Multi-state balance system reliability research considering load influence

In past studies of the balanced system, it was often assumed that there were multiple identical units in the system with symmetrically distributed. When a unit in the system fails, the system will shut down the unit, which was symmetrically distributed with the failure unit, to ensure that the system meets the balance requirements. However, in some real engineering cases, the redundancy of the system is low. If we continue completely closing the symmetrical position unit on the system, the system may fail due to insufficient redundancy and the existing model of the balance system is inapplicable. To solve this problem, a general balance system model in which the unit will change its output load rather than completely closing when the system needs to re-establish the balance is proposed in this paper. Furthermore, due to the different reliability of unit corresponding to different loads, using the accelerated failure time model and power law principle, the relationship between different output loads and unit reliability are studied. A new reliability evaluation method is established by using the Markov-Chain approach and obtaining a new multi-state balance system reliability model. Finally, taking the propulsion subsystem of a solar UAV as an example, the effectiveness of the evaluation method is verified.

Automated summary

Previous studies assumed a balanced system with multiple identical units symmetrically distributed. If a unit in the system failed, the symmetrically distributed unit would be shut down to maintain balance. However, in real engineering cases, system redundancy is often low, making the existing model inapplicable. To address this, a new balance system model is proposed, where units change their output load instead of completely shutting down to re-establish balance. Additionally, the relationship between different output loads and unit reliability is studied using accelerated failure time model and power law principle. A new reliability evaluation method is established using Markov-Chain approach, and its effectiveness is verified through a case study on a solar UAV's propulsion subsystem.