104 bearing the test bearing 102 at a standard velocity, in an embodiment of the present invention. In an embodiment of the present invention, the standard velocity may be 1000 units.
[0035] In an embodiment of the present invention, the accelerometer 110 may be adapted to capture vibration data in the test bearing 102 while in rotation. The accelerometer 110 may be a PCB 325c-03 accelerometer, in an embodiment of the present invention.
[0036] In an embodiment of the present invention, the microphone 112 may be adapted to capture noise data in the test bearing 102 while in the rotation.
[0037] In an embodiment of the present invention, the data acquisition card 114 may be adapted to receive the captured vibration data and the captured noise data from the accelerometer 110 and the microphone 112 respectively.
[0038] In an embodiment of the present invention, the computer device 116 may be connected with the data acquisition card 114. The computer device 116 may be adapted to receive the vibration data and the noise data from the data acquisition card 114, in an embodiment of the present invention. In an embodiment of the present invention, the computer device 116 comprises the LabVIEW software 118. In an embodiment of the present invention, the processing unit 120 may be installed in the computer device 116.
[0039] In an embodiment of the present invention, the power source 122 may be adapted to actuate the motor 106.
[0040] FIG. 1B illustrates the test bearing 102 as the healthy bearing, according to an embodiment of the present invention. The healthy bearing may be the test bearing 102 with no fault and may be suitable for usage in any application, in an embodiment of the present invention.
[0041] FIG. 1C illustrates the test bearing 102 with type-1 fault in the test bearing 102, according to an embodiment of the present invention. The type-1 fault in the test bearing 102 may be a crack fault, in an embodiment of the present invention. The test bearing with the type-1 fault may not be suitable for any application.
[0042] FIG. 1D illustrates the type-2 fault in the test bearing 102, according to an embodiment of the present invention. The type-2 fault in the test bearing 102 may be a hole or a vacancy fault, in an embodiment of the present invention. The test bearing 102 with the type-2 fault may not be suitable for any application.
[0043] FIG. 1E illustrates a first time-amplitude graph 124 of the healthy bearing, according to an embodiment of the present invention. The first time-amplitude graph 124 of the healthy bearing may depict a change in amplitude of the healthy bearing while in motion, with per unit passing time. The amplitude pulse for the first time-amplitude graph 124 of the healthy bearing may be concentrated.
[0044] FIG. 1F illustrates a second time-amplitude graph 126 of the test bearing 102 with the type-1 fault, according to an embodiment of the present invention. The second time-amplitude graph 126 of the test bearing 102 with the type-1 fault may depict the change in amplitude of the test bearing 102 with the type-1 fault while in motion, with per unit passing time. The amplitude pulse for the second time-amplitude graph 126 of the test bearing 102 with the type-1 fault may be distributed.
[0045] FIG. 1G illustrates a third time-amplitude graph 128 of the test bearing 102 with the type-2 fault, according to an embodiment of the present invention. The third time-amplitude graph 128 of the test bearing 102 with the type-2 fault may depict the change in amplitude of the test bearing 102 with the type-2 fault while in motion, with per unit passing time. The amplitude pulse for the third time-amplitude graph 128 of the test bearing 102 with the type-2 fault may be distributed.
[0046] FIG. 1H illustrates a table 130 comparing the healthy bearing, the test bearing 102 with the type-1 fault, and the test bearing 102 with the type-2 fault, according to an embodiment of the present invention. The table 130 may depict a comparison among the statistical parameters of the healthy bearing, the test bearing 102 with the type-1 fault, and the test bearing 102 with the type-2 fault. The statistical parameters may be selected from kurtosis, skewness, crest factor, mean, Root Mean Square (RMS), peak value, variance, standard deviation and organizing, and so forth.
[0047] In a preferred embodiment of the present invention, the Root Mean Square (RMS) for the healthy bearing may be 0.0234, for the test bearing 102 with the type-1 fault may be 0.0361, and for the test bearing 102 with the type-2 fault may be 0.0304.
[0048] In a preferred embodiment of the present invention, the mean for the healthy bearing may be 0.5235, for the test bearing 102 with the type-1 fault may be 0.1234, and for the test bearing 102 with the type-2 fault may be 0.1001.
[0049] In a preferred embodiment of the present invention, the peak value for the healthy bearing may be 0.0776, for the test bearing 102 with the type-1 fault may be 0.1434, and for the test bearing 102 with the type-2 fault may be 0.1410.
[0050] In a preferred embodiment of the present invention, the crest factor for the healthy bearing may be 4.2240, for the test bearing 102 with the type-1 fault may be 5.1250, and for the test bearing 102 with the type-2 fault may be 6.4403.
[0051] In a preferred embodiment of the present invention, the skewness for the healthy bearing may be -0.0012, for the test bearing 102 with the type-1 fault may be -0.1250, and for the test bearing 102 with the type-2 fault may be 0.0150.
[0052] In a preferred embodiment of the present invention, the kurtosis for the healthy bearing may be 2.7532, for the test bearing 102 with the type-1 fault may be 8.3250, and for the test bearing 102 with the type-2 fault may be 6.3250.
[0053] In a preferred embodiment of the present invention, the variance for the healthy bearing may be 0.0240, for the test bearing 102 with the type-1 fault may be 0.2459, and for the test bearing 102 with the type-2 fault may be 0.1540.
