Surface roughness prediction in fused deposition modelling by neural networks

Fused deposition modelling is a proven technology for the fabrication of both aesthetic and functional prototypes. The obtainable roughness is the most limiting aspect for its application. The prediction of surface quality is an essential tool for the diffusion of this technology, in fact at product development stage, it allows to comply with design specifications and in process planning it is useful to determine manufacturing strategies. The existing models are not robust enough in predicting roughness parameters for all deposition angles, in particular for near horizontal walls. The aim of this work is to determine roughness parameters models reliable over the entire part surface. This purpose is pursued using a feed-forward neural network to fit experimental data. By the utilisation of an evaluation function, the best solution has been found. This has been obtained using a feed-forward neural network for fitting the experimental data. The best solution has been founded by using an evaluation function that we constructed. The validation proved the robustness of the model found to process parameters’ variation and its applicability to different FDM machines and materials.     

Surface roughness prediction in turning of femoral head

This paper is dealing with the development of a surface roughness model for turning of femoral heads from AISI 316L stainless steel. The model is developed in terms of cutting speed, feed rate and depth of cut, using response surface methodology. Machining tests were carried out with TiN–Al₂O₃–TiC-coated carbide cutting tools under various conditions. First-order and second-order models predicting equations for surface roughness have been established by using the experimental results. The established equation shows that the depth of cut was the main influencing factor on the surface roughness. It increased with increasing the depth of cut and feed rate, respectively, but it decreased with increasing the cutting speed. In addition, analysis of variance for the second-order model shows that the interaction terms and the square terms are statistically insignificant. The predicted surface roughness of the samples was found close to the experimentally obtained results within a 95% confident interval.     

开花期稻田上方温、风垂直分布及动力粗糙度特征研究

本文根据开花期稻田微气象观测资料,分析了开花期稻田的一些微气象要素,旨在通过对温、风铅直方向上的分布变化及近地层动力粗糙度特征分析,揭示开花期稻田的一些潜在规律,并为实际稻田的农业生产管理提供指导。根据稻田风廓线符合对数规律这一特征,用最小二乘法求出摩擦速度和粗糙度。分析结果表明:气温铅直分布类型可以分成日射型、辐射型、早上过渡型和傍晚过渡型四类。风速有多种铅直分布模式,共同点是都考虑了因高度的增加,粗糙表面的影响减弱,风速将显著增大。     

Kinematics prediction and experimental validation of machined surface roughness

Online monitoring of surface roughness is a desirable capability for machining processes; however, 100 % inspection of all parts is not feasible unless it can be integrated into the machining process itself through real-time monitoring of cutting conditions. One strategy is to feed these conditions into a predictive modeling kernel which would in turn give the properties of the finished part. In the case of roughness, the surface resulting from turning can be largely represented as the trace of the passing tool geometry. The question addressed herein is whether computationally intensive modeling of the surface accounting for tool nose radius is necessary for online monitoring of surface roughness. This paper presents a predictive modeling methodology wherein the tool-workpiece contact position varies under a simple cutting model, and the resulting surface roughness is estimated. It presents the concept of calculating a “pseudo-roughness” value based only on tool tip locations and to compare this value to that determined by full predictive modeling of the tool geometry. Cutting experimental data has been presented and compared to predictions for model validation. It is found that the root mean square roughness calculation is dominated by tool geometry, rather than tool position deviations and surface roughness estimation could be implemented without a computationally intensive modeling component, thereby enabling online monitoring and potentially real-time control of the part finish.     

In-process surface roughness prediction system using cutting vibrations in turning

Machining is a complex process in which many variables can affect the desired results. Among them, surface roughness is a widely used index of a machined product quality and, in most cases, is a technical requirement for mechanical products since, together with dimensional precision, it affects the functional behavior of the parts during their useful life, especially when they have to be in contact with other materials. In-process surface roughness prediction is, thus, extremely important. In this work, an in-process surface roughness estimation procedure, based on least-squares support vector machines, is proposed for turning processes. The cutting conditions (feed rate, cutting speed, and depth of cut), parameters of tool geometry (nose radius and nose angle), and features extracted from the vibration signals constitute the input information to the system. Experimental results show that the proposed system can predict surface roughness with high accuracy in a fast and reliable way.     

