The　basic　idea　of　the　sequential　quadratic　programming　(SQP)　algorithm　is　to　approximate　and　initial　model　with　a　sequence　of　quadratic　programming　(QP)　sub-problems.　To　choose　rational　move　limits　box　(super-polyhedron)　for　QP　sub-problems,　a　new　SQP　algorithm,　SQPRML　is　presented.　Hence,　the　problem　of　search　direction　is　solved　in　a　more　rational　method,　and　it　facilitates　to　obtain　search　step　size.　Numerical　examples　are　presented　which　show　that　the　algorithm　is　feasible　and　efficient.