Boiler Combustion Optimization Research Of Circulating Fluidized Bed Based On Hybrid Chicken Swarm Optimization Algorithm

In recent years,although China's various new energy technologies have continued to develop,coal power generation is still the main mode of power supply in China.Since1980,China's circulating fluidized bed boiler combustion technology has matured over the past 30 years.CFBB technology is inexpensive and the combustion process is clean,and can meet almost all national requirements.But China still follows the world's highest pollution emission standards,which will be another challenge for CFBB technology.Therefore,this paper uses the artificial neural network modeling method and intelligent optimization algorithm that have been developed in recent years to optimize the CFBB combustion model.First of all,aiming at the disadvantage of the chicken swarm optimization algorithm(CSO)in its low precision and easy to fall into local optimum.The simulated annealing(SA)algorithm is combined with it to propose a hybrid chicken swarm optimization algorithm(SACOS).Then this algorithm is compared with differential evolution algorithm(DE),particle swarm optimization(PSO),and chicken swarm optimization algorithm.Then modeled using another new kernel extreme learning machine(KELM),KELM can overcome the shortcomings of the extreme learning machine for the hidden layer nodes is need to be manually set and the generalization ability is poor.Then use the grid search method to determine the superparameter of KELM,and establish the CFBB combustion model according to the adjusted parameters.After that this model is compared with the traditional BP neural network model and the ELM network model.Finally,based on the established model,the SACSO algorithm is used to optimize the parameters of the adjustable input space of the boiler.It will find the optimal operating parameters of the boiler to increase the combustion efficiency of the boiler and reduce the pollution emissions.The final experiment shows that the SACSO algorithm has fast convergence speed and high precision.KELM model has better fitting effect and higher stability,and can meet the actual needs.The combination of SACSO algorithm and CFBB combustionmodel can basically meet the requirements of the experimental target and achieve the expected results.