The China Building Materials Federation, Conch Group and Huawei recently held an event in Wuhu, China, to showcase their AI model for the cement building materials industry. More than 340 government leaders, industry experts, enterprise representatives and journalists attended the event. Attendees visited demonstration bases such as Baimashan Cement plant and Conch Wuhu, where the model is being implemented.
With the support of the China Building Materials Federation, Conch Group and Huawei began constructing an AI model for the cement building materials industry in April 2024. Since then, Conch Group and Huawei have identified over 200 promising AI application scenarios across 15 categories. These span the entire production process, from mining to packaging and shipment. Conch has set up an AI training centre using Huawei Cloud Stack. The centre is using Huawei Cloud Pangu prediction, and models to create an AI operating system that integrates central training, edge inference, cloud-edge synergy, continuous learning and ongoing optimisation.
The AI model leverages extensive cement industry data and industry expertise. Through real-time data analysis and autonomous learning, it has made significant breakthroughs in more than 40 scenarios in five categories – quality control, production optimisation, equipment management, safe production and intelligent Q&A. Where the model has been implemented so far, operators have benefitted from dynamic optimisation of process parameters, response to exception warnings in seconds and maximisation of resource utilisation, introducing a new intelligent engine for high-quality industry development.
Using the Huawei Cloud Pangu prediction model, real-time recommendations of key quality features enable accurate prediction of three-day and 28-day clinker strength. The predicted strength values closely match test results, with deviations within 1MPa and an accuracy rate exceeding 85 per cent. This allows for the optimisation of raw material mixtures and cement formulas, shifting from post-event adjustment to real-time control.
Global optimisation model
Regarding production optimisation, a global optimisation model for clinker burning is created by integrating data from multiple sources in the production process, studying the control strategies of the burning system, and utilising expert knowledge. This model provides real-time recommendations for key process parameter targets and automatically adjusts the optimal operational plan based on varying operating conditions. This enables a one per cent reduction in standard coal consumption beyond the level-1 energy efficiency baseline. For a 5000tpd clinker line, this leads to a reduction of over 4500tpa of CO2 emissions.
The AI model in the cement building materials industry represents not only a significant achievement for Conch Group in its digital transformation journey but also embodies the result of deep collaboration between Conch Group and Huawei. Conch Group and Huawei plan to continue to use advanced technologies like AI to fuel intelligent transformation, and foster steady and rapid growth in sectors such as cement, building materials and the wider manufacturing sector.
