Search

Balance is one of the most significant and commonly used measurement instruments in wind tunnels. It is thus critical that this instrument is fault-tolerant and reliable. To help in the design of dependable balance systems, one must model and analyze the system from this perspective. We achieved this by considering the structure of a typical balance system and the types of errors it may encounter during operation. We estimated the failure rates of the components of the balance system, such as the balance structure, strain gages, and power supply, using manufacturer technical documents and fatigue tests, and identified the effect of each component on the reliability of the balance system. On this basis, we presented a Markov model for the reliability of the balance system, assuming that the data acquisition of a typical flying model should be complete in less than 15 min in order for the balance system to be considered reliable. Simulation of the proposed model with SHARPE software showed that if we reduced the time of the experiments from 30 min to 15 min, the reliability of the balance system improved by about 30%. In addition, the cost of the experiments in the wind tunnel was reduced by 20%-30%. [ABSTRACT FROM AUTHOR]