Mitigating　portfolio　risk　and　enhancing　performance　are　key　goals　in　portfolio　optimization.　Investors　prioritize　the　long-term　implications　of　wealth　accumulation　and　the　benchmark　strategies　employed　to　mitigate　risk.　This　paper　presents　a　multi-period　mean-variance　model　which　includes　intertemporal　constraints　and　tracking-error　terms.　We　systematically　tackle　the　multi-period　mean-variance　optimization　problem,　integrating　these　constraints　and　considerations.　By　employing　dynamic　programming　and　the　Linear　Quadratic　control　framework,　we　derive　the　optimal　policy　for　our　model.　Using　a　numerical　example,　we　show　that　the　dynamic　mean-variance　model　effectively　increases　returns　while　reducing　variance　and　partial　risk.　This　model　helps　investors　maintain　wealth　stability,　lower　the　chance　of　significant　losses,　and　enhance　risky　returns　during　the　investment　period.