Optimal capacitor placement for line-loss reduction and importance of voltage reduction during reactive power compensation and its effects on load, line loss, and generation
A methodology to determine the optimal capacitor locations and sizes to minimize line loss on a radial distribution system was developed in this work. Both the power loss index (PLI)-based approach and the loss sensitivity coefficient-based approach were comparatively studied to determine the optimal capacitor location. The index-based approach combined with a genetic algorithm was used to determine the capacitor sizes. After reactive power compensation voltage-dependent loads consume more power because of the increase in node voltage; therefore, customers pay more for their electricity while utilities experience savings from line-loss reduction. Therefore, a rationale for the necessity of reducing voltage for load demand reduction during reactive power compensation is presented, and the optimal voltage setting at the substation regulator is determined. The joint effect of ambient temperature, price, size, and phase kVAr of the capacitor on load, line loss, and generation is analyzed using a 24 factorial design. How consumer energy consumption, line loss, and generation are affected by voltage reduction is also evaluated. Since bus voltage also depends on line resistance, which varies with ambient temperature, the impact of temperature on power consumption, line loss, and generation is discussed as well. At reduced voltage, variations in line loss need to be analyzed, because losses affect the cost-benefit analysis. A model is derived that explains, at reduced voltage, how line loss varies with the type of load. Also analyzed is the effect of line resistance on line loss for various types of loads. The results of this work will improve the effectiveness and reliability of future voltage-reduction programs. Finally, analyses for negative line loss, higher voltage at the downstream node, and the active and reactive current components of a capacitor are presented in this work.
Thesis (Ph.D.)--Wichita State University, College of Engineering, Dept. of Electrical Engineering and Computer Science