An Automatic Control Method for the Number of Simulation Experiments based on Multi-variable Steady-state Analysis

Different experiment times for the same simulation experiment affect the validity of the experimental results. In this paper, the basic principle of Monte Carlo method for determining simulation times is introduced, and its limitations in large-scale combat scenarios are analyzed. Then, a steady-state analysis method of variables is proposed, which takes the core index concerned by experiment-personnel as the investigation variable and passes normality test and significance analysis. In addition, the overall structure of automatic control of simulation experiment times is designed and the effectiveness of this method is verified. For the application scenarios where a large number of simulation experiment schemes need to be deduced, this method can realize unattended, and can greatly improve the experimental efficiency while ensuring the effectiveness of experimental results.