Internet of things based determination of machine reliability and automated maintainenace, repair and operation (MRO) logs

• US 10,599,982 B2
• Filed: 04/25/2015
• Issued: 03/24/2020
• Est. Priority Date: 02/23/2015
• Status: Active Grant

• Current Assignee: Machinesense LLC
• Original Assignee: Machinesense LLC
• Inventors: Pal, Biplab, Nagi, Neeraj, Chakrabarty, Amit
• Primary Examiner(s): Afshar, Kamran
• Assistant Examiner(s): Buss, Benjamin J
• Application Number:
  US14/696,402
• Publication Number:
  US 20170011298A1
• Time in Patent Office: 1,795 Days
• Field of Search: None
• CPC Class Codes: G05B 23/0283 Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks

Abstract

A computer implemented method and system for determining reliability of a machine includes receiving one of a machine data from one or more locations through an internet of things (IOT) based machine wearable sensor network. The method further includes storing the data in a distributed computer database communicatively coupled to an enterprise resource planning (ERP) system and extracting, through a computer server, one or more entity information from the data to compare against a pre-defined baseline. Further, mapping, though a big data machine learning engine, the extracted one or more entity information into a multi-classification model. The method includes indicating, through a machine learning engine coupled to a predictive analytics engine, on a user interface a set of analytical predictions for machine maintenance, repair and operation.

Claims

1. A computer implemented method for determining reliability of a machine, comprising:

a) receiving at least one of machine operational condition data, machine historical operational data and machine specific information data generated by at least one machine wearable sensor placed on a machine part from at least one location corresponding to said sensor through an internet of things based machine wearable sensor network;

b) storing the data in a distributed computer database communicatively coupled to an enterprise resource planning system;

c) extracting, through a computer server from the distributed computer database, the data for the machine to compare against a pre-defined baseline;

d) using cluster vector classification mapping, through a big data machine learning engine, the extracted data into a multi-classification model to classify the data into a root cause analysis engine;

e) using the root cause analysis engine mapping the data into one or more levels of predictive maintenance states associated with color schemes displayed as a gauge on a user interface of a mobile device, the color schemes including one of red, yellow and green where red indicates a bad maintenance condition, yellow indicates an intermediate maintenance condition, and green indicates a good maintenance condition;

f) analyzing the data mapped in steps (d) and (e) through a real-time data feed platform associated with a distributed real-time computation system;

g) determining reliability of the machine as defined by the results from making a set of analytical predictions for machine maintenance, repair and operation using the data analyzed in step (f) and the big data machine learning engine coupled to a predictive analytics engine;

h) updating machine historical operation data with data received on the sensor network, through a real-time data feed platform associated with the distributed real-time computation system and indicating, through the big data machine learning engine coupled to a predictive analytics engine, on the user interface displayed on a hand-held portable electronic device having wireless internet access capabilities, the set of analytical predictions for machine maintenance, repair and operation, as produced in step (g) determining machine reliability; and

i) performing sensor autocalibration.

2. The method of claim 1 wherein the internet of things sensor network is communicatively coupled to a communications network communicatively connected to a distributed computer database.

3. The method of claim 1 wherein the data is received through at least one industrial bus network.

4. The method of claim 1 further comprising scheduling, based on the set of analytical predictions, at least one of maintenance, repair and operation for the machine.

5. The method of claim 1 wherein at least one of maintenance and repair is scheduled over the communications network through notification on a mobile application, a telephone call, an email, or a combination thereof.

6. The method of claim 1, wherein the set of analytical predictions include at least one of a prediction of machine failure, mean time between failure, additional usage of electrical power and downtime of a machine.

7. A real-time industrial internet of things based system for determination of machine reliability and the root cause of ill health of the machine resulting from poor power quality, bad maintenance practice and abusive operations, comprising:

a) one or more machine wearable sensors for sensing vibration, current, and voltage, the sensors being operatively associated with a communications network;

b) the communications network including a data collection unit;

c) an enterprise resource planning system communicatively coupled to a distributed computer database over the communications network so that the distributed computer database can be utilized both from the cloud and a device communicating with the sensors via the communications network;

d) a real-time data feed platform associated with a distributed real-time computation system communicatively coupled to the sensors over the communications network; and

e) a big data machine learning engine coupled to a predictive analytics engine over the communications network which learns about oil state, filter alarm, oil level, blower temperature, and power factor and abusive operation over the communications network;

wherein the at least one of machine operational condition data, machine historical operational data and machine specific information data from at least one sensor is received through the communications network;

wherein the data is stored in a distributed computer database communicatively coupled to the enterprise resource planning system;

wherein at least one entity information is extracted from the data through a computer server to compare against a database of baseline features from sensor data or a physics based model to extract machine health state from the data;

wherein the at least one entity information is mapped onto a multi-classification model;

wherein the mapping includes at least one of classifying the data into a root cause analysis engine, and mapping the data into one or more predictive maintenance states;

wherein the predictive maintenance states are associated with color schemes including red, yellow and green;

wherein the data is analyzed through a real-time data feed platform associated with a distributed real-time computation system; and

wherein a set of analytical predictions defining machine reliability for machine maintenance is indicated, through a machine learning engine coupled to a predictive analytics engine, on a user interface for at least one of a repair, maintenance and operation of the machine.

