According to the optimum design model, a new algorithm based on design variable reconstruction is proposed for solving nonlinear constrained optimization, which can be used in the optimum design of dry-type air-core reactor. This algorithm constructs a set of linear equations. As a result, the relation of the reconstructed design variables and the original design variables is derived, the variable number of optimum design is decreased from m + n + 2 to 4, and the equality constraint optimization problem is converted into reduced- dimension no equality constraint optimization problem. In practical design, the proposed algorithm can be implemented by firstly using the traversal algorithm to make an optimization of the reconstructed design variables and then obtaining optimum value of the original design variables based on the relation of the reconstructed design variables and the original design variables. Moreover, on the premise of reconstructed design variables remaining constraint, a method of optimizing the reactor performance is proposed by fine tuning design parameters, and it makes up the defect of primary design results. Finally, a calculation example is carried out and the results show that the algorithm proposed has the advantages of quick calculation, high efficiency and minimal manual intervention. © 2009 Chin. Soc. for Elec. Eng.