| Beer production, one of traditional industry in China, contributes a lot to the development of civil economy. By far, however, low measurement and control precision caused by ordinary Pt100 and simple PID control strategy used in most brewery result in instable product quality. Therefore, it has practical meaning to research on how to improve and optimize measurement and control strategy in beer production process. On the basis of designing temperature monitoring & control system on the tank of fermentable liquid in the laboratory, this thesis investigates the improvement and optimization problem of measurement and control strategy on a class of long time delay and time varying plants, delegated by beer ferementation temperature. The following is the main contribution of the thesis.A. Summarize the characteristic of beer production technics and temperature control system, analyze the mathematic model of beer fermentation temperature system and compare with different structure and characteristic of control strategy.B. Because of limitation of experimental condition in the past, research on control strategy was usually on the basis of simulation, however, the model for simulation is often different from practical condition, which results in gap between theory and practice. Different from past, multipoint temperature measurement & control system based on the tank of fermentable liquid was built in the laboratory and computer monitoring system was designed in this thesis. As a platform, the system can be used to research on different control strategies for long time delay and time varying plants in lab and validate their practical control effect. The improvement and optimization for control strategies was able to go aimed at their practical deficiency. Also, the system can monitor multipoint temperature used for systematical research on temperature field distribution in the tank.C. The measurement error in the fermentation process will directly influence its control precision. Since fermentation temperature fluctuates just in a small range(-1-15℃, in general), the measurement error will be large if using ordinary Pt100 to measure fermentation temperature. Furthermore, a large amount of dew will yield in the protective tube and dew yielding may cause Pt100 instable. To solve this problem, a multipoint temperature measurementsystem based on AD590 was designed. The measurement precision in this system can reach ±0.1℃ in fermentation temperature range. Also, the system has stable quality and can prevent from dew efficiently. Not only the problem of high precision measurment required by experiment has been solved, but it also offers a kind of new idea for fermentation temperature measurement.D. Several control algorithms and strategies were investigated on the device. Firstly, fuzzy controller was designed. Compared with PID controller, experiments show that fuzzy controller operates more conveniently and has better dynamic performance, however, its static control precision is not so good. Therefore, aimed at fuzzy controller's deficiency, Fuzzy-PID controller was designed, it breaks through single control pattern, and experiments show that this control pattern can realize precise control in practical application.E. Beer fermentation is a dynamic process that must be guided along a temperature profile to obtain the desired results.The search of the optimal temperature profile in most domestic or international references is based on mathematic model in fermentation process, whose establishment is not easy beacuse model parameter is difficult to acquire and experiment to obtain model parameter must be done again in different production condition. Thus, an idea about dynamic search of the temperature profile by utilizing neural networks (NN) was brought forward. The main work is to search inherent nonlinear relation between temperature profile parameter and beer quality index after fermentation by utilizing NN and historical data in the beer production process, find out or forecast the optimal temperatu...
|
PDF Full Text Request