Papers by Keyword: Prediction Model

Abstract: Through the software defect prediction can effectively guide the rational distribution of software system development resources, so as to improve the quality of software and software reliability. In order to fully utilize the existing historical data to guide the software development of existing software system development, this paper based on an improved classification and regression tree (Classification and Regression, CART) algorithm software defect prediction models. The paper first principal component analysis of the data predicted correlation dimension (Principle Component Analysis, PCA) between data and reduce the data, and configured according to the theory and optimized CART decision tree algorithm, existing software defect prediction system, and with traditional defect prediction method, the experimental results show that the proposed prediction model has higher prediction accuracy and stability.

Abstract: To reduce warpage deformation of the differential pressure vacuum casting (DPVC) products and to improve product quality, One prediction method for process parameters based on support vector machine (SVM) and artificial fish-swarm algorithm (AFSA) is proposed.Firstly sample test data is abtained by using orthogonal experimental design and numerical simulation to construct models to forecast warpage of DPVC product based on SVM. Simultaneously to improve the predictive accuracy of the model, AFSA is introduced to optimize the SVM model. And then using this model recommends and adjusts the DPVC process in order to achieve quality control. Finally , through the analysis of a mouse shell , the validity of the method proposed is verified, providing a feasible method for DPVC product quality control

Abstract: A prediction method based on least square support vector machine is introduced into the surface roughness prediction model in low-frequency vibration cutting. The model is created with low-frequency vibration cutting experiment for the corresponding relationship between vibration parameters and cutting parameters and the workpiece surface roughness. The training sample set is constructed to train regression models of least square support vector machine through experimental data. Identification of training sample set is done to gain the regression parameters a and b. The amplitude of A, vibration frequency f, feed f₁ and spindle speed n are used as the input variable in X_i. Predicted values of surface roughness are forecasted with the model. Evaluation is made with the difference between the predicted value and experiment. Comparison with BP neural network and support vector machine method has shown that the least square support vector machine prediction model works faster than SVM method, the prediction error is about 29% of that by support vector machine, and the prediction accuracy is higher than the BP model.

Abstract: Accidents of coal mine have happened frequently in china and coal and gas outburst is a common and serious disaster. Outbursts are often accompanied by the release of gas and lead to gas suffocation and explosions. In this paper, we use BP neural network to predict outbursts. We selected mining depth, gas pressure, initial speed of gas emission, firmness coefficient and geological extent as the main factors of outbursts. By analyzing and comparing we established a prediction model which has 5 neurons in input layer, 14 in middle layer, 1 in output layer and the model is established based on Matlab7.8.0. Experimental results show that simulated curves have the same trend with actual curve, so the method is feasible.

Abstract: In human life signs, there are many cycle characteristics or similar cycles which change with external input stimulating factors based on the body's own feedback mechanism and influence. Based on alternating current characteristics and the rhythm of human body signal characteristics, according to the AC speed control system we design the simulation of model, and analyze the flux regulation principle. On the basis we construct acquisition and analysis system of the psychological factors and prediction model based on VR. We use MATLAB7.0Simulink software to simulate biological stimuli and obtain psychological change under different conditions. After the model specification, the results show that the model can make a scientific prediction of the psychological changes in a certain extent, with strong practicability. It can provide sufficient basis for regulation in psychological intervention.

Abstract: In order to ensure the success of the blasting projects, accurate prediction of blasting vibration is necessary. However, blasting vibration is affected by different blasting conditions. The paper analyzed the impacts of the conditions to vibration, and built a blasting vibration velocity prediction model based on BP neural network. Comparing the predicted results with measured data, there has good correlation between them; it can be well applied to predict blasting vibration velocity.

Abstract: This paper presents a fuzzy neural network model combining fuzzy mapping and artificial neural network for the prediction of sports results. The model for predicting track & field results of each individual event at the 27th Olympic Games is established. Through modeling and comparative validation it is shown that since 1950s the modeling of track & field results according to the sequence of number reflects the basic trends of the track & field result development with good precision.

Abstract: The paper presents rigorous experimental validation results of the algorithm to predict work-piece surface roughness in face milling operation as developed in the Part 1. The experimental verification system consisting of various devices was established according to the given experimental conditions. Experimental parameters are set for steady face milling. Experimental data are collected through four experiments. These experimental data include the axial and radial run-out errors of each square insert with flat edge, the modal parameters of the face milling system, the Z-axial milling force and the measured surface contour of the milled work-piece. The trajectory of cutting teeth is calculated by the MATLAB software based on the static surface roughness model. Z-axial dynamic relative displacement between the tooth and the work-piece is obtained as the predicted dynamic surface roughness. By integrating the prediction results of static and dynamic models, the surface contour is predicted. Predicted and measured results are compared in the same figure and basically consistent. The work-piece surface roughness prediction model will be useful and valid in high-speed face milling.

Abstract: The work-piece surface quality reflects the cutting performance of face-milling cutter. This paper presents the development of an algorithm to predict work-piece surface roughness in face milling operation. The prediction model is based on the face milling cutter fixed square inserts with flat edges. The static prediction model considers the effects of radial and axial run-out error of inserts, feed per tooth, tooth number, cutting edge length, nose radius, main lead angle, and axial depth of cut. The dynamic prediction model considers the effects of the Z-axial relative displacement between the work-piece and cutting teeth caused by forced vibration. By combining the prediction results of static and dynamic models, the surface roughness of the work-piece in face milling is predicted.

Abstract: When machining the complex parts of aircraft engines, the milling force for the circular contour must be accurately predicted to reduce machining vibration. In this paper, the prediction model of the mean milling force per tooth during machining circular contour is developed. Firstly, the formulas of the entry angle, the exit angle and the equivalent feed per tooth are established through the analysis of circular contour milling process. Then, the equation of the mean milling force per tooth is deduced based on mechanistic force model during the circular contour machining process. Finally, the prediction model of mean milling force per tooth during machining circular contour is developed using MATLAB programming. The relationship between the milling force per tooth and surface curvature radius of the machined workpiece is also analyzed in this paper.