I. Background and Objective
II. Background of the Problem
III. Formulation of the Solution using Linear Programming (LP)
IV. Solution and Description of the Analysis Tool Used
V. Results
VI. Discussion of the Results and Conclusion

VI. Discussion of the Results and Conclusion

As expected base from the conditions and constraints in the dispatch of MERALCO IPP’s stated in the earlier discussions, the result of LP optimization for both cases shows that all of the contracted capacity must be dispatched in as full as possible before dispatching power that is above the contracted capacity. This is because whether the MERALCO consumed the contracted power or not, they will be still paying for it. Thus, the best way to minimize the cost of electricity is to fully utilize the contracted capacity first. And this was proven in the LP solution.

It was also showed in the result for both cases, that once the amount of the contracted capacity has been fully utilized, the first generator that will dispatched power beyond its contracted capacity is CC3MM (P5) (as observed from Hour 9 to Hour 21 in Case 1 results). This is because this generator has the lowest variable cost.

Comparing the resulting total cost of generation for both case 1 and case 2, it is very clear that for 23 hours, the generation cost paid by MERALCO is indeed lower if their purchased power from its own IPP’s will not be limited to 40%. The amount of money that could be saved is shown in table 5. However, at a certain magnitude of peak demand, as showed in the peak hour at hour 11, the savings (or the difference) is zero, because at this amount of load, it is more economical to buy power from NPC than from MERALCO IPP’s, and the constraint,

does not matter anymore. This scenario was highlighted in the case 3, which results are shown in table 6.

VII. Summary and Conclusion

It was verified in the study, that at certain load level, particularly, at low load demand, it is more economical for MERALCO to purchase its power from its own IPP’s and not to be bound by the 40% limit. But for instance, when the load demand is high, the result of the LP solution shows that it is now more economical to buy power from NPC and allow the amount of power to be purchase from their own IPP to fall below 40% limit.

However, it is very important to note that all of the conclusion and recommendations discussed above may be hypothetical because of the inaccuracy of the data used, though computationally correct using the suggested methodology. What is more important to emphasize here is the effectivity of the methodology which is Linear Programming.