8. The system of claim 7, wherein the sensor data collection unit is one of a mobile device or a wireless enabled device or a wired gateway device.

9. The system of claim 7, wherein based on the set of analytical predictions, scheduling is arranged for at least one of machine maintenance, repair and operation.

10. The system of claim 7, wherein the set of analytical predictions include at least one of machine failure, mean time between failure, additional usage of electrical power, and downtime of the machine.

11. The system of claim 7, wherein the set of analytical predictions include at least one of a prediction of machine failure, days to turn red for a particular predictive maintenance state, mean time between failure, additional usage of electrical power, and downtime of the machine.

12. The system of claim 7, wherein the predictive maintenance states are represented on the user interface through a graphical representation associated with color schemes to represent state for the machine conditions.

13. A computer implemented method for determining reliability of a machine comprising:

a) receiving machine operational condition data, machine historical operational data and machine specific information data generated by at least one machine wearable sensor operatively associated with a part of the machine, through an internet of things based machine wearable sensor network, each sensor sensing a physical parameter during machine operation;

b) storing the data in a distributed computer database communicatively coupled to an enterprise resource planning system;

c) using a big data machine learning engine for extracting data from the distributed computer database for comparison against a pre-defined baseline;

d) mapping, using the big data machine learning engine, the extracted data into a multi-classification model using clustered vector classification to classify the data into a root cause analysis engine, the mapping placing the data into levels of predictive maintenance state having associated red, yellow and green colors where red indicates a poor maintenance condition, yellow indicates an intermediate maintenance condition, and green indicates a good maintenance condition;

e) analyzing the mapped data, through a real-time data feed platform associated with a distributed real-time computation system and an operatively connected predictive analytics engine;

f) using the mapped data and the big data machine learning engine coupled to the predictive analytics engine, displaying on a mobile device user interface having visible fields of red, yellow, and green, a set of analytical predictions for machine maintenance, repair and operation defining reliability of the machine; and

g) updating machine historical operation data with data received on the sensor network as used in performing step (f).

14. A computer implemented method for determining reliability of a machine comprising:

a) receiving machine operational condition data, machine historical operational data and machine specific information data generated by at least one machine wearable sensor operatively associated with a part of the machine, through an internet of things based machine wearable sensor network, each sensor sensing a physical parameter during machine operation;

b) storing the data in a distributed computer database communicatively coupled to an enterprise resource planning system;

c) using a big data machine learning engine for extracting data from the distributed computer database to compare against a pre-defined baseline;

d) using the big data machine learning engine, mapping the extracted data into a multi-classification model using clustered vector classification to classify the data into a root cause analysis engine, the mapping placing the data into levels of predictive maintenance state having associated red, yellow and green colors where red indicates a poor maintenance condition, yellow indicates an intermediate maintenance condition, and green indicates a good maintenance condition;

e) analyzing the mapped data, through a real-time data feed platform associated with the distributed real-time computation system and an operatively connected predictive analytics engine to create a set of analytical predictions for machine maintenance, repair, and operation;

f) using the mapped data and the big data machine learning engine coupled to the predictive analytics engine, displaying the set of analytical predictions for machine maintenance, repair and operating defining the reliability of the machine;

g) updating the machine historical operation data with data used in performing step (f); and

h) performing sensor autocalibration using the data from step “

f”

.

15. A real-time internet of things-based system for determining reliability of machines of interest as defined by analytic predictions of machine need for repair, maintenance and operation, comprising:

a) a plurality of machine wearable sensors operatively connected to machines of interest;

b) a communications network operatively connected to the sensors;

c) an internet of things sensor network communicatively coupled to the communications network and receiving machine operational condition data, machine historical operational data, and machine specific information data from the machine wearable sensors via the communications network;

d) an enterprise resource planning system communicatively coupled to the communications network;

e) a distributed database, connected to the enterprise resource planning system via the communications network, for storing machine operational condition data, machine historical operational data, and machine specific information data;

f) a real-time data feed platform communicatively coupled to the internet of things sensor network via the communications network;

g) a big data machine learning engine;

h) a predictive analytics engine coupled to the big data machine learning engine via the communications network;

i) a distributed real-time computation system, communicatively connected to the real-time data feed platform and to the enterprise resource planning system for storing the machine operational condition data, the machine historical operational data and the machine specific information data received from the sensors over the communications network and the internet of things sensor network;

with the machine operational condition data, machine historical operational data and machine specific information data from the machines of interest being received by the distributed database through the internet of things sensor network;

with the machine operational condition data, the machine historical operational data and the machine specific information data being stored in the distributed database communicatively coupled to the enterprise resource planning system, and the operational condition data, historical operation data, and specific information for at least one machine being extracted from the data to compare against a pre-defined baseline, with the extracted information being mapped into a supervised multi-classification machine learning model and being classified into a root cause analysis engine and into one or more predictive maintenance states associated with a color scheme including red, yellow and green;

wherein the extracted information is analyzed, through the real-time data feed platform associated with the distributed real-time computation system to create a set of analytical predictions for machine repair, maintenance; and

wherein the set of analytical predictions for machine maintenance is indicated, through the big machine learning engine coupled to the predictive analytics engine, on a user interface for at least one of a repair, maintenance and operation, defining reliability of the machine.