Design and Application of an Edge Computing-Based Condition Monitoring and Fault Diagnosis System for Coal Mining Machinery

DOI: https://doi.org/10.62517/jes.202502425

Author(s)

Kunpeng Ge1, Fei Dong2,*, Yaqi Xu1

Affiliation(s)

1College of Information Engineering, Suqian University, Suqian, Jiangsu, China 2School of Internet, Anhui University, Hefei, Anhui, China *Corresponding Author

Abstract

Addressing the challenges of poor real-time data processing and high bandwidth consumption inherent in traditional cloud computing for coal mining machinery condition monitoring, alongside the limited generalization capability of existing fault diagnosis models, this paper proposes a distributed intelligent monitoring and diagnostic framework integrating edge computing, wireless sensor networks (WSN), and deep learning-based transfer learning techniques. Initially, a multi-source data acquisition system grounded on WSN is established, utilizing the STM32F405 microcontroller, ADXL1005 MEMS sensor, and nRF24L01+ modules to enable high-precision vibration data collection and robust transmission. Subsequently, an edge computing terminal powered by the RK3588 processor is designed, featuring heterogeneous communication and multi-source data aggregation capabilities, effectively reducing data transmission latency. Finally, the fault diagnosis model is refined through transfer learning to filter effective deep features and minimize distribution discrepancies between source and target domains. Experimental results demonstrate that the system achieves 16-bit data acquisition precision, wireless transmission success rates exceeding 98%, and a 12.3% improvement in fault diagnosis accuracy compared to conventional deep learning methods, thereby fulfilling the real-time monitoring and fault diagnosis requirements of equipment operating in the complex environments of coal mines.

Keywords

Edge Computing; Coal Mining Machinery; Fault Diagnosis; Wireless Sensor Network (WSN); RK3588 Processor; Transfer Learning

References

[1] Qin Tao, Yang Xingzhou. Research on Fault Diagnosis of Coal Mine Electromechanical Equipment Based on Domain Adaptation Method. Coal Technology, 2025, 44(11): 1-6. [2] Zhu Hongyu. Application Research of Remote Online Fault Diagnosis System in Coal Mine Shearer Operation. Mechanical Management and Development, 2025, 40(10): 203-204+207. [3] Wang Shuo, Ji Xinxin, Han Qi, et al. Online Monitoring and Fault Diagnosis Technology for Transformers and Capacitive Devices Based on Integrated Regulation Automation System. Science and Technology Innovation, 2025, (21): 57-60. [4] Zhao Xingwei. Application and Optimization of 5G Edge Computing in Mobile Video Surveillance Platforms. Information and Computers, 2025, 37(21): 90-92. [5] Lin Chuan, Liu Rongfeng, Su Yan, et al. Underwater image enhancement and crack quantification driven by deep learning and transfer learning. Journal of Hydroelectric Engineering, 2025, 44(10): 73-84. [6] Zhang Wei, Tian Hongxun, Sun Yang, et al. Research on Clustering Routing Strategy of Wireless Sensor Networks in Intelligent Substations Based on Energy Consumption Balancing. Electrical Measurement and Instrumentation, 2025, 62(11): 176-181. [7] Zhou Sikang, Zhang Mingzhu, Wang Ningning, et al. Research on Wear State Recognition of Grinding Wheel Based on ECA-VGG16 and Transfer Learning. Machine Tool & Hydraulics, 2025, 53(19): 103-109. [8] Liu Ziliang. Terminal Design of Shearer Vibration Signal Acquisition System Based on STM32. Industrial and Mining Automation, 2025, 51(S1): 94-97. [9] Guo Rui, Shi Xuezhu. Design and Implementation of Intelligent Patrol Car Based on Raspberry Pi and Image Recognition Technology. Internet of Things Technologies, 2025, 15(15): 150-153. [10] Huang Xinyi, Guo Zile, Huang Lizhi, et al. Vehicle Access Control System Based on Raspberry Pi. Industrial Control Computer, 2025, 38(07): 125-127.