Abstract:
Based on the ideas of infeasible interior-point methods and predictor-corrector algorithms,two interior-point predictor-corrector algorithms for second-order cone programming (SOCP) were presented.They use the Newton direction and the Euler direction as the predictor directions,respectively.The corrector directions belong to the category of the Alizadeh-Haeberly-Overton (AHO) directions.The two new algorithms were suitable to cases of feasible and infeasible interior iterative points.A simpler neighborhood of central path for the SOCP was proposed,which was the pivotal difference from other interior-point predictor-corrector algorithms.Under some assumptions,the algorithms possess global,linear and quadratic convergence.The complexity bound O (rln (ε0/ε)) was obtained,where r denotes the number of second-order cones in SOCP problem.The numerical results show that the proposed algorithms are effective.