US20160245279A1

US20160245279A1 - Real time machine learning based predictive and preventive maintenance of vacuum pump

Info

Publication number: US20160245279A1
Application number: US14/628,322
Authority: US
Inventors: Biplab Pal; Steve Gillmeister; Amit Purohit
Original assignee: Prophecy Sensors LLC
Current assignee: Machinesense LLC
Priority date: 2015-02-23
Filing date: 2015-02-23
Publication date: 2016-08-25
Also published as: US11002269B2; US20190154032A1

Abstract

A method and system of a machine learning architecture for predictive and preventive maintenance of vacuum pumps. The method includes receiving one of a motor sensor data and a blower sensor data over a communications network. The motor sensor data is classified into one of a vacuum state sensor data and break state sensor data. The vacuum state sensor data is analyzed to detect an operating vacuum level and an alarm is raised when the vacuum state sensor data exceeds a pre-defined safety range. Vacuum break data is classified into one of a clean filter category and clogged filter category and an alarm is raised if an entry under the clogged filter category is detected. The blower sensor data in association with the motor sensor data is analyzed based on machine learning to detect one of a deficient oil level and a deficient oil structure.

Description

FIELD OF TECHNOLOGY

The present invention generally relates to Internet of Things (IoT), and more particularly relates to an IoT-based system for predictive and preventive maintenance of machines that uses a blower and a pump, through machine learning and physics based modeling of physical parameters like vibration, sound, temperature monitored by machine wearable and other related sensors.

BACKGROUND

Internet of Things (IoT) is a network of uniquely-identifiable and purposed “things” that are enabled to communicate data over a communications network without requiring human-to-human or human-to-computer interaction. The “thing” in the Internet of Things may virtually be anything that fits into a common purpose thereof. For example, a “thing” could be a person with a heart rate monitor implant, a farm animal with a biochip transponder, an automobile comprising built-in sensors configured to alert its driver when the tire pressure is low, or the like, or any other natural or man-made entity that can be assigned with a unique IP address and provided with the ability to transfer data over a network. Notably, if all the entities in an IoT are machines, then the IoT is referred to as a Machine to Machine (M2M) IoT or simply, as M2M IoT.
It is apparent from the aforementioned examples that an entity becomes a “thing” of an M2M IoT especially, when the entity is attached with one or more sensors capable of capturing one or more types of data pertaining to: segregation of the data (if applicable); selective communication of each segregation of data to one or more fellow “things”; reception of one or more control commands (or instructions) from one or more fellow “things” wherein, the control commands are based on the data received by the one or more fellow “things”; and execution of the commands resulting in manipulation or “management” of an operation of the corresponding entity. Therefore, in an IoT-enabled system, the “things” basically manage themselves without any human intervention, thus drastically improving the efficiency thereof.
EP Patent No. 1836576 B1 discusses a diagnostic method of failure protection of vacuum pumps. Based on comparison of the currently evaluated diagnostics analysis results and an initial data, maintenance engineers would decide the replacement of the considered vacuum pump, according to the evaluated pump performance indicators. However, in this prior art invention, there is no mention of machine learning or use of machine wearable sensors. Also, the remedial decisions are left to the maintenance engineers.
US Patent application 20120209569 A1 discusses a method for predicting a failure in rotation of a rotor of a vacuum pump. The prior art invention fails to disclose machine learning capabilities and also is dependent on an observation time prediction window. Further, the prior art fails to disclose machine wearable sensors and Internet of things.
U.S. Pat. No. 7,882,394 B2 discusses fault diagnostics through a data collection module. The prior art discloses a system for condition monitoring and fault diagnosis that includes: a data collection function that acquires time histories of selected variables for one or more of components; a pre-processing function that calculates specified characteristics of time histories; an analysis function for evaluating the characteristics to produce one or more hypotheses of a condition of the components, and a reasoning function for determining the condition of the components from one or more hypotheses. The prior art invention however, fails to suggest the concept of IoT. Further, the prior art invention does not mention machine learning for effective predictive or preventive maintenance of vacuum pumps or similar devices.
It is evident from the discussion of the aforementioned prior art that none of them discloses or suggests regarding predictive and preventive maintenance of vacuum pumps through machine learning. Therefore, there is a need in the art for a solution to the aforementioned problem.

