Fuji Wang

Characteristics forecasting of hydraulic valve based on grey correlation and ANFIS

Accurate prediction is crucial for the synthesis characteristics of the hydraulic valve in industrial production. A prediction method (G-ANFIS for short) based on grey correlation and adaptive neuro-fuzzy system (ANFIS) to forecast synthesis characteristics of hydraulic valve is devised and the utilizing of the method can help enterprises to decrease the repair rate and reject rate of the product. Grey correlation model is used first to get the main geometric elements affecting the synthesis characteristics of the hydraulic valve and thus simplifies the system forecasting model. Then use ANFIS to build a prediction model based on the above mentioned main geometric elements. To illustrate the applicability and capability of the devised prediction method, a specific hydraulic valve production was used as a case study. The results demonstrate that the prediction method was applied successfully and could provide high accuracy. The method performed better than artificial neural networks (ANN) to forecast the synthesis characteristics of hydraulic valve.

Hybrid of simulated annealing and SVM for hydraulic valve characteristics prediction

Accurate prediction for the synthesis characteristics of hydraulic valve in industrial production plays an important role in decreasing the repair rate and the reject rate of the product. Recently, Support Vector Machine (SVM) as a highly effective mean of system modeling has been widely used for predicting. However, the important problem is how to choose the reasonable input parameters for SVM. In this paper, a hybrid prediction method (SA-SVM for short) is proposed by using simulated annealing (SA) and SVM to predict synthesis characteristics of the hydraulic valve, where SA is used to optimize the input parameters of SVM based prediction model. To validate the proposed prediction method, a specific hydraulic valve production is selected as a case study. The prediction results show that the proposed prediction method is applicable to forecast the synthesis characteristics of hydraulic valve and with higher accuracy. Comparing with Adaptive Neuro-Fuzzy Inference System (ANFIS) and Artificial Neural Networks (ANN) are also made.

Improved camera calibration method based on perpendicularity compensation for binocular stereo vision measurement system

High-precision calibration of binocular vision systems plays an important role in accurate dimensional measurements. In this paper, an improved camera calibration method is proposed. First, an accurate intrinsic parameters calibration method based on active vision with perpendicularity compensation is developed. Compared to the previous work, this method eliminates the effect of non-perpendicularity of the camera motion on calibration accuracy. The principal point, scale factors, and distortion factors are calculated independently in this method, thereby allowing the strong coupling of these parameters to be eliminated. Second, an accurate global optimization method with only 5 images is presented. The results of calibration experiments show that the accuracy of the calibration method can reach 99.91%.

The interaction between the cutting force and induced sub-surface damage in machining of carbon fiber-reinforced plastics:

Carbon fiber-reinforced plastics (CFRPs) have the characteristics of non-homogeneity and anisotropy. Damage occurs frequently in machining of CFRPs, and it can seriously influence the performance of work piece. This study builds a finite element model for machining of CFRPs based on the constitutive relation with damage, the Hashin failure criterion, and the damage evolution. The continuous cutting processes of unidirectional CFRPs with various fiber orientations are simulated. Cutting forces and sub-surface damage are determined from simulations. Furthermore, machining experiments on unidirectional-CFRPs are performed. Cutting processes are monitored, and cutting forces are measured. An artificial neural network (ANN) force model is proposed by using the experimental data, and then simulation results of the cutting forces are validated by these of the ANN model. Cutting force increases when the fiber orientation varies from 0° to 135°. Fiber orientation is the critical factor affecting the cutting force and the sub-surface damage. More sub-surface damage occurs in a fiber orientation range of 90–135°. The primary reasons for the induced sub-surface damage include the damage evolution and the crack propagation of matrix caused by the cutting force. In addition, the effects of cutting parameters and tool geometries on the cutting force and the damage are discussed by simulations. The cutting force thus can be reasonably controlled to reduce the damage.

Spindle Speed Selection for High-Speed Milling of Titanium Alloy Curved Surface

The development of high-speed milling technology provides an effective processing method for titanium alloy curved surface with high quality, and the spindle speed is an important machining parameter for the high-speed milling of titanium alloy curved surface. The variation of the geometric features of the titanium alloy curved surface results in the sharp fluctuation of the cutting force as well as the vibration of machine tool, which not only makes a severe impact on the surface machining quality and the tool life but also greatly affects the efficiency of the high-speed milling. An experimental study is carried out to determine the spindle speed for high-speed milling of the titanium alloy curved surface based on the cutting force. The experimental results indicate that in high-speed milling process, the cutting force is associated with the geometric feature of the curved surface and the change of cutting force is relatively smooth when the spindle speed is in the range from 9000 to 13,000 rpm for the machining of titanium alloy curved surface.

