Description:
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a system for online health monitoring of conveyor belts using computer vision and machine learning.

BACKGROUND OF THE INVENTION
Conveyor belts are responsible for the transportation of material from one location to another. Conveyor belts have made material handling and transport, a smooth and convenient process and seamless functioning without conveyor belts in any material handling unit in or beyond steel plants is unimaginable making the role of conveyor belts very vital. Due to the complicated force conditions of the conveyor belt, it is very prone to failures such as tearing, deviation, and surface damage, which poses serious threats to the safe and efficient production of enterprises. When a conveyor belt failure occurs, it usually results in an increase in conveyor running resistance, the speed of wear and tear increases, and even leads to downtime. Monitoring the health of conveyor belts by human involvement is a very tedious and difficult task owing to spread and approach of conveyor belts.

Hence there is a need for the introduction of a novel system that aims to overcome these drawbacks and provide a more efficient solution. This patent application proposes a novel system designed for online health monitoring of running conveyor belts.

SUMMARY OF THE INVENTION
The following disclosure presents a simplified summary of the invention in order to provide a basic understanding of some aspects of the invention. This summary is not an extensive overview of the present invention. It is not intended to identify the key/critical elements of the invention or to delineate the scope of the invention. Its sole purpose is to present some concept of the invention in a simplified form as a prelude to a more detailed description of the invention presented later.

An object of the present invention is to provide a novel system designed for online health monitoring of running conveyor belts.

Another object of the present invention is to provide man-less monitoring of conveyor belts during operation and prevent unplanned breakdowns of the conveyor system.

Another object of the present invention is to eliminate production delays and ensure preventive maintenance of conveyor belts along with available planned shutdowns.

In one aspect of the present invention, a system is provided for online health monitoring of conveyor belts, using computer vision and deep learning, wherein the system comprises:
one or more conveyor belts;
one or more servers;
one or more video cameras;
a buzzer alarm;
a belt sway detection module, for identifying lateral sway in conveyor belts;
a foreign material detection module, for identifying the presence of foreign materials on conveyor belts in addition to the primary material being transported;
a surface defect detection module, for identifying various defects on the surface of conveyor belts both on the loading side and the non-loading side.

Other aspects, advantages, and salient features of the invention will become apparent to those skilled in the art from the following detailed description, which, taken in conjunction with the annexed drawings, discloses exemplary embodiments of the invention.

BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
The above and other aspects, features and advantages of the embodiments of the present disclosure will be more apparent in the following description taken in conjunction with the accompanying drawings, in which:

Figure 1 illustrates the schematic of Installed hardware for the developed system.

Persons skilled in the art will appreciate that elements in the figures are illustrated for simplicity and clarity and may not have been drawn to scale. For example, the dimensions of some of the elements in the figure may be exaggerated relative to other elements to help to improve understanding of various exemplary embodiments of the present disclosure.

DETAILED DESCRIPTION OF THE PRESENT INVENTION
The following description with reference to the accompanying drawings is provided to assist in a comprehensive understanding of various embodiments of the present disclosure as defined by the claims and their equivalents. It includes various specific details to assist in that understanding, but these are to be regarded as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the various embodiments described herein can be made without departing from the scope and spirit of the present disclosure. In addition, descriptions of well-known functions and constructions may be omitted for clarity and conciseness.

All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which various embodiments belong. Further, the meaning of terms or words used in the specification and the claims should not be limited to the literal or commonly employed sense but should be construed in accordance with the spirit of the disclosure to most properly describe the present disclosure.

The terminology used herein is for the purpose of describing particular various embodiments only and is not intended to be limiting of various embodiments. As used herein, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising" used herein specify the presence of stated features, integers, steps, operations, members, components, and/or groups thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, members, components, and/or groups thereof.

The present disclosure will now be described more fully with reference to the accompanying drawings, in which various embodiments of the present disclosure are shown.

A Raw Material Handling Plant (RMHP) is generally equipped with bedding and blending facilities for raw materials along with facilities of unloading of all raw materials at one location. Major equipment used in RMHP are Wagon Tippler, Side Arm charger, Paddle feeder, Stacker, Barrel reclaimer, transfer car, screen, rod mill, impact crusher, Conveyor belts etc. Out of all the major equipment’s mentioned above conveyors belts are responsible for transportation of material from one location to another. Conveyor belts have made material handling and transport, a smooth and convenient process and a seamless functioning without conveyor belts in any material handling unit in or beyond steel plants is unimaginable making the role of conveyor belts very vital. Due to the complicated force conditions of the conveyor belt, it is very prone to failures such as tearing, deviation, and surface damage, which poses serious threats to the safe and efficient production of enterprises. When a conveyor belt failure occurs, it usually results in an increase in conveyor running resistance, the speed of wear and tear increases, and even leads to downtime. In the existing system, monitoring the health of conveyor belts by human involvement is a very tedious and difficult task owing to spread and approach of conveyor belts.

