An automatic cutting planes inducing method accelerating the solution of network-constrained unit commitment

Improving the computational efficiency of Network-Constrained Unit Commitment (NCUC) is a critical challenge in power systems and electricity market operation. Tightening the formulation of NCUC with cutting planes is a major method to accelerate NCUC without sacrificing solution accuracy. However, prevailing cutting plane methods for NCUC are mainly generated based on manual mathematical derivation, which is case-by-case and difficult to adapt to massive heterogeneous resources. This paper proposes a universal framework for automatically generating cutting planes, applicable to general NCUC formulated as a mixed-integer linear program (MILP). The proposed method divides the NCUC model into subsystems and employs inducing functions, thereby obtaining high-quality cutting planes through the solution of subsystem models. Simulation results reveal that the proposed CPIM can accelerate the solution of NCUC, especially when the problem scales up or becomes complex.