Soft-Sensoring And Researching Of The Output Of Ball Mill Pulverizing System

As one of important parts of pulverizing system, ball mill is used widely in coal-fired power plants. However, most pulverizing system can not be operated economically. One main reason is that the output of ball mill pulverizing system is hardly measured. Without an effective and precise method of measurement, operators have to make ball mill operated in low output and not to consider about the economy problem, in case of the accidents of heaped load and coal powder running over. Therefore, an effective measurement method of the output of ball mill pulverizing system would be valuable.In order to solve the problems above, soft-sensoring, and researching of output of ball mill pulverizing system are studied in this paper. The main contents of this dissertation are as follows:1.Summarized current status of instrument of the output of ball mill pulverizing system. And it is put forward that using of soft-sensoring technology is a inevitable trend of output measurement.2.This paper researches the dynamic mathematical models of the ball mill and analyzes the ball mill work specialities. This paper also integrates the ball mill mechanisms and its work characteristics to analyse the main factors which impact the output of pulverizing system of the ball mill.3. This paper analyses the main factors which impact the output of pulverizing system, then uses grey entropy relation theory to analyse the degree of importance of many factors that impact the output of pulverizing system ,namely contribution degrees; This paper uses the theory of the Chaos reconstruction to restructure a major contribution factor, press diff , and then comes the input dimensions of the soft sensoring assistant variables. This paper integrates contribution degrees and input dimensions , and then confirms the assistant variables which the output of ball mill pulverizing system soft sensoring modeling needs. This measure has theory bases. It puts forward a new method to choose soft sensoring assistant variables from another point.4. This paper uses many methods such as wave let theory to preprocess the datum from the assistant variables choosed , then puts forward the datum marked error checkout and revision based the observation variables change rate . After simulation research, this method is proved applicable.5.Trying to find a good soft sensoring method for ball mill, this paper presents a brief overview of existing modeling methods of soft sensoring ,and use three typical soft sensoring modeling methods among them to process modeling of the output of the ball mill pulverizing system in many ways. Through the design of experiments this paper obtains the necessary modeling reliable samples, and uses it to check preprocess ability of preprocess model. Through the model preprocess analysis, this paper concludes that among three methods soft sensoring modeling based CMAC net has its advantages.The problems and direction of further study are pointed out at the end of this dissertation.