School of Coal Engineering, Datong University Shanxi Province, Datong, China.
School of Mines, China University of Mining and Technology, Xuzhou, China.
PLoS One. 2024 Oct 18;19(10):e0312229. doi: 10.1371/journal.pone.0312229. eCollection 2024.
Addressing the challenges of current scraper conveyor health assessments being influenced by expert knowledge and the relative difficulty in establishing degradation models for equipment, this study proposed a method for assessing the health status of scraper conveyors based on one-dimensional convolutional neural networks (1DCNN). The approach utilizes four preprocessed monitoring signals representing different health states of the scraper conveyor as input sources. Through multiple transformations of the data using a constructed one-dimensional convolutional neural network model, it extracts effective features from the data and establishes a mapping relationship between input data and equipment health status. This enables the recognition of the health status of the scraper conveyor. Comparative experimental analysis indicates that the proposed method can effectively identify the health status of the scraper conveyor, achieving an accuracy rate of 98.9%. This method provides an effective means and technical support for the subsequent health management of scraper conveyors in coal mining fully mechanized workfaces.
针对当前刮板输送机健康评估受专家知识影响的挑战，以及为设备建立退化模型的相对困难，本研究提出了一种基于一维卷积神经网络（1DCNN）的刮板输送机健康状况评估方法。该方法利用四个预处理监测信号作为输入源，代表刮板输送机的不同健康状态。通过对构建的一维卷积神经网络模型对数据进行多次变换，从数据中提取有效特征，并建立输入数据与设备健康状况之间的映射关系，从而实现对刮板输送机健康状况的识别。对比实验分析表明，所提出的方法可以有效识别刮板输送机的健康状况，准确率达到 98.9%。该方法为后续煤矿综采工作面刮板输送机的健康管理提供了有效的手段和技术支持。
