Research And Application Of Boiler Main Steam Temperature Optimization Control System Based On Hybrid Intelligent Algorithm

Thermal power generation is the main way of power generation in China,and its unit parameters have been increasing with the growing national productivity and demand in recent years.The main-steam temperature control is one of the important parts of the power boiler control system,which has the characteristics of long time-delay,uncertainty and strong coupling.The quality of the main-steam temperature control directly affects the safe and economic operation of the unit,and the main steam-temperature control deviation is usually required to be within ±5℃ of the set value.The conventional PID control algorithm is widely used in most power plants in China because of its simple control structure and easy operation.But with thermal power units moving in the direction of large capacity and high parameters,it is difficult for conventional PID control algorithm to meet the needs of field control.In view of the above research background,this paper improves the traditional algorithm and the existing intelligent optimization algorithms and makes a hybrid optimization.Moreover,this paper designs a simple-structured and easy-to-apply controller with strong anti-jamming and good robustness to meet the requirements of industrial field control,which has important research significance and application value to realize the safe,economic and stable operation of the generator set.This paper studies the Energy Plant’s No.7 gas-fired power boiler of Hunan Valin Lianyuan Iron and Steel Company Limited.Firstly,the structure of the boiler and the field process of the steam water system are analyzed,and the main influencing factors and design points of the main steam temperature control are found.Through comprehensive analysis,the basic control scheme is designed.Then,a neural network with high performance in dealing with nonlinear and uncertain objects is proposed as the identification tool of the boiler’s main steam temperature model.After analyzing and comparing the performances of various neural networks that can be applied to the field control system,it is concluded that RBF neural network needs to be used as the identification tool of the boiler’s main steam temperature system,and the accuracy of the identified model is verified by MATLAB.In addition,in this paper,the performance of each intelligent optimization algorithm is simulated and analyzed.After comprehensive comparison,it is decided to apply the least square method as the hybrid optimization algorithm to the optimization controller based on the simulated annealing algorithm,and the main steam temperature model is identified to verify the effectiveness of the hybrid optimization algorithm.Finally,the optimized control system is designed to meet the requirements of industrial application,and the software and hardware parameters and programs are designed by the relevant software configuration.The practical application results show that the optimal control system has high control performance after being put into the field application,pplication,...