Soft-measuring Technologies On Ferment Process

Ferment process is a strong nonlinear, time varying and correlating process. To carry out further optimizing and control strategy, more information must be gained first. However, the process will face many problems when being measured. Nowadays, only those parameters such as physical and chemical parameters can be measured on-line, and other parameters, such as bio-parameters which are more complex and important, can not be measured yet. A way-out is to build a soft-sensor.This thesis mainly studies on the ferment process. With the mechanism and the soft-measuring method, several schemes to solve the problem of the soft-sensor are taken out. And in the end, this thesis also studies the hardware and the software designing where software design patterns for the soft-sensor are applied.The main contributions of the thesis are as follows:1. An overview is made on the ferment process, and the characteristics of the process are studied, then the problems in the plants are given. To solve the problems, soft-sensor must be used, and then the soft-measuring methods are briefly introduced. At the end of chapter one, the strategy of the control device for ferment process has been presented.2. The mathematical methods for soft-measuring models are shown in the thesis, including linear tools as Kalman Filter, Least Square Approximation, and the nonlinear tools as Wavelet Networks. A comparison of the Wavelet Networks and other networks are also made.3. Following a deeper study on the ferment process, several important parameters are listed and introduced. Also correlations of these parameters are analyzed. Several mechanical models studied by the predecessor are also presented.4. By using the soft-measuring tools, three models are built for deferent cases including the cases of definite models, vague models and ignorance models.According to the cases, three algorithms have been presented. With the comparison of emulating results, a new idea of multi-models for soft-measuring is taken out.5. Control devices for ferment process are the hardware and software platform of the soft-measuring strategy. In this thesis, by studying the DCS, and considering the characteristics of soft-sensor, a solution of soft-measuring software designing is carried out. The architecture of the software is also demonstrated.At the end of the thesis, an conclusion is made, and future research directions are proposed.