Papers by Keyword: Predictive Model

Abstract: Cutting forces is one of the important physical phenomena in metal cutting process. It directly affects the surface quality of machining, tool life and cutting stability. The orthogonal experiments of cutting forces and influence factors with indexable and solid end mill were accomplished and the predictive model of milling force was established during high speed end milling 7050-T7451 aluminum alloy. The paper makes research mainly on the influence which the cutting speed, cutting depth and feed have on the cutting force. The experimental results of single factor showed that the cutting forces increase earlier and drop later with the increase of cutting speed, and the cutting speed of inflexion for 7050-T7451 is 1100m/min. As axial cutting depth, radial cutting depth and feed rate increase, the cutting force grows in different degree. The cutting force is particularly sensitive to axial cutting depth and slightly to the radial cutting depth.

Abstract: This works through the establishment of model of vehicle model and the road to development based on cooperative system of heavy road vehicles under the condition of the speed side prediction model,for the initiative to prevent heavy semi - trailer rollover risk, improving the reliability of prediction, forecasting, prediction of universality and stability.The driver ahead of access to information that is of potential rollover risk for heavy vehicles rollover warning time for the TTR, in turn, the driver enough time to take appropriate action to avoid a vehicle rollover accident happened.

Abstract: According to the process characteristics of the medium thickness steel plate temperature control in the heat treatment furnace, a new steel temperature predictive model of heat treatment furnace based on particle swarm optimization algorithm and genetic algorithm to optimize the parameters of regression support vector machine (PSO-SVR and GA-SVR) is established. Based on the new model, the design steps are given. The very good forecast effect is obtained when on-site production process data are taken as the training samples to train the model, and then the data samples of model test are selected to simulate it. The comparison of the PSO-SVR and GA-SVR forecast results indicates that GA-SVR is able to obtain better regression result and stronger predictive ability.

Abstract: Titanium alloys have been applied to aerospacemedical and other fields. The surface roughness of titanium alloy about these areas is very high. Based on the results of orthogonal test, belt grinding surface roughness prediction model of TC4 Titanium alloy is established using linear regression method. The significant tests of regression equation are conducted and proved that the prediction model has a significant. The results indicate that the model has reliability on the prediction of surface roughness, abrasive belt grinding pressure has certain influence on the surface roughness, and grain size of belt and the belt linear speed have high significant influence on surface roughness and the influence coefficient are-0.9378 and-0.2317. While the contact wheel hardness and workpiece axial feeding speed have no significant influence on surface roughness.

Abstract: The plate camber is one of the thorny problems in the plate rolling process. The characteristic variables of plate camber at the delivery and at the entry sides of the mill were illustrated based on the primary concepts of camber. The relationship between the plate characteristic variable and velocity distribution in the deformation area of the plate was also determined. This paper focuses on the features of asymmetry in the transverse direction during rolling, an elastic deformation mathematical model of four-high mill has been developed to optimize the predictive model of plate camber, which ensures the theory of influence factors of plate camber to be applied in plate rolling.

Abstract: For the prediction of yarn quality, this paper presents method to predict the quality of the spinning by a support vector machine. The input parameters to support vector machines including density of coarse yarn, roving twist factor, yarn linear density, yarn twist factor , the output variable is the CV values of spinning, breaking strength, establishment prediction model of CV values, breaking strength SVM. The results showed that: 11 groups of training samples randomly selected from 13 groups samples, two groups as predict sample, forecast errors are below 5% with high accuracy. This research provides a new approach for the spinning process design and quality control.

Abstract: Micro abrasive air jet (MAAJ) cutting is a promising technology for the fabrication of three-dimensional microstructures in hard and brittle materials. In this paper, a study on the cross-sectional shape of the kerf cut with MAAJ is presented. It shows that the machining depth and slope of the sidewall increase with an increase in air pressure, abrasive flow rate and jet incidence angle, while decrease with an increase in nozzle traverse speed. Using a dimensional analysis technique, predictive model for cross-sectional profile is developed. The research results may be meaningful to the highly precision three-dimensional micro-structural cutting.

Abstract: Leaching rate is one of the key parameters in the nickel stir leaching process of sulfuric acid and it is hard to online measure directly due to a lot of uncertain facts. In this paper, the prediction model of nickel leaching rate is established by least squares identification method. A controller combining predictive control(PFC) and PID control is designed to control nickel leaching rate in stir leaching process of sulfuric acid and better results of leaching rate control is proved by computer simulation.

Abstract: The critical amount of corroded steel that causes concrete cover cracking can be readily calculated based on thick-walled cylinder theory. However, the results may vary significantly depending on how the rust deposition is considered. There are several rust deposition hypothesis proposed in the literature for modelling concrete cover cracking of RC structures due to reinforcement corrosion. Among them, three are considered representative ones and have been widely cited in the literature. They are: (i) assumes a certain amount of rust product carried away from the rust layer and deposited within the open cracks proposed by Pantazopoulou and Papoulia; (ii) assumes all of the rust products build up around the bar and all of them are responsible for the expansive pressure proposed by Bazant; (iii) assumes certain amount of rust products deposited into a porous zone around the bar/concrete interface proposed by Liu and Weyers. In this paper, all three rust deposition hypotheses were examined for the critical amount of corrosion to induce cover cracking. When compared to the test data available from the literature, it showed that the porous zone model proposed by Liu and Weyers gives the best predictions. Thus it may be concluded that assuming a porous zone around the steel/concrete interface would be reasonable and may be adopted in developing concrete cover cracking predictive model.

Abstract: Due to the complicated behavior of the dispensing fluid and have no effective method to measure the micro-fluid value dispensed online, controlling the micro-fluid dispensing process has proven to be a challenging task in achieving a high degree of consistency in the amount of fluid dispensed. To enhance the dispensing performance and get a high consistency, this paper presents the development of a model for prediction the amount of fluid dispensed. Based on the model, a vision based control strategy is then developed to improve the consistency in the amount of fluid dispensed. Experiments were conducted to verify the effectiveness of the control strategy.