Application of one-step singular correlation in random-across probability modeling

The relationship between the position for data pretreatment and modeling accuracy for the process of random-across was effectively studied based on the singular correlation law of one-step differential variance identification and treatment methods. Based on the practically measured data from a barrel´s random-across shot door, the accuracy of statistical model and theoretical model of density function of residence time, awaiting time and crossing period were analyzed through the simulations. The results show that, through this method, the accuracy of the model for across series becomes much higher. The method is also an effective treatment way for online modeling in maneuver condition.