Faster identification of redundant security constraints in SCUC

Abstract

One of the main reasons why it has long been difficult to solve the security-constrained unit commitment (SCUC) problem is that there are a large number of security constraints in SCUC such as transmission constraints in the base case and N-1 criterion constraints, and adding all of them to the model will make the solution of the problem very challenging. Related research shows that most of these security constraints are redundant for solving SCUC, so identifying and eliminating most of the redundant security constraints in the model will greatly improve the speed of solving SCUC. In this paper, a fast and efficient algorithm is proposed for identifying redundant security constraints in SCUC. The proposed algorithm introduces parameters in the basic transmission constraints to additionally limit the transmission capacity. In the N-1 constraints based on Line Outage Distribution Factor (LODF), more redundant constraints are identified and removed in the model by comparing the upper limit of transmission line capacity with the maximum transmission power flow that may occur in the line when the system is running. More importantly, the proposed algorithm does not need to solve additional auxiliary optimization problems or iteratively solve multiple models, and only needs to identify a large number of redundant security constraints through a quick calculation and logical judgment in the early stage of model establishment. It has been tested on the IEEE 30-bus system and the IEEE 118-bus system. The numerical experimental results show that the algorithm can identify 98.52% and 99.3% of redundant security constraints, and the computational speed is 1.63 times and 17.85 times of the full model, respectively. Compared with other methods in the literature, it has more advantages in computational efficiency.