The invention relates to the technical field of
flue gas desulfurization of a
coal-fired power
plant, and particularly discloses an oxidation fan operation optimization
algorithm based on
particle swarm optimization, which comprises the following steps: S1, finding out influence factors influencing the operation efficiency of an oxidation fan according to historical operation data, and establishing a
data set, S2, carrying out correlation preprocessing on the
data set in S1 by adopting a Pearson
correlation analysis method, s3, selecting operation parameters related to the
oxidation rate and the
gypsum quality as input, taking the
oxidation rate and the
gypsum quality as output, and establishing an
oxidation rate LSTM prediction model and a
gypsum quality LSTM prediction model, S4, taking the oxidation rate and the field operation parameters as references, establishing a mechanism model, obtaining a comprehensive cost model of the desulfurization
system under various working conditions, and calculating the comprehensive cost of the desulfurization
system. S5, performing single-target optimization on the oxidation
air volume by adopting a
particle swarm optimization algorithm on the basis of the LSTM prediction model, the gypsum quality LSTM prediction model and the desulfurization
system comprehensive cost model; and S6, searching the optimal oxidation
air volume and providing the optimal oxidation
air volume to an oxidation fan
control system.