Optimum surface roughness prediction in face milling by using neural network and harmony search algorithm

This paper presents a new approach to determine the optimal cutting parameters leading to minimum surface roughness in face milling of X20Cr13 stainless steel by coupling artificial neural network (ANN) and harmony search algorithm (HS). In this regard, advantages of statistical experimental design technique, experimental measurements, analysis of variance, artificial neural network and harmony search algorithm were exploited in an integrated manner. To this end, numerous experiments on X20Cr13 stainless steel were conducted to obtain surface roughness values. A predictive model for surface roughness was created using a feed forward neural network exploiting experimental data. The optimization problem was solved by harmony search algorithm. Additional experiments were performed to validate optimum surface roughness value predicted by HS algorithm. The obtained results show that the harmony search algorithm coupled with feed forward neural network is an efficient and accurate method in approaching the global minimum of surface roughness in face milling.     

Construction of a surface roughness prediction model for high speed machining

In manufacturing environment prediction of surface roughness is very important for product quality and production time. For this purpose, the finite element method and neural network is coupled to construct a surface roughness prediction model for high-speed machining. A finite element method based code is utilized to simulate the high-speed machining in which the cutting tool is incrementally advanced forward step by step during the cutting processes under various conditions of tool geometries (rake angle, edge radius) and cutting parameters (yielding strength, cutting speed, feed rate). The influences of the above cutting conditions on surface roughness variations are thus investigated. Moreover, the abductive neural networks are applied to synthesize the data sets obtained from the numerical calculations. Consequently, a quantitative prediction model is established for the relationship between the cutting variables and surface roughness in the process of high-speed machining. The surface roughness obtained from the calculations is compared with the experimental results conducted in the laboratory and with other research studies. Their agreements are quite well and the accuracy of the developed methodology may be verified accordingly. The simulation results also show that feed rate is the most important cutting variable dominating the surface roughness state.     

The role of stepover ratio in prediction of surface roughness in flat end milling

Surface roughness prediction studies in end milling operations are usually based on three main parameters composed of cutting speed, feed rate and depth of cut. The stepover ratio is usually neglected without investigating it. The aim of this study is to discover the role of the stepover ratio in surface roughness prediction studies in flat end milling operations. In realising this, machining experiments are performed under various cutting conditions by using sample specimens. The surface roughnesses of these specimens are measured. Two ANN structures were constructed. First of them was arranged with considering, and the second without considering the stepover ratio. ANN structures were trained and tested by using the measured data for predicting the surface roughness. Average RMS error of the ANN model considering stepover ratio is 0.04 and without considering stepover ratio is 0.26. The first model proved capable of prediction of average surface roughness (Ra) with a good accuracy and the second model revealed remarkable deviations from the experimental values.     

Position-varying surface roughness prediction method considering compensated acceleration in milling of thin-walled workpiece

Machined surface roughness will affect parts’ service performance. Thus, predicting it in the machining is important to avoid rejects. Surface roughness will be affected by system position dependent vibration even under constant parameter with certain toolpath processing in the finishing. Aiming at surface roughness prediction in the machining process, this paper proposes a position-varying surface roughness prediction method based on compensated acceleration by using regression analysis. To reduce the stochastic error of measuring the machined surface profile height, the surface area is repeatedly measured three times, and Pauta criterion is adopted to eliminate abnormal points. The actual vibration state at any processing position is obtained through the single-point monitoring acceleration compensation model. Seven acceleration features are extracted, and valley, which has the highest R-square proving the effectiveness of the filtering features, is selected as the input of the prediction model by mutual information coefficients. Finally, by comparing the measured and predicted surface roughness curves, they have the same trends, with the average error of 16.28% and the minimum error of 0.16%. Moreover, the prediction curve matches and agrees well with the actual surface state, which verifies the accuracy and reliability of the model.     

A neural-network-based methodology for the prediction of surface roughness in a turning process

A neural-network-based methodology is proposed for predicting the surface roughness in a turning process by taking the acceleration of the radial vibration of the tool holder as feedback. Upper, most likely and lower estimates of the surface roughness are predicted by this method using very few experimental data for training and testing the network. The network model is trained using the back-propagation algorithm. The learning rate, the number of neurons in the hidden layer, the error goal, as well as the training and the testing dataset size, are found automatically in an adaptive manner. Since the training and testing data are collected from experiments, a data filtration scheme is employed to remove faulty data. The validation of the methodology is carried out for dry and wet turning of steel using high speed steel and carbide tools. It is observed that the present methodology is able to make accurate prediction of surface roughness by utilising small sized training and testing datasets.     

