Industrial Production Of Software Measurement Techniques Modeling Study

With the project of "soft-sensing technology used in polymer plant and advance control and management software develop" which was set up by Changzhou Worldbest Radici Co.,Ltd and East China University of Science and Technology, this paper is written in an software application based on soft-sensing models and achieve the goal of display the on-line quality parameter. First, the analysis of the DCS utility in industry areas is provided, the emphasis in this chapter is discussing how to control the polymer plant with CENTUMCS-3000 system produced by YOKOGAWA JAPAN and the existing problems. Based on these, this paper is described how to set up a dynamics soft-sensing model for the esterification process, the model parameters are estimated and validated using unconstrained nonlinear optimization method, such as DFP. BP neural network assistant model is used to amend the error of the mechanic model. The running result of the soft-sensing system shows that it can help the operators in real-time monitoring and meet the requirement and accuracy of engineering. Finally, the construction and function of the whole software system is introduced.