Research On Combustion Optimization Of Circulating Fluidized Bed Boiler Based On Artificial Intelligenc

With the development of China’s socialist modernization and industrialization,the social demand for coal has also increased significantly.At present,the main form of boilers used in China’s industry is circulating fluidized bed boilers,coal-fired boilers are important thermal power equipment,which are widely used in the industrial field.However,coalfired boilers in factories will produce a large number of pollutants in the process of use,including sulfur dioxide,carbon dioxide,nitrogen oxides and soot,which will not only cause pollution to the ecological environment,but also threaten the health of the people,which is not in line with the concept of sustainable development advocated by our country,and improving the combustion efficiency of coal-fired boilers and reducing nitrogen oxide emissions has become the primary goal,so the factory currently needs to solve the The problem is energy saving and emission reduction,and the optimal control of boilers has become a hot topic of research for scholars at home and abroad.At present,there are limitations in the traditional modeling research methods,which do not consider the time-series of data,ignore the logical relationship between data,and simply map the historical data,making it difficult to establish an accurate mathematical model,so the prediction error is large,the generalization ability is low,and the effect produced is unsatisfactory,while the optimization research of boilers through artificial intelligence algorithms,taking into account the correlation between the combustion mechanism and the algorithm principle,has advantages for dealing with nonlinear data.In this paper,we propose an artificial intelligence-based approach to optimize the boiler control by using a long and short-term memory network for modeling,applying the firefly algorithm to optimize the established model hyperparameters,finding the optimal hyperparameter combinations,and obtaining the optimal boiler combustion prediction model to achieve the prediction of NOx emissions,thermal efficiency,and steam flow.Since reducing NOx emissions and improving thermal efficiency is a contradictory process,multi-objective optimization is used for comprehensive modeling.The modeling process takes into account that each enterprise has different optimization priorities for both,so different weights are set for NOx and thermal efficiency.In this paper,the optimization model is established by particle swarm algorithm to find the optimal combination of boiler operating parameters under the condition of satisfying normal boiler operation to achieve This paper establishes an optimization model by particle swarm algorithm to find the optimal combination of boiler operating parameters to achieve the optimization of boiler NOx emission and thermal efficiency under different working conditions.The experimental results show that the prediction model established based on long and short-term memory network and firefly algorithm has high accuracy,strong generalization ability and good performance,achieving accurate prediction of boiler thermal efficiency,NOx and steam flow rate.The boiler optimization model established based on particle swarm optimization algorithm improves boiler thermal efficiency by about 2%,and reduces NOx emission by about 5% under the premise of ensuring normal plant production,making the boiler achieve optimal operation under different working conditions,which also shows the feasibility and importance of artificial intelligence-based methods in circulating fluidized bed boiler combustion optimization.