SUMMARY

A method of machine learning architecture according to the present invention includes a step of: receiving data from machine wearable sensors placed on a motor (henceforth motor sensor data) and a blower (henceforth a blower sensor data) over a communications network. The machine wearable sensors can be selected from a group consisting of vibration sensors, temperature sensors, magnetic field sensors, gyroscope and its combinations thereof. The sensor type can be single silicon or MEMS (Micro-electromechanical systems) type. The motor or blower sensor data is classified into one of a vacuum state sensor data and vacuum break state (where rotor is switched off and the revolution of the rotor is gradually damping in vacuum medium) sensor data, wherein the vacuum state sensor data is further analyzed to detect an operating vacuum level and an alarm is raised when the vacuum state sensor data exceeds a pre-defined safety range. The vacuum break state sensor data is then classified into clean filter category and clogged filter category and an alarm is raised if the real time data of sensors belonging to clogged filter category is detected. Vacuum state data is further classified based on a multi-class learning model which classifies a pump running with machine oil into clean, old, leaked or overfilled classes. If the sensor data suggest neither, it is classified under uncategorized bearing issues. The blower sensor data in association with the motor sensor data is analyzed based on machine learning to detect deficient oil level and deficient oil structure.
An IoT based machine learning architecture according to an embodiment of the present invention includes: a vacuum pump associated with a blower and a motor coupled with one or more machine wearable sensors; a communications network; and a mobile application associated with a mobile device. The mobile application is communicatively coupled to one or more machine wearable sensors, over the communications network. The mobile application can be replaced by a PC based communication such as a PC based app, as well. The machine learning architecture receives sensor data from the blower and the motor and classifies the motor sensor data into vacuum state sensor data and break state sensor data. Also, the machine learning architecture analyzes the vacuum state sensor data to detect an operating vacuum level and an alarm is_raised when the vacuum state sensor data exceeds a pre-defined safety range. The machine learning architecture classifies vacuum break data into clean filter category and clogged filter category and an alarm is raised if an entry under the clogged filter category is detected, and the machine learning architecture further analyzes the blower sensor data in association with the motor sensor data through a machine learning algorithm in order to detect at least one of a deficient oil level and a deficient oil structure.
The present invention relates to an Internet of Things (IoT) based system for overseeing process control and predictive maintenance of a machine or a network of machines by employing machine wearable sensors. The IoT based system includes a plurality of machine-wearable sensors, secured to the exterior of the machine. These sensors can be any combination of Temperature sensors, Vibration sensors, Magnetometer, Gyroscope. Each sensor is capable of transmitting captured data wirelessly over a communications network. The IoT based system further includes a sensor network for receiving and transmitting the captured data over a communications network. The system also includes: a machine learning algorithm engine capable of receiving data from the sensor network and processing the received data to recognize one of a pattern and a deviation to issue an alarm and appropriate control commands pertaining to the machine. The system further includes one or more control modules disposed in operative communication with the control panel of the machine, wherein the control module is capable of receiving the control commands over a communications network and executing the control commands.
The methods and systems disclosed herein may be implemented in any means for achieving various aspects, and may be executed in a form of a machine-readable medium embodying a set of instructions that, when executed by a machine, cause the machine to perform any of the operations disclosed herein. Other features will be apparent from the accompanying drawings and from the detailed description that follows.

BRIEF DESCRIPTION OF THE DRAWINGS

The embodiments of this invention are illustrated in a non-limiting but in a way of example, in the figures of the accompanying drawings, in which like references indicate similar elements and in which:

FIG. 1 is a diagrammatic representation of a machine learning architecture, according to one or more embodiments.

