Abstract: Abstract： 【 Objective 】 To solve the problem that directly invoking the optimization algorithm to analyze the worst-case of three-span beam structures under multiple patch loads has the possibility of trapping the local optimal solution rather than the global solution, 【 Method 】 A analysis method embedded domain knowledge with the general black-box optimization algorithm for the worst-case analysis of the beam is proposed. On the one hand, the position of each wheel patch load is defined as a design variable, and there is no need to specify the relative position of the group of wheel patch loads in advance, which is more universal. On the other hand, by integrating the knowledge of ship structural mechanics, such as large stress resulting from the close aggregation of loads in order of magnitude, large bending moment and shear force usually generated by the load in the mid span of the beam and near the support, into the optimization algorithm, a strategy of generating dangerous initial population individuals in genetic algorithm and the global translation moving strategy of wheel patch load are proposed respectively to reduce the possibility of falling into the local optimal solution. The theoretical bending moment and shear force distribution of a three-span beam under single wheel patch load are derived respectively. The theoretical most dangerous positions of multiple wheel patch loads are found out by enumerating all possible combinations, which are employed to verify the correctness of the proposed algorithm. 【 Results 】 Compared with the optimization algorithm without domain knowledge and under the same computational resource, the most dangerous bending normal stress and shear stress increase by 5.9% and 14.7% under six wheel patch loads respectively, and the error between the calculation results in five times and the theoretical solution is less than 0.5%. 【 Conclusion 】 The numerical results show that the proposed method can accurately, stably and quickly obtain the most dangerous load positions.