275 Figure 10-(a) Construction mechanism of digital twin model, (b) schematic diagram of real-time fault diagnosis, (c) schematic diagram of modularization The system takes the digital twin model of multi-source fusion as the core, and builds a three-dimensional digital twin platform for belt conveyor, to realize the intelligent control of the whole process of equipment, with multimodal perception and early warning functions for major faults such as fire and belt breakage, so that the normalized operation rate of equipment can be increased to 95%, and the unplanned shutdown and maintenance costs can be reduced. A high speed synchronous acquisition device based on edge calculation is developed, which has the ability of 2.7G floating point operation, and supports real time processing and diagnosis analysis of multi-source signals. The principle of real time fault diagnosis is shown in Figure 10(b). The multimodal information knowledge base is constructed, and the intelligent diagnosis expert system of key components is developed, to realize health status assessment and early fault warning. The accuracy of fault prediction is 90%. Figure 10(c) shows the components of each module in the intelligent diagnosis expert system for belt conveyor. INDUSTRIAL APPLICATION VALIDATION In Figure 11, the industrial tests is conducted on the main inclined shaft of the Kekegai coal mine for this study. The core parameters of the entire machine are shown in Table 2. The operation results show that all indexes of the system meet the design requirements, meeting the annual production capacity of 25-30 Mt of raw coal. The start stop is stable, the operation is reliable, and the core parameters such as power balance, dynamic tension, and key component vibration are all within the safety threshold. Table 2-Comparison of key performance indicators Content International Chinese level This paper Comparison
RkJQdWJsaXNoZXIy MTM0Mzk2