An analytical approach on stochastic model for cutting force prediction in milling ceramic matrix composites. (15th February 2020)

• Record Type: Journal Article
• Authors: Zhang, Xuewei; Yu, Tianbiao; Zhao, Ji
• Is Part Of: International journal of mechanical sciences. Volume 168(2020)
• Publication Date: 2020-02-15
• Subjects: Fiber-reinforced ceramic matrix composites -- Milling -- Cutting forces -- Stochastic distribution of carbon fibers -- Variable tool wear; Mechanical engineering -- Periodicals; Génie mécanique -- Périodiques; Mechanical engineering; Maschinenbau; Mechanik; Zeitschrift; Periodicals; 621.05
• DOI: 10.1016/j.ijmecsci.2019.105314
• Languages: English
• ISSNs: 0020-7403

Abstract

Highlights: An analytical cutting force model with stochasticity of fiber distribution and tool wear is presented for milling ceramic matrix composites. The instantaneous relative content of fibers is introduced based on the Student-t distribution to construct the cutting mechanism algorithm. The cutting forces model is formulated in the shear deformed region, the friction deformed region and the ploughing region. A probabilistic approach based on the particle filter is used to predict the stochastic tool wear by measurement data of cutting forces. The empirical uncertain components of cutting forces considering tool wear are determined with a radial basis function neural network. Abstract: Fiber-reinforced ceramic matrix composites are increasingly applied in the aerospace, energy and electronic industries. Nevertheless, the milling process of ceramic matrix composites is considerably difficult owing to the presence of heterogeneous, anisotropic and brittle nature. This paper presents a novel stochastic model of cutting forces in milling process of ceramic matrix composites. The algorithm model of randomly distributed carbon fibers is developed by incorporating the cutting mechanism. The obtained instantaneous relative content of fibers is introduced for dividing the resultant cutting forces into fiber and matrix components in the shear deformed region, the friction deformed region and the ploughing region, respectively. In addition, a probabilistic approach based on the particle filter is used to predict the random tool wear progression, linking online measurement data with the state of tool wear. Then, the empirical uncertain components of cutting forces considering tool wear can be established by using a radial basis function (RBF) neural network. The predicted cutting forces are in good agreement with the measured values. The effects of stochastic fiber distributions, tool wear and machining parameters on cutting forces are investigated by the proposed model. Graphical abstract: Image, graphical abstract