FIG. 2 is a diagrammatic representation of a data processing system capable of processing a set of instructions to perform any one or more of the methodologies herein, according to one embodiment.

FIG. 3 is a process flow diagram detailing the operations of a method of a machine learning architecture, according to one or more embodiments.

FIG. 4 is an exemplary representation of data on a mobile application associated with the machine learning architecture, according to one or more embodiments.

FIG. 5 is an exemplary representation of a mobile status dashboard for a dryer. The mobile status dashboard displays a status of a process anytime, anywhere through a connection with an internet.

FIG. 6 is an exemplary representation of historical records of alarms/issues any time as displayed on a Mobile application associated with a mobile device generated automatically by a Dryer, according to one or more embodiments.

FIG. 7 is an exemplary representation of historical records of alarms/issues any time as displayed on a Mobile application associated with a mobile device generated automatically by a Vacuum pump, according to one or more embodiments.

FIG. 8 is a representation of a mobile application tracking abusive operations for preventive maintenance so that pumps may last longer from real time vibration data, according to one embodiment.

FIG. 9 is a representation of a real time status of a vacuum pump as seen on a mobile application, according to one embodiments.

FIG. 10 is a representation of real time trend of drying-indicating clean dryer against dryers with clogged filter, according to one embodiment.

FIG. 11 is a representation of normal oil levels against low oil levels, according to one embodiment.

FIG. 12 is a clustering diagram showing pump temperature and vibration indicating different possible cluster of operation based on oil levels, according to one embodiment.

FIG. 13 is a representation of normalized vibration against time, according to one embodiment.

FIG. 14A through FIG. 14C illustrates graphical representations of vibration fault detection using transformed vibrational data based on PCA (Principal component analysis) transformation of X, Y and Z axis components, respectively of the vibrational data at a first vacuum level.

FIG. 15A through FIG. 15C illustrates graphical representations of vibration fault detection using transformed vibrational data based on PCA (Principal component analysis) transformation of X, Y and Z axis components, respectively of the vibrational data at a second vacuum level.

FIG. 16A through FIG. 16D illustrates an exemplary graphical representation of pressure state alarm, filter alarm, oil state alarm, and blower alarm when a bad oil is detected with respect to a silencer.

FIG. 17A through FIG. 17D illustrates graphical representations of pressure state alarm, filter alarm, oil state alarm, and blower alarm, respectively, when a legacy pump has a clogged filter.

FIG. 18 is a graphical representation of overfill alarm, according to one embodiment. Other features of the present embodiments will be apparent from the accompanying drawings and from the detailed description that follows.

