Currently, a key metric for evaluating the value of cutting-edge technologies, such as AI models, is how much they can be integrated into production processes, so as to become a catalyst for change in traditional industries. The global mining industry, in particular, is now embracing the change that comes with AI.
In this critical period of transformation, Sichuan Zigong Conveying Machine Group Co Ltd (ZGCMC) has deepened its cooperation with Huawei by establishing a joint innovation centre through new company Huayun Zhiyuan (Chengdu) Technology Co Ltd. Together, they are driving the deep integration of AI within the traditional mining industry, creating next-generation intelligent solutions for global customers.
The traditional mining industry has long faced challenges such as high equipment maintenance costs, difficult fault prediction, ensuring production continuity, and managing personnel safety. Taking the mining conveyor system as an example, a conveyor belt that is several kilometres long often contains thousands of rollers. As well as being prone to failure, these rollers are difficult to inspect. Once they fail, they not only risk damaging or even tearing the expensive conveyor belt but may cause fires, leading to significant production losses and safety risks.
In response to these challenges, Huayun Zhiyuan collaborated with Huawei to incubate a range of scenario-based solutions, products, and services focused on the intelligent mining conveyor system. At its core, it aims to create a closed-loop O&M system based on Huawei Cloud Stack hybrid cloud that combines sensing, analytics, decision making, and coordination.
The system integrates many cutting-edge technologies, including distributed fiber optic sensing, machine vision, the Internet of Things, and a predictive maintenance model. In this way, it enables real-time monitoring of equipment status, early fault diagnosis, and predictive maintenance. Ultimately, it achieves cost reduction, efficiency improvement, and safe production through inspections with little or no human involvement along the ore workflow.
In a pilot project at a copper mine in Lhasa, Xizang, a distributed fibre optic-based roller anomaly monitoring system has been deployed. The system uses distributed fibre optic acoustic sensing technology to transform communications cables into a ‘stethoscope’ that covers several kilometres. The cables can capture subtle vibrations and acoustic changes during roller operations in real time. In the back end, the AI model analyses massive amounts of data to locate rollers likely to fail with a precision of within 4 m, as well as categorise faults. In addition, O&M engineers can remotely ‘listen’ to verify the faults. This has transformed reactive emergency repairs into proactive planned maintenance, ensuring the continuous safety of the production ‘arteries’ 24/7.
In an intelligent O&M project for belt conveyors at a mine in Chongqing, temperature and vibration sensors have been deployed on the conveyors’ electromechanical drive equipment. Data generated by these sensors is uploaded in real time via IoT gateways. Through Huawei Cloud Stack, a hybrid cloud platform, the Pangu prediction model learns the normal operating patterns of the equipment. Once the system detects subtle deviations from the normal state, it provides early warnings of potential faults and predicts the optimal maintenance window.
In addition, HD self-cleaning cameras can capture video in real time, and the computer vision (CV) model is used to monitor the belts’ tracking deviation, damage, and foreign objects, as well as personnel safety 24/7, creating a multi-dimensional protection network. This solution effectively reduces unplanned downtime, lowers spare parts inventory, and reduces O&M labor costs.
In fact, equipment intelligent transformation goes beyond optimising a single device or process. In the predictive maintenance solution developed for a mining company group, the two parties established data links across five core industrial systems, bringing over 1,600 critical pieces of equipment into the ‘sight’ of AI. Through analysis by the Pangu prediction model, machine inspections have replaced a large number of manual inspections, enabling precise predictions of equipment trend anomalies, early faults, and energy efficiency changes. This has laid the foundation for the future application of this solution to tens of thousands of equipment in a mining company group.
In the future, Huayun Zhiyuan and Huawei are to continue to advance their cooperation, from the incubation and commercialisation of CV and prediction models to the adoption of natural language processing (NLP) and multimodal models. The scope of the cooperation will extend from the ore transportation system to the entire ore workflow. Through technological innovation and ecosystem co-building, the two parties will invigorate traditional industries like mining by driving safer, smarter, and more efficient production and O&M, setting a new benchmark for the digital and intelligent transformation of global mining operations.