Extrusion printing processes allow for manufacturing complex shapes in a relatively cheap way with low-cost machines. The present study analyzes the effect of printing parameters on dimensional error, roughness, and porosity of printed PLA parts obtained with grid structure. Parts are obtained by means of the fused filament fabrication (FFF) process. Four variables are chosen: Layer height, temperature, speed, and flow rate. A two-level full factorial design with a central point is used to define the experimental tests. Dimensional error and porosity are measured with a profile projector, while roughness is measured with a contact roughness meter. Mathematical regression models are found for each response, and multi-objective optimization is carried out by means of the desirability function. Dimensional error and roughness depend mainly on layer height and flow rate, while porosity depends on layer height and printing speed. Multi-objective optimization shows that recommended values for the variables are layer height 0.05 mm, temperature 195 ºC, speed 50 mm/min, and flow rate 0.93, when dimensional error and roughness are to be minimized, and porosity requires a target value of 60%. The present study will help to select appropriate printing parameters for printing porous structures such as those found in prostheses, by means of extrusion processes.

In the present paper an indirect model based on neural networks is presented for modelling the rough honing process. It allows obtaining values to be set for different process variables (linear speed, tangential speed, pressure of abrasive stones, grain size of abrasive and density of abrasive) as a function of required average roughness Ra. A multilayer perceptron (feedforward) with a backpropagation (BP) training system was used for defining neural networks. Several configurations were tested with different strategies, number of layers, number of neurons and transfer function. Best configuration for the network was searched by means of two different methods, trial and error and Taguchi design of experiments (DOE). In both cases, best configuration corresponds to a single network with two hidden layers. Once best configuration was found, a network was defined for obtaining honing parameters as a function of required roughness parameters related to Abbott-Firestone curve, Rk, Rpk and Rvk.

This study focuses on obtaining regression models for material removal rate and tool wear in rough honing processes. For this purpose, experimental tests were carried out according to a central composite design of experiments. Five different parameters were varied: grain size or particle size of abrasive, density of abrasive or abrasive concentration, pressure of the stones against the cylinder internal surface, tangential speed (in this case, corresponding to the rotation speed of the cylinder), and linear speed of the honing head. In addition, multi-objective optimization was carried out with the aim of maximizing the material removal rate and minimizing tool wear. The results show that, within the range studied, the material removal rate depends mainly on tangential speed, followed by grain size and pressure. Tool wear is directly influenced by density of abrasive, followed by pressure, tangential speed, and grain size. According to the multi-objective optimization, if the two responses are given the same importance, it is recommended that high grain size, high density, high tangential speed, and low pressure be selected. Linear speed has less influence on both responses studied. If the material removal rate is considered to be more preponderant than tool wear, then the same values should be considered, except for high pressure. If tool wear is preponderant, then lower grain size of 128 (ISO 6106) should be selected, and lower tangential speed of approximately 166 min−1. The other variables, density and pressure, would not change significantly from the first situation.

No abstract

In present study, three roughness parameters defined in the Abbott-Firestone or bearing area curve, Rk, Rpk and Rvk, were modeled for the rough honing processes by means of Artificial Neural Networks (ANN). Input variables were grain size and density of abrasive, pressure of abrasive stones on the workpiece's surface, tangential or rotation speed of the workpiece and linear speed of the honing head. A back propagation algorithm was used for training the networks. Two strategies were considered, use of either one network for modeling the three parameters at the same time and use of three networks, one for each parameter. Best network was chosen among different structures, having either one or two hidden layers. When one network is considered, best solution corresponds to two hidden layers having 26 and 11 neurons. However, overall best neural network consists of three networks, one for each roughness parameter, with one hidden layer having 25, 9 and 5 neurons respectively.