DETAILED DESCRIPTION

Example embodiments, as described below, may be used to provide a method, an apparatus and/or a system of real time machine learning based predictive and preventive maintenance of a vacuum pump. Although the present embodiments have been described with reference to specific exemplary embodiments, it will be evident that various modifications and changes may be made to these embodiments without departing from the broader spirit and scope of the invention.
FIG. 1 is a system diagram of a machine learning architecture, according to one embodiment. The machine learning architecture 100 may include a vacuum pump 106 associated with a blower 110 and motor 108. The architecture 100 may include one or more machine wearable sensors 114, 112 coupled respectively to the blower 110 and the motor 108 of the vacuum pump, a communications network 102, and a mobile application 118 associated with a mobile device. The mobile application 118 may be communicatively coupled to the machine wearable sensors 112, 114 over the communications network 102. The machine learning architecture 100 may receive sensor data from the blower 110 and the motor 108 and classifies the motor sensor data into vacuum state sensor data and break state sensor data. The machine learning architecture 100 analyzes the vacuum state sensor data to detect an operating vacuum level and an alarm is raised when the vacuum state sensor data exceeds a pre-defined safety range. A computer database 116 in communication with the mobile application 118 through the communication network 102 and machine learning algorithm engine 104. The machine learning architecture 100 classifies vacuum break data into clean filter category and clogged filter category and an alarm is raised if an entry under the clogged filter category is detected. The machine learning architecture 100 analyzes the blower sensor data in association with the motor sensor data through a machine learning algorithm in order to detect a deficient oil level and a deficient oil structure.
In one or more embodiments, the motor sensor data may be determined from a machine wearable sensor placed on the motor. Similarly, the blower sensor data may also be determined from a machine wearable sensor placed on the blower. The communications network may include WiFi, 2G, 3G, 4G, GPRS, EDGE, Bluetooth, ZigBee, Piconet of BLE, Zwave, or a combination thereof.
In one or more embodiments, the machine learning architecture may be associated with a machine learning algorithm. The motor sensor data and blower sensor data may be received over a communications network onto a mobile application coupled to a mobile device. The alarm may be raised over the communications network through one of a notification on the mobile application including Short message service (SMS), email, or a combination thereof.
In one or more embodiments, machine learning of a vibrational data may comprise of information related to shape factor of the vibration calculated as a ratio of moving RMS (Root mean square) value to moving average of absolute value.
FIG. 2 is a diagrammatic representation of a data processing system capable of processing a set of instructions to perform any one or more of the methodologies herein. FIG. 2 shows a diagrammatic representation of a machine in the exemplary form of a computer system 200 within which a set of instructions, for causing the machine to perform any one or more of the methodologies discussed herein, may be executed. In various embodiments, the machine operates as a standalone device and/or may be connected (e.g., networked) to other machines.
In a networked deployment, the machine may operate in the capacity of a server and/or as a client machine in server-client network environment, and or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal-computer (PC), a tablet PC, a set-top box (STB), a Personal Digital Assistant (PDA), a cellular telephone, a web appliance, a network router, switch and or bridge, an embedded system and/or any machine capable of executing a set of instructions (sequential and/or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term “machine” shall also be taken to include any collection of machines that individually and/or jointly execute a set (or multiple sets) of instructions to perform any one and/or more of the methodologies discussed herein.
The exemplary computer system 200 includes a processor 202 (e.g., a central processing unit (CPU) a graphics processing unit (GPU) and/or both), a main memory 204 and a static memory 206, which communicate with each other via a bus 208. The computer system 200 may further include a video display unit 210 (e.g., a liquid crystal displays (LCD) and/or a cathode ray tube (CRT)). The computer system 200 also includes an alphanumeric input device 212 (e.g., a keyboard), a cursor control device 214 (e.g., a mouse), a disk drive unit 216, a signal generation device 218 (e.g., a speaker) and a network interface device 220.
The disk drive unit 216 further includes a machine-readable medium 222 on which one or more sets of instructions 224 (e.g., software) embodying any one or more of the methodologies and/or functions described herein is stored. The instructions 224 may also reside, completely and/or at least partially, within the main memory 204 and/or within the processor 202 during execution thereof by the computer system 200. The main memory 204 and the processor 202 also constituting machine-readable media.
The instructions 224 may further be transmitted and/or received over a network 226 via the network interface device 220. While the machine-readable medium 222 is shown in an exemplary embodiment to be a single medium, the term “machine-readable medium” should be taken to include a single medium and/or multiple media (e.g., a centralized and/or distributed database, and/or associated caches and servers) that store the one or more sets of instructions. The term “machine-readable medium” shall also be taken to include any medium that is capable of storing, encoding and/or carrying a set of instructions for execution by the machine and that causes the machine to perform any one or more of the methodologies of the various embodiments. The term “machine-readable medium” shall accordingly be taken to include, but not be limited to, solid-state memories, optical and magnetic media, and carrier wave signals.
