Uncertainty Assessment For Roundness Error Particle Swarm Optimization(PSO) Intelligent Evaluation

Roundness as a mechanical product of the basic geometric elements in common, is an important evaluation indicator in rotary parts, and is frequently assessed in the equipment processing. Roundness is very common in mechanical parts, it is an important indicator of the accuracy of part geometry. Ability to correctly measure and assess the value of roundness error is essential for improving the machinery and ensuring the quality of the product. Articles with " uncertainty assessment for roundness error particle swarm optimization(PSO) intelligent evaluation " in the title, this paper study a particle swarm optimization to calculate roundness error and uncertainty assessment, which will promote the development of measurement geometry and has important academic value and the actual significance.This paper has had a wide analysis on the of the classical numerical methods, computational geometry algorithms and intelligent algorithm for roundness error assessment in domestic and foreign research. This paper determined using a particle swarm algorithm to calculate the error of roundness error for increasing accuracy and reducing computing time. This paper has has studied the uncertainty for intelligent assessment, and given the mathematical model for uncertainty assessment given. Its main tasks are:(1)According to ISO international standards for roundness error definition, the mathematical model of roundness error was derived. Based on particle swarm optimization theory and geometric properties of roundness error, initial conditions of the particle, the objective function and shutdown conditions were structured.(2)This paper studied the dispersion of PSO: firstly uncertainty source for the dispersion was analysis, then the use ofβ-distribution system Analysis intelligent evaluation results show the probability distribution of types. The implementation uncertainty Roundness measuring instruments for exploratory study, the mathematical model of uncertainty was structured based on measurement precision components perform.(3)The uncertainty assessment for intelligent evaluation was investigated. Intelligent assessment of the results of research in the case of small samples, using an unbiased estimate of the maximum entropy method to calculate the probability distribution function by which the function the fitting uncertainty is estimated . Monte Carlo study based on combined uncertainty for intelligent evaluation was studied, including measurement error and error simulation intelligent algorithm.This paper measured a ring by CMM to obtain surface measurement point data, then used PSO algorithm to calculate the ring roundness error, respectively, using the maximum entropy method and Monte Carlo to assessment the uncertainty. The results show that the proposed PSO algorithm has the global optimum performance and fast convergence, shorter time-consuming calculations. The maximum entropy method and Monte Carlo uncertainty assessment methods are able to rationally and effectively evaluate the intelligence evaluation uncertainty, and easy to manipulate.