Sun H.,Hohai University |
Sun H.,Jiangsu Key Laboratory of Power Transmission and Distribution Equipment Technology |
Sun H.,Changzhou Key Laboratory of Sensor Networks and Environmental Sensing |
Wan P.,Hohai University |
And 4 more authors.
Journal of Computational and Theoretical Nanoscience | Year: 2014
Thermal nanoimprint lithography (NIL) has the advantage of high resolution and low cost, however it requires a significant amount of processing time. Finite element analysis (FEA) simulation can save research time and cost and help to understand the behavior of the polymer flow. ANSYS was employed to set up a nonlinear hyperelastic FEA model. The model was applied to analyze the resist flow during the NIL process. When the width of the stamp cavity increases or the resist thickness increases, the external pressure needed for resist flow decreases and the resist flows more easily. While the cavity depth becomes deeper, the resist needs bigger external pressure in order to fill in the cavity fully. Copyright © 2014 American Scientific Publishers.
Ranran L.,Jiangsu University |
Ranran L.,Changzhou Liu Guojun Vocational Technology College |
Zhengming L.,Jiangsu University |
Tianhong P.,Jiangsu University
Sensors and Transducers | Year: 2013
The industrial data has the characteristic of clustering and migrating with the operating point. The accuracy and generalization ability of the single-model prediction are poor because of the large amount of information lost in single-model modeling. In order to overcome these problems, a modeling method of multimodel soft sensor was proposed based on adaptive affinity propagation clustering (ADAP) and Bayesian filtering. ADAP algorithm was utilized in this method to realize the clustering and tracking of multiple operating points. The sub-models of various types of samples were established utilizing Bayesian filtering method, and the joint output and estimation were carried out based on the model of the subclass of current working point. The soft sensor models of CO and CO2 in PX oxidation side reaction were utilized in the method. The simulation results show that the estimation and generalization ability of soft sensing model is significantly improved by the method. © 2013 IFSA.
Guo A.-Y.,Changzhou Liu Guojun Vocational Technology College |
Shuai W.,Changzhou Liu Guojun Vocational Technology College |
Liu T.-S.,Changzhou Liu Guojun Vocational Technology College |
Xu T.-T.,Changzhou Liu Guojun Vocational Technology College |
Xiong J.-H.,Changzhou Liu Guojun Vocational Technology College
Wool Textile Journal | Year: 2015
Grinding machine is one of main equipment of textile post-processing. In order to improve the efficient of grinding wool fabric production process, a control system of grinding machine which was based on Mitsubishi FX3G-40 Programmable controller was designed. It contained the HLP-NV series inverter which can communicate with the master station by MODBUS protocol and DOP-B10S411 touch screen. The system was successfully used on eight roller grinding machines. Eventually, it has proved that this system can run and adjust tension quickly and smoothly with fewer faults. Also the overall performance and the efficiency of production have been improved compared with the older grinding machine.