FIG. 3 is a process flow diagram detailing the operations of one of a method of the machine learning architecture, according to one or more embodiments. In one embodiment, the method of machine learning architecture includes: receiving a motor sensor data and a blower sensor data over a communications network 302; classifying the motor sensor data into one of a vacuum state sensor data and break state sensor data 304; and analyzing the vacuum state sensor data to detect an operating vacuum level 306 and an alarm is raised when the vacuum state sensor data exceeds a pre-defined safety range. Further, the method includes classifying vacuum break data into clean filter category and clogged filter category 308 and an alarm is raised if an entry under the clogged filter category is detected. Also, the method includes analyzing the blower sensor data in association with the motor sensor data based on machine learning to detect deficient oil level and deficient oil structure 310.
In one or more embodiments, the method of machine learning architecture may also include determining motor sensor data from machine wearable sensor placed on the motor and determining blower sensor data from machine wearable sensor placed on the blower.
In one or more embodiments, the communications network may include WiFi, 2G, 3G, 4G, GPRS, EDGE, Bluetooth, ZigBee, Piconet of BLE, Zwave, or a combination thereof. The machine learning architecture may be associated with a machine learning algorithm.
In one or more embodiments, the motor sensor data and blower sensor data may be received over a communications network onto a mobile application associated with a mobile device and an alarm may be raised over the communications network through one of a notification on the mobile application, Short message service (SMS), email, or a combination thereof.
In an exemplary embodiment, the Internet of Things (IoT) based system may include machine wearable sensors. Further, the IoT system may be used for overseeing process control and predictive maintenance of a machine or a network of machines. The system may include a plurality of machine-wearable sensors, each of which secured to the exterior of the machine. Each sensor may be capable of transmitting captured data wirelessly over a communications network. The system may further include a sensor network for receiving and transmitting the captured data over a communications network and a machine learning algorithm engine capable of receiving data from the sensor network. The machine learning algorithm engine may process the received data to recognize one of a pattern and a deviation to issue control commands pertaining to the machine. Lastly, the system may include one or more control modules disposed in operative communication with a control panel of the machine, wherein the control module is capable of receiving control commands over a communications network and executing the control commands.
In an exemplary embodiment, the machine learning algorithm engine may raise an alarm when one of a filter is clogged and deficient oil is detected, wherein the deficient oil may be one of a low oil level and an overused oil structure. The plurality of machine wearable sensors may include motor sensors and blower sensors. The machine learning algorithm engine associated with the IoT based system may issue commands based on a learning outcome from the motor sensor data and the blower sensor data. The learning outcome may be dependent on recognition of one of a pattern and deviation by the machine learning algorithm engine.
In an exemplary embodiment, the machine learning algorithm engine may include three layers which may be used for predictive and preventive maintenance of vacuum pump.
The machine learning algorithm engine may deploy three layers of supervised machine learning for predictive and preventive maintenance. Layer one of the supervised machine learning may receive vibration data from motor and/or blower, the vibration data may be classified into vacuum state and vacuum break state. In one or more embodiments, the vacuum break state may be a state in which vacuum is released periodically.
In layer two, a motor vibration data from vacuum state may be classified to detect operating vacuum level such as −8 inch or −12 inch of mercury. Then depending on a safety range (E.g.: between −7 to −12 inch of mercury), an alarm may be raised and conveyed to the users via a mobile application.
In layer three, vacuum break data may be classified into clean filter category and clogged filter category using a supervised machine learning. If the clog filter category is detected then an alarm may be raised.
Blower Temperature and blower vibration data may be used for classification of a bad and/or low oil level. Bad oil level may increase the friction and thereby raise the surface temp of a blower. The machine learning based classification includes an oblique and/or support vector machine. Support vector machines may be supervised learning models with associated learning algorithms that analyze data and recognize patterns. The supervised learning models may be used for classification and regression analysis.
Motor vibration data may not be affected by bad oil. However, blower vibration data may get affected by bad oil. Therefore motor vibration data may be indicative of a particular vacuum pressure level. By comparing the blower data for good and bad oil using supervised machine learning, operation with bad oil may be detected.
FIG. 4 is an exemplary representation of data on a mobile application associated with the machine learning architecture, according to one or more embodiments. FIG. 4 shows a sensor data representation on a mobile application associated with a mobile device.
FIG. 5 is an exemplary representation of a mobile status dashboard for a dryer. The mobile status dashboard displays a status of a process anytime, anywhere through a connection with an internet.
FIG. 6 is an exemplary representation of historical records of alarms/issues any time as displayed on a Mobile application associated with a mobile device, generated automatically by a Dryer, according to one or more embodiments.
FIG. 7 an exemplary representation of historical records of alarms/issues any time as displayed on a Mobile application associated with a mobile device, generated automatically by a Vacuum pump, according to one or more embodiments.
FIG. 8 is a representation of a mobile application tracking abusive operations for preventive maintenance, according to one embodiment.
FIG. 9 is a representation of a real time status of a vacuum pump as seen on a mobile application, according to one embodiment.
FIG. 10 is a representation of Real time trend of drying-indicating clean dryer against dryers with clogged filter, according to one embodiment.
FIG. 11 is a representation of normal oil levels against low oil levels, according to one embodiment.
