Fast and optimized hydro-thermal coordination scheduling

In this paper, Tenaga Nasional Berhad (TMB), Malaysian Giant Power Company, real generation data has been undertaken, refined, applied and implemented, the size of the TNB generation system problem is relatively large. The system comprises 53 thermal units and 10 hydro plants of 37 generators. Lagrangian relaxation (LR) technique, Dantzig-Wolf linear programming, sub-gradient method (SGM) and branch and bound (B&B) have been used to relax the LR multipliers for demand and reserve requirements, and decompose the problem into unit-wise initially and thermal and hydro sub-problems finally. The multipliers are then updated at the lower bound level by SGM. B&B along some heuristics is also used to get a feasible schedule in the hydro-thermal coordination system. The heuristics used here are based on programming with C language code that simplifies on/off statuses of units for commitment (UC) and de-commitment scheduling. The required optimum cost and computation time have been achieved.