Data-Driven and Model-Driven Integration Approach for Optimizing Equipment Safety Investment in Digital Twin Coal Mining Enterprises

Abstract

In coal mining companies, investment in equipment safety plays a crucial role in improving equipment safety and ensuring worker safety. To address issues such as subjective and uncertain equipment safety investment methods leading to irrational resource allocation and poor safety and economic outcomes in coal mining enterprises, a data- and model-driven approach based on digital twin technology is proposed for optimizing safety investment and predicting accident losses in coal mine equipment. The effectiveness of the investment optimization plan is validated by predicting accident losses post-implementation, ensuring maximized safety and economic benefits of the investment plan. Finally, using S company’s equipment safety investment as a case study, the proposed method is validated. Experimental results demonstrate that the optimized investment plan reduces accident losses by 11.73% compared to traditional coal mine equipment safety investment schemes. Furthermore, in accident loss prediction, the IPSO-BP model (R2 = 0.99) outperforms traditional PSO-BP (R2 = 0.96) and BP (R2 = 0.93) models, showing higher accuracy and suitability for accident loss prediction.

Keywords

• coal mine equipment
• safety investment
• digital twin
• optimization
• accident loss prediction