Abstract
In order to strengthen the safety management of chemical enterprises, improve the quality of operators and management personnel, a dynamic linear modeling method combining data mining with BP neural network is proposed for coal gasification furnace. The coal gasification furnace can indirectly reflect the upstream temperature, the down temperature and the sum of up and down temperature of the gasification layer. On the one hand, the dynamic description of the system model is used to accurately reflect the magnitude and variation trend of the gasification layer temperature. On the other hand, the variables that can be controlled in the system is effectively utilized. The results show that the modeling method proposed in this paper has higher accuracy than other models, and it reduces the training time of neural network. At the same time, the variable of the modeling method is the variable that can be controlled by the gas making system, so the model can be used in practical application. Therefore, it is concluded that the research work in this paper is beneficial to the prediction and control of the coal gasification system, and establishes the prerequisite for the simulation optimization of the control system in the plant.
