Topographical monitoring of conveyor belt surface and root cause determination
A system for monitoring conveyor belts includes an emitter, a sensor and electronic circuitry. The emitter is configured to emit light at the surface of a conveyor belt. The sensor is configured to measure the light reflected from the belt surface. The circuitry is configured to identify a defect in or on the surface of the belt based on the light received from the sensor, and determine the source or cause of the observed surface defect. Machine learning including data from outside sources may be used in the root cause analysis.
1 . A system for monitoring surface defect(s) of a conveyor belt to determine potential root cause(s), the system comprising:
at least one sensor configured to obtain information associated with at least one defect in or on a surface of a conveyor belt; and
electronic circuitry operatively coupled to the at least one sensor to receive the information associated with the at least one defect, the electronic circuitry being configured to:
(i) Identify the at least one defect based at least upon the information received from the at least one sensor, the information comprising topographical variations in the surface of the conveyor belt;
(ii) identify at least one defect pattern based at least upon the identified at least one defect, and
(iii) identify one or more root causes based at least upon the identified at least one defect pattern by using empirical data collected on a specific conveyor process that had specific defect patterns which could be associated with a specific defect event, by applying analysis tools to historical data sets containing conveying applications with known conveyor design and processing criteria and resulting defect patterns, or by studying development of the identified at least one defect pattern over time.
2 . The system according to claim 1 , wherein:
the system includes at least one light emitter configured to emit light that is reflected from the surface of the conveyor belt;
the at least one sensor includes at least one light detection sensor that is configured to receive the light reflected from the surface of the conveyor belt; and
the electronic circuitry is configured to identify the at least one defect based upon variations in the reflected light received by the at least one light detection sensor.
3 . The system according to claim 1 ,
wherein the topographical variations are determined using a plurality of light reflections that form datapoints which form a three-dimensional topographical profile of the surface of the belt.
4 . The system according to claim 1 , wherein:
the identifying the at least one defect pattern includes identifying a spatial relation of the at least one defect relative to a predefined belt location and/or relative a position of an accessory of the system as related to its position relative to the belt; and/or
the identifying the at least one defect pattern includes identifying a plurality of defects and a spatial relation of the plurality of defects relative to each other and/or relative to a predefined belt location; and/or
the identifying the at least one defect pattern includes identifying a progression of the at least one defect or a plurality of defects over time.
5 . The system according to claim 1 , wherein:
the circuitry is configured to classify the at least one defect pattern into a defect category based upon the identification of the at least one defect pattern.
6 . The system according to claim 5 , wherein:
the identifying the at least one defect pattern includes topographically mapping a plurality of defects, and the topographically mapping includes determining a transverse location and a longitudinal location of each identified defect, and
the circuitry is configured to track the at least one defect pattern for the conveyor belt over time and use information associated with the tracking to identify the one or more root causes.
7 . The system according to claim 1 , wherein:
the circuitry is configured to utilize machine learning in the identifying the at least one defect; and/or
the circuitry is configured to utilize machine learning in the identifying the at least one defect pattern; and/or
the circuitry is configured to utilize machine learning in the identifying the one or more root causes associated with the at least one defect.
8 . The system according to claim 7 , wherein:
the machine learning is at least partially based upon prior defect events of the system; and/or
the machine learning is at least partially based upon prior defect events of one or more other systems that are remote from the system.
9 . The system according to claim 7 , wherein:
the machine learning is at least partially based upon structural design data of the system that impact one or more regions of the conveyor belt; and/or
the machine learning is at least partially based upon structural design data of one or more systems tat are remote from the system.
10 . The system according to claim 7 , wherein:
the identifying the at least one defect includes generating at least one image of the at least one defect, and
the machine learning is at least partially based upon image classification and pattern recognition of information in the at least one image that is generated.
11 . The system according to claim 10 , wherein:
the circuitry is configured to utilize a tachometer to provide longitudinal location data associated with the identification of the at least one defect.
12 . The system according to claim 1 , wherein:
the circuitry is configured to utilize a proximity sensor to measure the position of the edge of the belt to provide transverse belt position associated with the identification of the at least one defect.
13 . The system according to claim 1 , wherein:
the circuitry is configured to utilize an RFID reader to identify and track specific defects relative to a unique traceable belt position.
14 . The system according to claim 1 , wherein:
the circuitry evaluates the one or more root causes and classifies the one or more root causes according to severity, and the circuitry generates an alarm based on the classified one or more root causes.
15 . A conveyor comprising:
a conveyor belt,
at least one structure that interacts or causes an interaction with the conveyor belt, and
the system according to claim 1 ,
wherein the circuitry is configured to identify the one or more root causes associated with the at least one structure based at least upon the identification of the defect pattern.
16 . A method comprising:
running a conveyor belt; and
using machine learning to perform one or more of:
(i) identifying at least one defect in or on the conveyor belt surface based at least upon topographical variations in the surface of the conveyor belt;
(ii) identifying at least one defect pattern based at least upon the identified at least one defect, and
(iii) identifying one or more root causes of the at least one defect based at least upon the identified at least one defect pattern by using empirical data collected on a specific conveyor process that had specific defect patterns which could be associated with a specific defect event, by applying analysis tools to historical data sets containing conveying applications with known conveyor design and processing criteria and resulting defect patterns, or by studying development of the identified at least one defect pattern over time.
17 . The method according to claim 16 , wherein the machine learning includes data from one or more systems that are remoteoperating different conveyor belt designs from a system having the conveyor belt.
18 . A non-transitory computer readable medium storing program code which when executed by one or more processors performs at least the steps:
(i) Identifying at least one defect in or on the conveyor belt surface based at east upon topographical variations in the surface of the conveyor belt;
(ii) identifying at least one defect pattern based at least upon the identified at least one defect, and
(iii) identifying one or more root causes of the at least one defect based at least upon the identified at least one defect pattern by using empirical data collected on a specific conveyor process that had specific defect patterns which could be associated with a specific defect event, by applying analysis tools to historical data sets containing conveying applications with known conveyor design and processing criteria and resulting defect patterns, or by studying development of the identified at least one defect pattern over time.