[0054] In a preferred embodiment of the present invention, the standard deviation for the healthy bearing may be 0.0240, for the test bearing 102 with the type-1 fault may be 0.0451, and for the test bearing 102 with the type-2 fault may be 0.0350.
[0055] FIG. 1I illustrates a prediction class, according to an embodiment of the present invention. The prediction may represent a machine learning algorithm classifying the fault in the test bearing 102 as the type-1 fault or the type-2 fault.
[0056] FIG. 2 illustrates a block diagram of the processing unit 120 of the framework 100, according to an embodiment of the present invention. The processing unit 120 may comprise the computer-executable instructions in form of programming modules such as a data receiving module 200, a data analysis module 202, a data feeding module 204, and a graph plotting module 206.
[0057] In an embodiment of the present invention, the data receiving module 200 may be configured to receive the vibration data and the noise data from the data acquisition card 114.
[0058] In an embodiment of the present invention, the data analysis module 202 may be configured to conduct the time domain analysis on the received vibration data and the noise data to compute statistical parameters.
[0059] In an embodiment of the present invention, the data feeding module 204 may be configured to feed the statistical parameters into the machine learning algorithm with fault classification. The machine learning algorithm may classify the fault in the test bearing 102 as the Type-1 fault or the Type-2 fault.
[0060] In an embodiment of the present invention, the graph plotting module 206 may be configured to plot the time-amplitude graph using the computed statistical parameters for the test bearing 102.
[0061] FIG. 3 depicts a flowchart of a method 300 for detecting bearing faults via vibrational signals by leveraging the framework 100, according to an embodiment of the present invention.
[0062] At step 302, the framework 100 may receive the vibration data and the noise data from the data acquisition card 114.
[0063] At step 304, the framework 100 may conduct the time domain analysis on the received vibration data and the noise data to compute statistical parameters.
[0064] At step 306, the framework 100 may feed the statistical parameters into the machine learning algorithm with fault classification to classify faults in the test bearing 102 as the Type-1 fault or the Type-2 fault.
[0065] At step 308, the framework 100 may plot the time-amplitude graph using the computed statistical parameters for the test bearing 102.
[0066] While the invention has been described in connection with what is presently considered to be the most practical and various embodiments, it is to be understood that the invention is not to be limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the scope of the appended claims.
[0067] This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined in the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements within substantial differences from the literal languages of the claims. , Claims:CLAIMS
I/We Claim:
1. A leveraging decision tree machine learning framework (100) to detect bearing faults via vibrational signals, the framework (100) comprising:
a test bearing (102) worn on a shaft (104), wherein the shaft (104) is connected to a motor (106) using a coupler (108), and the motor (106) is adapted to rotate the shaft (104) bearing the test bearing (102) at a standard velocity;
an accelerometer (110) adapted to capture vibration data in the test bearing (102) while in rotation;
a microphone (112) adapted to capture noise data in the test bearing (102) while in the rotation;
a data acquisition card (114) adapted to receive the captured vibration data and the captured noise data from the accelerometer (110) and the microphone (112) respectively;
a computer device (116), connected with the data acquisition card (114), and adapted to receive the vibration data and the noise data from the data acquisition card (114), wherein the computer device (116) comprises a LabVIEW software (118); and
a processing unit (120) installed in the computer device (116), characterized in that the processing unit (120) is configured to:
receive the vibration data and the noise data from the data acquisition card (114);
conduct a time domain analysis on the received vibration data and the noise data to compute statistical parameters;
feed the statistical parameters into a machine learning algorithm with fault classification, wherein the machine learning algorithm classifies fault in the test bearing (102) as a Type-1 fault or a Type-2 fault; and
plot a time-amplitude graph using the computed statistical parameters for the test bearing (102).
2. The framework (100) as claimed in claim 1, wherein the accelerometer (110) is a PCB 325c-03 accelerometer.
3. The framework (100) as claimed in claim 1, wherein the machine learning algorithm classifies the test bearing (102) as a healthy bearing.
4. The framework (100) as claimed in claim 1, wherein the data acquisition card (114) is a 4-channel NI 9234 DAQ card.
5. The framework (100) as claimed in claim 1, wherein the statistical parameters are selected from kurtosis, skewness, crest factor, mean, Root Mean Square (RMS), peak value, variance, standard deviation, and organizing, or a combination thereof.
6. The framework (100) as claimed in claim 1, wherein the standard velocity of the motor (106) for rotating the test bearing (102) is 1000 units.
7. The framework (100) as claimed in claim 1, wherein the motor (106) is actuated using a power source (122).
8. A method (300) for detecting bearing faults via vibrational signals by leveraging a decision tree machine learning framework (100), the method (300) is characterized by steps of:
receiving a vibration data and a noise data from a data acquisition card (114);
conducting a time domain analysis on the received vibration data and the noise data to compute statistical parameters;
feeding the computed statistical parameters into a machine learning algorithm with fault classification; and
plotting a time-amplitude graph using the computed statistical parameters for a test bearing (102).
9. The method (300) as claimed in claim 8, wherein the machine learning algorithm classifies fault in the test bearing (102) as a Type-1 fault or a Type-2 fault.
10. The method (300) as claimed in claim 8, wherein the data acquisition card (114) is a 4-channel NI 9234 DAQ card.
Date: May 22, 2024
Place: Noida

Dr. Keerti Gupta
Agent for the Applicant
(IN/PA-1529)