Influence of Measurement Parameters on the Accuracy of Determining the Capacity of a Vertical Tank

摘要

Vertical tanks are commonly used for oil storage. The accuracy of measurement of the tank's capacity depends directly on the method used for processing the experimental data. When the Monte Carlo method is applied for evaluating the capacity of a vertical tank, the measurement result is determined by the factors of sensor point distribution, sensor point density, and sample point density. Therefore, in this study, we investigate the impact of these factors and their influence on the evaluated capacity of a vertical tank based on the Monte Carlo method. The research results show that the absolute and relative errors of the orderly distribution are similar to those of the rotary distribution, their absolute errors do not exceed +/- 0.1 m(3), and the relative errors can be converged to within 0.1. When the sensor points are randomly distributed, the stable value of the absolute error differs from 0, and the relative error cannot be converged to within 0.1, thus, the random distribution is not acceptable for measuring the capacity of vertical tanks. When the sensor point density is higher than or equal to 1.44 /m(2), the absolute error of the measurement result does not exceed +/- 0.1 m(3), and the relative error can be converged to within 0.1, which is acceptable for measuring the capacity of vertical tanks. The higher the sample point density, the smaller the absolute error range and the faster the relative error convergence speed. When the sample point density is higher than or equal to 1745.23 /m(3), the relative error can be converged to within 0.1.