Belt conveyors are used for transporting bulk materials over distances. The core of the belt, by transferring the longitudinal stresses and ensuring proper frictional coupling of the belt, enables belt movement and transportation of materials on its surface. As the belt cover and edges are used, the belt becomes abraded, and the core is subject to fatigue. The result is the development of cracks in rubber covers across the belt, which leads to the development of damage not only along the cables (the natural direction of water migration and corrosion) but also in the direction transverse to the belt axis. Conducting a series of scans of the St-type belt operating in one of the underground copper ore mines in Poland allowed identifying the number of failures as well as their size and changes over time. These data were in turn used to determine the measures defining the condition of the belt such as the density of defects (the number of defects per 1 m of the belt), the density of the area of damage (the area of damage per 1 m of the belt) and the change in the average area of a single defect over time. By determining the regression of these measures in time and the rate of damage development in both directions (along the axis of the belt and across the belt), it was possible to forecast future states of the belt, as well as to evaluate the costs of different belt replacement strategies and the economic rationalization of the decision to replace them. This research has become possible owing to the development of the DiagBelt system for two-dimensional imaging of the damage to the core of steel-cord belts with resolution sufficiently high to allow tracking the development of single core defects.

The authors propose a new method of predictive replacements of belt sections in a conveyor belt loop with their refurbishment based on their condition recognized by specially designed diagnostic devices: DiagBelt and BeltSonic. Selection of the moment for belt replacement is made on the basis of forecasted state of its wear, when steel cable failure density exceeds the selected threshold or the percentage of belt covers abrasion exceeds the permissible level. The use of double belt refurbishment extends the total belt operation time, but without diagnostics, the reconditioning success rate is relatively low. Visual inspection allows only an uncertain assessment of cable condition. New devices give precise assessment of their condition and allow the selection of a non-linear rate of belt condition based on consecutive scans. The assessed condition of the cables is the key factor in successful belt refurbishment and obtained cost savings. Additionally, diagnostics can help extend belt life. Identified failures can be repaired in order to prevent accelerated belt degradation. The use of diagnostics in the current replacement strategy significantly improves the economic efficiency of belt reconditioning. The reuse of the belt core ensures efficient use of resources and reduces the level of waste, which fits into the circular economy.

Non-invasive conveyor belt diagnostics in damage detection allows significant reductions of the costs related to belt replacement, as well as the evaluation of belt usability and wear degree changes over time. As a result, it increases safety in the location where the belt is used. Depending on the location of a belt conveyor, its length or the type of the transported material, the belt may undergo wear at different rates, albeit the wear process itself is inevitable. This article presents an artificial intelligence-based approach to the classification of conveyor belt damage. A two-layer neural network was implemented in the MATLAB programming language, with the use of a Deep Learning Toolbox set. As a result of the optimization of the created network, the effectiveness of operation was at the level of 80%.

Diagnostics of conveyor belts used in horizontal transport without the need to take the belt off the conveyor and test it in laboratory conditions is an important aspect in mining plants (Jurdziak et al., Adv Intell Syst Comput, 835:645–654, 2019). Current testing, and thus obtaining knowledge about the current thickness of the conveyor belt covers, allows for control accelerated changes. It also avoids emergency stoppages in the operation of the conveyor and enables economically justified planning of a break in its operation. The article presents the concept of the first in Poland mobile device for measuring the thickness of conveyor belts in motion, implemented as part of the NCBR project (No. 0227 / L-10/2018 [LIDER program, Transport Przemysłowy i Maszyny Robocze, 1(47)/2020, pp. 60–61]), and also presents the measurement results obtained thanks to the laboratory version of the device.

The initial set of four measurements (scans) captures the failures occurring in the core of steel cable conveyor belt segments used in a lignite coal mine in Poland. These measurements utilize the DiagBelt+ magnetic system, enabling the analysis of averaged damage sizes. The scanning results allowed the analysis of complete changes in the failure sizes to a maximum of six different time intervals from 1 to 19 months. The cumulative increase in the length and width of the failure (their total extent in both directions) is a linear function of time. However, the inclination of the selected lines (growth speed) depends on the age of the belt segment at the time of its first scan. It is possible to find the best regression function of slope coefficients for the age of the belt segment. The obtained results are crucial for rational belt management, including reliable short-term forecasts of belt needs with the flow of time, indications of the best moments for belt segment replacements, and estimating the budget of the belting department.