Damian Bzinkowski

This paper presents the experimental results of a Machine Learning application for the health monitoring of a conveyor belt. The real-time analysis of the rubber belt condition is a crucial issue in achieving safety and avoiding critical failures and related expenses. The measuring system based on strain gauges was applied to identify the actual state of the belt. Using the Classification Lerner application from MATLAB platform, 22 algorithms were tested, and using the Diagnostic Feature Designer application, the analysis was performed. Three tested ML algorithms were able to classify the states of the conveyor belt with preset damages correctly, exhibiting 100% prediction accuracy. The k-nearest neighbors (KNN) classifiers and neural networks failed to achieve that level of accuracy.

The article describes an innovative method for detecting conveyor belt damage using a strain gauge system and deep learning techniques. The strain gauge system records measurement data from conveyor belts. A major challenge encountered during the research was the insufficient amount of measurement data for effectively training deep neural networks. To address this issue, the authors implemented a hybrid data augmentation method that combines generative artificial intelligence techniques and signal analysis. The TimeGAN model, based on Generative Adversarial Networks (GANs), was used to augment data from undamaged belts. Meanwhile, the superposition of one-dimensional observation sequences was applied to generate data representing damages by combining signals from randomly selected undamaged runs with strain gauge system responses to damage, effectively increasing the number of samples while accurately replicating defect conditions. For damage diagnosis, a Long Short-Term Memory Network with an attention mechanism (LSTM-AM) was employed, enabling anomaly detection in strain gauge signals. The application of the LSTM-AM algorithm allows for real-time monitoring of conveyor operation and facilitates precise localization and estimation of damage size through data synchronization.

There are different deep neural network (DNN) architectures and methods for performing augmentation on time series data, but not all the methods can be adapted for specific datasets. This article explores the development of deep learning models for time series, applies data augmentation methods to conveyor belt (CB) tension signal data and investigates the influence of these methods on the accuracy of CB state classification. CB systems are one of the essential elements of production processes, enabling smooth transportation of various industrial items, therefore its analysis is highly important. For the purpose of this work, multi-domain tension data signals from five different CB load weight conditions (0.5 kg, 1 kg, 2 kg, 3 kg, 5 kg) and one damaged belt condition were collected and analysed. Four DNN models based on fully convolutional network (FCN), convolutional neural network combined with long short-term memory (CNN-LSTM) model, residual network (ResNet), and InceptionTime architectures were developed and applied to classification of CB states. Different time series augmentations, such as random Laplace noise, drifted Gaussian noise, uniform noise, and magnitude warping, were applied to collected data during the study. Furthermore, new CB tension signals were generated using a TimeVAE model. The study has shown that DNN models based on FCN, ResNet, and InceptionTime architectures are able to classify CB states accurately. The research has also shown that various data augmentation methods can improve the accuracy of the above-mentioned models, for example, the combined addition of random Laplace and drifted Gaussian noise improved FCN model’s baseline (without augmentation) classification accuracy with 2.0 s-length signals by 4.5% to 92.6% ± 1.54%. FCN model demonstrated the best accuracy and classification performance despite its lowest amount of trainable parameters, thus demonstrating the importance of selecting and optimizing the right architecture when developing models for specific tasks.

This paper is devoted to the real-time monitoring of close transportation devices, namely, belt conveyors. It presents a novel measurement system based on the linear strain gauges placed on the tail pulley surface. These gauges enable the monitoring and continuous collection and processing of data related to the process. An initial assessment of the machine learning application to the load identification was made. Among the tested algorithms that utilized machine learning, some exhibited a classification accuracy as high as 100% when identifying the load placed on the moving belt. Similarly, identification of the preset damage was possible using machine learning algorithms, demonstrating the feasibility of the system for fault diagnosis and predictive maintenance.

Conveyors play a very important role in modern manufacturing processes, and one of the most popular types is the belt conveyor. The main elements of a conveyor include a conveyor belt, roller sets, a supporting frame and a drive and control system. The reliable operation of the conveyor depends on the strength and durability of individual elements (especially the belt). Conveyor belts are made from various materials and have received a lot of attention in the scientific and research community. This article presents tests of the strength of the rubber belt material and its damage under load. The belt consists of two internal layers covered with a PVC coating on the outside, and the nominal belt thickness was 2 mm. In the experiment, various configurations of longitudinal and transverse damage were verified, and statistical methods were used to analyze the results. The obtained test results provided a new understanding of the propagation of conveyor belt damage and helped to improve the strain gauge-based monitoring system.

The paper presents a novel system for monitoring of the work of industrial belt conveyor. It is based on the strain gauges placed directly on the roller surface that measure pressing force of the belt on the roller. Automatical operation of the measurement system minimizes impact of an operator on the measurement results. Experimental researches included the stability of indications during 5 days, Type A uncertainty estimation and equipment variation EV calculations. Expanded uncertainty calculated for the level of confidence 95% was below 0.1% of the actually measured value, and percentage repeatability %EV = 9.5% was obtained. It can be considered satisfactory, since usually it is required %EV < 10% for new measurement systems.

In this paper, long short-term memory (LSTM) and Transformer neural network models were developed for classification of different conveyor belt conditions (loaded and unloaded). Comparative shallow models such as logistic regression, support vector machine and random forest were also developed and summarized. Six different-length belt pressure signals were analyzed: 0.2, 0.4, 0.8, 1.6, 3.2, and 5.0 s. Both LSTM and Transformer models achieved 100% accuracy using pressure raw signal. Furthermore, LSTM model reached the highest classification level with the shortest signals. Accuracy and F1-score of 98% and 100% were reached using only 0.8 and 1.6 s-length signals, respectively. Also, LSTM model performed training and testing procedures faster than Transformer. Random forest model demonstrated the best classification level using aggregated signal data with accuracy of 85% and F1-score for loaded and unloaded conditions of 85% and 69%, respectively. Loaded conveyor belt condition was significantly easier to classify than the unloaded one in all models. Only LSTM showed better classification recall for unloaded conveyor belt condition using short signal. Experimental research dataset CORBEL (Conveyor belt pressure signal dataset) and models are open-sourced and accessible on GitHub https://github.com/TadasZvirblis/CORBEL .

In the paper, the issue of the supervision of belt tension and wear in industrial feeder is addressed. The designed system is based on strain gauges that are built into the roller and are subject to the belt pressure at each revolution. In order to assess the effectiveness of this system, calibration and uncertainty analysis was performed. As a result, it was demonstrated that the main source of uncertainty was the function of approximation, while the others were orders of magnitude smaller. The final function provided results with accuracy of ca. 10% of actually measured value, which was assumed to be a good result for this particular industrial application.

In the paper, an issue of CMM measurement strategy of the sleeves and pins designed for wear tests in the cycloidal drive. The measurement strategy was proposed, based on initial out-of-roundness measurement in scanning mode. Proposed approach ensured that the pin was measured along its entire 40 mm length with only small area of fixation ca. 5 mm left out, and all probing points for cylindricity deviation assessment were collected in one fixation. It was demonstrated that the cylindricity and roundness measurement results based on 8, 16 and 48 probing points provided sufficient data for further wear analysis. In some cases, the circles calculated from 4 points gave additional insights allowing to accept the part that otherwise might be possibly rejected.