In this paper, an algorithm is proposed for finding a quasi‐optimal schedule for the short‐term thermal unit commitment problem taking LNG fuel constraints into account.
In recent years, LNG fuel has been used increasingly. As a result, LNG fuel constraints should be considered in making a unit commitment schedule. Generally, unit commitment is a nonlinear combinatorial problem including discrete variables. To solve the problem, a two‐step algorithm is developed using mathematical programming methods. First a linear programming problem is solved to determine the amount of LNG fuel to be consumed by each LNG unit, then a Lagrangian relaxation approach is used to obtain a unit commitment schedule. This two‐step algorithm simplifies the problem and thus has good convergence characteristics.
To test the effectiveness of the proposed algorithm, a numerical simulation was carried out on a 46‐unit thermal system over a 24‐hour period. A result with a dual gap of 0.00546 was obtained. © 1998 Scripta Technica, Electr Eng Jpn, 125(3): 22–30, 1998