Unit commitment by enhanced adaptive Lagrangian relaxation

This paper proposes an enhanced adaptive Lagrangian relaxation (ELR) for a unit commitment (UC) problem. ELR consists of adaptive LR (ALR) and heuristic search. The ALR algorithm is enhanced by new on/off decision criterion, new initialization of Lagrangian multipliers, unit classification, identical marginal unit decommitment, and adaptive adjustment of Lagrangian multipliers. After the ALR best feasible solution reached is obtained, the heuristic search consisting of unit substitution and unit decommitment is used to fine tune the solution. The proposed ELR is tested and compared to conventional Lagrangian relaxation (LR), genetic algorithm (GA), evolutionary programming (EP), Lagrangian relaxation and genetic algorithm (LRGA), and genetic algorithm based on unit characteristic classification (GAUC) on the systems with the number of generating units in the range of 10 to 100. ELR total system production costs are less expensive than the others especially for the large number of generating units. Furthermore, the computational times of ELR are much less than the others and increase linearly with the system size, which is favorable for large-scale implementation.