扫描方式对熔融挤压快速成型影响的数值模拟

利用ANSYS有限元软件中的“单元生死”技术，对熔融挤压快速成型的成型过程进行模拟。分析了扫描方式以及长宽比对温度与应力分布的影响，数值模拟的结果与实验现象相符，为合理的选择工艺参数提供了理论依据。     

Effect of surface roughness parameters on mixed lubrication characteristics

In this paper, a computer program was developed to generate non-Gaussian surfaces with specified standard deviation, autocorrelation function, skewness and kurtosis, based on digital FIR technique. A thermal model of mixed lubrication in point contacts is proposed, and used to study the roughness effect. The area ratio, load ratio, maximum pressure, maximum surface temperature and average film thickness as a function of skewness and kurtosis are studied at different value of rms. Numerical examples show that skewness and kurtosis have a great effect on the contact parameters of mixed lubrication.     

Study on the prediction model of surface roughness for side milling operations

The paper presents a feasibility study on prediction of surface roughness in side milling operations using the different polynomial networks. A series of experiments using S45C steel plates is conducted to study the effects of the various cutting parameters on surface roughness. The different polynomial networks for predicting surface roughness are developed using the abductive modeling technique and based on the F-ratio to select their input variables. The results show that the developed models achieve high predicting capability on surface roughness, especially for the case of smaller flank wear of peripheral cutting edge. Hence, it can be concluded that the developed polynomial-network models posses promising potential in the application of predicting surface roughness in side milling operations.     

基于人工神经网络的超声加工表面粗糙度预测

利用BP神经网络良好的非线性映射能力,建立了普通珩磨和超声珩磨条件下的磨削表面粗糙度预测模型,经过多次网络训练,得到了具有良好性能的BP神经网络。对超声珩磨加工钕铁硼材料表面粗糙度进行了预测,并取得了理想的预测结果。     

Practical identification of non-linear characteristics of elastomeric couplings in engine assemblies

In many engineering applications, the connection between a combustion engine and the load is often done using elastomeric materials. However, these couplings far from behaving in a linear way, show a complex behavior that sometimes is difficult to evaluate. In this sense, with the present work it is intended to develop an easy methodology to identify the coupling characteristics to validate dynamic models of engine assemblies with this kind of connections, as well as to identify possible malfunction or damage behavior in a future. The method is based on static and dynamic tests, non-linear models and techniques for parameter identification. The process has been applied using two different couplings, mounted on a single-cylinder diesel engine (DE) and a three-cylinder spark ignition engine (SIE). Results demonstrate the validity of the method for any kind of engine (DE or SIE) and number of cylinders.     

Study on micro-interacting mechanism modeling in grinding process and ground surface roughness prediction

In the research on grinding process modeling, the stochastic nature of grain sizes and locations need to be considered. A new numerical model was developed which will describe the micro-interacting situations between grains and workpiece material in grinding contact zone. The model was established based on a series of reasonable assumptions, the critical conditions of starting points of plowing and cutting stages, and the redefined grinding contact zone. It indicated that there are four types of grain existing in grinding contact zone: uncontact, sliding, plowing, and cutting grains. The number of grains per unit wheel volume (N _v ) and the undeformed chip thickness (h _cu,max), which are key parameters in grinding process modeling, were firstly obtained. The numbers and distributions of different grain types along grinding contact zone were then obtained and analyzed. Calculation results showed that only a small fraction of grains participate in cutting interactions and the changing laws of each grain types along grinding contact length are very different from each other, which gives a deeper insight into grinding process and can be a good foundation for more precise grinding force prediction and thermal analysis. Another important application of this model is for ground surface roughness prediction and a new method on this purpose was developed. At last, two comparisons were made between calculation results and existing experimental data for validating the work on paper. Comparison results showed that the roughness of ground surface can be well predicted and gave the method theoretically to reduce ground surface roughness.     

CBN刀具车削淬硬钢表面粗糙度的试验与预测

利用正交设计方法,对立方氮化硼(CBN)刀具硬态干式车削淬硬钢Cr12Mo V时,切削用量三要素(切削速度、进给量和切削深度)对加工表面粗糙度的影响进行了分析,运用响应曲面法(RSM)建立了加工表面粗糙度的预测模型。研究结果表明:CBN刀具车削淬硬钢Cr12Mo V时对加工表面粗糙度影响最大的加工参数是切削速度,其次是进给量,切削深度对加工表面粗糙度的影响较小;预测模型能够高精度地对表面粗糙度进行预测,平均误差不超过9.7%。