FIG. 12 is a clustering diagram showing pump Temperature and vibration indicating different possible cluster of operation based on oil levels, according to one embodiment.
Pumps may run into failure very often due to abusive operation coupled with poor maintenance. Vacuum pumps may report one or more of temperature, vibration, pressure and sound. These data may be used by a platform to check against a baseline pump database and the platform offers early warning for pump failure and/or real time alarm for abusive operation. Similarly, a blower temperature along with vibration may also be tracked. From machine learning algorithms of data, the platform sends out early indication of clogging of safety filters and/or low oil level indication.
FIG. 13 is a representation of normalized vibration against time. Vibration data may help to identify clogged safety filter in a vacuum pump to stop abusive operation. Clogged safety filters may lead to malfunctioning of pump within months.
In an exemplary embodiment, a drying process with a check on health is available tracking temperature and flow data at inlet, outlet and on site glass of a dryer. A recorded database may be created for normal and/or baseline operation with a clean filter, By comparing with the baseline operation, a mobile application may indicate degradation of filters and drying process. The mobile application may also offer recommended operation for optimal temperature to save energy and may also act as a platform for dryer maintenance.
In one or more embodiments, a machine learning architecture may be associated with a machine learning algorithm where normal states of the vacuum pumps with operational range, clean filer and clean oil may be learned with a baseline reading. Further, anomalous readings from one of a clogged filter, a bad operation, a bad oil, a low oil level and an over filled oil level are also recorded. The baseline reading and the anomalous readings may be used as a training database for the machine learning algorithm.
In one or more embodiments, data from multiple vacuum pumps associated with machine wearable sensors may also be acquired. A mobile or web or desktop application may act as a mobile middleware to scale the machine learning architecture to a single data collection unit. The single data collection unit may be one of a mobile device and a wireless device.
In one or more embodiments, the machine learning may be used on a transformed vibrational data based on PCA (Principal component analysis) transformation of X, Y and Z axis components of the vibrational data to transpose an acceleration into reference frame of the rotor of the vacuum pump. FIG. 14A through FIG. 14C illustrates graphical representations of vibration fault detection using transformed vibrational data based on PCA (Principal component analysis) transformation of X, Y and Z axis components, respectively of the vibrational data at a first vacuum level (Vacuum=8).
FIG. 15A through FIG. 15C illustrates graphical representations of vibration fault detection using transformed vibrational data based on PCA (Principal component analysis) transformation of X, Y and Z axis components, respectively of the vibrational data at a second vacuum level (Vacuum=6).
Machine learning of the vibrational data may comprise a transfer of vibrational energy from one axis of rotation to other axis in order to determine the extent of oldness of the oil used in the blower bearings for smooth rotation. Machine learning of the vibrational data may also comprise information related to instability and wobbling of rigid rotational axis which aids in determining an extent of oldness of oil used in bearings of the blower.
In one or more embodiments, a predictive and preventive maintenance system for a vacuum pump may include one or more machine wearable sensors associated with the vacuum pump, a tracking module associated with a computing device, a machine learning module associated with a database and a communications network. A changing condition of vacuum pump may be tracked through the tracking module over the communications network. The tracking module may receive one of a temperature, a vibration and a sound data from the one or more machine wearable sensors. The machine learning module associated with the tracking module may identify a pattern from the temperature, the sound and the vibration data and may raise an alarm based on an analysis of the pattern.
FIG. 16A through FIG. 16D illustrates an exemplary graphical representation of pressure state alarm, filter alarm, oil state alarm, and blower alarm when a bad oil is detected with respect to a silencer. FIG. 17A through FIG. 17D illustrates graphical representations of pressure state alarm, filter alarm, oil state alarm, and blower alarm, respectively, when a legacy pump has a clogged filter. FIG. 18 is a graphical representation of overfill alarm, according to one embodiment.
In one or more embodiments, a wearable sensor may be one of a MEMS or a single silicon sensor.
Although the present embodiments have been described with reference to specific example embodiments, it will be evident that various modifications and changes may be made to these embodiments without departing from the broader spirit and scope of the various embodiments. For example, the various devices and modules described herein may be enabled and operated using hardware circuitry, firmware, software or any combination of hardware, firmware, and software (e.g., embodied in a machine readable medium). For example, the various electrical structure and methods may be embodied using transistors, logic gates, and electrical circuits (e.g., application specific integrated (ASIC) circuitry and/or in Digital Signal Processor (DSP) circuitry).
In addition, it will be appreciated that the various operations, processes, and methods disclosed herein may be embodied in a machine-readable medium and/or a machine accessible medium compatible with a data processing system (e.g., a computer devices), and may be performed in any order (e.g., including using means for achieving the various operations). The medium may be, for example, a memory, a transportable medium such as a CD, a DVD, a Blu-ray disc, a floppy disk, or a diskette. A computer program embodying the aspects of the exemplary embodiments may be loaded onto the retail portal. The computer program is not limited to specific embodiments discussed above, and may, for example, be implemented in an operating system, an application program, a foreground or background process, a driver, a network stack or any combination thereof. The computer program may be executed on a single computer processor or multiple computer processors.
Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense.

