Optimal Combustion Control Of Calciner Based On Improved Particle Swarm Optimization And Fuzzy Neural Network

Precalciner kiln is a key part of cement production process.Precalciner is the core equipment of pre-decomposition system.It is responsible for the calcination,heat exchange and decomposition of carbonate in the whole process.The performance of the calciner directly affects the output,quality,heat and power consumption of the firing system.Precalciner combustion process has the characteristics of delay,large delay,non-linearity,uncertainty and multi-disturbance.At present,most cement plants adopt manual operation for their combustion control,supplemented by automatic control,which has high energy consumption and waste gas.High emissions,large temperature fluctuations,and poor clinker quality.This paper aims at this kind of situation,establishes the central temperature control model of precalciner,proposes a comprehensive optimization control algorithm at the same time,realizes the system design of the optimized combustion control of the decomposer through implementing and controlling the coal feeding amount and the air intake amount reasonably,in order to realize energy conservation and emission reduction The objective of improving cement quality is as follows:1.It establishes a calciner optimized combustion control model.In this paper,a kind of calciner combustion control model is designed by analyzing the factors affecting the calciner combustion control,the difficulty of the calciner's combustion control,the key parameters of the calciner's combustion control,and the dynamic temperature mechanism model of the calciner.The experimental results show that with the amount of coal fed and the third air volume as the control quantity,the temperature and oxygen content are taken as the controlled quantity,and a reasonable and accurate control effect can be obtained.2.It proposes a random weighted particle swarm fuzzy neural network optimization algorithm.According to the characteristics of non-linear,multi-perturbation and dynamic changes in the calciner combustion control,the fuzzy neural network controller is adopted,and the basic fuzzy neural network model easily falls into local minimum and premature convergence,so it is combined with the random weighted particle swarm algorithm to perform Optimized,a stochastic weighted particle swarm optimization fuzzy neural network model was proposed to optimize the calciner combustion control.The simulation results show that the optimization algorithm greatly reduces the training error,improves the stability of the control system,reduces the adjustment time and temperature error,creates great economic benefits for the enterprise and has environmental significance.