Research Of Working-status Recognition System To NSP Cement Calcining And Cooling Processes

Calcining and cooling processes of clinker, whose working-status determines the cement quality and cost of energy sources directly, play an extremely important role in NSP cement production. However, these processes have the non-linearity, the multivariable, the close coupling, the big time lag, time-variable and so on the characteristics which cause the working-status of the cement rotary kiln system to be extremely complex, so that the working-statue is very difficult to be stabilized by traditional control mode. At present, the research on the production process control of kiln system usually focuses on one part. Sometimes it maybe takes a good control effort; however, once the working-status has a large change, the control impact will fail. In order to guarantee the stabilization of working-statue, operators mainly depend on adjusting craft parameters frequently according to their experience and the changes of the parameters. It is an urgent problem that how to realize intellective recognition of working-statue instead of the manual control mode, and finally realize the integrated automatic control of the whole process with the guide of working-status result. Deeply research of this problem has a far-reaching significance which can improve the integrated automatic level of our cement industry, guarantee the quality of cement production, and improve the productivity of the factories. The technology of artificial intelligence which has a high speed development gives a solid basis for dealing with this problem.As an important application zone of artificial intelligence, the theory and technology of expert system is more and more mature with long time research. Applications of expert system also take a fast developing speed. Recent years, the technology of fuzzy reasoning was used in expert system successfully. This causes that the expert system has an accurate method to deal with imprecise problems.The characteristics of fuzzy expert system give a great academic and technical support on complex working-statue intellective recognition. According to this situation, this dissertation is relied on the National High-Tech Research Development Plan (863): Soft measurement, optimization unit-level control system design, development and engineering applications issues which is a subtopic of"The application of NSP cement production integrated process control". Aiming at the cement rotary kiln and cooler of Sunnsy Group 4# production line, a solution of working-status intellective recognition based on fuzzy expert system is made and put into application after craft analysis, parameter selection and working-status sum-up according to the experience from craft engineer. Five parameters are choosed to realize the characters recogniton which includes the trend and the real-time value characters according to the caft and working-status of cement production.The solution of working-status recognition based on fuzzy expert system is researched from two sides. Firstly, the expert rules are shown by the form of fuzzy matrix in order to improve the efficiency of reasoning. Secondly, weight fuzzy reasonning theory which takes parameter characters as input and working-status as output is used in order to improve the traditional fuzzy reasoning.The Software of this system of working-status recognition is made using C++. The operation results demonstrate the effectiveness and practicality of this solution.