The invention discloses a fuel cell optimization modeling method for a fusion krill behavior artificial bee colony algorithm. The fuel cell optimization modeling method comprises the following steps: 1) acquiring fuel cell current input and voltage output sampling data through field operation or an experiment; 2) taking the error sum of squares of voltage estimation output and actual voltage output of a fuel cell model as a target function; 3) by virtue of the inspiration of a krill behavior under the guide effect of other individuals in a population, proposing an employed bee phase position improvement rule, and proposing the fuel cell optimization modeling method for the fusion krill behavior artificial bee colony algorithm; 4) setting algorithm operating parameters; 5) estimating unknown parameters in the fuel cell model by operating the fusion krill behavior artificial bee colony algorithm, obtaining a parameter estimation value of the fuel cell model by minimizing the target function, and substituting the parameter estimation value into the fuel cell model to form a mathematical model. The modeling method disclosed by the invention has the characteristics of high optimization precision and high convergence rate, and is also suitable for modeling of other complex chemical reaction processes.