Research On The DSP Based Soft Measurement For Pulp Flow Via Time-Frequency Spectral Estimation

At present, on the production line of papermaking industry, the pulp flow meter and concentration sensor have been widely used in order to get accurate information of flow and concentration for automatic control of paper production. The fast and accurate measurements of pulp concentration and flow not only provide key information in the automation process of paper production, but also can improve the effect of production processes significantly, to ensure the quality of the finished products of paper. With the promotion and popularization of wide and high speed paper machine, the current paper production puts forward higher requirements on the measurement accuracy of pulp flow and concentration. It is very urgent to develop new flow meter to replace traditional flow meter to meet the development requirements of current paper production.Since the development of concentration measurement technology already tends to be mature now, through the research, the stable high frequency components related to flow are contained in pulp concentration noise signal. From this angle, this paper puts forward to a new method of flow measurement, which uses the high-speed digital signal processing ability of DSP to achieve the technique of time-frequency spectral estimation of concentration noise signal, to get the information of pulp flow and so on. Through the soft measurement system based on 2000-series DSP, the author can realize this new method. The main contents of this paper are summarized as follows:(1) The research of relevant theoretical foundation about the noise and pulp. Through the research on the mathematical model of noise and the related characteristics, combining with analys is of time-frequency features of the concentration noise, it provides a theoretical basis for the analysis algorithm research of time-frequency spectral estimation.(2) The research about analysis algorithms of time-frequency spectral estimation. On the basis of theoretical analys is, for the real-time analysis capability and the anti-interference of the algorithm, we do the simulation analys is and verification for the time-frequency analysis algorithm such as CWT by using Matlab tools.(3) The application research of virtual instrument in soft measurement of flow. By using the advantages of virtual instrument of high performance, low cost and short cycle, we build an online measurement and test platform of flow to further verify the correctness of the algorithms.(4) The development of software and hardware of soft measurement system. In order to realize and guarantee the high speed operation of algorithm, achieve the expected measuring accuracy, this paper uses the C2000-series F2812 chip as the core processor of this system to process time-frequency algorithm, and completes the design of schematic and fabrication of circuit board, and completes the development and debugging of the software of DSP by using CCS programming environment.Through the above work, we gradually improve the system software of data acquisition, information processing, results display, and other functions on the basis of existing system of hardware platform. The test results of system show that this new method of flow measurement in this paper not only can increase the pulp flow measurement accuracy to 0.5%, but also greatly reduces the hardware cost of papermaking produce line, and reaches the anticipative goal.