Identification of the superheated steam temperature system based on the variable step improved fruit fly optimization algorithm

In order to establish an accurate model of the superheated steam temperature, the experimental data of a certain power plant was obtained by the method of field experiment which was preprocessed, and the paper proposed variable step improved fruit fly optimization algorithm to identify the model of superheated steam temperature. The ability of global search and local optimization is enhanced through search step size of the algorithm based on the iterative error and adding relaxation factor to the algorithm, and so the identification accuracy of the algorithm is improved. Finally, identification results of the simulation was compared with the simulation results which do not improve fruit fly optimization algorithm that in different range of random initial position and step length, and compared the simulation results with the field experimental data at the same time. The results show that variable step improved fruit fly optimization algorithm have the advantages that the demand for random initial position and step length is not high, identification error is small and highly precision; the simulation result of variable step improved fruit fly optimization algorithm was consistent with actual characteristic of unit which verified the practical effectiveness of the algorithm.