Papers by Author: Long Shan Wang

Abstract: A dynamic size control model during cylindrical grinding is built. The model consists of Elman neural network, fuzzy control subsystem and deformation optimal adaptive control subsystem. To improve the size prediction accuracy, the first and the second derivative of the actual amount removed from the workpiece are added into the Elman network input; To self-adapt and adjust the quantification factor and scale factor in the fuzzy control, the flexible factor is introduced to the fuzzy control model. Simulation and experiment verify that the developed prediction control model is feasible and has high prediction and control precision.

Abstract: In this paper, a NC compliant abrasive polishing system was developed based on a magnetorheological torque servo (MRT), using a bonnet as the polishing tool. The MRT was uesd to provide a load acting on bonnet tool to produce the polishing force. In order to find the factors affecting the polishing quality, the models of the applied load, polishing force and contact pressure between the bonnet and the part were established based on Hertz theory, and the factors affecting the contact pressure were analyzed. Theoretical analysis indicated that there was a significant effect of the applied load, the initial pressure of the bonnet and the radius of curvature of the part on contact pressure. Experiments were carried out to verify the validity of the models, and experiment results showed that the roughness of the part was improved greatly and the polishing efficiency could be controlled by changing the initial pressure of the bonnet and applied load.

Abstract: In this paper, a compliant tool was designed for abrade polishing parts with surface of revolution on general NC lathe. The tool can adapt itself automatically to the part’s surface. A magnetorheological torque servo driver (MRT) was developed to control independently the polishing force between the tool and the surface which determines the contact pressure and affects the polishing quality. The models of polishing force and contact pressure were established based on Hertz theory. The model of tool head posture was deduced. The impact of tool head posture and position on contact pressure was analyzed. The experiment results verified the validity of the compliant tool and models.

Abstract: A dynamic intelligent prediction control system is built in slender cylindrical grinding. Elman network is used in the dynamic size prediction control model, and the first and the second derivative of the actual amount removed from the workpiece are added into the network input, which can greatly improve the size dynamic prediction accuracy. Moreover, a surface roughness equation with vibration data is proposed. Based the equation, the surface roughness dynamic fuzzy neural network prediction subsystem is built. Experiment verifies that the developed prediction control system is feasible and has high prediction and control accuracy.

Abstract: A surface roughness intelligent prediction control system during grinding is built. The system is composed of fuzzy neural network prediction subsystem and fuzzy neural network controller. In the fuzzy neural network prediction subsystem, the vibration data are added to the inputs besides the grinding condition, such as feed and speed, so as to improve the dynamic performance of the prediction subsystem. The fuzzy neural network controller is able to adapt grinding parameters in process to improve the surface roughness of machined parts when the roughness is not meeting requirements. Experiment verifies that the developed prediction control system is feasible and has high prediction and control accuracy.

Abstract: A real-time model of two-direction grinding force for external plunge grinding process has been built based on the research of grinding system dynamics equation. The two-direction grinding force can be measured on-line by a force cell sensor developed in this paper. And the grinding depth and grinding force, which depend on grinding time, were dynamic simulated by computer simulation. The experimental results are in good agreement with the simulation results, proved the real-time model’s exactitude. The real-time models showed in this paper not only can predict and evaluate grinding behave and quality, but also provide the precondition for grinding optimum, intelligential control, and virtual grinding research etc.

Abstract: During the process of plunge grinding, a test piece can be axially grooved with thermocouple embedded, thermo-electromotive force will be conducted through collecting ring and then after A/D converting, data will be digitally acquired and online test will be realized. Based on the thermal theories, this paper presents a simple model to calculate the workpiece’s contact zone temperature during plunge grinding process. This model can be used to predict and calculate the grinding temperature easily. An empirical formula calculating the temperature of the grinding contact surface had been obtained through multivariate linear regression analysis experiments. The thermometric system showed in this paper had solved the knotty problem that the grinding temperature was very difficult to measure for plunge grinding. Compared with traditional method, the testing method adopted by the author will give more accurate result and more closely represent the test piece situation. The results of the computer simulation and the experiments proved the exactitude of the thermal model and the feasibility of the thermometric system.

Abstract: A size intelligent prediction control model during traverse grinding is constructed. The model is composed of the neural network prediction model, the deformation optimal adaptive control system and fuzzy control model. Dynamic Elman network is used in the prediction model. The first and the second derivative of the actual amount removed from the workpiece are added into the network input, which can greatly improve the prediction accuracy. The flexible factor is introduced to the fuzzy control model, which can self-adapt and adjust the quantification factor and scale factor in the fuzzy control. Simulation and experiment verify that the developed prediction control model is feasible and has high prediction and control precision.

Abstract: Two kinds of ZnO nanowires with different morphologies and characters were produced on Si (100) using a new PVD approach by heating and oxidizing Zn powders directly without any catalysts. The nanowires have an average diameter of 80nm (deposited at the evaporation source) and 30nm (deposited downstream of the evaporation source) respectively, and are evenly distributed over an area of > 20mm2. The two nanowires have complete different photoluminescence properties. One has a strong green emission (510nm) with a weak UV emission (380nm). The other has a strong and sharp UV emission (385nm), showing the capability for UV laser emitters.