The purpose of this innovation is to create a predictive maintenance system for conveyor belts. This groundbreaking invention demonstrates the effectiveness of applying machine vision to identify common issues that arise during the operation of conveyor belts. Given the novelty of incorporating machine vision, which combines computer vision with machine/deep learning, the development strives to encompass a wide range of materials transported by conveyor belts.

A significant hurdle encountered in this development involves preserving the cleanliness of camera lenses while the conveyor belt is in operation, given the substantial amount of dust in the surrounding environment. To address this challenge, a lens cleaning system has been devised utilizing the plant's compressed air. The system takes in compressed air from the standard plant supply lines, with dedicated pipelines established for each of the nine cameras originating from the nearest connection point of the existing compressed air line. The compressed air undergoes moisture separation to yield dry compressed air, which is then directed onto the camera lenses.

In the software architecture of the Conveyor Belt Health Monitoring System, real-time video streams are collected by the media server. The media server then subdivides the streams into a streaming server and a session manager. The session manager extracts frames from the video streams, invoking video analytics through the Image probe. These video analytics employ a combination of algorithms, incorporating image processing, machine learning, and/or deep learning. Events are generated based on the logic embedded in the video analytics algorithms. For instance, if the conveyor belt exhibits swaying beyond a predefined threshold, the corresponding frame capturing the exceeding limit is visually recorded. Simultaneously, such events are logged in Excel sheets. Buzzer alarms are triggered by these events, and digital signals are sent to the PLC to stop the conveyor belt.

The current system has been developed for monitoring of health of conveyor belts utilizing the techniques of machine vision (computer vision and deep learning). This system is beneficial to cover widely spread conveyor belt network. The online health monitoring system has been designed utilizing the concepts of image and video processing along with machine learning applied to collected data from practical operation of conveyor belts. To achieve this, real time scenario of operational belts is recorded using video cameras. These video cameras are interfaced with a server computer server where software modules for serving different purposes of this system are built and running.

Software modules have been developed for separation of different frames of video, extraction of information from these images and machine learning for estimation of surface condition of conveyor belt and operational condition like swaying of conveyor belt and presence of foreign material on conveyor belt.

Belt Sway Detection Module:
The module has been designed to identify lateral sway in conveyor belts, both within and beyond safe limits. The analytics incorporate two distinct limits: the soft limit and the hard limit. The soft limit signals the onset of a belt sway tendency in a specific belt, and the limit values are adjustable, allowing user control with input easily provided from the workstation. On the other hand, the hard limit, situated at the end of belt rollers, signifies a critical point where the belt extending beyond may result in material spillage. If not prevented, it could lead to belt slippage, causing significant downtime. Consequently, hard limit values are set as fixed parameters.

Foreign material detection Module:
This analytical module is employed to identify the presence of foreign materials on conveyor belts in addition to the primary material being transported. The module operates in two stages: first, it classifies the material actively being transported, and in the second step, based on the characteristics of the primary material, the system identifies foreign materials such as rods, plates, boulders, and sacks. Distinguishing foreign materials among various types, including dry and wet materials, mixed materials, as well as half-filled and fully filled materials, has resulted in diverse combinations of varieties. While defining foreign materials as strictly limited to rods, plates, and boulders might seem simpler, the practical reality is that rods and plates can come in various types, and boulders can exhibit variations in colors and sizes. Consequently, the definition of foreign materials has evolved to encompass any material other than the precise materials the conveyor belt is currently transporting or has transported previously.

Surface defect detection module:
This analytics has been created to identify various defects on the surface of conveyor belts, both on the loading side (with a frontal view) and the non-loading side (with a slant view) using the previously mentioned cameras. This analytics module focuses on detecting surface defects such as wear and tear.

Initially, the software module was designed to monitor the percentage of the surface identified as healthy, with a manually set threshold indicative of deteriorating belt conditions. As the system operated in practice and insights into the process were gained, the module underwent a redesign. It now monitors the percentage of the surface identified as unhealthy, and the threshold setting has been automated to enhance module performance. Before the system was operationalized, data/videos of defective belt conditions were limited, making the learning phase of the module time-consuming. This learning process continues during system operation. The deep learning capabilities, coupled with ongoing manual data collection, are utilized to develop updates for the software module, aligning with user requirements.

The server is interfaced with a Human Machine Interface (HMI) workstation displaying the real time videos, the health status of belts, and alarms in case of deviation being observed on the conveyor belt by the developed online health monitoring system. An inbuilt buzzer alarm in the system to provide alerting sounds in case of observance of conditions requiring immediate attention. This whole system has also been interfaced with the existing PLC system of operational conveyor belts to bring the belts to immediate stoppage if an identified extreme situation has occurred.

