An optimal dynamic underfrequency load shedding scheme

• Main Author: Soomro, Dur Muhammad
• Format: Thesis
• Language: English
• Published: 2011
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/36657/1/DurMuhammadSoomroPFKE2011.pdf; http://eprints.utm.my/id/eprint/36657/

Electric power system network is highly sensitive to the supply and demand of power at generation as well as at user level. Erratic power demand under prevailing generation conditions may cause decay in power system frequency that can lead power system network towards cascading, islanding or blackouts. To avoid this undesirable situation and further streamline the system, load shedding is one of the safe alternative to restore the frequency from further decay. Numerous researches have been conducted on this aspect of the problem; however, there is a potential for another provision through optimization of the load shedding. Therefore, the main purpose of this project is to devise and present an optimal dynamic underfrequency load shedding scheme. The parameters studied in this study include: the implementation of developed dynamic underfrequency load shedding relay, the modified and simplified primary controllers (turbine governor and automatic voltage regulator) and the application of dynamic loads (especially frequency dependent loads) to enhance the load shedding optimization using power system simulation tool. The developed algorithm for underfrequency load shedding relay has considered load priority whereby the load with least priority will be shed first. The algorithm has been tested on some IEEE standard systems and one utility system. These test systems include the IEEE 9, 39 bus systems and one 27 bus utility system. The results of these test cases confirm the achievement of the objectives of this thesis such as; saving of load shedding amount of 1 MW, 2 MW and 0.01 MW in IEEE 9, 39 bus systems and 27 bus utility system respectively. Other achievement includes reduction in load shedding steps i.e. for each test case, the complete load shedding was achieved in 3 steps compared to 4 or more steps in other researches and the frequency converged to its nominal value in less time i.e. 3 sec, 5 sec and 10 sec in each test case respectively, compared to greater than or equal to 20 seconds in other researches.