Claims

What is claimed is:

1. A method of a machine learning architecture comprising:

i) receiving at least one of a motor sensor data and a blower sensor data over a communications network,

wherein one of the motor sensor data and the blower sensor data comprises at least one of a vibration, a magnetometer, a gyroscope, a sound and a temperature;

ii) classifying at least one of the motor sensor data and the blower sensor data into one of a vacuum state sensor data and break state sensor data,

wherein at least one of the motor sensor data and the blower sensor data are classified by one of individually and in combination,

wherein the break state sensor data is received when a rotor of a vacuum pump is malfunctioning;

iii) analyzing the vibration data of the vacuum state sensor data to detect an operating vacuum level,

wherein an alarm is raised when the vacuum state sensor data of one of a vibration and a temperature exceeds a pre-defined safety range; and

iv) classifying vacuum break data into one of a clean filter category and clogged filter category,

wherein the alarm is raised if an entry under the clogged filter category is detected; and analyzing the blower sensor data in association with the motor sensor data based on machine learning to detect at least one of a deficient oil level and a deficient oil structure.

2. The method of claim 1, further comprising:

receiving sensor data from at least one machine wearable sensor placed on one of a motor and a blower.

3. The method of claim 1, wherein the communications network between sensor and the data collection unit, comprises one of WiFi, 2G, 3G, 4G, GPRS, EDGE, Bluetooth, ZigBee, Piconet of BLE, Zwave, or a combination thereof, wherein the data collection unit is one of a mobile device and a wireless enabled device.

