Aiming at solving the problem that the thermal resistance of fouling on condenser is difficult to be predicted

the Elman neural network optimized by the improved particle swarm optimization(PSO) algorithm was used to establish the condenser fouling thermal resistance prediction model. According to the individual and global cognitive ability of the particles

the inertia weight was dynamically adjusted to improve the convergence precision and rate of the PSO algorithm. Then

the improved PSO algorithm was applied to optimize the weights and thresholds of the neural network so as to improve the predictive ability of the model. Taking the condenser in a 300 MW unit as an example

the model for the running condenser after cleaning was built up

the predicted thermal resistance of the fouling was compared with the actual value to verify the accuracy of the model. The results show that

the improved Elman prediction model has better precision and adaptability

which provides a theoretical foundation for predicting the condenser fowling resistance and determining the clean-up period.