In　this　paper,　a　self-centering　joint　between　the　concrete-filled　double　steel　tubular　(CFDST)　column　and　the　steel　beam　is　proposed.　The　joint　was　self-centered　by　prestressed　strands　when　unloading,　and　energy　was　dissipated　during　loading　by　friction.　This　work　is　a　numerical　and　theoretical　exploratory　study　based　on　a　modeling　approach　already　validated　by　experiments,　aiming　to　investigate　the　mechanical　behavior　of　the　novel　self-centering　joint　and　optimize　its　design　parameters.　The　reasonableness　of　the　joint　model　was　verified　by　numerical　simulation　with　ABAQUS　finite　element　software,　and　the　damage　modes,　hysteresis　curves,　and　energy　dissipation　capacity　of　the　joint　were　analyzed.　A　mechanism-based　semi-theoretical　restoring　force　model　was　established.　Performed　a　parametric　analysis　on　the　primary　factors　influencing　the　mechanical　behavior　of　the　joint.　The　results　indicated　that　there　was　a　contradiction　between　self-centering　performance　and　energy　dissipation　in　the　joint,　and　the　ratio　β　between　the　moment　resistance　provided　by　the　strands　and　that　provided　by　the　friction　devices　in　the　decompression　moment　of　the　joint　played　a　key　role　in　moderating　the　relationship　between　the　two.　A　value　of　β　between　1　and　1.5　was　recommended　for　a　balance　between　the　two.　The　restoring　force　model　of　the　joint　provided　a　well-predicted　mechanical　behavior　of　the　joint.　The　modified　rigid　model　was　more　consistent　with　the　numerical　results.　This　study　provided　theoretical　support　and　optimized　design　parameters　for　the　design　of　CFDST　column-steel　beam　joints　with　favorable　self-centering　performance　and　energy　dissipation　capability.　The　parameter　analysis　showed　that　increasing　strand　prestress　increased　the　bearing　capacity　but　potentially　resulted　in　stress　loss;　an　increase　in　friction　force　improved　the　bearing　capacity　and　energy　dissipation　capacity,　yet　it　augmented　residual　deformation;　and　an　enlargement　of　the　cross-sectional　area　bolstered　both　the　bearing　capacity　and　stiffness,　albeit　it　also　led　to　stress　loss.　©　2025　Institution　of　Structural　Engineers.　Published　by　Elsevier　Ltd.　All　rights　are　reserved,　including　those　for　text　and　data　mining,　AI　training,　and　similar　technologies.