Powdered Explosives Ingredients Predictive Control System

Burden is an important process for power dynamite production. The quality of the burden not only influences the performance of the dynamite, but also influences the security of its producing system. However, the automatic control technology is not used in the current dynamite product line, this makes workers industrial. In the producing process, the many kinds of factor influencing it couple mutually always, and a big delay comes from the process parameter measurement. This will result in the parameters changing their value in a large range. All mentioned above make the recipe adjust poorly and a real-time control of whole process of production to be difficultly realized. Therefore, it is significant for realizing a steady and high production of power dynamite and enhancing the competitive power of enterprises to research how to implement real-time adjustment on the recipe through establishing a predict model on dynamite quality.The product process of dynamite is researched and the key factors which influence dynamite burden are analyzed. It is sure that the system turnes out the typical non-linearity and big lag characteristics. Accurate mathematics model is hard to establish. So a predict model of dynamite quality based on BP (Back Propagation) NN (Neuron Network) optimized by GA(Genetic Algorithm) is proposed. The burden process can be controlled by predicting the dynamite quality. The repeating model is adopted, which keeps to the following steps: "predict, compare, conseq -uence,repredict,recompare,reconsequence"until quality index is obtained. The guide line on the dynamite burden is real-time controlled.A burden optimizing system based on the predictive model of quality is developed. The structure and function of system are presented, in which the technologies such as data communication, database and report are introduced in detail. The run result from the factory indicated that the model satisfies with the precision request on burden optimizing control. Through the control method mentioned above, the burden real-time control is realized, the quality of dynamite is improved, the production process is stabilized and the security is strengthened.