Author(s)

Liu Zhaofeng ¹, Qi Ziyuan ¹, Cui Kaibo ¹, Guo Zixi ¹

Affiliation(s)

¹ Shijiazhuang Campus, Army Engineering University of PLA, Shijiazhuang, Hebei, China

Corresponding Author

Qi Ziyuan

ABSTRACT

The performance and reliability of complex equipment operating in extreme environments, particularly at high altitudes, are critical for operational success and safety. Traditional assessment methods often struggle with the dynamic and coupled nature of environmental stressors and equipment responses. This paper proposes a multi-source data-driven framework to address these challenges. The framework integrates data from equipment sensors, environmental monitoring sources (including satellite and ground-based systems), and operational/maintenance logs using a spatio-temporal alignment approach. It employs a dynamic weighting method combining the Entropy Weight Method (EWM) and Analytical Hierarchy Process (AHP) for adaptability quantification, adjusting parameter importance based on real-time conditions. A novel hybrid machine learning architecture, HybridML-ADAPT, combining Random Forest and LSTM, is introduced for modeling complex interactions and temporal dependencies to predict equipment adaptability levels and performance degradation. Enhanced anomaly detection mechanisms incorporating environmental context are used to improve reliability. The framework was validated through a case study involving high-altitude deployed electronic monitoring systems. Results demonstrate significant improvements, including a 34.2% increase in power degradation prediction accuracy, an 85.7% increase in Mean Time Between Failures (MTBF), and a 61.9% reduction in fault detection delay following framework-guided optimizations. This research provides a robust methodology for assessing and enhancing equipment resilience in challenging high-altitude conditions.

KEYWORDS

Environmental Adaptability, Multi-Source Data-Driven, Dynamic Optimization, High-Altitude Environments, Equipment Reliability, Hybrid Machine Learning, Predictive Maintenance

CITE THIS PAPER

Liu Zhaofeng, Qi Ziyuan, Cui Kaibo, Guo Zixi, A Multi-Source Data-Driven Framework for Environmental Adaptability Assessment and Dynamic Optimization of Complex Equipment. Journal of Engineering Mechanics and Machinery (2025) Vol. 10: 95-115. DOI: http://dx.doi.org/10.23977/jemm.2025.100111.

REFERENCES

[1] Scott M. Lundberg, S.L. A Unified Approach to Interpreting Model Predictions. Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017,4768-4777.
[2] Shi, W.; Cao, J.; Zhang, Q.; Li, Y.; Xu, L. Edge Computing: Vision and Challenges. IEEE Internet Things J. 2016, 3, 637-646.
[3] Saaty, T.L. Decision making with the analytic hierarchy process. International Journal of Services Sciences. 2008, 1, 83-98.
[4] Liu Wei; Xue Dong; Wang Pin. Research on Weapon System Guarantee of a Certain Type of Missile under High Plateau and High Cold Battlefield Environment. Science and Technology Wind.2018, 191-193.
[5] Yang Q.Federated Learning and Artificial Intelligence.Software and Integrated Circuits.2019, 52-53.
[6] Zhao, Z. M.; Liu, F.; Cai, Z. P.; Xiao, N..Edge computing: platforms, applications and challenges.Computing Research and Development.2018, 55, 327-337.
[7] Wang, X.W.; Fan, D.F.; Lei, J.D.; Li, H.M.; Chen, K.L.Combat test evaluation of equipment based on multi-source information fusion. Firepower and Command and Control.2024, 49, 192-198. 
[8] Yang Q. Research on image classification algorithm based on heterogeneous federated learning. University of Electronic Science and Technology, 2023.
[9] Levy, R.C.; Mattoo, S.; Munchak, L.A.; Remer, L.A.; Sayer, A.M.; Patadia, F.; Hsu, N.C. The Collection 6 MODIS aerosol products over land and ocean. Atmos. Meas. Tech. 2013, 6, 2989-3034.
[10] Fei Tony Liu, K.M.T.Z. Isolation Forest. 2008 Eighth IEEE International Conference on Data Mining, 2008,413-422. 
[11] BREIMAN, L. Random Forests. Machine Learning:Kluwer Academic Publishers, 2001:5-32.
[12] Sepp Hochreiter, J.S. Long Short-Term Memory. Neural Comput. 1997, 9, 1735-1780. 
[13] Arcas, H.B.M.E. Communication-Efficient Learning of Deep Networks from Decentralized Data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, 2017,1273-1282.

Sponsors, Associates, and Links

---

• Cybernetics and Mechatronics
  
  
• Digital Manufacturing and Process Management
  
  
• Ultra-Precision Machining Process
  
  
• Journal of Robotics and Biomimetics
  
  
• Prognostics, Diagnostics and Health Management
  
  
• Micro-Electro-Mechanical Systems
  
  
• Journal of Precision Instrument and Machinery
  
  
• Engineering and Solid Mechanics
  
  
• Fracture and Damage Mechanics
  
  
• Frontiers in Tribology
  
  
• Fluid and Power Machinery
  
  
• Chemical Process Equipment
  
  
• Journal of Assembly and Manufacturing
  
  
• Mechanical Vibration and Noise