| Powdery Emulsion Explosive(PEE) is a high-tech product with independent intellectual property rights in China. As a new waterproofing industry explosive, it has both advantages of powdered explosive and emulsion explosive, which is widely applied in many areas of the national economic construction, such as traffic, city construction, mining and oil exploration. In order to adapt changing market, the tend of process control is to improve product quality, reduce energy consumption and change the traditional process control at present.In this paper, the mechanism of emulsifying process and control requirements are considered from engineering application. The quality of emulsified bases is a large impact on explosive performance. Due to characters of spray powdering process, it is demanded higher. However, the fact is that importance of viscosity is ignored. And it is just concerned about the performance of the final explosive. The viscosity of emulsified bases is only measured by sampling, so there is a serious delay of emulsifying process control. It is also too blind to waster energy.Therefore, a method based on soft-sensor technology and information fusion is proposed. At present there is no paper about this. Firstly, the viscosity is predicted by soft sensor modeling. Secondly, according to the result obtained, the frames of discernment in D-S algorithm is selected then data is fused. Finally, integrate the results of fusion and the expert experience, in order to make decision to adjust key parameters timely, just as flow, temperature, speed and so on. This purpose is to have an optimal control strategy to enhance the control efficiency and reduce energy consumption. And simulation results show that the method applied is viable. It provides a new approach to enhance the quality of emulsion explosive.This paper has seven chapters, which is divided into four parts. First, aiming at factors on the quality of PEE and problems in production control, an optimal control program is proposed. Second, soft-sensor technology and engineering design are studied. By the simulation in soft sensor of viscosity of emulsified bases, an excellent soft-sensor model is constituted. Third, information fusion theory and an improved algorithm of evidence theory are studied for emulsifying process fusion. Fourth, the decision-making system of PEE process control is designed. The features and innovations in this paper are as following:(1) It is first time to apply soft-sensor technology in measuring viscosity of emulsified bases, using improved BP algorithm for modeling and simulating. By contrasts it can be seen clearly that predicted results is within acceptable limits from the actual value. Thus, the model can be used in system development, and the predicted result is an important reference for information fusion. (2) Aiming at over reliance on expert and lack of theoretical support about the traditional method in emulsifying process control, a new method is proposed based on information fusion using D-S evidence theory. Due to weakness of the evidence theory, which can not distinguish set size, general evidence theory can not be applied in the frames of discernment proposed in this paper. So the algorithm is improved by defines a new combination rule and it has been verified through the simulation. Fusion result can reflect the actual situation accurately. And it can show which parameter should be adjusted. These efforts have not seen in other papers.
|
PDF Full Text Request