The present system is a novel system designed for online health monitoring of running conveyor belts. This facilitates man-less monitoring of conveyor belts during operation and prevents unplanned breakdowns of the conveyor system. This system eliminates production delays and ensures preventive maintenance of conveyor belts along with available planned shutdowns. The system design has been conceptualized taking into consideration its adaptability under different shop requirements and logistics.

The diagram illustrating the hardware incorporated into the developed system is depicted in the figure 1.

The system is installed at Raw Material Handling Plant (RMHP) of ISP, Burnpur. Further, to accomplish this goal three belts were carefully chosen from the Raw Material Handling Plant in collaboration with the plant.The system for each of the three conveyor belts consists of three strategically placed cameras that serve specific purposes:

Location 1, positioned near the Vertical Gravity Take-Up, monitors surface defects on the loading side of the empty belt and detects belt sway. It captures a frontal view of the conveyor belt to monitor the empty surface on the loading side.

Location 2, near the bend pulley of each belt, monitors surface defects on the non-loading side, providing a slant view from a distance greater than 1 meter.

Location 3, near the tail end of the conveyor belt, around 15 to 20 meters in the direction of belt motion, detects belt sway and foreign material on the transported material.

To process the acquired videos, two tower servers and workstations are installed—one set in the Ore Handling Plant (OHP) control room and another in the Base Mix Plant (BMP) control room. The videos are processed using image processing techniques and machine learning-based artificial intelligence as outlined in Table 1. The servers handle video analytics and deep learning software, while the workstations serve as the input and output interface between the software and operators in the respective control rooms.

Table 1: Purpose of respective cameras of conveyor health monitoring system

S.No. Belt Name Camera at Location No. Purpose being served
1 J1C1’ 1 Monitoring of belt condition on loading side of belt and generating an alarm & digital output when unhealthy surface condition is detected.
2 J1C1’ 2 Monitoring of belt condition on non- loading side of belt and generating an alarm & digital output when unhealthy surface condition is detected.
3 J1C1’ 3 Monitoring of belt sway at tail end and identification of foreign material namely rods, plates, boulders and sacks
4 SH1C1 1 Monitoring of belt condition on loading side of belt and generating an alarm & digital output when unhealthy surface condition is detected.
5 SH1C1 2 Monitoring of belt condition on non- loading side of belt and generating an alarm & digital output when unhealthy surface condition is detected.
6 SH1C1 3 Monitoring of belt sway at tail end and identification of foreign material namely rods, plates, boulders and sacks
7 FSC3 1 Monitoring of belt condition on loading side of belt and generating an alarm & digital output when unhealthy surface condition is detected.
8 FSC3 2 Monitoring of belt condition on non- loading side of belt and generating an alarm & digital output when unhealthy surface condition is detected.
9 FSC3 3 Monitoring of belt sway at tail end and identification of foreign material namely rods, plates, boulders and sacks

Usefulness of the Invention:
The system for online health monitoring of running conveyor belts facilitates man-less monitoring of conveyor belts during operation and prevents unplanned breakdowns of the conveyor system. This system eliminates production delays and ensures preventive maintenance of conveyor belts along with available planned shutdowns.

Industrial Applicability:
Similar system for online health monitoring of running conveyor belts can be implemented in industry in areas where such conveyors are installed for transport of material and the section is unmanned.
, Claims:
1. A system for online health monitoring of conveyor belts, using computer vision and deep learning, wherein the system comprises:
one or more conveyor belts;
one or more servers;
one or more video cameras;
a buzzer alarm;
a belt sway detection module, for identifying lateral sway in conveyor belts;
a foreign material detection module, for identifying the presence of foreign materials on conveyor belts in addition to the primary material being transported;
a surface defect detection module, for identifying various defects on the surface of conveyor belts both on the loading side and the non-loading side.

2. The system as claimed in claim 1, wherein video cameras are interfaced with the server computer.

3. The system as claimed in claim 1, wherein at least one camera is positioned near the Vertical Gravity Take-Up, monitors surface defects on the loading side of the empty belt and detects belt sway.

4. The system as claimed in claim 1, wherein at least one camera is placed near the bend pulley of each belt, monitors surface defects on the non-loading side.

5. The system as claimed in claim 1, wherein at least one camera is placed near the tail end of the conveyor belt, detects belt sway and foreign material on the transported material.

6. The system as claimed in claim 1, wherein the modules adapted to monitor the percentage of the surface identified as unhealthy, and the threshold setting has been automated to enhance module performance.

7. The system as claimed in claim 1, wherein the inbuilt buzzer alarm provides alerting sounds in case of observance of conditions requiring immediate attention.
8. The system as claimed in claim 1, wherein the server displays real time videos, the health status of belts, and alarms.

9.The system as claimed in claim 1, wherein the videos are processed using image processing techniques and machine learning-based artificial intelligence.