4. The method of claim 1, further comprising:

associating the machine learning architecture with a machine learning algorithm where, normal states of the vacuum pumps with operational range, clean filer and clean oil are learned with a baseline reading and anomalous readings from one of a clogged filter, a bad operation, a bad oil, a low oil level and an over filled oil level, wherein the baseline reading and the anomalous readings are used as a training database.

5. The method of claim 1, further comprising:

acquiring data from multiple vacuum pumps associated with machine wearable sensors,

wherein at least one of a mobile, a web and a desktop application acts as a mobile middleware to scale the machine learning architecture to a single data collection unit, and

wherein the single data collection unit is at least one of a mobile device and a wireless device.

6. The method of claim 1, wherein the alarm is raised over the communications network through one of a notification on the mobile application, Short message service (SMS), email, or a combination thereof.

7. A machine learning architecture comprising:

a vacuum pump including a motor and a blower;

a motor associated with a machine wearable sensor;

a blower associated with a another machine wearable sensor,

wherein at least one of a motor sensor data and a blower sensor data is received over a communications network,

wherein at least one of the motor sensor data and the blower sensor data is at least one of a vibration, a sound and a temperature,

wherein at least one of the motor sensor data and the blower sensor data is classified into one of a vacuum state sensor data and break state sensor data,

wherein the break state sensor data is received when a rotor of a vacuum pump is malfunctioning,

wherein the vibration data of the vacuum state sensor data is analyzed to detect an operating vacuum level,

wherein an alarm is raised when the vacuum state sensor data of one of a vibration and a temperature exceeds a pre-defined safety range,

wherein vacuum break data is classified into one of a clean filter category and clogged filter category,

wherein the alarm is raised if an entry under the clogged filter category is detected; and

wherein the blower sensor data is analyzed in association with the motor sensor data based on machine learning to detect at least one of a deficient oil level and a deficient oil structure.

8. The architecture of claim 7, wherein machine learning is used on a vibrational data transformed by PCA (Principal component analysis) transformation of X, Y and Z axis components of the vibrational data to transpose the acceleration into reference frame of the rotor of the vacuum pump.

9. The architecture of claim 7, wherein machine learning of the vibrational data which comprises transfer of vibrational energy from one axis of rotation to other axis to determine extent of oldness of the oil used in the blower bearings for smooth rotation.

10. The architecture of claim 7, wherein machine learning of the vibrational data comprises of information related to instability and wobbling of rigid rotational axis to determine an extent of oldness of oil used in bearings of the blower.

11. The architecture in claim 7, wherein machine learning of the vibrational data comprises of information related to shape factor of the vibration calculated as a ratio of moving RMS value to moving average of absolute value.

12. A predictive and preventive maintenance system for a vacuum pump comprising:

one or more machine wearable sensors associated with the vacuum pump;

a tracking module associated with a computing device;

a machine learning module associated with a database; and

a communications network,

wherein at least one changing condition of vacuum pump is tracked through the tracking module over the communications network,

wherein the tracking module receives at least one of a temperature, a vibration and a sound data from the one or more machine wearable sensors,

wherein the machine learning module associated with the tracking module identifies a pattern from the temperature, the sound and the vibration data, and

wherein the machine learning module raises an alarm based on an analysis of the pattern.

13. The system of claim 12, wherein the machine learning algorithm engine raises an alarm when at least one of a filter is clogged and deficient oil is detected.

14. The system of claim 12, wherein the deficient oil is one of a low oil level and an overused oil structure.

15. The system of claim 12, wherein the one or more machine wearable sensors include at least one of a motor sensor and a blower sensor.

16. The system of claim 12, wherein the machine learning module engine is associated with an IoT based system,

wherein the machine learning module issues commands based on a learning outcome from the at least one changing condition.

17. The system of claim 16, wherein the learning outcome is dependent on recognition of at least one of a pattern and deviation by the machine learning module.