1: \begin{abstract} \noindent

2: We study T-semidefinite programming (SDP) relaxation for constrained polynomial optimization problems (POPs).

3: T-SDP relaxation for unconstrained POPs was introduced by Zheng, Huang and Hu in 2022.

4: In this work, we propose a T-SDP  relaxation for POPs with polynomial inequality constraints and show that the resulting T-SDP relaxation

5: formulated with third-order tensors can be transformed into the standard SDP relaxation with block-diagonal structures.

6: The convergence of the T-SDP relaxation 

7: to the optimal value of a given constrained POP

8:  is established under moderate assumptions as the relaxation level increases. 

9: Additionally, the feasibility and optimality of the T-SDP relaxation are  discussed.

10: Numerical results illustrate that the proposed T-SDP  relaxation enhances numerical efficiency.

11: \end{abstract}

12: