Mingheng Xu

Intelligent tool condition monitoring system for turning operations

In order to predict tool wear accurately and reliably under different cutting conditions, a new monitoring methodology for turning operations is proposed. Firstly, the correlation coefficients approaches were used to indicate the dependencies between the different sensed information features and tool wear amount, and the most appropriate features were selected. Secondly, B-spline neural networks were introduced to model the non-linear relationship between extracted features and tool wear amount, and multi-sensor information were fused by an integrated neural network. Lastly, the final result of tool wear was given through fuzzy modeling. Experimental results have proved that the monitoring system based on the methodology is reliable and practical.

The Investigation of A Self-adjusting Tool Wear Monitoring System

The structure of a self-adjusting tool wear monitoring system was proposed to improve classifying accuracy of tool wear and solve the problems of high design cost of tool condition monitoring system under multi machining modes and different machining condition. The monitoring features extracted from various sensor signals and selected automatically by synthesis coefficients change with difference in cutting conditions, tool quality, workpiece properties, etc, and the nonlinear relation between tool wear amounts and features were built through a novel sensor-integration strategy including localized neural networks that optimized by an adaptive learning algorithm, and integrated neural networks that fuse the outputs of subnets, the final results of monitoring system was given by decision algorithm that compare tool wear values at different time intervals. As demonstrated by examples of tool wear monitoring in milling and in turning, the self-adjusting monitoring system proposed in the paper has a number of advantages over the existing methods, provided with high classifying precision, high reliability and short design periods, so it is good for popularization in industry

Experimental study of tool wear monitoring based on neural networks

The influence of experimental design on modeling of tool condition monitoring system based on different neural networks was investigated. The orthogonal experiments and complete parameter combination experiments were carried out on a Vertical Machining Centre, and BPNN and CSGFFNN were adopted to model the mapping relations between tool condition and features extracted from different sensor signals by using experimental data. The research results show the orthogonal experiments canpsilat meet the need of modeling of TCMS based on different NN and the classifying error is high above 87%, complete parameter combination experiments can provide enough data for modeling of NN for TCMS and realize reliable monitoring of tool condition or tool wear.

Tool wear monitoring using FNN with compact support gaussian function

A novel approach of tool wear monitoring based on localized fuzzy neural networks with compact support Gaussian basis function (CSGFFNN) was proposed to improve classification accuracy of tool states and solve the problems of slow computing speed of BP neural networks. By analyzing cutting forces signals, acoustic emission signals and vibration signals in time domain, frequency domain, and time-frequency domain, a series of features that sensitive to tool states were selected as inputs of neural networks according to synthesis coefficient. The nonlinear relations between tool wear and features were modeled by using CSGFFNN that constructed and optimized through fuzzy clustering and an adaptive learning algorithm. The experimental results show that the monitoring system based on CSGFFNN is provided with high precision, rapid computing speed and good multiplication.

Artificial neural network for screw life prediction

The life change of the screw in High-end CNC machine tool in the process has some features such as non-linear, dynamic and uncertainty. A screw online life prediction system was designed to monitor the performance of lead screw by vibration sensors and temperature sensors which installed at different locations of Lead Screw Pair and reflected the trend of changes of different processing conditions, lifting wavelet transform was used to extract the most sensitive characteristics of screw performance. RBF neural network was used to build the non-linear relationship between screw vibration signal changes and screw life. Eventually constructed screw life prediction model based on RBF neural network bring into effect of effective assessment of residual life of lead screw. The results show that the performance degradation model can predict the remaining life of screw effectively.

CNC screw life prediction based on DFNN performance model

After considering a variety of factors to the performance degradation of screw-nut pair, a novel performance degradation model based on dynamic fuzzy neural network (DFNN) was proposed to predict screw life in this paper. The model can assessed the life of Roller Lead Screw-nut pair of the CNC feed drive system effectively, and can resolved the problems of screw life changing with the machining conditions which impacted the CNC performance. The vibration signals from screw pair parts were processed by analyzing the empirical mode decomposition (EMD) algorithm, time-series features extracted from vibration signals defined as the DFNN input vector, and the rated life under different machining condition used to be the expectancy life. Finally, a physical model of the actual residual life about screw was built, and the online prediction of screw life was achieved. The experimental results show that the model can accurately predict the dynamic life of the screw. It is conducive to the active maintenance of screw.

Tool wear monitoring based on novel evolutionary artificial neural networks

In order to improve the accuracy and speed of on-line tool wear monitoring system, an evolutionary neural network using variable string genetic algorithm (VGA) was developed to construct the relations between tool wear and signal features extracted from cutting forces, vibrations, and acoustic emission by different signal processing methods. The system could automatically evolve the appropriate architecture of neural network and find a near-optimal set of connection weights globally. Then the conformable connection weights for model could be found with back-propagation (BP) algorithm, the multi-model finally completed calculation of tool wear. The experimental results show that the system proposed in the paper has higher classification precision and calculating speed.

Screw performance degradation model based on novel neural networks

A screw performance degradation model based on neural network which was optimized by improved genetic algorithm was proposed to predict screw life accurately and provide active maintenance proof. Key factors which related to screw life were analyzed by screw motion mechanism. Three vibration sensors were installed on different position of screw and vibration signal were processed by EMD, time domain analysis, frequency domain analysis and wavelet packet analysis. The most sensitive features to screw life were selected by correlation coefficient and evaluation index. The relation between screw life and features was built by neural network that constructed by BP training algorithm, and screw life was calculated. The long practical results show that the screw life prediction model can meet the need of active maintenance and reduce maintenance cost.

Study on the e-HUB system model based on workflow

To analyzed the characteristic of the e-hub and the whole project life cycle. The e-HUB can support the lifecycle business process. Based on this, the paper established the e-HUB system model based on workflow. The model divides project collaboration into the five stages. The five stages are job solicitation, finding the right partner, planning the collaboration, executing the project, and partnership closing. The business workflow was carried on the detailed description using the UML technology, especially the project planning process. Then, construct the workflow management prototype system in the e-HUB system.

The Study on Intelligent Tool Wear Monitoring Techniques

A novel method of tool wear monitoring based on variation features was proposed to increase recognized precision of monitoring system and solve the problem of feature selection under multi machining conditions. The trend of different signal features with the change of tool wear amounts were investigated through analyzing vibration signal, acoustic emission signal and cutting forces signal in time domain, frequency domain and time-frequency domain, the most sensitive features to tool wear were selected by means of synthesis coefficient, and the nonlinear relation between tool wear values and features was built by B-spline neural networks. The experimental results indicate that the proposed method can improve classifying accuracy and self-adjusting ability of tool wear monitoring system