A New Method for Discharge State Prediction of Micro-EDM Using Empirical Mode Decomposition

The property of high frequency in micro-EDM (electrical discharge machining) causes the discharge states to vary much faster than in conventional EDM, and discharge states of micro-EDM have the characteristics of nonstationarity, nonlinearity, and internal coupling, all of this makes it very difficult to carry out stable control. Thus empirical mode decomposition is adopted to conduct the prediction of the discharge states obtained through multisensor data fusion and fuzzy logic in micro-EDM. Combined with the autoregressive (AR) model identification and linear prediction, the mathematical model for EDM discharge state prediction using empirical mode decomposition is established and the corresponding prediction method is presented. Experiments demonstrate that the new prediction method with short identification data is highly accurate and operates quickly. Even using short model identification data, the accuracy of empirical mode decomposition prediction can stably reach a correlation of 74%, which satisfies statistical expectations. Additionally the new process can also effectively eliminate the lag of conventional prediction methods to improve the efficiency of micro-EDM, and it provides a good basis to enhance the stability of the control system.

Research on the carbon fibre-reinforced plastic (CFRP) cutting mechanism using macroscopic and microscopic numerical simulations

The essence of cutting carbon fibre-reinforced plastic (CFRP) composites is a process of material failure and chip formation. The mechanism of cutting CFRPs can be explained from the perspective of local removal of material on the microscopic level. The morphology of the chips resulting from the cutting process can be determined from the perspective of the overall failure of the material on the macroscopic level. To reveal the mechanism of cutting CFRPs at both levels, a macroscopic model and a microscopic model are established in this study. Orthogonal cutting is applied in both of the models to illuminate the removal process. Combined with experimental observations, the results that obtained from both the macroscopic and microscopic level revealed the different mechanics of cutting CFRPs for different fibre orientations. For example, the forms of fracture that occur at 0° fibre orientation are primary interface cracking and fibre bending; the resulting chips have long shapes.

Precision calibration method for binocular vision measurement systems based on arbitrary translations and 3D-connection information

Binocular vision systems play an important role in computer vision, and high-precision system calibration is a necessary and indispensable process. In this paper, an improved calibration method for binocular stereo vision measurement systems based on arbitrary translations and 3D-connection information is proposed. First, a new method for calibrating the intrinsic parameters of binocular vision system based on two translations with an arbitrary angle difference is presented, which reduces the effect of the deviation of the motion actuator on calibration accuracy. This method is simpler and more accurate than existing active-vision calibration methods and can provide a better initial value for the determination of extrinsic parameters. Second, a 3D-connection calibration and optimization method is developed that links the information of the calibration target in different positions, further improving the accuracy of the system calibration. Calibration experiments show that the calibration error can be reduced to 0.09%, outperforming traditional methods for the experiments of this study.

Using support vector machine for characteristics prediction of hydraulic valve

Accurate prediction for the synthesis characteristics of a hydraulic valve plays an important role in decreasing the repair and reject rate of the hydraulic product. Recently, intelligence system approaches such as Artificial Neural Network (ANN) and neuro-fuzzy methods have been used successfully for system modelling. The major shortcomings of these approaches are that a large number of training data sets are needed or the training time is too long. Using Support Vector Machine (SVM) approaches would help to overcome these issues. In this study, the SVM approach was used to construct a hydraulic valve characteristics forecasting system. To illustrate the applicability and capability of the SVM, a specific hydraulic valve production was selected as a case study. The prediction results showed that the proposed prediction method was more applicable and has higher accuracy than adaptive neuro-fuzzy inference system (ANFIS) and ANN in predicting the synthesis characteristics of hydraulic valve.

Machining Parameter Optimization in High-Speed Milling for Inconel 718 Curved Surface

Machining technology for nickel-based alloy Inconel 718 is a hotspot and difficult problem in industrial fields and the high-speed milling (HSM) shows obvious superiority in difficult-to-process material machining. As the machining parameters are crucial in processing of Inconel 718 and the study of chip is important in metal cutting, there is an urgent need for deep research into the machining parameter optimization based on chip variation in HSM for Inconel 718 curved surface, so as to further increase the productivity of Inconel 718 in aerospace field. Regarding Inconel 718 curved surface, an experimental study about the machining parameter optimization based on chip variation in HSM is conducted. The relationship between chip shape and machining parameters is studied, and the roughness is measured and discussed for the machined curved surface. Results indicate that the chip area relates to geometric feature of curved surface, the optimal range for spindle speed is from 9000 to 11000 rpm based on chip variation, the feed per tooth should be large in case that condition permitted, and the cutting depth can be selected according to other constraint conditions. This study is significant for improving the machining quality and efficiency of Inconel 718 curved surface.